41results about How to "Break through dependence" patented technology

The invention relates to the technical field of disaster early warning processing, in particular to a mountain torrent disaster early warning system based on deep learning and air-space-ground integrated perception. The system comprises the following modules: a space-air-ground sensing module, an edge calculation preprocessing module, a deep learning prediction module and an intelligent decision and early warning module. Wherein the air-space-ground sensing module is used for collecting multi-source data; wherein the multi-source data comprises earth surface deformation data obtained through a spaceborne synthetic aperture radar, high-precision topographic data obtained through an unmanned aerial vehicle laser radar, hydrological sensing data obtained through a ground internet-of-things distributed sensor, and meteorological numerical forecasting data obtained through a meteorological data interface. Through air-space-ground integrated perception, physical constraint deep learning and edge intelligence technologies, mountain torrent multi-source data collaborative acquisition and accurate mountain torrent disaster prediction are realized, so that the comprehensiveness, reliability and timeliness of mountain torrent disaster early warning are improved.

The application discloses a novel full-line fast distance protection method under a power system, which comprises the following steps: S1, continuously monitoring the voltage and current on one side of the line, and starting the protection when detecting that the current mutation variable exceeds a set threshold; S2, establishing an accurate line differential equation based on a pi type equivalent model; S3, constructing the differential equation in the time domain by using the collected voltage and current data, and directly and real-timely solving the equivalent transition resistance of the fault point; S4, establishing a relationship model of the additional measurement impedance and the equivalent transition resistance, accurately compensating the apparent impedance measured by the protection device, and accurately calculating the real distance from the protection installation to the fault point; and S5, comparing the calculated fault distance with a preset protection section, and realizing the fast isolation of the fault; the application has the advantages of weak feed adaptability improvement, strong anti-transition resistance capability, long line anti-maloperation optimization and fast action speed, and can provide effective technical support for the main protection configuration of new energy sending-out lines and high-proportion power electronic power grids.

The utility model relates to a greening structure suitable for a steep soil slope. The method is suitable for the technical field of slope greening engineering. According to the technical scheme, according to the greening structure suitable for the steep soil slope, the surface of the steep soil slope is excavated to form a slope protection construction face, the slope protection construction face is covered with a slope protection structure, and the greening structure comprises drainage holes which are drilled in a slope body of the steep soil slope, and the multiple drainage holes are arranged at intervals in a multi-layer mode along the slope face of the slope; the planting holes are drilled in the slope surface of the steep soil slope, the multiple planting holes communicate with the multiple drainage holes in a one-to-one correspondence mode, and the ends, away from the drainage holes, of the planting holes penetrate through the slope protection structure; the planting tanks are arranged in the planting holes, are used for planting green plants and can discharge excessive water inside; and the drainage floral tube is arranged in the drainage hole and can collect surplus water in the slope body and guide the water into the planting tank so as to irrigate green plants.

This invention discloses an online inspection device and method for the light conversion film of photovoltaic modules. The online inspection device includes a conveyor roller, an inspection cover, multiple top excitation light source units, multiple top signal acquisition units, multiple edge excitation light source units, multiple edge signal acquisition units, and a processing unit. The conveyor roller is used to transport photovoltaic modules along a first horizontal direction; the inspection cover has an openable and closable inlet / outlet on its side wall along the first horizontal direction; the top excitation light source units and top signal acquisition units are both located on the top inner wall of the inspection cover and are used to project a first wavelength of light onto the photovoltaic modules inside the inspection cover; the edge excitation light source units and edge signal acquisition units are both located on the inner side wall of the inspection cover along a second horizontal direction and are used to collect the second wavelength of light projected by the photovoltaic modules inside the inspection cover. This online inspection device achieves accurate inspection of non-transparent photovoltaic modules without blind spots, breaking through the dependence of traditional through-type inspection on transparent photovoltaic modules.

The invention provides a Hall memristor based on an aged insulator and storage and calculation integrated computing equipment, and the method comprises the steps: firstly, through symmetry engineering design, employing a quasi-two-dimensional aged insulator which breaks the central symmetry and rotational symmetry, supporting a quantum abnormal Hall state at the same time, and generating track magnetization and the coupling of an in-plane electric field and out-of-plane magnetization; secondly, a quantum geometric mechanism is introduced into the quantum Hall state without the Fermi surface, intrinsic track magnetoelectric coupling is achieved, and the dependence of a traditional mechanism on the Fermi surface is broken through; and the domain wall is directly driven by the in-plane electric field to move to realize magnetization overturning, so that the participation of body current is avoided, and the energy consumption is remarkably reduced. And finally, the device combines the overturning of the magnetic state with the hysteresis behavior of the Hall resistor, and the inherent non-volatility and memory characteristics of the device provide a key device basis for constructing novel calculation hardware integrating storage and calculation.

The invention discloses a Tibetan type cultural creative graph automatic generation method and system based on artificial intelligence, and relates to the field of Tibetan type cultural creative graphs, and the method comprises the steps: carrying out the semantic analysis of a natural language description, and extracting a feature vector and a core keyword of a user intention; constructing a Tibetan culture symbol gene pool; based on the vector similarity, an initial culture symbol set conforming to the design intention of the user is screened out; eliminating symbol combinations with culture conflicts, and recommending symbol combination schemes conforming to culture specifications; extracting the shape feature of each symbol as an invariant constraint, and generating a symbol graph set after shape self-adaptive adjustment; and based on the diffusion model framework, generating a final Tibetan cultural and creative pattern. And outputting a structure report of the current Tibetan cultural and creative pattern according to the generated Tibetan cultural and creative pattern. The method has the advantages that natural language description can be automatically converted into Tibetan cultural and creative patterns conforming to culture specifications and aesthetic requirements, and deep fusion of culture inheritance and intelligent generation is achieved.

This invention relates to the field of wireless communication signal processing technology, and particularly to an interference cancellation method and device based on a convolutional neural network. The method involves receiving a raw radio frequency (RF) signal, converting it into a baseband in-phase quadrature signal, buffering the baseband in-phase quadrature signal into a sequence of preset length, inputting it into a pre-trained lightweight one-dimensional convolutional neural network, and outputting an interference modeling signal. The baseband in-phase quadrature signal is subtracted from the interference modeling signal to obtain a useful signal after interference cancellation. This useful signal is then converted into an RF transmission signal and transmitted, yielding the transmitted RF signal. Based on the transmitted RF signal, the current interference cancellation performance is evaluated. This enables high-precision automatic modeling and cancellation of interference signals in complex time-varying channels, improving the robustness and generalization ability of interference cancellation.

This invention discloses a short-window gamma-gamma turbulence parameter prediction method for space-to-ground laser communication, belonging to the technical field of space-to-ground laser communication and atmospheric turbulence channel parameter prediction. This method addresses the problems of traditional methods, such as strong dependence on long observation windows, significant degradation in prediction accuracy under short-window scenarios, and insufficient robustness under low signal-to-noise ratio conditions. It establishes a short-window observation model for the space-to-ground optical link and a gamma-gamma channel statistical model, constructs time-dependent short-window training data, and designs a short-window gamma-gamma network. Temporal features are extracted through a convolutional backbone, and a scintillation exponential physical regularization auxiliary head is used to achieve joint parameter prediction under physical constraints, outputting predicted gamma-gamma distributed parameters. This method can achieve high accuracy and good stability in parameter prediction under short observation windows and low signal-to-noise ratio conditions, and can be used for turbulence channel state characterization at the space-to-ground laser communication receiver.

The invention relates to the technical field of craniocerebral injury biomechanical analysis, discloses a semi-automatic craniocerebral injury index CSDM extraction method, and aims to improve CSDM extraction efficiency, applicability and flexibility. The method comprises the following steps: a brain tissue area display state pre-selection step: importing a head model simulation result file into post-processing software, screening and retaining brain tissue entity unit components, hiding irrelevant components and recording parameters; a strain and volume frame-by-frame output step: compiling an automatic script based on a post-processing software secondary development function to realize frame-by-frame output of unit data; and an automatic operation step: automatically calculating the CSDM value of each frame based on the output file through a Matlab calculation program. According to the method, the problems that manual extraction is low in efficiency and prone to errors and a full-automatic method is poor in universality are solved through a manual preprocessing and automatic calculation fusion mode, and the method is suitable for various head models and supports craniocerebral injury mechanism research and clinical assistance.

This invention relates to the field of marine engineering material preparation technology, and particularly to an eco-friendly marine engineering material and its preparation method. The preparation method includes: leaching waste concrete fine powder with ammonium chloride solution to dissolve calcium ions, separating the calcium ion solution and the leachated waste concrete fine powder; mixing waste concrete aggregate, sea sand, and the leachated waste concrete fine powder to prepare a porous framework; mixing urea, reactive bacteria, calcium ion solution, and seawater to obtain a reaction solution; the reactive bacteria include urease hydrolyzing bacteria and carbonic anhydrase bacteria; injecting the reaction solution into the porous framework, and mineralizing the porous framework in a carbon dioxide gas environment to obtain the eco-friendly marine engineering material. This invention breaks through the dependence of traditional cementitious materials on cement, achieving the synergistic development of construction solid waste resource utilization and green marine engineering.

The present application relates to the technical field of engineering machinery, and provides a centering steering system, a front steering axle, a rear steering axle, a switching valve assembly and a steering gear, the front steering axle is connected with a first reversing valve and a second reversing valve for changing the flow direction of hydraulic oil into and out of first and second oil cylinders, the rear steering axle is connected with a third reversing valve and a fourth reversing valve for changing the flow direction of hydraulic oil into and out of third and fourth oil cylinders, the switching valve assembly is connected with a hydraulic pump for providing a hydraulic oil source for centering and locking the front steering axle or / and the rear steering axle at an execution inlet end, the switching valve assembly is electrically connected with an electric controller for switching the hydraulic pump and the front steering axle or / and the rear steering axle on at a control end, and the steering gear provides hydraulic oil for driving the front steering axle or / and the rear steering axle to steer, through the arrangement of the switching valve assembly and the steering gear, the limitation of traditional steering systems on relying on sensors and complex programming is broken through, and low-cost, high-precision and high-reliability steering control is realized.

The present application relates to the technical field of molecular retrosynthesis prediction, and discloses a kind of molecular retrosynthesis prediction method and system based on multi-step information fusion.The present application is directed to the problem that the existing single-step prediction model only relies on the current molecular structure characteristics and lacks the synthesizability guidance at the multi-step path level, the method first uses knowledge graph embedding to model the semantics of single-step reaction;Then a directed graph of multi-step reaction is constructed, and a relational graph convolution network is used to aggregate neighbor information to generate path fusion features;Finally, in the decoding stage of the prediction model based on the Transformer architecture, the path fusion features are introduced, and the dynamic interaction between the path fusion features and the current synthesis path encoding features is realized through an external attention mechanism, and the global synthesizability information is injected into the generation process of the reactants. The method can fully utilize the context constraints of the synthesis path and the prior knowledge of chemical reactions, significantly improving the accuracy of the retrosynthesis path prediction and the landing feasibility of the final route.

The utility model relates to the technical field of metering calibration, in particular to a metering calibration adaptation device which comprises a protective shell, a screen groove is formed in one side of the top end of the protective shell, a display screen is arranged in the screen groove, a protective cover is installed on the side, away from the display screen, of the protective shell through bolts, and the protective shell and the protective cover are assembled to form a cavity and provided with a detection assembly. An exhaust connector is installed on the side, away from the display screen, of the protective shell in a penetrating mode, a switch is fixedly installed on the side, adjacent to the exhaust connector, of the protective shell, and a wiring slot is installed on the side, adjacent to the switch, of the protective shell in a penetrating mode. A traditional multi-device calibration process is integrated into a portable device, the operation complexity is reduced, the efficiency is improved, meanwhile, a lightweight protective shell and an integrated layout are adopted, pressure adjustment, data acquisition and power supply units are highly integrated, calibration can be achieved without external connection, environmental dependence is broken through, and the mobile scene operation requirement is met.

The invention discloses a gas meter digital measurement uncertainty evaluation method based on deep learning, and relates to the field of metering, and the method comprises the steps: collecting verification data of a gas meter, carrying out the preprocessing of the verification data to obtain a known data set, and dividing the known data set into a training set and a verification set; training the deep learning model by using the training set and verifying by using the verification set; calculating the mean value and the standard uncertainty of the input quantity based on the known data set; in combination with the Monte Carlo method and the prediction uncertainty probability distribution of the model instance, calculating the probability distribution of the measurement through random sampling and probability propagation; based on the measured probability distribution, the measured mean value and the standard uncertainty are calculated, and evaluation of the digital measurement uncertainty of the gas meter is completed. The evaluation process is full-automatic, and subjective errors caused by manual introduction are avoided; and the requirements of measurement development planning are met, digital measurement construction is promoted, and verification credibility is improved.

This invention relates to the field of water treatment technology, specifically to a non-aeration, low-current-density electrocatalytic oxidation water treatment method and electrocatalytic membrane, comprising the following steps: S1, placing a cobalt phthalocyanine-functionalized carbon nanotube electrocatalytic membrane into a membrane module as the cathode of the electrocatalytic reaction device; wherein, the cobalt phthalocyanine-functionalized carbon nanotube electrocatalytic membrane is obtained by the following preparation method: carbon nanotubes are acid-washed, functionalized and modified, then composited with cobalt phthalocyanine and calcined, then mixed with a polyvinyl alcohol solution and sprayed onto a base membrane material and crosslinked; S2, connecting the positive and negative electrodes of a power supply to the counter electrode and cathode respectively, and transporting the water sample to be treated into the electrocatalytic reaction device, controlling the current density ≥ 0.15 mA / cm². 2 No aeration is required during the treatment process. The electrocatalytic oxidation water treatment method of this invention is applicable to a wide range of water qualities and can stably treat novel antibiotic pollutants by generating hydroxyl radicals to achieve the oxidative degradation of pollutants.

The utility model discloses a melon interspecific hybrid tissue culture seedling domestication device relates to plant seedling domestication device technical field. Its structure includes nutrient solution groove, matrix groove, double -deck light -transmitting light -supplementing cover and ventilation adjustment assembly, and the matrix groove is installed in the nutrient solution groove, and ventilation adjustment assembly is installed on the nutrient solution groove and matrix groove, and ventilation adjustment assembly divides matrix groove and is divided into a plurality of seedling domestication area, and every seedling domestication area installs a double -deck light -transmitting light -supplementing cover. The utility model discloses through double -deck light -transmitting light -supplementing cover and temperature feedback system realizes environment accurate control, adopts M shape cotton line, filter cotton section constructs microcirculation nutrition supply system, and combines the partition formula modular design and promotes space utilization rate and operating efficiency, and the device is verified through the test, can make interspecific hybrid domestication survival rate to improve, and domestication period shortens, provides efficient technical solution for melon interspecific hybrid breeding.

This invention relates to the fields of biomedical imaging and spatial metabolomics, specifically to a method and system for registering spatial metabolomics images with HE-stained images. The method includes the following steps: acquiring two adjacent frozen sections of the same tissue sample; performing image preprocessing using the frozen sections; obtaining the extraction and matching results of key point features in the images; performing rigid image registration and obtaining a rigid registration result; performing non-rigid image registration and obtaining a non-rigid registration result; and obtaining the spatial location information of the spatial metabolic imaging using the non-rigid registration result. This invention, by combining advanced image processing technology and machine learning algorithms, automatically registers metabolic imaging images with HE-stained images, significantly improving registration accuracy and efficiency. It effectively solves the problems of large errors, low efficiency, and strong operational dependence in existing manual registration methods, meeting the needs of high-throughput, high-precision spatial metabolomics analysis and supporting large-scale data processing.

The application discloses a surgery room infection management system based on big data, which comprises a data acquisition module, a preoperative surgery room infection risk assessment module, an intraoperative surgery room infection dynamic monitoring module, a postoperative surgery room infection terminal assessment module and a surgery room infection intelligent management module. The application relates to the technical field of data processing, in particular to a surgery room infection management system based on big data. The application innovatively combines preoperative, intraoperative and postoperative surgery room infections to realize dynamic monitoring and whole-process control. The application introduces a segmented convergence factor and an iterative level reverse learning optimization population strategy to improve the optimization algorithm, thereby improving the model hyperparameter optimization precision and the accuracy of model output results. The application designs a bidirectional convolution path, combines a gated recurrent unit and simultaneously adopts a model semi-supervised loss function with a supervised loss, a consistency regularization loss and a pseudo label loss, thereby significantly improving the precision and real-time performance of the monitoring model.

This invention discloses a biomarker combination, a test kit, and their applications for assessing ovarian reserve function. The biomarker combination includes the protective biomarker melatonin, dehydroepiandrosterone sulfate, and the damage biomarker 8-hydroxydeoxyguanosine. A protection-damage balance system is constructed to differentiate between physiological and pathological ovarian reserve decline. The test kit contains a saliva sampling component, a pretreatment solution, a lateral chromatography test strip, and an age-stratified colorimetric card. The test strip has three test lines and one control line, simultaneously detecting the three indicators using a colloidal gold method. The test involves morning fasting sampling, pretreatment to remove interference, color development, and comparison with the age-stratified colorimetric card, outputting three types of results: age-consistent, physiological decline, or pathological decline. This enables non-invasive, instrument-free, and visually interpretable home testing, solving the problems of high invasiveness, high platform dependence, difficulty in eliminating age interference, and unintuitive results in existing technologies.

The invention belongs to the technical field of negative ion coatings, and particularly relates to a graphene negative ion coating which comprises the following raw materials in percentage by weight: 15-25% of titanium dioxide, 4-6% of kaolin, 10-20% of high-calcium powder, 20-40% of natural resin emulsion, 1-2% of tourmaline, 1-2% of cryolite powder, 25-30% of water and 1-2.5% of single-layer graphene solution.

The utility model discloses a mosquito killer lamp of many energy paths, including main part structure, charging system, mosquito killer, safety device and collection device, charging system includes solar photovoltaic board, external power source interface and energy storage battery, and solar photovoltaic board, external power source interface are connected with energy storage battery electricity respectively, mosquito killer includes light source module and power grid component, and light source module and power grid component are connected in main part structure, and light source module is located in the inside center of power grid component, and light source module is used to send light ray and attract mosquito, and power grid component is used for electric shock and kill, safety device surrounds mosquito killer, and forms the physical isolation barrier, collection device is detachably installed in the bottom of main part structure, and is used for concentrating mosquito debris, through the double path cooperation mechanism of solar energy and commercial power, break through the dependence of traditional equipment to single energy, both solve the power supply interruption problem in the rainy environment, and reduce user's electricity cost, realize "uninterrupted mosquito killer" from the energy logic level.

The application discloses a method and system for identifying farmland runoff pollution in surface water by using microbial fingerprints. Microbial samples of multiple types of water bodies are collected, and microbial composition data are obtained through 16S rDNA sequencing; based on a sensitivity-specificity screening index, combined with specific anaerobic characteristics, a farmland runoff pollution microbial fingerprint is determined, a double-model integrated judgment mechanism is established: an artificial neural network is used to process microbial category data, an XGBoost model is used to analyze microbial relative abundance data, and a control variable method is used to extract the classification decision basis of the double models; and the double models are used to make result judgment based on logical rules. The application innovatively integrates microbial fingerprints and machine learning classification models, can realize accurate and simple identification of farmland runoff pollution in surface water, and improves the identification accuracy and efficiency of farmland runoff pollution.

This invention discloses a self-activated solid waste slurry and its preparation method, comprising waste slag, a solid waste cementitious material, and water. The components are expressed as follows by weight percentage or parts: 100 parts waste slag; 15-60 parts solid waste cementitious material, wherein the solid waste cementitious material consists of a silicon-aluminum rich precursor, a strongly alkaline activation source, and a rheology-modifying supplementary source, with a mass ratio of 1:(0.2-0.8):(0.1-0.4); and water. This invention eliminates the highly corrosive liquid chemical activator required in traditional alkaline activation materials, achieving self-activation through dry powder premixing of a strongly alkaline activation source, a silicon-aluminum rich precursor, and a rheology-modifying supplementary source.

This invention belongs to the field of display interaction technology and provides a pointing interaction method and system based on hidden QR code embedding. Several QR code images are hidden and embedded in the display screen of the target screen. These QR code images utilize the human eye's flicker fusion effect, displaying frames alternately on the target screen according to a preset refresh rate, making the QR code images invisible to the human eye. In response to scanning and capturing by a smartphone camera, the QR code images are extracted and recognized. Based on the extracted and recognized information, the coordinates of the smartphone camera's optical center on the screen are calculated, obtaining the corresponding pointing position of the user on the screen. A cursor pattern that moves with the smartphone is displayed at the corresponding coordinate position, realizing pointing positioning of the smartphone on the display screen. This invention realizes pointing positioning function based on a smartphone on the display screen.

The utility model relates to heat pump technical field, concretely relates to a double heat source high temperature heat pump steam generation system. In order to make up for the deficiency of prior art, the present application provides a double heat source high temperature heat pump steam generation system. Double heat source high temperature heat pump steam generation system includes air source heat pump, waste heat source heat pump, water vapor compressor, flash tank, heat storage water tank, first circulating pump and second circulating pump. The flash tank, the first circulating pump and the air source heat pump are communicated and form the first circulating water circuit. The flash tank, the second circulating pump and the waste heat source heat pump are communicated and form the second circulating water circuit. The outlet of the flash tank is communicated with the water vapor compressor inlet. The heat storage water tank is communicated with the waste heat source heat pump and forms the third circulating water circuit. In view of the intermittency and fluctuation of the waste heat source, the double heat source high temperature heat pump steam generation system can improve the waste heat supply stability by configuring the heat storage water tank, and has good carbon reduction efficiency and investment income.

This invention proposes a fault detection method and system for industrial control computer control systems. It belongs to the field of industrial control system fault diagnosis technology. The method includes: real-time acquisition of multi-source heterogeneous data through a sensor network deployed in various modules of the industrial control computer; preprocessing the acquired multi-source heterogeneous data, which includes time-series data, topological relationship data, and event edge representations of real-time interaction relationships between components, forming a multimodal dynamic graph dataset; modeling the complex interaction relationships of the industrial control system through the dynamic graph structure, and achieving deep fusion of multimodal features by combining a spatiotemporal graph neural network, thus breaking through the dependence of traditional methods on static rules.

The invention discloses a method for preparing spherical nano barium fluoride, spherical barium fluoride smaller than 5nm is synthesized by a microemulsion reaction method, and the microemulsion is temperature-responsive and does not contain a surfactant. Controllable preparation of the spherical barium fluoride with the ultra-small size smaller than 5 nm under the surfactant-free condition is achieved for the first time. A ternary surfactant-free microemulsion system composed of diethylene glycol monohexyl ether, octane and water is constructed, the amphiphilic solvent characteristic of diethylene glycol monohexyl ether is used for replacing a traditional surfactant, and the dependence of an existing template method on the surfactant is successfully broken through. The particle size of the obtained BaF2 product is only smaller than 5 nm and enters a strong quantum confinement scale range, and the preparation blank of a barium fluoride material in the field of sub-5-nm ultra-small solid spheres is filled.

The invention relates to a communication signal receiving method and device based on an attention mechanism, equipment and a medium. The method comprises the steps that a multi-source coupling signal separation network composed of an end-to-end encoder-separation module-decoder is constructed and trained, in the network, short-time overlapping segments in mixed signals are mapped into high-dimensional feature representation according to an encoder, and the high-dimensional feature representation is spliced into mixed signal feature vectors along a time axis; a mask of a target signal is extracted from the mixed signal feature vector according to an attention mechanism process modeled by a separation module, and a target signal feature representation is separated by using the mask; and finally, a decoder receives the target signal feature representation to perform segment-by-segment reconstruction, and outputs a complete target signal by overlapping reconstructed signal segments. According to the invention, efficient and accurate reconstruction of a target signal in a complex electromagnetic environment can be realized, accurate receiving of a communication signal is completed, and the robustness of a communication system is enhanced.

The application discloses a permafrost water and heat state simulation method and system based on a physical model, and relates to the technical field of permafrost regions engineering. The method comprises the following steps: obtaining meteorological driving data and land surface characteristic data; constructing a numerical model of a permafrost water and heat process by taking into account the freezing and thawing process of permafrost and the soil ice action by using the land surface characteristic data; preparing boundary constraint conditions of the numerical model according to the meteorological driving data and existing temperature field data; and driving the numerical model to simulate the permafrost water and heat state of the land surface characteristic data by using the meteorological driving data, constraint condition data and key vegetation and soil parameters, so as to obtain a sample set composed of key permafrost water and heat information. The application breaks through the dependence of the existing model on data and structure, meets the multi-scale application demand of permafrost regions, and finally can generate a high-quality meteorological driving and boundary constraint condition data set suitable for permafrost simulation in unobserved regions.

This invention discloses a miRNA detection system, kit, and detection method based on DNA aggregation-induced precipitation as a signal output. The detection system comprises: a DNA triangle, a hairpin-structured probe H1, a hairpin-structured probe H2, and a DNA double-stranded linker with sticky ends. Probes H1 and H2 can generate DNA double-stranded strands H1-H2 with sticky ends at both ends through a catalytic hairpin self-assembly reaction. Probe H1 contains a sequence complementary to the miRNA sequence to be detected. The DNA triangle has sticky ends at its first and second corners, complementary to both ends of the DNA double-stranded linker, and a sticky end at its third corner, complementary to both ends of DNA double-stranded strands H1-H2. This invention achieves highly sensitive and specific detection of miRNAs without the need for complex instruments and equipment, making it suitable for on-site, real-time detection.