Dense Mode Research Articles

Regulation and Liquid Sensing of Electromagnetically Induced Transparency-like Phenomena Implemented in a SNAP Microresonator

Optical microresonators supporting whispering-gallery modes (WGMs) have become a versatile platform for achieving electromagnetically induced transparency-like (EIT-like) phenomena. We theoretically and experimentally demonstrated the tunable coupled-mode induced transparency based on the surface nanoscale axial photonics (SNAP) microresonator. Single-EIT-like and double-EIT-like (DEIT-like) effects with one or more transparent windows are achieved due to dense mode families and tunable resonant frequencies. The experimental results can be well-fitted by the coupled mode theory. An automatically adjustable EIT-like effect is discovered by immersing the sensing region of the SNAP microresonator into an aqueous environment. The sharp lineshape and high slope of the transparent window allow us to achieve a liquid refractive index sensitivity of 2058.8 pm/RIU. Furthermore, we investigated a displacement sensing phenomenon by monitoring changes in the slope of the transparent window. We believe that the above results pave the way for multi-channel all-optical switching devices, multi-channel optical communications, and biochemical sensing processing.

Time-domain structural model updating following the Bayesian approach in the absence of system input information

The modal-domain structural model updating and damage detection methods own the advantage that they perform the whole computation without the information of system input, when compared to the time-domain. The system input required in the time-domain methods is hardly measured in realistic field tests of civil engineering structures. However, the time-domain methods own the advantage that they carry out model updating based on raw measured dynamic responses but not the processed (and compressed) modal parameters. The modal-domain methods only uses the compressed modal parameters, and some useful system information may be lost in the compressing process. Thus, higher errors and uncertainties are induced in the model updating results from the modal-domain methods. To take advantages and avoid disadvantages from the traditional time-domain and modal-domain model updating methods, a novel time-domain MCMC-based Bayesian model updating method without the information of the system input is developed in this paper by incorporating a newly developed dynamic response reconstruction technique and extending the formulation of the posterior probability density function (PDF) and the corresponding bridge PDF in the MCMC sampling process. A new modal decomposition algorithm is proposed in this paper for accurate modal response decomposition even in case of dense or sparse modes, and filters are not required for higher computation accuracy. Experimental case studies on a two-story steel frame were conducted to illustrate the proposed methodology in both structural model updating and damage detection. According to the analysis results, it is concluded that the proposed methodology outperforms the modal-domain methods especially in the cases with small number of measured DOFs.

Broadband tunable single-frequency Yb-doped fiber laser with dual ring subcavity and Sagnac ring filter

We demonstrate a broadband tunable ultra-narrow linewidth single-frequency (SF) Yb-doped fiber laser (YDFL). A dual ring subcavity (DRS) as a filtering component to eliminate the dense longitudinal mode inherent in YDFL and theoretically analyzed its filtering characteristics. By inserting a 0.8 m unpumped polarization-maintaining Yb-doped fiber in the Sagnac ring, a dynamic narrowband grating (DNG) based on the saturable absorption (SA) effect is constructed to ensure SF laser output over the entire wavelength tuning range. We achieve a wide wavelength tunable SF laser output in the range of ∼74 nm, covering 1021.06 nm to 1094.98 nm. The laser has an optical signal-to-noise ratio (OSNR) of more than 62.5 dB, a slope efficiency of 3.57 %, and an ultra-narrow linewidth of 622 Hz over the entire tunable range. In addition, the stability of the wavelength and power fluctuations below 0.018 nm and 0.67 dB, respectively. This tunable SF fiber laser has a compact structure, lower cost, and excellent performance, and is expected to be used in the fields of coherent communications, precision metrology, and so on.

Cerberus: Triple Mode Acceleration of Sparse Matrix and Vector Multiplication

The multiplication of sparse matrix and vector (SpMV) is one of the most widely used kernels in high-performance computing as well as machine learning acceleration for sparse neural networks. The design space of SpMV accelerators has two axes: algorithm and matrix representation. There have been two widely used algorithms and data representations. Two algorithms, scalar multiplication and dot product, can be combined with two sparse data representations, compressed sparse and bitmap formats for the matrix and vector. Although the prior accelerators adopted one of the possible designs, it is yet to be investigated which design is the best one across different hardware resources and workload characteristics. This paper first investigates the impact of design choices with respect to the algorithm and data representation. Our evaluation shows that no single design always outperforms the others across different workloads, but the two best designs (i.e., compressed sparse format and bitmap format with dot product) have complementary performance with trade-offs incurred by the matrix characteristics. Based on the analysis, this study proposes Cerberus, a triple-mode accelerator supporting two sparse operation modes in addition to the base dense mode. To allow such multi-mode operation, it proposes a prediction model based on matrix characteristics under a given hardware configuration, which statically selects the best mode for a given sparse matrix with its dimension and density information. Our experimental results show that Cerberus provides 12.1× performance improvements from a dense-only accelerator, and 1.5× improvements from a fixed best SpMV design.

Switchable Dual-Wavelength Thulium-Doped Fiber Laser Based on Polarization-Maintaining Fiber Bragg Grating and Compound Cavity Filter

This paper presents experimental evidence regarding a novel switchable dual-wavelength thulium-doped fiber laser (TDFL). Wavelength switching is achieved by combining a polarization-maintaining fiber Bragg grating (PM-FBG) with a polarization controller (PC). The three-coupler double-ring compound cavity (TC-DRC) structure, acting as a mode-selection filter, is designed to select a single longitudinal mode (SLM) from the dense longitudinal modes. This paper introduces the design and fabrication method of the TC-DRC filter and analyzes, in detail, the mechanism for SLM selection. The experimental results demonstrate that the designed filter exhibits excellent performance. By adjusting the PC, the TDFL achieves stable SLM operation at the wavelengths of 1940.54 nm and 1941.06 nm, respectively. The optical signal-to-noise ratio (OSNR) is superior to 65 dB. When the TDFL is tested at room temperature, there is no significant wavelength drift, and power fluctuations are less than 1.5 dB. The operation of the SLM is verified through the self-heterodyne method, and the laser maintains stable SLM states for both wavelengths after continuous operation for an hour. Furthermore, based on the phase noise demodulation method, the linewidths of both wavelengths are measured to be less than 10 kHz at the integration time of 0.001 s.

Vertical crack identification of arch dam under underwater explosion based on mode transition

The failure process of concrete arch dam subjected to underwater explosion has been the research hotspot in recent years, however, there is a lack of an effective method to identify the crack in dam body. To solve this problem, the crack identification of arch dam caused by underwater explosions has been studied in this paper. Firstly, an analysis method to identify the vertical crack in dam body is proposed, according to a comparison of the modal characteristics of the arch dam with and without crack. Secondly, the reliability of the proposed method is verified by a small-scale arch dam that cracking along the crown beam attacked by foundation explosion. The modal change of the dam induced by cracking is captured in both experiment and numerical simulation. Thirdly, this method is applied to analyze the influence of vertical crack on the modal of a 240 m high arch dam. Some findings need be noted as follows: vertical cracks lead to modal transition of arch dam of two adjacent dense modes, such as orders between 4 and 5, 7 and 8, and 11 and 12 respectively in this case, the ratio of modal transition is 6.04% in the first 8 modes and it is 14.1% in the first 12 modes; The effect of cracks on the mode of arch dam is related to the spatial position of the vibration mode peak, the nearer the two are, the more significant the change in that order of mode is. This finding indicates that the vertical cracks may lead to mode transition or mode loss of arch dams, it should be considered that the change of arch-cantilever system vibration mode sequence in the identification of the crack of arch dam caused by underwater explosion.

Tetraphenylethylene-Based Hydrogen-Bonded Organic Frameworks (HOFs) with Brilliant Fluorescence.

By synergistically employing four key strategies: (I) introducing tetraphenylethylene groups as the central core unit with aggregation-induced emission (AIE) properties, (II) optimizing the π-conjugated length by extending the building block branches, (III) incorporating flexible groups containing ethylenic bonds, and (IV) applying crystal engineering to attain dense stacking mode and highly twisty conformation, we successfully synthesized a series of hydrogen-bonded organic frameworks (HOFs) exhibiting exceptional one/two-photon excited fluorescence. Notably, when utilizing the fluorescently superior building block L2, HOF-LIFM-7 and HOF-LIFM-8 exhibiting high quantum yields (QY) of 82.1 % and 77.1 %, and ultrahigh two-photon absorption (TPA) cross-sections of 148959.5 GM and 123901.1 GM were achieved. These materials were successfully employed in one and two-photon excited lysosome-targeting cellular imaging. It is believed that this strategy, combining building block optimization and crystal engineering, holds significant potential for guiding the development of outstanding fluorescent HOF materials.

Tetraphenylethylene‐Based Hydrogen‐Bonded Organic Frameworks (HOFs) with Brilliant Fluorescence

AbstractBy synergistically employing four key strategies: (I) introducing tetraphenylethylene groups as the central core unit with aggregation‐induced emission (AIE) properties, (II) optimizing the π‐conjugated length by extending the building block branches, (III) incorporating flexible groups containing ethylenic bonds, and (IV) applying crystal engineering to attain dense stacking mode and highly twisty conformation, we successfully synthesized a series of hydrogen‐bonded organic frameworks (HOFs) exhibiting exceptional one/two‐photon excited fluorescence. Notably, when utilizing the fluorescently superior building block L2, HOF‐LIFM‐7 and HOF‐LIFM‐8 exhibiting high quantum yields (QY) of 82.1 % and 77.1 %, and ultrahigh two‐photon absorption (TPA) cross‐sections of 148959.5 GM and 123901.1 GM were achieved. These materials were successfully employed in one and two‐photon excited lysosome‐targeting cellular imaging. It is believed that this strategy, combining building block optimization and crystal engineering, holds significant potential for guiding the development of outstanding fluorescent HOF materials.

Design and experiment of brush-roller ginkgo leaf picker for the dwarf dense planting mode

At present, ginkgo leave are still picked manually. A brush roller ginkgo leaf picker has been designed to improve harvesting efficiency and reduce losses caused by manual failure to pick leaves in time under large-scale planting areas. The ginkgo leaf picker is mainly composed of crawler chassis, gantry frame, brush roller picking parts, and collecting box. The kinematics of the brush roller is analyzed for the picking omission situation. An experimental platform for picking ginkgo biloba leaves was established. Three parameters, namely roller speed, moving speed, and roller inclination, were selected for optimization. Then the maximum net harvest rate and the minimum damage rate were achieved. The orthogonal test showed that when the roller speed was 130 rev/min, the moving speed was 0.25 m/s, and the roller inclination was 32°, the picking effect was the best, the net harvest rate was 93.32%, the damage rate was 1.42%, and the damage degree of the trunks was slight. The experiment proved that the brush-roller ginkgo leaf picker has a good picking effect, which can provide a reference for the optimization design of ginkgo leaf harvesting machinery under the dwarf dense planting mode.

Method and Experimental Study of Oscillator Frequency Optimization of Distributed Tuned Mass Dampers for Broadband Multimodal Vibration Mitigation of Reinforced Concrete Wall

Distributed tuned mass dampers (dTMD) can effectively mitigate the broadband vibration of a structure. However, when the vibration frequency in question reaches several hundred hertz, traditional optimization methods represented by fixed point theory are difficult to apply due to dense modal density, complex boundary conditions, and vibration inputs. This paper proposes the minimax method based on modal damping to optimize the oscillator’s frequency. Two typical wall panel specimens are tested to evaluate the proposed method. The mode shape of the uncontrolled wall and the vibration mitigation effect of the stacked sandwich-damped TMD under single-point bidirectional excitation is tested. The correlation between the modal damping and the vibration mitigation effect is evaluated. The results show that the RC wall panel has a dense mode when the frequency of interest reaches 300 Hz and above; the distributed stacked sandwich-damped TMDs can effectively mitigate the vibration of the RC wall panel in the frequency range of 200~450 Hz; and that the idea of optimizing the frequency of dTMD based on modal damping is feasible.

Mobile crowdsensing framework for drive-by-based dense spatial-resolution bridge mode shape identification

Moving vehicles equipped with various types of sensors can efficiently monitor the health conditions of a population of transportation infrastructure such as bridges. This paper presents a mobile crowdsensing framework to identify dense spatial-resolution bridge mode shapes using sparse drive-by measurements. The proposed method converts mode shape identification into a physical-informed optimization problem with two objective function terms. The first objective minimises the mode shape identification error based on the fact that the ratio of a specific order mode shape value at any two locations is time-invariant. Since the bridge mode shape should be globally smooth even when the local stiffness is discontinuous, the smoothness of the identified mode shape is introduced as the second objective. The feasibility and advantages of the proposed model are verified numerically and through large-scale experimental studies. Numerical results demonstrate that the proposed method can efficiently identify bridge mode shapes with a desirable accuracy. The adverse effects of road roughness and measurement noise on the mode shape identification accuracy are substantially suppressed by introducing crowdsensing and making use of collected responses over multiple trips. The applicability of the proposed method for bridges having varying cross sections and multiple spans is also studied. A series of drive-by tests with different vehicle masses and speeds are conducted on a large-scale footbridge. The experimental results verify that the proposed method can accurately identify the bridge mode shapes and is robust to vehicle mass and speed variation. The identification accuracy of large-scale bridge mode shapes using crowdsensing drive-by measurements is demonstrated in this study.

Experimental design and optimization of a novel solids feeder device in energy efficient pneumatic conveying systems

Pneumatic conveying of powders is an engineering process used for conveying dry granulate or powder material where energy consumption is a significant cost factor and contributes to greenhouse gas emissions. In this RD&I project, work was conducted to model pneumatic conveying and bulk characteristics of the particulate product being conveyed. Because pneumatic conveyance is highly empirical, general models are difficult to establish. Due to these limitations, evaluating energy efficiency is usually limited to a specific experimental range of conditions. This work is based on engineering optimization of a workflow with data from an industrial operation commanded by a Programmable Logic Controller (PLC) with a control algorithm, performing logical, sequential, and timed tasks for plant control. The PLC communicates with a Human–Machine Interface and a Supervision and Control System, which are the means of interaction through a graphical environment interface with the process operator. By applying mathematics to introduce a systematic method to select the gas (air) pressure and flow necessary to operate a pneumatic conveying system in dense phase mode, it has been shown, on an industrial scale of 10 t/h, the feasibility of controlling an efficient pneumatic conveying system manipulating only two input parameters. This allows operation at pre-determined conveying rates with lower operational expenditures. The same methodology can be explored for several other systems.

Results of dose-dense neoadjuvant chemotherapy in the treatment of stages IB2—IIB cervical cancer

Background. Applied standard methods of cervical cancer (CC) treatment have several disadvantages in terms of local and systemic toxicity and the risks of local recurrence. Our study proposed the use of neoadjuvant chemotherapy (NACT) in a dose-dense mode, followed by radical hysterectomy, in order to improve the results of treatment of locally advanced CC (IB2-IIB stages) (FIGO)/T1b2-2bN0M0 (TNM).Aim. Improvement of immediate and long-term results in treatment of locally advanced CC IB2-IIB stages (FIGO), determining the immediate effectiveness of dose-dense NACT, based on magnetic resonance imaging (MRI) data, and the identification of prognostic factors for tumor response to chemotherapy.Materials and methods. Between April 2016 to November 2019 120 women with IB2-IIB stages (FIGO) were included in this study. Of these, 58 patients underwent 3 cycles of intensified NACT according to the AP scheme (cisplatin 75 mg/m2, doxorubicin 35 mg/m2) and 62 patients underwent NACT according to the TR scheme (cisplatin 60 mg/m2, paclitaxel 60 mg/m2). The assessment of the effect of intensified NACT was performed according to the RECIST 1.1 criteria according to MRI data. Also 2 retrospective groups of patients were recruited: group I - surgical treatment without preoperative chemotherapy (25 patients with IB2-IB3 stages (FIGO) from November 2009 to August 2019), group II - concurrent chemoradiotherapy (44 patients with stage IIB cervical cancer (FIGO) from January 2013 to June 2018).Results. Complete clinical response was detected in 12 (10.0 %) patients, partial response - in 69 (57.5 %) patients, stable disease - in 35 (29.2 %) patients, progression was observed in 4 cases (3.3 %). After NACT in a dose dense mode, due to an objective clinical response (complete response, partial response and stable disease for stage IIA according to RECIST 1.1 criteria), 99 (82.5 %) patients underwent surgical treatment. For 21 (17.5 %) patients due to the absence achievement of objective clinical response (stable disease for stage IIB according to RECIST 1.1 criteria or progression), performed concurrent chemoradiotherapy. During the observation period, 113 (94.2 %) patients remained alive. Mortality for the first year was 1.7 % (2 cases). There was a coincidence of the results of a complete response according to MRI in 11 (91.7 %) cases out of 12 cases with the pathology data. According to multivariate analysis, it was revealed that parametrial invasion, according to the histological examination of the surgical material, is an independent prognostic factor for recurrence. According to comparative evaluation of disease-free survival curves between groups of IB2-IIB stages (FIGO) patients who received and did not receive dose-dense chemotherapy, it was found that disease- free survival in the group with NACT was higher compared to the group without NACT (p = 0.03, the log-rank criterion).Conclusions. The use of dose-dense NACT is an effective treatment mode for patients with IB2-IIB CC (FIGO), which in most cases allows achieving a tumor response with subsequent improvement in immediate and long-term results and can be considered as an alternative to standard treatment for locally advanced CC.

Single-Longitudinal Mode Ytterbium-Doped Fiber Laser with Ultra-Narrow Linewidth and High OSNR Using a Double-Ring Passive Subcavity

A continuous wave (CW) ultra-narrow linewidth single-longitudinal mode (SLM) ytterbium-doped fiber laser (YDFL) based on narrowband fiber Bragg grating (NB-FBG) and double-ring passive subcavity (DR-PS) was studied. The filtering characteristics of the double-ring passive subcavity are analyzed theoretically, and it is used as a high-precision mode filter to eliminate the dense longitudinal mode and mode hopping of YDFL and ensure that the laser operates in the SLM state. Experimental results show that the laser has a central wavelength of 1030.052 nm at room temperature, an optical signal-to-noise ratio of up to 73 dB, and an ultra-narrow linewidth of 355 Hz. In addition, we measured the short-term and long-term stability of the laser, with wavelength and power fluctuations of less than 0.008 nm and 0.19 dB, respectively, over 120 min. As a result, we obtain SLM YDFL with high stability, ultra-narrow linewidth, and a high optical signal-to-noise ratio.

Dynamically reconfigurable variable-precision sparse-dense matrix acceleration in Tensorflow Lite

In this paper, we present a dynamically reconfigurable hardware accelerator called FADES (Fused Architecture for DEnse and Sparse matrices). The FADES design offers multiple configuration options that trade off parallelism and complexity using a dataflow model to create four stages that read, compute, scale and write results. FADES is mapped to the programmable logic (PL) and integrated with the TensorFlow Lite inference engine running on the processing system (PS) of a heterogeneous SoC device. The accelerator is used to compute the tensor operations, while the dynamically reconfigurable approach can be used to switch precision between int8 and float modes. This dynamic reconfiguration enables better performance by allowing more cores to be mapped to the resource-constrained device and lower power consumption compared with supporting both arithmetic precisions simultaneously. We compare the proposed hardware with a high-performance systolic architecture for dense matrices obtaining 25% better performance in dense mode with half the DSP blocks in the same technology. In sparse mode, we show that the core can outperform dense mode even at low sparsity levels, and a single-core achieves up to 20x acceleration over the software-optimized NEON RUY library.

Sub-kHz Narrow-Linewidth Single-Longitudinal-Mode Thulium-Doped Fiber Laser Utilizing Triple-Coupler Ring-Based Compound-Cavity Filter

This paper proposes and demonstrates a single-longitudinal-mode thulium-doped fiber laser using a passive triple-coupler ring-based compound-cavity filter (TCR-CC) and a uniform fiber Bragg grating. For the first time, the TCR-CC filter is used to select a single mode from dense longitudinal modes. Experimental results show that laser in the wavelength of 1941.28 nm can maintain exceptional stability with an optical signal-to-noise ratio of 74.1 dB. The measured maximum wavelength drift and power fluctuation are 0.01 nm and 0.45 dB, respectively. Meanwhile, the measured linewidth of the laser is 910 Hz, and the relative intensity noise is below −125.82 dB/Hz above 2 MHz frequencies.

Adversarial learning-based multi-level dense-transmission knowledge distillation for AP-ROP detection.

The Aggressive Posterior Retinopathy of Prematurity (AP-ROP) is the major cause of blindness for premature infants. The automatic diagnosis method has become an important tool for detecting AP-ROP. However, most existing automatic diagnosis methods were with heavy complexity, which hinders the development of the detecting devices. Hence, a small network (student network) with a high imitation ability is exactly needed, which can mimic a large network (teacher network) with promising diagnostic performance. Also, if the student network is too small due to the increasing gap between teacher and student networks, the diagnostic performance will drop. To tackle the above issues, we propose a novel adversarial learning-based multi-level dense knowledge distillation method for detecting AP-ROP. Specifically, the pre-trained teacher network is utilized to train multiple intermediate-size networks (i.e., teacher-assistant networks) and one student network by dense transmission mode, where the knowledge from all upper-level networks is transmitted to the current lower-level network. To ensure that two adjacent networks can distill the abundant knowledge, the adversarial learning module is leveraged to enforce the lower-level network to generate the features that are similar to those of the upper-level network. Extensive experiments demonstrate that our proposed method can realize the effective knowledge distillation from the teacher to student networks. We achieve a promising knowledge distillation performance for our private dataset and a public dataset, which can provide a new insight for devising lightweight detecting systems of fundus diseases for practical use.

Stable narrow-linewidth single-longitudinal-mode thulium-doped fiber laser by exploiting double-coupler-based double-ring filter

A stable single-longitudinal-mode (SLM) thulium-doped fiber laser (TDFL) was combined with a phase-shift fiber Bragg grating (PS-FBG), uniform fiber Bragg grating (UFBG), and a double-coupler-based double-ring (DCDR) filter. Mode selection was achieved by using a sub-ring cavity design. For the first time, the structure of the DCDR filter was used to select a single mode from dense longitudinal modes in a ring cavity TDFL. The theoretical analysis of the DCDR filter and the principle for achieving SLM operation are presented in detail. The experimental results demonstrated the exceptional performance of the proposed DCDR filter. Within the observation period of an hour, the maximum fluctuations of center wavelength and optical power are 0.02 nm and 0.47 dB, respectively. The linewidth of the lasing wavelength is 4.8 kHz and the relative intensity noise is below −129.48 dB/Hz at frequency above 2 MHz. The power scaling, slope efficiency, and threshold power of the TDFL were also investigated. The proposed SLM TDFL shows great potential for free-space optical communication applications.

Three-Dimensional Vibrational Characteristics of Functionally Graded Rotating Microbeams Based on the Higher-Order Refined Shear Deformation Theory

This study establishes a novel three-dimensional dynamic model for functionally graded rotating microbeams simultaneously considering flapwise and chordwise shear effects based on a refined shear deformation theory and the modified couple stress theory. Five kinematic variables are introduced to describe the rotating microbeam’s axial, flapwise, and chordwise motions using a higher-order refined shear deformation theory. The related strain energy and kinetic energy functionals are discretized by the assumed mode method. Applying the Euler–Lagrange variational formulation, the governing equations for the rotating microbeam are obtained. Selective examples are provided to demonstrate the convergence and effectiveness of the proposed model. Finally, the impacts of slenderness ratio, rotation speed, material index, material length scale parameter (MLSP), and Poisson’s effect on the vibrational characteristics of the microbeam are examined. Numerical results indicate that the mode shape transition and frequency loci veering phenomena could easily occur under slight changes in the above factors when dense modes appear in the rotating microbeam; moreover, there is a positive/negative competitive relationship among the effects of rotation speed, MLSP, and shear deformation; Poisson’s effect seriously affects the coupling between axial and non-axial motions of the rotating microbeam. The results obtained are expected to provide data accumulation and methodological basis for the failure analysis and vibration shape determination of micro air vehicles.

Optimal Design and Testing of a Crawler-Type Flax Combine Harvester

China is a large flax-growing country, with planting area and production ranking among the top three in the world. However, the cultivation range of flax in China is very broad, complex, and diverse, resulting in different planting scales and patterns, making it difficult to apply foreign large combine harvesters, and China lacks a dedicated flax combine harvester. This research improved the design of the 4LZ-4.0 crawler-type flax combine harvester for the regional features and flax cropping patterns in China. First, the structure, technical parameters, and working principles of the machine were introduced; second, the theoretical analysis and optimization of key components were performed; and finally, with the advancing speed of the machine, the speed of the threshing drum, and the speed of the suction fan as independent variables and the rate of removal and the total loss rate as response values, a three-factor, three-level response surface analysis method was used. For each component and response value, a mathematical model was created, and the factors and their interactions were evaluated and confirmed. The results demonstrated that the three parameters impact the threshing drum speed, advancing speed, and centrifugal fan speed in that order of priority, as well as the total loss rate in that order of priority. The machine’s optimal operating settings were 1.5 m·s−1 advancing speed, 788.49 r·min−1 threshing drum speed, and 885.34 r·min−1 centrifugal fan speed, and the validation test results indicated that under the typical dryland dense flax cultivation mode, it had a 97.46% threshing rate and 2.99% total loss rate after the test. This demonstrated that optimizing operational parameters may decrease losses in the process of mechanical flax harvesting, enhance harvesting efficiency, and satisfy the marketable flax harvesting standards.