Fusion Conditions Research Articles

Pump depletion and the Raman gap in ignition-scale plasmas.

Laser-plasma instabilities under ignition conditions for direct-drive inertial confinement fusion are studied using two-dimensional particle-in-cell simulations with a combination of in-plane (PP) and out-of-the-plane (SP) lasers. The results show that stimulated Raman side-scattering can induce significant pump depletion and form a gap in the Raman scattered-light spectra that have been observed in experiments.

Height-resolved emission spectroscopy and high-speed imaging of the TiAl6V4 vapor plume under laser powder bed fusion conditions.

Optical emission spectroscopy is increasingly used as an in situ monitoring technique during laser powder bed fusion (LPBF) because plume emission holds elemental information not found in other in situ sensing techniques. This work explores the shape, stability, and temperature of the emission plume above the melt pool of Ti6Al4 V undergoing laser scans under LPBF-like processing conditions, using high-speed video and height-resolved spectroscopy to study the details of plume dynamics. Optical emission spectroscopy is conducted in the 480 nm to 525 nm region, where Ti emission is strong, with 0.3 mm vertical resolution above the baseplate. The Boltzmann plot method is used to determine temperature at each probed elevation, which indicates that the hottest location of the plume is occasionally elevated 0.3 mm to 0.6 mm above the scanning surface. The results show that the plume shape, stability, temperature, and spectra are highly dependent on the processing conditions. We highlight some of the complexities of optical emission spectroscopy and discuss potential challenges for implementing optical emission spectroscopy across an entire build.

Efficient extraction of rare earth elements from coal-series kaolin by combining alkali fusion and sulfamic acid leaching

As a solid waste produced by coal mining, the comprehensive utilization of coal-series kaolin is the focus of attention. In this paper, the extraction of rare earth elements (REEs) from coal-series kaolin was carried out by alkali fusion and sulfamic acid leaching. The occurrence modes of REEs in coal-series kaolin was analyzed by means of inductively coupled plasma-mass spectrometry analyse and sequential chemical extraction. The influence of alkali fusion conditions on the leaching of REEs was investigated, and the activation mechanism of alkali fusion was revealed by X-ray diffraction (XRD), Fourier transform infrared (FT-IR) and field emission scanning electron microscope (FESEM) analysis. In addition, the effect of key operating parameters on the leaching efficiency of REEs from coal-series kaolin was further investigated. The results show that REEs mainly existed in silicate & aluminosilicate state of coal-series kaolin. The mineral structure was destroyed by alkali fusion, and the leaching efficiency of REEs was increased by 59.08 %. At the sulfamic acid concentration of 0.5 mol/L, H2O2 concentration of 0.2 mol/L, solid to liquid ratio of 1/50, the leaching efficiencies of total rare earth elements, light rare earth elements and heavy rare earth elements were 84.91 %, 88.43 % and 71.03 %, respectively.

Reasons and strategies of reoperation after oblique lateral interbody fusion

To summarize the reasons and management strategies of reoperation after oblique lateral interbody fusion (OLIF), and put forward preventive measures. From October 2015 to December 2019, 23 patients who underwent reoperation after OLIF in four spine surgery centers were retrospectively analyzed. There were 9 males and 14 females with an average age of (61.89±8.80) years old ranging from 44 to 81 years old. The index diagnosis was degenerative lumbar intervertebral dics diseases in 3 cases, discogenic low back pain in 1 case, degenerative lumbar spondylolisthesis in 6 cases, lumbar spinal stenosis in 9 cases and degenerative lumbar spinal kyphoscoliosis in 4 cases. Sixteen patients were primarily treated with Stand-alone OLIF procedures and 7 cases were primarily treated with OLIF combined with posterior pedicle screw fixation. There were 17 cases of single fusion segment, 2 of 2 fusion segments, 4 of 3 fusion segments. All the cases underwent reoperation within 3 months after the initial surgery. The strategies of reoperation included supplementary posterior pedicle screw instrumentation in 16 cases;posterior laminectomy, cage adjustment and neurolysis in 2 cases, arthroplasty and neurolysis under endoscope in 1 case, posterior laminectomy and neurolysis in 1 case, pedicle screw adjustment in 1 case, exploration and decompression under percutaneous endoscopic in 1 case, interbody fusion cage and pedicle screw revision in 1 case. Visual analogue scale (VAS) and Oswestry disability index (ODI) index were used to evaluate and compare the recovery of low back pain and lumbar function before reoperation and at the last follow-up. During the follow-up process, the phenomenon of fusion cage settlement or re-displacement, as well as the condition of intervertebral fusion, were observed. The changes in intervertebral space height before the first operation, after the first operation, before the second operation, 3 to 5 days after the second operation, 6 months after the second operation, and at the latest follow-up were measured and compared. There was no skin necrosis and infection. All patients were followed up from 12 to 48 months with an average of (28.1±7.3) months. Nerve root injury symptoms were relieved within 3 to 6 months. No cage transverse shifting and no dislodgement, loosening or breakage of the instrumentation was observed in any patient during the follow-up period. Though the intervertebral disc height was obviously increased at the first postoperative, there was a rapid loss in the early stage, and still partially lost after reoperation. The VAS for back pain recovered from (6.20±1.69) points preoperatively to (1.60±0.71) points postoperatively(P<0.05). The ODI recovered from (40.60±7.01)% preoperatively to (9.14±2.66)% postoperatively(P<0.05). There is a risk of reoperation due to failure after OLIF surgery. The reasons for reoperation include preoperative bone loss or osteoporosis the initial surgery was performed by Stand-alone, intraoperative endplate injury, significant subsidence of the fusion cage after surgery, postoperative fusion cage displacement, nerve damage, etc. As long as it is discovered in a timely manner and handled properly, further surgery after OLIF surgery can achieve better clinical results, but prevention still needs to be strengthened.

Compatible multi-model output fusion for complex systems modeling via set operation-based focus data identification

Is an inferior model completely useless in complex systems modeling? This study proposes a novel multi-model output fusion approach that makes the best of the outputs from multi-models, i.e., using the limited superior results from inferior models to compensate for certain inferior results from even superior models. For fusing the outputs from multi-models, the weight assigned to each output is calculated based on two factors, namely the accuracy of each model and the similarity between the testing input and the focus data used for constructing the respective model. Specifically for similarity calculation, the focus data list is identified based on set operations. There are three theoretical contributions of this study, namely accuracy-and-similarity-based weight calculation, the set-operation-based similarity calculation which is an addition to traditional distance-based calculation, and the high compatibility of the proposed approach which is independent from any baseline approach. A practical case of overall reconnaissance capability evaluation of the Unmanned Aerial Vehicle (UAV) swarm is studied for validation. Primary results indicate that the proposed approach can outperform two single models: the backpropagation neural network (BPNN) and the Radial Basis Function (RBF) neural network. Further validations demonstrate that the proposed approach also outperforms multi-model output fusion with equal weights, without model accuracy, with varied focus data percentages ranging from 0.1 to 0.9. More importantly, the proposed approach remains effective in four different conditions of multi-model outputs fusion with improvements from 9.58 % to 38.03 %.

A Method of Multiple Targets and Sensors Track Association Analysis

In order to reduce the probability of leakage and improve detection accuracy, multiple radar equipment is required to perform multiple detections on the same airspace. Due to the overlap of airspace, there are multiple local tracks reported by various radar equipment that belong to the same target. The local tracks reported by each radar require track fusion to form a system track, and track association is a prerequisite for track fusion. This paper proposes a multi-target and multi-sensor track association analysis method, which can distinguish whether local tracks come from the same target and associate local tracks with the same target, providing conditions for track fusion to form system tracks. This method divides track association into two steps: Coarse association and fine association, which greatly improves the accuracy of track association.

Comparison between EAST and ITER tokamak

This paper provides an overview of nuclear fusion, which is considered as an ideal energy source for the future, with emphasis on its theoretical foundation and tokamak device. In nuclear fusion, the mass of products is always less than the mass of reactants, and this mass defect is converted into enormous released energy. Scientists have proposed several ideas to reach such extreme conditions for nuclear fusion. Among these concepts, the tokamak device which uses magnetic fields to generate heat and confine high temperature plasma is probably the most promising one. Over the last half century, with the unremitting endeavor of building and operation a series of international tokamak projects, magnetic confinement fusion has made a lot of satisfactory successes. In this paper, two representative tokamak devices are discussed, which are International Thermonuclear Experimental Reactor (ITER) and Experimental Advanced Superconducting Tokamak (EAST), about their major difference and similarities between them. ITER is constructed with the cooperation of many countries, and it is the largest tokamak project which targets to prove nuclear fusion is commercially feasible. EAST is another tokamak device which is located in China and made great achievements of magnetic confinement nuclear fusion, such as the record-breaking plasma temperature.

Development of a Robust Saccharomyces cerevisiae Strain for Efficient Co-Fermentation of Mixed Sugars and Enhanced Inhibitor Tolerance through Protoplast Fusion.

The economical and efficient commercial production of second-generation bioethanol requires fermentation microorganisms capable of entirely and rapidly utilizing all sugars in lignocellulosic hydrolysates. In this study, we developed a recombinant Saccharomyces cerevisiae strain, BLH510, through protoplast fusion and metabolic engineering to enhance its ability to co-ferment glucose, xylose, cellobiose, and xylooligosaccharides while tolerating various inhibitors commonly found in lignocellulosic hydrolysates. The parental strains, LF1 and BLN26, were selected for their superior glucose/xylose co-fermentation capabilities and inhibitor tolerance, respectively. The fusion strain BLH510 demonstrated efficient utilization of mixed sugars and high ethanol yield under oxygen-limited conditions. Under low inoculum conditions, strain BLH510 could completely consume all four kinds of sugars in the medium within 84 h. The fermentation produced 33.96 g/L ethanol, achieving 84.3% of the theoretical ethanol yield. Despite the challenging presence of mixed inhibitors, BLH510 successfully metabolized all four sugars above after 120 h of fermentation, producing approximately 30 g/L ethanol and reaching 83% of the theoretical yield. Also, strain BLH510 exhibited increased intracellular trehalose content, particularly under conditions with mixed inhibitors, where the intracellular trehalose reached 239.3 mg/g yeast biomass. This elevated trehalose content contributes to the enhanced stress tolerance of BLH510. The study also optimized conditions for protoplast preparation and fusion, balancing high preparation efficiency and satisfactory regeneration efficiency. The results indicate that BLH510 is a promising candidate for industrial second-generation bioethanol production from lignocellulosic biomass, offering improved performance under challenging fermentation conditions. Our work demonstrates the potential of combining protoplast fusion and metabolic engineering to develop superior S. cerevisiae strains for lignocellulosic bioethanol production. This approach can also be extended to develop robust microbial platforms for producing a wide array of lignocellulosic biomass-based biochemicals.

Multidimensional information fusion and broad learning system-based condition recognition for energy pipeline safety

Mechanical activities near energy pipelines pose a significant threat to energy transportation safety and energy system supply. The distributed acoustic sensing (DAS) system, which is an emerging intelligent sensing technology, can help to identify threat activities outside the pipeline. However, existing research has not distinguished fine-grained threat conditions because of insufficient information utilization and real-time requirements in the DAS system dataset, resulting in a large number of false alarms and time-consuming training in practice. To address these issues, this study proposes a pipeline radial threat condition recognition model based on multidimensional information fusion and a broad learning system (MIFBLS). First, the signal is preprocessed to improve the signal-to-noise ratio and generate the original features. Then, the extracted multidimensional temporal features are dimensionally reduced through information entropy, and the time–frequency map is fused and extracted based on a pretrained deep module, thereby achieving information fusion. Finally, the BLS incremental learning strategy is adopted to enhance the updating ability of the model and reduce the burden of data increments on the model training. Comparative experiments on a real-world natural gas pipeline dataset demonstrate the effectiveness and efficiency of the proposed method, indicating the potential for real-time monitoring and intelligent identification in pipeline transportation scenarios.

Measurements of laser-plasma instabilities in double-cone ignition experiments relevant to the direct-drive conditions at Shenguang-II Upgrade laser facility

Experiments have been performed to investigate laser-plasma instabilities (LPIs) relevant to the direct-drive inertial confinement fusion (ICF) conditions at the Shenguang-II Upgrade (SG-II UP) laser facility during the double-cone ignition (DCI) experimental campaigns, with the overlapped laser intensity ranging from 6×1014Wcm−2 to 1.8×1015Wcm−2 . An overall LPI scenario has been built from the collection and investigation of angularly distributed scattered light. Across broad ranges of laser and plasma parameters, relevant to the direct-drive ICF conditions, the dominance of the stimulated Raman side scattering (SRSS) was confirmed, and its coexistence with the two-plasmon decay (TPD) was also observed. Significant SRSS scattered light was observed across an extremely wide range of emission angles, concentrated at large angles, and demonstrated to be robust throughout the parameter space. Time-resolved spectral measurements show distinctly different SRSS behaviors along different angles and a slightly higher threshold compared to the TPD. The overall SRSS energy loss was measured, indicating a scattered light energy fraction of up to 6%. These results are of crucial importance for the precise assessment of the LPIs in the DCI as well as other laser directly driven ICF schemes.

Analysis of the Success of the "Central Kitchen" All-media Platform in the Context of Media Fusion

Media fusion has become an important trend in the contemporary news and communication industry, which benefits from the rapid development of technology and the push for digital transformation. Media fusion has not only changed the way and form of news dissemination, but also profoundly influenced the way and habits in which the public obtains information. Drawing on the context of media fusion, this paper examines the practical conditions of media fusion from the perspectives of positive and negative factors, and integrates the specific concept of a "Central Kitchen" to explore its innovative performance on the all-media platform. Through analysis, it is found that innovative performance encompasses cross-media integration, reengineering of news production processes, data mining, and intelligent processing, as well as socialization and decentralization. This paper also synthesizes the experiences and insights gained from the all-media platform "central kitchen" in today's media integration. As media convergence progresses, traditional audience identities are gradually transformed through interactions between production and consumption, impacting news content production methods. Furthermore, with deepening levels of media integration, there is an increasing demand for higher capabilities among media practitioners, necessitating collaboration across all forms of media to minimize news production costs while enhancing communication effectiveness.

A Passive Wireless Smart Washer for Locking Force Monitoring on the Orthopedic Pedicle Screw

A pedicle screw is a component for fixation in spine fusion surgery, often for patients with osteoporosis. Spine fusion condition highly depends on whether the pedicle screw is tightly fixed on spine, if not, spine fusion will not work properly. After the surgery, the first 3 months is the most crucial period, and bone healing situation cannot be shown through X-ray before the first radiologic images taken around the sixth week. Therefore, it is helpful to have a nonradiative method to monitor the locking force of the pedicle screw after surgery, especially during the early stage. Here a passive wireless force sensor is developed for monitoring the locking force of the pedicle screw, as we call it a smart washer. By integrating a capacitive ring-shape force sensor with an inductor, a passive LC sensor can be built by measuring the resonant frequency wirelessly. The smart washer is designed and calibrated to establish the relation between the locking force and resonant frequency, and then it is fixed with a screw in a porcine femur, with and without medium between the reader and inductor. When the locking force decreases from 8.3 to 0.9 N, the error is less than 0.5 N, and the maximum wireless sensing distance is 72 mm. However, the medium between the reader and the sensor inductor will affect the resonant frequency, but not the sensitivity. Therefore, the locking force variation can still be calculated by the resonant frequency shift accurately. Furthermore, by designing another five LC sensors with different operating resonant frequency ranges, it is possible to identify locking forces at different locations for more pedicle screws. To our knowledge, no LC force sensor was proposed to monitor the locking force of pedicle screws after surgery in the past.

GrainGNN: A dynamic graph neural network for predicting 3D grain microstructure

We propose GrainGNN, a surrogate model for the evolution of polycrystalline grain structure under rapid solidification conditions in metal additive manufacturing. High fidelity simulations of solidification microstructures are typically performed using multicomponent partial differential equations (PDEs) with moving interfaces. The inherent randomness of the PDE initial conditions (grain seeds) necessitates ensemble simulations to predict microstructure statistics, e.g., grain size, aspect ratio, and crystallographic orientation. Currently such ensemble simulations are prohibitively expensive and surrogates are necessary.In GrainGNN, we use a dynamic graph to represent interface motion and topological changes due to grain coarsening. We use a reduced representation of the microstructure using hand-crafted features; we combine pattern finding and altering graph algorithms with two neural networks, a classifier (for topological changes) and a regressor (for interface motion). Both networks have an encoder-decoder architecture; the encoder has a multi-layer transformer long-short-term-memory architecture; the decoder is a single layer perceptron.We evaluate GrainGNN by comparing it to high-fidelity phase field simulations for in-distribution and out-of-distribution grain configurations for solidification under laser power bed fusion conditions. GrainGNN results in 80%–90% pointwise accuracy; and nearly identical distributions of scalar quantities of interest (QoI) between phase field and GrainGNN simulations compared using Kolmogorov-Smirnov test. GrainGNN's inference speedup (PyTorch on single x86 CPU) over a high-fidelity phase field simulation (CUDA on a single NVIDIA A100 GPU) is 150×–2000× for 100-initial grain problem. Further, using GrainGNN, we model the formation of 11,600 grains in 220 seconds on a single CPU core.

Maximizing the Processing of Polymetallic Concentrates via Actinide Separation and Rare Earth Retrieval

During the last decades, the growing demand for rare earth elements (REEs) has led to numerous recent studies to recover these elements from various bearing ores and wastes. Therefore, the recovery of REEs from Ras Baroud polymetallic concentrate has been investigated in the current study. Physical beneficiation for the Ras Baroud pegmatite sample was carried out, yielding a concentrate for euxenite (Y), fergusonite (Y), xenotime (Y), monazite (Ce), allanite, thorite, uranothorite, and Hf-zircon, which resulted in raising the concentrations of rare earth elements, Th, Zr, U, and Ti in the sample. Fusion digestion processes with sodium hydroxide were studied using the Conceived Predictive Diagonal (CPD) technique. The three experimental digestion groups proved the dissolution of 99.9, 95.6, 99.9, 52.5, and 0.47% for REEs, Th, U, Ti, and Zr, respectively, under fusion conditions of 723 K, 120 min, 1/1.5 ore/alkali ratio, and − 100-μm particle sizes. Fusion kinetics, isotherms, and thermodynamics were investigated using several suggested models, namely, pseudo reversible first order, uptake general model, and shrinking core model which matched well with the experimental digestion results. Selective recovery of actinide content from REE content of the digested concentrate chloride solutions was accomplished using solvent extraction with di-2-ethyl hexyl phosphoric acid. About 99.9, 99.9, and 4.2% extraction efficiencies for Th, U, and REEs were performed, respectively, using 0.3 mol/L solvent concentration in kerosene as a diluent, 1/2 organic to aqueous ratio, an aqueous pH of 0.2, and 15-min contact time. Thorium and uranium ions were stripped with sulfuric acid solution 2.5 mol/L with 94 and 98% stripping efficiency, respectively. A highly purified REE precipitate was obtained from the raffinate solutions. Zircon mineralization tailings were obtained as a by-product through the alkaline digestion process.

High-strength and high-conductivity additively manufactured Cu-O alloy enabled by cellular microstructure

Printing Cu alloys with simultaneously high strength and high electrical conductivity by laser powder bed fusion remains a challenge. In conventionally manufactured Cu alloys, nanoprecipitates form through alloying elements to impart significant strengthening effects while having a lesser impact on electrical conductivity. However, the ultrahigh cooling rate of laser powder bed fusion promotes the supersaturated solid solution of alloying elements, resulting in severe lattice distortion in the Cu matrix and a significant compromise in electrical conductivity. Therefore, it is vital to select Cu alloying elements that substantially prevent the formation of a supersaturated solid solution in laser powder bed fusion to achieve outstanding strength while maintaining sufficient electrical conductivity. In this study, we fabricated a Cu-O alloy with high strength (491.6 MPa) and high electrical conductivity (68.0 % IACS) using laser powder bed fusion. The alloy features a cellular microstructure in which Cu2O nanoprecipitates are configured orderly as the cellular boundary. This cellular microstructure can promote strength by impeding dislocation motion and maintain a high electrical conductivity by preserving a longer free path for conductive electrons. This study explores the potential of combining the alloy composition and extreme process conditions of laser powder bed fusion to provoke unique microstructures and overcome the dilemma between strength and electrical conductivity in Cu alloys.

Demonstration of low-mode shape control in indirect-drive double shell implosions at the NIF

Double shell inertial confinement fusion is a concept for achieving robust thermonuclear burn that uses dense metal shells to compress deuterium-tritium (DT) fuel to fusion conditions. Double shell implosions are typically indirectly driven and involve a target that consists of a low-Z ablator, a foam layer, and a high-Z pusher surrounding the DT fuel. The goal of the campaign is to achieve a volumetric burn as radiation losses from the DT fuel are trapped by the opaque high-Z shell. The overall performance of double shell implosions relies on the efficient collisional transfer of kinetic energy between layers. The efficiency of this transfer (and therefore the overall performance of a given implosion) is degraded by the presence of low-mode asymmetries. P2 asymmetries are often observed in spatially resolved 2D radiographs of nominal double shell implosions. This work discusses three such experiments: one with an oblate P2 asymmetry, one with a prolate P2 asymmetry, and one with an approximate spherical symmetry. After performing a shape analysis of the oblate and prolate implosions to quantify asymmetries, these experimental results were compared with the results of hydrodynamic simulations for the two experiments. Differences between the experiment and simulation were then used to design an approximately spherical implosion by altering the incident laser cone fraction. Radiographs from the experiment that implemented the modified cone fraction show evidence of an implosion that is approximately spherical until bang time. This design is intended to serve as a point design for future studies that will seek to optimize various aspects of the double shell target.

Separation and determination of percentage level niobium in gadolinium zirconate using bifluoride fusion and spectrophotometry

A simple and reliable method for the determination of percentage level niobium (3.5-8 %) in highly refractive gadolinium zirconate (Gd2Zr2O7) doped with niobium, a burnable poison material of nuclear reactor has been proposed. A fusion procedure using ammonium bifluoride (ABF) with Gd2Zr2O7 was developed. Fusion conditions such as time, temperature and ABF to sample ratio were optimised for realizing quantitative recovery of Nb. The fused masses obtained under different conditions were characterized using SEM-EDX and XRD to comprehend the conversion of the refractive oxide form to soluble fluoro compounds and their identification. The real samples fused under optimized conditions were dissolved in 2 % tartaric acid solution and the same was used for quantifying Nb employing spectrophotometry using 5-bromo pyrogallol red as chromogenic agent. Precision of the measurements was better than 3 % (RSD) for Nb(V) at 0.5 ppm and the accuracy of the method is 99.8 %.

Elevated Electron Temperature Coincident with Observed Fusion Reactions in a Sheared-Flow-Stabilized Z Pinch.

The sheared-flow-stabilized Z pinch concept has been studied extensively and is able to produce fusion-relevant plasma parameters along with neutron production over several microseconds. We present here elevated electron temperature results spatially and temporally coincident with the plasma neutron source. An optical Thomson scattering apparatus designed for the FuZE device measures temperatures in the range of 1-3keV on the axis of the device, 20cm downstream of the nose cone. The 17-fiber system measures the radial profiles of the electron temperature. Scanning the laser time with respect to the neutron pulse time over a series of discharges allows the reconstruction of the T_{e} temporal response, confirming that the electron temperature peaks simultaneously with the neutron output, as well as the pinch current and inductive voltage generated within the plasma. Comparison to spectroscopic ion temperature measurements suggests a plasma in thermal equilibrium. The elevated T_{e} confirms the presence of a plasma assembled on axis, and indicates limited radiative losses, demonstrating a basis for scaling this device toward net gain fusion conditions.

Imaging the dynamics of initial laser-driven shocks and blowoff plasmas in polystyrene under laser-direct-drive fusion conditions

The dynamics of laser-driven shock propagation in solid ablator material, along with the concomitant processes of shinethrough and blowoff plasma plume expansion, at conditions relevant to laser-direct-drive fusion are measured using a tabletop system capable of supporting rapid dataset development with significantly higher resolution compared to existing methods. The experimental results are directly compared with current radiation-hydrodynamic simulations using the two-dimensional code . Discrepancies between experiments and simulations are evidenced and attributed to limitation of simulation to account for important processes involved in the solid-to-plasma transition. Published by the American Physical Society 2024

MamlFormer: Priori-experience guiding transformer network via manifold adversarial multi-modal learning for laryngeal histopathological grading

Pathologic grading of laryngeal squamous cell carcinoma (LSCC) plays a crucial role in diagnosis, prognosis, and migration. However, the grading performance and interpretability of the intelligent grading model based on LSCC low magnification images are poor. This is because it lacks the delicate nuclear information and information more relevant to grading contained in the high magnification images labeled by pathologists. Yet, low magnification images have information such as tissue texture and contours. Thus, we proposed an end-to-end transformer network with manifold adversarial multi-modal learning (MamlFormer). It effectively fuses and learns LSCC high and low magnification pathology image modalities. Firstly, we demonstrate the feasibility and sufficient conditions for modal fusion of LSCC high and low magnification images from Hoeffding's inequality and multimodal co-regularization. Secondly, we design a new manifold block. It constructs the manifold subspace by some principles. Those principles are divisibility, recoverability, and local distance closest of the feature matrix before and after the mapping of the LSCC each magnification image modalities. Meanwhile it can well solve the problems of redundant feature matrix information and weak modal semantic consistency after multimodal learning. Thirdly, we utilize the encoder and the adversarial loss function to implement adversarial block. It can adaptively learn the latent metrics of the modal distributions of LSCC high and low magnification images. Therefore, it also enhances the complementarity of LSCC high and low magnification image modalities. Then, numerous experiments show that MamlFormer outperforms other SOTA models in both grading performance and interpretability. Finally, we also performed generalization experiments on highly prevalent cervix squamous cell carcinoma. The MamlFormer over is superior to other SOTA models in terms of grading performance and interpretability. This indicates its excellent generalization performance and clinical practicability.