191results about How to "Achieve compression" patented technology

The invention discloses a respiratory rehabilitation training data management method and system based on artificial intelligence, and relates to the technical field of data processing, the respiratory rehabilitation training data management method based on artificial intelligence comprises the following steps: obtaining training data of a respiratory rehabilitation person, and preprocessing the training data to obtain a multi-source data set; constructing a teacher network based on the high-performance model, training the teacher network through the multi-source data set, constructing a student network, and training the student network based on the trained teacher network; optimizing the trained student network by using a differential evolution algorithm to obtain a student network optimized model; and training a reinforcement learning agent by using data generated by the model after student network optimization, and automatically adjusting rehabilitation training parameters of the respiratory rehabilitation personnel through the reinforcement learning agent. According to the invention, computing resource and storage requirements are reduced, data-driven decision is realized, and a personalized rehabilitation training scheme is provided.

The invention discloses construction waste recycling and filling equipment with a dust falling function, and belongs to the technical field of construction waste recycling. According to the construction waste recycling device, the movable dust falling mechanism, the pushing and compressing mechanism, the first recycling and smashing roller and the second recycling and smashing roller are arranged in a matched mode, through the design, effective smashing in the construction waste recycling process can be achieved, meanwhile, effective filling of the construction waste can be achieved, and the construction waste can be recycled in the treatment process. Effective and all-directional dust falling treatment on dust is achieved, the sanitation degree of a working area is guaranteed, meanwhile, the dust falling range can be effectively expanded, the actual application effect of the whole device is greatly improved, meanwhile, construction waste can be pushed to the two sides, it can be effectively avoided that waste is accumulated in a certain position of the device in a concentrated mode, and the service life of the device is prolonged. And meanwhile, construction waste can be compressed, waste of storage space in the device is avoided, and follow-up centralized treatment of the recycled waste is facilitated.

The invention relates to a rail corrugation detection and identification method, device, equipment, medium and program, and the method comprises the steps: obtaining an original vibration response signal, decomposing the original vibration response signal, and obtaining a modal function component; screening out a target component from the modal function components according to a preset correlation coefficient criterion, and reconstructing the target component to obtain a preprocessed signal; performing time-frequency analysis on the preprocessed signal to obtain a time-frequency ridge line, and intercepting a corrugation section according to a vibration response signal positioned by the time-frequency ridge line; and extracting time domain features of the corrugation section, and generating a wave depth grade result according to the time domain features.

The invention relates to a wall-climbing robot positioning system and method based on a space structure circular descriptor. A depth image sensor (RGB-D) camera outputs a depth image; collecting three-dimensional point cloud data of three-dimensional space wall surface points; and conformally mapping the contour of the wall-climbing robot to a minimum bounding ball (MBS). A space structure circular descriptor (SSCD) algorithm module maps the three-dimensional point cloud data to a two-dimensional positioning navigation plane, and an SSCD image is formed on the two-dimensional positioning navigation plane; and point cloud features of the SSCD images from the k-1 moment to the k moment are matched, and positioning of the wall-climbing robot on the three-dimensional space wall surface at the current moment is completed. An airborne inertial measurement unit (IMU) measurement module acquires the angular velocity and acceleration of the wall-climbing robot; the attitude information of the wall-climbing robot at the current moment is obtained by carrying out pre-integration on the collected angular velocity and acceleration. Therefore, three-dimensional data are mapped to a two-dimensional plane through a space structure circular descriptor (SSCD) algorithm, and the wall-climbing robot is accurately positioned and navigated.

The invention provides a mine video semantic communication transmission method and system based on deep learning, which can realize the recovery of observable data and the execution of pragmatic tasks. According to the method, a semantic communication system model under a mining working face is constructed, the semantic communication system model comprises a semantic encoder and a semantic decoder which are realized based on a deep neural network, and the semantic encoder is located at a transmitting end and used for extracting data containing observable data and semantic information; the semantic decoder is located at the receiving end and used for reconstructing data related to empirical data and pragmatic tasks. According to the method, a semantic encoder and a semantic decoder are trained based on a DL-SC algorithm, image restoration and pragmatic task execution are considered in the aspect of design of a loss function, appropriate distortion measure functions are selected for the image restoration and pragmatic task execution respectively, and then the weighted sum of the distortion measure functions is solved to form a semantic distortion measure function to serve as the loss function. According to the invention, more efficient information interaction can be realized with lower data transmission quantity.

The invention discloses a binocular video compression method and device based on deep learning and a readable medium, and relates to the field of video processing, and the method comprises the steps: constructing and training a binocular video compression model, and obtaining a trained binocular video compression model, the method comprises the following steps: respectively acquiring an original frame of a current moment and an original frame of a previous moment of a left viewpoint and an original frame of a current moment and an original frame of a previous moment of a right viewpoint in a binocular video to be compressed, and inputting the original frames into a trained binocular video compression model; and reconstructing to obtain a reconstructed frame sequence of the left viewpoint and a reconstructed frame sequence of the right viewpoint in an iteration mode, and obtaining a compressed binocular video. According to the method, the problems of low redundancy utilization rate and low compression performance between viewpoints of the existing binocular video are solved.

The invention discloses an aero-engine flexible spline coupling with an elastic clamping tongue, and belongs to the technical field of aero-engines. The coupler comprises a fan shaft, a low-pressure turbine shaft, a centering compensation part, a locking part and an elastic clamping tongue, the fan shaft is a hollow shaft, an annular boss is integrally formed in the fan shaft, one end of the low-pressure turbine shaft is inserted into the fan shaft, is in sliding connection with the annular boss and is in transmission connection with the fan shaft, and the other end of the low-pressure turbine shaft is in transmission connection with the locking part. The centering compensation part is located in a radial gap between the fan shaft and the low-pressure turbine shaft, the locking part is arranged on the low-pressure turbine shaft in a sleeving mode, the centering compensation part and the annular boss jointly conduct axial limiting on the centering compensation part, and the elastic clamping tongue is located in a radial gap between the locking part and the low-pressure turbine shaft and conducts rotation stopping on the locking part and the low-pressure turbine shaft. The elastic clamping tongue can bear larger shearing force, and locking of the locking piece is more reliable and effective. And the centering compensation piece is not easy to wear and long in service life. The whole assembling process of the coupler is simple.

The invention discloses a high-power air compressor and a using method thereof, and relates to the technical field of air compressor equipment, the high-power air compressor comprises a main tank body, a main motor, a main rotating shaft, a first air compressor, a second air compressor, an auxiliary motor and an auxiliary rotating shaft, the first air compressor and the second air compressor are arranged on the two sides of the main motor and are in transmission connection with the main rotating shaft, the auxiliary rotating shaft is arranged at the side end of the main rotating shaft, the axis of the main rotating shaft and the axis of the auxiliary rotating shaft are located on the same straight line, and the main rotating shaft and the auxiliary rotating shaft are connected through a coupler. The auxiliary motor is arranged on the auxiliary rotating shaft and is in transmission connection with the auxiliary rotating shaft, the air outlet end of the first air compressor and the air outlet end of the second air compressor are communicated to the main tank body, and the main motor and the auxiliary motor simultaneously drive the first air compressor and the second air compressor to operate, so that the operation power of equipment is improved.

The invention discloses a carbon emission monitoring method based on the Internet of Things, and relates to the technical field of data monitoring. Comprising the steps of collecting multi-dimensional carbon emission data through sensor networks deployed on different carbon emission sources; an edge computing device is configured at each monitoring node, and primary analysis of real-time multi-dimensional carbon emission data is carried out; the multi-dimensional carbon emission data processing module is used for processing the multi-dimensional carbon emission data, uploading the processed data to a cloud end through the Internet of Things communication module and managing by adopting a distributed storage system, and a cloud end server adopts a distributed node storage mode and dynamically schedules storage resources; multi-dimensional carbon emission data are integrated at the cloud end, secondary analysis is performed on carbon emission based on primary analysis, and a monitoring result is obtained. Through cooperative work of the sensor network, the edge computing device and the cloud server, real-time acquisition, compression, uploading and deep analysis of carbon emission data are realized, and the problems of uneven computing resource distribution, insufficient data analysis and the like in the prior art are effectively solved.

The embodiment of the utility model relates to the technical field of elastic sheet connectors, in particular to an elastic sheet connector and electronic equipment, which comprise a shell and an elastic sheet main body, the shell is provided with an accommodating cavity with a first opening, the elastic sheet main body comprises a first elastic arm, a second elastic arm and a contact part, the first elastic arm, the second elastic arm and the contact part are connected in sequence, at least part of the first elastic arm is located in the containing cavity, the end, away from the second elastic arm, of the first elastic arm is connected with the shell, the contact part stretches out of the first opening, and the first elastic arm and the second elastic arm are arranged in an inclined mode. Through the above mode, the embodiment of the utility model can enable the elastic sheet connector to be compressed by the external component in the forward direction and also to be compressed by the external component in the lateral direction.

The utility model discloses a two-way output air compressor which comprises a driving machine shell, a motor shaft is fixedly connected with two eccentric wheels, each eccentric wheel is internally provided with a rolling groove, the rolling grooves are internally connected with rotating shafts in a rolling mode, through holes are fixedly connected with pressurizing pipes which are coaxially aligned with connecting rods left and right, and the directions of the two pressurizing pipes are opposite. The pressurizing pipes are slidably connected with the connecting rods, the connecting rods are fixedly connected with pistons located in the pressurizing cavities, the pistons are slidably connected with the pressurizing pipes in a sealed mode, and each pressurizing pipe is fixedly connected with a one-way air outlet valve and a one-way air inlet valve which are communicated with the corresponding pressurizing cavity. When the eccentric wheel rotates, the rotating shaft rolls along the rolling groove, so that the piston is driven to move, air enters from the one-way air inlet valve and flows out from the one-way air outlet valves to the air storage pipe, air compression is achieved, two-way output is achieved through the two one-way air outlet valves, and air compression of the air storage tank in two directions is achieved.

The utility model discloses a medical garbage can which comprises a can body and a can cover, a garbage inlet is formed in the top of the can body, the can cover is arranged at the garbage inlet, a containing cavity used for storing medical garbage is formed in the can body, and the can cover is used for opening or closing the containing cavity. The extrusion device is configured to be provided with the closed part and the movable part, the closed part can be blocked in the longitudinal space where the garbage inlet is located, the movable part is used for driving the closed part to move, then medical garbage in the barrel body can be compressed, and meanwhile the barrel body and the extrusion device are integrated for use; the time of searching external tools for compression work by a user is shortened, and manpower is saved.

The invention discloses a hydrogen energy battery lamination device and method, and relates to the technical field of battery lamination, the hydrogen energy battery lamination device comprises a constant temperature bin for battery lamination, the bottom of the constant temperature bin is provided with a support for supporting the constant temperature bin, and the bottom of the constant temperature bin is provided with a hydraulic rod and a battery lamination mechanism; the top of the hydraulic rod is connected with the bottom of the battery laminating mechanism, the battery laminating mechanism comprises a plurality of laminating frames arranged at equal intervals from bottom to top, lifting assemblies are installed on the two short side edges of each laminating frame, and the laminating frame located at the topmost portion in the constant-temperature bin is provided with a fixing rod fixedly connected with the inner top wall of the constant-temperature bin. By means of the hydrogen energy battery laminating machine, multiple layers of materials can be laminated at the same time, and the processing efficiency of hydrogen energy batteries is improved.

The invention relates to an index prefix compression processing method and device, computer equipment, a computer readable storage medium and a computer program product, and relates to the technical field of databases. According to the method, intelligent compression of the index prefix can be realized, the compression efficiency is improved, the space occupation is reduced, and the storage overhead of the index is reduced. The method comprises the following steps: generating a prefix symbol for simplifying and representing a corresponding repeated prefix part according to the repeated prefix part of each original index through a symbol generation model; the symbol generation model generates a prefix symbol corresponding to the original index based on the use condition of a historical index containing a historical prefix symbol and / or the use condition of an index sample containing a prefix symbol sample; registering each prefix symbol corresponding to each repeated prefix part in a prefix symbol table; and for the target index, replacing the repeated prefix part of the target index with the corresponding prefix symbol according to the prefix symbol registered in the prefix symbol table.

The invention belongs to the technical field of aquaculture equipment, and particularly relates to an aquaculture net cage mechanical fish collecting device which comprises a net cage frame, an external vertical net, an internal vertical net and a fish collecting channel. The movable gathering assembly is installed in the net cage frame, and the movable gathering assembly is used for actively gathering fish schools; according to the device, the moving gathering assembly is arranged, the moving net moves to drive fish schools in the culture cavity designed in an annular mode, the space where the fish schools are located is compressed in cooperation with the separation net and the bottom drop net, then the gathering density of the fish schools is increased, and at the moment, the difficulty of capturing the fish schools through a fish collecting pump is small, and the effect is good.

The invention discloses an edge computing encryption communication method and device, relates to the technical field of data transmission, and mainly aims to improve the security and efficiency of edge computing encryption communication. According to the main technical scheme, the method comprises the steps of collecting target data of edge equipment; encrypting the target data according to a preset encryption rule to obtain corresponding encrypted data, the preset encryption rule being used for representing an encryption rule for codebook transposition of the target data; and sending the encrypted data to a server, so that the server decrypts and stores the encrypted data based on a decryption mode corresponding to the preset encryption rule. The method and the device are used for edge computing encryption communication.

The invention discloses an index sequence generation method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring first gene expression data and at least one codebook entry, encoding the first gene expression data, extracting local information in the first gene expression data, and mapping the local information into a potential space to obtain potential representation of the local information; performing quantization processing on the potential representation based on the at least one codebook entry to generate a first codebook index matrix and at least one quantization vector; decoding and reconstructing the first codebook index matrix, outputting second gene expression data, performing adversarial training on a decoder and at least one quantization vector, and outputting optimization parameters; and performing autoregression modeling processing on at least one new codebook entry to obtain a second codebook index matrix, and processing the second codebook index matrix by using the optimized decoder to obtain a data sample. According to the method, the calculation burden of high-dimensional data modeling is reduced, and the data capturing capability of the model is improved.

The present invention provides an intelligent energy-saving bottle garbage recycling bin and method, belonging to the technical field of garbage recycling, comprising a box body, a delivery mechanism, a compression mechanism, a detection unit, a loading platform and a transport trolley; the delivery mechanism is installed on the top of the box body, the loading platform is installed in the middle of the box body, the compression mechanism is installed on the upper surface of the loading platform, and the transport trolley is arranged below the loading platform; a delivery port is provided on the outer side surface of the box body, a detection unit is provided on one side of the box body, the detection unit is located above the compression mechanism and below the delivery port; the compression mechanism adopted by the present invention adopts a screw drive to realize a "four-in-one" drive, and compression can be performed in four directions in one movement, and the efficiency is improved by four times; a hollow compression hole is adopted on the compression slider to discharge the air in the bottle through the air hole, thereby reducing resistance and achieving high efficiency and energy saving.

The utility model discloses a motor shaft rust cleaning equipment for motor, it includes the U type mounting seat, the top of one side of U type mounting seat is installed motor, the output of motor is equipped with the rotating rod no. 2 that extends to the inside of U type mounting seat. Through rotating rod no. 2 and motor can realize the rotation of motor shaft, and cooperate bevel gear no. 2, bevel gear no. 1, rotating rod no. 1, rotating rod no. 1, the structure design of tooth block and connecting rod, still can drive sliding block, internal thread sleeve, threaded rod and arc rust cleaning plate move in the inside of sliding slot, to realize the compression of spring, and through the movement of arc rust cleaning plate and then cooperate the rotation of motor shaft, can realize the polishing rust cleaning work at different positions of motor shaft outside, and through the structure design of this part, need not operator hand -held polisher, effectively reduced the labor intensity of later period worker, increased the convenience and practicality of this equipment later period use.

The invention provides a yellow rod feeding system of a biomass ash rod furnace, and mainly relates to the field of feeding systems. A yellow rod feeding system of a biomass ash rod furnace comprises a main conveying line, a mounting frame is arranged above the downstream of the main conveying line, at least two compression boxes are arranged on the mounting frame, compression cavities are formed in the tops of the compression boxes, conical cavities are formed in the bottoms of the compression cavities, compression assemblies are arranged on the two sides in the compression cavities, and the sides, facing each other, of the compression assemblies are arc faces. A compression power unit is arranged between the compression assembly and the compression cavity, a power frame is arranged above the compression box, a spiral feeding shaft is rotationally arranged on the power frame, a conical feeding piece matched with the conical cavity is arranged at the bottom of the spiral feeding shaft, a linear feeding mechanism is arranged on the adjusting frame, an auxiliary conveying line is arranged above the feeding conveying belt, and a receiving bin is arranged below the mounting frame. A leading-out conveying belt is arranged at the bottom of the receiving bin. The fuel boiler has the beneficial effects that uniform proportioning and constant-speed conveying of yellow-stem fuel and grey-stem fuel are achieved, and the temperature uniformity of the fuel boiler is guaranteed.

The invention discloses a mobile bearing device of a nondestructive testing X-ray machine, which comprises a vehicle body, a first sliding groove is fixedly connected above the vehicle body, a starting column is rotatably connected to one side of the first sliding groove, a bottom plate is rotatably connected to one end of the starting column, an inclined spring is arranged above the bottom plate, and a sliding groove plate is fixedly connected to the top end of the inclined spring. A sliding matching block is slidably connected to the upper portion of the sliding groove plate, a splitting plate is fixedly connected to the upper portion of the sliding matching block, a fixing rope is fixedly connected to one end of the splitting plate, a second pawl and a ratchet wheel are arranged on the grip, the grip is manually operated to enable the second pawl to push the ratchet wheel to rotate, and locking of the ratchet wheel is achieved through the first pawl. And the inclined spring arranged at the bottom of the sliding groove plate can provide a damping effect, shaking of the X-ray machine in the moving or working process is reduced, and the precision and safety of the X-ray machine are protected.

This utility model relates to the technical field of elevator car top guardrail equipment, and discloses a retractable guardrail for elevator car tops, including an elevator car top with a rectangular structure. First supports are fixedly connected to the four right angles of the elevator car top. First limiting cylinders are fixedly connected to the upper sides of multiple first supports. Threaded sleeves are fixedly connected to the bottom inner sides of multiple first limiting cylinders. Each threaded sleeve is threaded with a threaded rod, and a first support tube is rotatably connected to each threaded rod. This utility model allows the threaded rods to rotate by rotating a handwheel, causing the first support tubes to rise or fall from the first limiting cylinders. When the elevator is paused for maintenance, each first support can be raised; when the elevator is running normally, each first support can be retracted into the first limiting cylinder, thereby reducing the space occupied by the guardrail in the elevator shaft.

The invention belongs to the technical field of cast-in-place pile hole collapse prevention, and particularly relates to a silt layer self-balancing hole collapse prevention cast-in-place pile structure which comprises a reinforcement cage and a pile casing, the pile casing is detachably installed at an opening of a cast-in-place pile hole, and the reinforcement cage penetrates through an inner cavity of the pile casing and makes contact with the bottom of the inner cavity of the cast-in-place pile hole. The fixing part is detachably mounted at the top of the reinforcement cage, and the top of the reinforcement cage mounted closest to an opening of a cast-in-place pile hole can be continuously pressed through the fixing part; the adjusting part is arranged outside the fixing part, and the position of the reinforcement cage can be adjusted in time through the adjusting part, so that the central axis of the reinforcement cage coincides with the central axis of the cast-in-place pile hole all the time. According to the concrete foundation pile, through mutual cooperation of the parts, stirrup deformation and main reinforcement bending of the reinforcement cage caused by collapse of a silt layer during concrete pouring can be avoided, the bearing capacity of the built concrete foundation pile is guaranteed, and meanwhile the influence on lifting of the guide pipe and normal construction of a cast-in-place pile hole is avoided.

The invention belongs to the technical field of molecular oil refining, and particularly relates to a crude oil information compression, storage and decompression method, device and equipment, and the crude oil information compression method comprises the following steps: obtaining structure-oriented lumped data of crude oil; traversing the structure-oriented lumped data line by line according to a first sequence, and identifying each field value in each line and the continuous occurrence frequency thereof according to a second sequence; combining the field values in each row and the continuous occurrence times of the field values into a plurality of compressed data subsequences of the row according to a third sequence; splicing the plurality of compressed data subsequences of each row into a compressed data sequence of the row according to the second sequence; and combining the compressed data sequence of each row according to the first sequence to obtain compressed data of the structure-oriented lumped data. The problems of high crude oil information storage cost and high transmission pressure are solved.

The utility model discloses a novel double-screen KVM (Keyboard Video Mouse) transceiver, which comprises a KVM transmitter and a KVM receiver, the KVM transmitter comprises a first machine body, a first SoC chip, an audio analog-to-digital conversion chip and two HDMI receiving chips, wherein the first SoC chip, the audio analog-to-digital conversion chip and the two HDMI receiving chips are installed in the first machine body. The KVM receiver comprises a second machine body, a second SoC chip, an audio digital-to-analog conversion chip and two HDMI sending chips, wherein the second SoC chip, the audio digital-to-analog conversion chip and the two HDMI sending chips are installed in the second machine body. The novel double-screen KVM transceiver can support transmission of two paths of HDMI video signals at the same time, one sending terminal can be connected with HDMI signals of double display cards of a computer, one receiving terminal can be connected with two display screens, cost is reduced, and the double-screen display function is achieved. Therefore, a user can connect the two displays and check and operate the content on the remote computer at the same time, and the working efficiency is improved.

The invention relates to a video file compression method and device, computer equipment and a readable storage medium. The method comprises the following steps: acquiring video files respectively acquired by different types of video terminal equipment in a power grid field operation process; for each video file, extracting video data and audio data from the video file; determining the resolution of the video data and the code rate of the video data; under the condition that the resolution of the video data is greater than a resolution threshold and the code rate of the video data is greater than a first code rate threshold, compressing the resolution and the code rate of the video data to obtain compressed video data; and combining the compressed video data and the audio data to obtain a compressed video file corresponding to the video file. By adopting the method, the compression effect can be improved.

A lunar base energy storage system based on a lunar soil-filled flywheel relates to the field of space energy storage. The present invention is to solve the problem that chemical storage systems are limited by battery life and charge-discharge depth, and that electrical storage, heat storage, and flywheel energy storage systems are too heavy, resulting in high launch costs. Based on the in-situ manufacturing concept, a lunar base energy storage system based on a lunar soil-filled flywheel is designed, including multiple parallel energy storage flywheel groups, fuel cells, multiple parallel solar cell groups, and wires. The energy storage flywheel group and the fuel cell are located in a lunar crater. The multiple energy storage flywheel groups and fuel cells are connected to the multiple solar cell groups through wires. The energy storage flywheel group includes a foldable flywheel box that can be filled with lunar soil on site to achieve the counterweight of the flywheel body. The flywheel energy storage device can install and disassemble the flywheel unit according to the energy storage demand through the series / parallel arrangement of the flywheel energy storage unit to achieve the expansion and reduction of energy storage. The present invention is mainly used for nighttime energy storage in a lunar base.

The invention relates to the technical field of voice recognition, in particular to a voice recognition method and device, electronic equipment and a storage medium, and aims to improve the voice recognition efficiency. The method comprises the following steps: acquiring to-be-recognized voice, encoding the to-be-recognized voice through a target encoder, and acquiring candidate encoding features corresponding to the to-be-recognized voice; obtaining a blank frame probability corresponding to each voice frame in the to-be-recognized voice according to the frame distribution information of the to-be-recognized voice, and screening out at least one target frame from the to-be-recognized voice based on the obtained blank frame probability; the frame distribution information represents probability distribution of each voice frame in a dictionary space; screening out a target candidate coding feature corresponding to at least one target frame from the candidate coding features; and decoding based on the target candidate coding features to obtain a text recognition result corresponding to the to-be-recognized voice. According to the invention, the frame rate compression is carried out through the frame distribution information, only part of important frames are decoded, and the speech recognition efficiency can be effectively improved.

The present invention discloses a binocular video compression method, device and readable medium based on deep learning, which relates to the field of video processing, including: constructing and training a binocular video compression model to obtain a trained binocular video compression model, respectively obtaining the original frame of the left viewpoint at the current moment and the original frame of the previous moment and the original frame of the right viewpoint at the current moment and the original frame of the previous moment in the binocular video to be compressed and inputting them into the trained binocular video compression model, reconstructing the reconstructed frame sequence of the left viewpoint and the reconstructed frame sequence of the right viewpoint in an iterative manner, and obtaining a compressed binocular video. The present invention solves the problems of low redundancy utilization and low compression performance between viewpoints of the existing binocular video.

The application belongs to the technical field of artificial intelligence, and discloses a question and answer identification method and device, electronic equipment and readable storage medium, the method comprises the following steps: receiving a question and answer request for a target question; based on the question and answer request for the target question, a pre-trained question and answer model is used to determine the answer corresponding to the target question; wherein the question and answer model is a student model obtained based on a second sample data set and a teacher model, and the teacher model is constructed based on a first sample data set and a transformer-based bidirectional encoding representation model. In this way, when question and answer identification is performed, the response speed of question answering can be improved and resource consumption can be reduced.