107results about How to "Improve dynamic response performance" patented technology

The invention discloses a harmonic suppression and dynamic reactive compensation control method for an energy router, which is suitable for a micro-grid and a distributed energy access system. According to the method, three-phase current and voltage signals of an alternating current port of the energy router are collected, five-time and seven-time main harmonic components are extracted through fast Fourier transform, harmonic compensation current instructions equal in amplitude and opposite in phase are generated and injected into an inverter to be output, and active counteracting of main harmonics is achieved. Meanwhile, the bus voltage deviation and the change rate thereof are detected, a reactive current instruction is adjusted on line by adopting a fuzzy self-adaptive PI controller, rapid and smooth dynamic reactive compensation is realized, and the bus voltage is stabilized. Decoupling control of active / reactive and harmonic components is realized in combination with Park conversion, and virtual impedance is introduced to enhance system damping and harmonic suppression capabilities. Compared with the prior art, harmonic suppression and voltage support can be simultaneously realized under a single control framework, the response speed is high, the stability is high, and the method is suitable for high-performance grid-connected operation of the multi-port energy router.

The invention discloses a cold-chain pork quality dynamic evaluation method and system based on multi-source data fusion, and relates to the technical field of data processing and intelligent evaluation, and the method comprises the following steps: collecting multi-source signal information of a whole cold-chain transportation process, sorting according to a unified time sequence, continuously arranging all signals on a time axis, and carrying out data processing on all the signals; a complete dynamic record is formed; and extracting each time interval and signal amplitude change track from the formed dynamic record, analyzing a signal section with repeated wave crests in a short time, and generating a preliminary data manuscript for reflection interference judgment. According to the invention, through time sequence arrangement and dynamic association of multi-source signals, a wave crest track analysis mechanism is constructed, real-time identification and correction of reflection interference are realized, and continuity and accuracy of a gas concentration curve are ensured; and by generating a reflection source position list and adjusting numerical value distribution of repeated fragments, false fluctuation is eliminated, and the stability and reliability of cold-chain pork quality evaluation are improved.

The invention discloses a pneumatic muscle driving joint force position control method and device based on fuzzy admittance and a medium. According to the method, firstly, through dynamic adjustment of the fuzzy admittance model, accurate cooperative control over the joint position and the output force is effectively achieved; secondly, the self-adaptive capacity of the system to external disturbance is enhanced by monitoring the influence of the external disturbance on the joints in real time; and the fuzzy admittance model dynamically adjusts admittance parameters according to the data, so that the control requirements of force and position are balanced, and the system can still maintain a predetermined control effect under a disturbance condition. Finally, through closed-loop feedback control, in combination with real-time monitoring data and an expected control instruction, admittance parameters are continuously adjusted, and the stability and robustness of the system in a complex dynamic environment are enhanced.

This utility model discloses a screw nut internal thread grinding machine, comprising: a bed, a fixed worktable on the front side of the bed, a headstock fixedly mounted on the left side of the worktable, and a movable center support on the right side of the worktable; a chuck for clamping the workpiece is mounted at one end of the headstock; a movable dual-axis worktable is mounted on the rear side of the bed, a movable grinding wheel head is mounted on one side of the dual-axis worktable, and a grinding wheel dresser is mounted on the other side; a retractable and rotatable grinding wheel grinding device is mounted at the front end of the grinding wheel head, and a grinding wheel drive device is connected to the grinding wheel grinding device. The bed of this utility model adopts a rectangular structure and is a grinding wheel head movable structure, which, compared with the traditional T-shaped structure machine tool with a movable worktable, has advantages such as high structural rigidity, fast dynamic response speed, good thermal stability, high machining accuracy, and small footprint.

The application provides a kind of general servo system control method and system based on special equipment application, it is related to special equipment control technical field, first receive the operation instruction sequence corresponding to special equipment application scene, then according to operation instruction sequence construction job flow timing diagram, the time sequence correlation of each operation instruction unit is clear, then obtain the real-time running state data set of special equipment, based on job flow timing diagram and real-time running state data set generate predicted motion trajectory curve, cover the expected position and speed parameter of each time point in future time window, finally call the trajectory tracking controller of general servo system to the deviation correction processing of predicted motion trajectory curve and real-time running state data set, generate the control signal sequence of driving component, drive special equipment to move according to predicted motion trajectory curve, effectively improve the motion precision and dynamic response capability of special equipment, enhance the adaptability and reliability of equipment in complex application scene.

This invention relates to the field of power management technology, and particularly to an intelligent integrated power system and control method based on multi-source collaboration, including a main power module, a backup power cluster, an energy management unit, a voltage adaptation module, a status monitoring unit, a communication module, and a load interface. Compared to existing technologies that typically use a single type of battery pack as a backup power source, which suffers from slow response speed and difficulty in withstanding instantaneous high-power surges, this invention employs a heterogeneous energy storage system composed of parallel lithium battery packs and supercapacitor modules as the backup power cluster, and uses a bidirectional DC / DC converter for intelligent control. This solution fully leverages the different advantages of lithium batteries (high energy density) and supercapacitors (high power density and rapid response), enabling the system to provide both long-term backup power support and instantaneous high-power discharge, thereby significantly improving the dynamic response capability, power support reliability, and overall service life of the backup power source.

The application relates to the technical field of power supply circuits, and particularly discloses a self-adaptive constant-current control method and system for a DC-DC direct-current conversion power supply, wherein the output voltage is acquired in real time and compared with a plurality of preset voltage thresholds, the interval where the output voltage is located is identified, the input current constant-current given value is reduced according to a first step value in response to the output voltage rising in the voltage limiting zone, the input current constant-current given value is reduced according to a second step value in response to the output voltage rising in the current sharing zone, and the first step value is greater than the second step value, so that the voltage overshoot caused by load shedding is quickly inhibited, the stable control of the output voltage is brought, the dynamic response capability of the system is improved, and the risk of damage of power devices due to overvoltage is reduced.

The invention discloses a forklift path tracking control method based on MPC and ARC cooperation, and belongs to the technical field of industrial vehicle automatic control. Comprising the steps that an upper-layer MPC is designed based on a forklift kinematics model, and an expected steering angle and an expected speed instruction are obtained through solving; a lower-layer ARC is designed based on the pump control steering system dynamics model to track an expected steering angle, and a PID controller is designed to track an expected speed; a time delay compensation mechanism is integrated in the upper-layer MPC, and a high-frequency execution strategy is adopted. According to the method, interference of the steering system is accurately observed and compensated, parameter uncertainty is adaptively processed through parameters, nonlinear uncertainty is processed through robust items, a time delay compensation mechanism is integrated in MPC, all control rates are display expressions, online solution is not needed, and the method is suitable for large-scale popularization and application. The influence of steering resistance dramatic change, external interference, unmodeled dynamics and system time delay on the control precision is effectively reduced, and high-precision and low-time-delay path tracking performance is achieved.

A transmission characteristic parameter identification method for a ball screw feeding system comprises the following steps: in the normal operation process of the ball screw feeding system, acquiring voltage, current, rotating speed and other signals through a servo motor and a driving system thereof, and realizing online identification of key electromagnetic parameters and mechanical parameters in stages; under the static working condition of the motor, controlled voltage excitation is applied, stator current is measured, and identification of stator winding resistance and stator inductance is completed; under the steady-state operation condition, the flux linkage amplitude of the permanent magnet is calculated based on a q-axis voltage equation; under a dynamic operation condition, a friction torque model is established through segmented constant-speed operation, a rotor motion equation is combined, a recursive least square algorithm with a forgetting factor is introduced, and online identification of the equivalent rotational inertia of the system is realized; according to the method, complete and accurate transmission characteristic parameters can be obtained in the ball screw feeding system, the identification process is stable, the calculation complexity is low, and the method is suitable for being embedded into a servo control system.

This application provides a converter circuit and a control method for the converter circuit. The converter circuit provided in this application includes a converter module and a control module. The control module is used to detect the voltage parameters of the converter module and determine the voltage deviation based on the voltage parameters; determine the current zero-sequence component coefficient based on the voltage deviation; and determine the modulation waveform based on the current zero-sequence component coefficient, so as to drive the converter module using the modulation waveform. The converter circuit of this application associates the current zero-sequence component coefficient with the voltage parameters. When voltage changes are caused by system startup, shutdown, load abrupt changes, etc., only the current zero-sequence component coefficient is adjusted to quickly and smoothly switch between SVPWM and DPWM under these conditions, improving system switching efficiency and dynamic response capability.

The invention provides a hybrid drive robot joint and a working method, relates to the field of robots, and aims to solve the problems of joint return errors, system vibration and dynamic precision reduction caused by transmission backlash of a precision speed reducer in existing robot joint drive. The first motor and the main driving unit of the RV speed reducer are combined with the compensation unit of the torque motor, so that the output shaft is simultaneously subjected to the synergistic effect of hybrid driving, and the torque motor outputs real-time controllable reverse compensation torque, so that gears in the RV speed reducer are always kept in tight engagement, the transmission backlash is effectively eliminated, and the service life of the RV speed reducer is prolonged. Compared with a traditional mechanical pre-tightening method, the defect that the compensation effect is attenuated due to long-term operation is overcome, stability and effectiveness can be kept in the whole process through a compensation mechanism of a torque motor, the transmission precision and position consistency are improved, the rigidity and dynamic response performance of robot joints are improved, vibration and track disturbance in the movement process are reduced, and the reliability is improved. And the robot is more stable and reliable in performance when executing work such as precise assembly and high-precision positioning.

The application discloses a gas turbine regulation and control method and system based on a decentralized control system, and particularly relates to the field of gas turbine regulation and control, and comprises the following steps: S1, synchronously collecting gas turbine operation parameters by using a distributed sensor array; S2, performing feature engineering processing on local perception data by an edge computing node; S3, based on a decentralized control system architecture, adopting an improved model predictive control algorithm to realize control quantity decoupling; S4, after completing each control cycle, evaluating the control effect; S5, establishing a regulation and control threshold, judging the evaluation according to the regulation and control threshold, and issuing a corresponding signal according to the judgment; the application synchronously collects multi-field coupling parameters by using a distributed sensor array and realizes real-time processing of local data through an edge computing node, thereby improving the dynamic response capability of the gas turbine under multi-dimensional working conditions and reducing the occurrence rate of combustion instability events.

The invention discloses a hybrid control method and system for an interleaved flyback photovoltaic micro inverter. The method comprises the following steps: S1, acquiring a real-time photovoltaic panel voltage, a power grid voltage effective value, a power grid phase and the average output power of the inverter; s2, calculating a gain factor based on the voltage of the photovoltaic panel and the phase of the power grid; s3, according to the gain factor, the network voltage effective value, the transformer excitation inductance, the average output power and the transformer turn ratio, the duty ratio and the switching frequency under the BCM are calculated; s4, comparing the switching frequency with a preset limiting frequency of a BCM mode and a preset fixed working frequency of a DCM mode, and selecting a working mode according to a comparison result; and S5, generating a corresponding control signal according to the working mode, wherein the control signal is used for driving a flyback switching tube and a power frequency flip circuit switching tube. The method has the advantages of reducing the calculation burden of mode switching and the like.

This invention discloses a simulated loading system, a simulated loading method, and a storage medium, relating to the field of excavator power matching and debugging technology. The simulated loading system includes a hydraulic main pump, a control valve, a power source, a proportional relief valve, a speed sensor, a first pressure sensor, a second pressure sensor, a flow meter, and control components. The hydraulic main pump is driven by the power source. The speed sensor is used to detect the speed value of the power source. The first pressure sensor and the flow meter are located at the output end of the hydraulic main pump to detect the output pressure and output flow of the hydraulic main pump, respectively. The inlet of the proportional relief valve is connected to the outlet of the control valve. The proportional relief valve is used to simulate the actual load on the actuator by adjusting its own opening. The second pressure sensor is located at the inlet of the proportional relief valve to detect the pressure value at the load end. This system can controllably simulate the actual load on the actuator during the excavator power matching and debugging stage and improve the power matching accuracy between the hydraulic main pump and the power source.

The invention provides a commodity recommendation method and system based on consumer behaviors, and the method comprises the steps: firstly extracting explicit behavior features and implicit preference features of consumers in different context scenes, and determining node preference indexes of consumption behaviors in different context scenes; generating a short-term intention chain and a long-term preference chain of the consumer according to the node preference index, and further constructing an element feature map reflecting consumer consumption behavior preference according to a mapping association relationship between the short-term intention chain and the long-term preference chain; extracting the preference weight of each preference node in the element characteristic spectrum under different context scenes; and detecting the matching degree between the element characteristic spectrum and the current consumption behavior characteristics, and when the matching degree is lower than a preset threshold value, performing preference updating on the element characteristic spectrum based on the preference weight corresponding to each preference node and the current consumption behavior characteristics, thereby generating a commodity recommendation list. By adopting the scheme of the invention, personalized commodity recommendation based on different context scenes can be realized in an internet e-commerce platform.

This invention relates to the field of social network analysis technology and proposes a social network sentiment evolution model that integrates user influence and activity. To address the shortcomings of existing technologies, such as inaccurate simulations and inflexible predictions in current network sentiment analysis and sentiment evolution processes, this invention provides a social network sentiment evolution model that integrates user influence and activity. The model analysis method involves: collecting text sentiment data from target users; collecting user influence data from each user; collecting the activity level of the target social network at time t; collecting the forgetting probability of each user at time t as a measure of sentiment evolution capability; establishing a dynamic model for the sentiment evolution of the social network; optimizing the evolution probability of the information propagation model based on text sentiment; and analyzing the evolution process of the dynamic model for the sentiment evolution of the social network based on the optimized information propagation model. This model is suitable for application in social network analysis.

The utility model discloses an AC-AC chopping electronic transformer, which comprises a bus power supply unit, a chopping unit, a protection output unit, a current sampling unit, a voltage sampling unit, a control unit and a frequency zero crossing point sampling unit, the bus power supply unit is connected with the chopping unit; the chopping unit is connected with the protection output unit; the current sampling unit and the voltage sampling unit are respectively connected with the protection output unit; the current sampling unit and the voltage sampling unit are respectively connected with the control unit; the bus power supply unit is connected with the frequency zero crossing point sampling unit; the frequency zero crossing point sampling unit is connected with the control unit. By means of the transformer, the control precision can be improved, dynamic real-time monitoring is achieved, and the overload protection capacity is enhanced.

The present application relates to a kind of stator coil structure and magnetic levitation plane motor for magnetic levitation plane motor.The stator coil structure includes multiple groups of shaping coils and multiple groups of long strip coils, each group of shaping coils and corresponding group of long strip coils form a group of coil combination units;Multiple groups of coil combination units are arranged according to predetermined coil period, for interacting with the mover permanent magnet array located above or below when energized, generate horizontal direction driving force and / or vertical direction suspension force;Each group of shaping coils and long strip coils are arranged alternately, and the shaping coil is used to fill the space gap formed by the arrangement of long strip coil, to improve the effective conductor ratio and coil slot fill factor in stator plane.The present application increases the effective conductor area involved in electromagnetic force under the premise that the heat dissipation condition and thermal power remain basically unchanged, so as to improve the output force density of magnetic levitation plane motor.

The application discloses a kind of secondary frequency control method, device and storage medium based on neural network, belong to microgrid control technical field, method includes: obtaining the rated angular frequency of distributed microgrid preset, and the active power of each distributed power generation unit;Active power is respectively input into the corresponding pre-trained neural network, and the frequency compensation of each distributed power generation unit is output, the frequency deviation caused by droop control is compensated, so that the angular frequency of distributed microgrid output recovers to the rated angular frequency preset;Corresponding pre-trained neural network refers to the neural network that each distributed power generation unit corresponds one training process and is added privacy protection processing, so that attacker cannot infer whether single sample is used for the training of neural network according to neural network output result.The application can improve the security and stability of microgrid, solve the problem that traditional secondary frequency control method exists privacy leakage risk and security risk.

The application discloses a kind of unmanned plane flight control method, device, unmanned plane and storage medium, the method includes: obtaining IMU data, laser radar data, determine unmanned plane current position and attitude information;Obtain the depth map data of camera, convert it into point cloud data, and construct unified environment model;Based on the dynamics constraint of environment model and unmanned plane, the feasible path from current position to target point is planned out using path planning algorithm;According to the path point sequence on feasible path, generate corresponding control instruction, drive unmanned plane to fly along feasible path autonomously;Use visual image to identify target, dynamically adjust the speed and attitude of unmanned plane according to the position deviation between unmanned plane and target point, realize accurate following to target point.The above scheme significantly improves the autonomy and robustness of unmanned plane in complex dynamic environment, so that it can complete positioning, obstacle avoidance, navigation and tracking tasks stably in variable scene.

The utility model discloses a floating fan gas-liquid linkage ballast balancing device with a bidirectional flow channel, which mainly comprises a gas-liquid linkage ballast cylinder and an annular mounting frame, and the gas-liquid linkage ballast cylinder comprises a central gas-liquid linkage ballast cylinder and an outer side gas-liquid linkage ballast cylinder; the gas-liquid linkage ballast cylinder is mainly composed of a cylinder, a plurality of sealing partition plates are arranged in the cylinder in the axial direction, the sealing partition plates divide a plurality of sealing cabins in the cylinder, a compression bin is formed between the bottommost sealing partition plate and the movable cover plate, and a compression air bag is installed in the compression bin. An electronic water inlet valve and an electronic drain valve are mounted on the inner wall of the sealed cabin; an air pipe is installed on the top face of the cylinder, and the output end of the air pump extends to each sealed cabin and the compression cabin through an air supply pipe. According to the device, gas and liquid can be dynamically adjusted through the sealed cabin, so that the position of the generator set on the sea surface is dynamically adjusted through mutual cooperation of the outer side gas-liquid linkage ballast cylinder and the center gas-liquid linkage ballast cylinder, impact of wind power and sea waves is effectively reduced, and the resistance of the device to the sea waves is remarkably improved.

The application relates to the technical field of spatial information processing and data fusion, and discloses a structure perception driven long-time sequence point-surface data fusion method and equipment. The method extracts spatial structure information in precipitation products, elevation model products and slope data by using a spatial feature extraction module, and combines a sequence feature analysis module to model time sequence change characteristics, so that the nonlinear relationship between product data and measured stations is comprehensively captured, and high-quality enhanced station data is generated. Finally, the measured and enhanced stations are fused, long-time sequence sparse observation data is high-precision converted into continuous spatial information, the spatial coverage capability and dynamic response capability of the data are effectively improved, and more continuous, accurate and reliable data support is provided for disaster monitoring and early warning of mountainous basins.

The application discloses an improved deadbeat control method for reducing switching loss, and is implemented according to the following steps: step 1, modeling a single-phase circuit of an MMC, and meanwhile, establishing a single-phase upper and lower bridge arm node current equation; step 2, discretizing the established equation, and taking a ring current as a reference quantity to establish a deadbeat control system; step 3, collecting voltages of upper and lower bridge arms of each phase of the MMC and discriminating voltage discretization; step 4, operating the upper and lower bridge arm voltages output by step 2 to obtain the number of sub-modules that should be turned on for each phase of the upper and lower bridge arms of the MMC in the next control period; and step 5, sorting each sub-module voltage of each phase after being processed in step 3, and meanwhile, turning on or turning off specific sub-modules according to the number of sub-modules that should be turned on for each phase of the upper and lower bridge arms in the next control period obtained in step 4. The application can improve system dynamic performance, improve output voltage quality and reduce the number of times of turning on sub-modules of the system.

Provided are a high-frequency dynamic response type LVDT displacement sensor and a displacement measurement method thereof, and belong to the field of precision displacement measurement. The sensor comprises a coil framework, a primary coil, a secondary coil, a core, a shell and a signal processing circuit. The primary coil and the secondary coil adopt a layered and segmented winding structure, an insulating layer is arranged between the layers, and a gap is arranged between the segments, so as to reduce the distributed capacitance and the eddy current loss; the core adopts a nanocrystalline alloy material and has high magnetic permeability and low loss; the signal processing circuit comprises a high-frequency excitation source of 10 kHz to 20 kHz, a lock-in amplifier and an LMS adaptive filtering module, and is used for suppressing noise. The measurement method comprises generating a high-frequency excitation, converting displacement into a differential signal, lock-in demodulation, adaptive filtering, temperature compensation and digital output. The dynamic response frequency of the present application is greater than or equal to 10 kHz, the response time is less than 1 ms, the precision is better than ±0.1% F.S., and the present application is suitable for high-frequency dynamic displacement measurement of aircraft control surfaces, engine vibration monitoring and the like.

This application relates to a method, system, device, and medium for torque distribution of hub motors, including: acquiring the robot's current total required torque and the real-time speed of each hub motor; inputting the total required torque and the real-time speed into a quadratic programming solver; constructing a quadratic programming objective function with the distributed torque of each hub motor as the decision variable; setting initial distributed torque components that satisfy the planning constraints and initializing the effective set, wherein the effective set includes all inequality constraints that are exactly on the boundary in the current solution and all equality constraints; performing an iterative loop subproblem solution based on all inequality constraints and equality constraints in the effective set until the optimality condition is met, and outputting the optimal distributed torque vector; using each element in the optimal distributed torque vector as the target output torque of the corresponding hub motor, and controlling the corresponding hub motor to perform torque output.

The application discloses a maximum torque current ratio control method and system for a dual three-phase permanent magnet synchronous motor based on virtual signal injection, injects a bipolar square wave signal into a current vector angle, calculates d-q axis currents after the injection of the bipolar square wave signal by using the current vector angle of the injected bipolar square wave signal, substitutes the d-q axis currents after the injection of the bipolar square wave signal into a modified torque equation, calculates positive polarity torque T e up and negative polarity torque T e down of the dual three-phase permanent magnet synchronous motor, and according to Taylor series expansion, when the difference between T e up and T e down is equal to 0, T e up and T e down are superposed with an initial current vector angle β after PI controller control, an optimal current vector angle is obtained, and the required d-q axis reference currents of the dual three-phase permanent magnet synchronous motor are calculated. The application realizes accurate control of the maximum torque current ratio.

The invention relates to the field of pump control, in particular to a joint control method and system for a motor-pump all-in-one machine, and the method comprises the steps: collecting multi-source sensor data and equipment configuration parameters of the motor-pump all-in-one machine, carrying out the preprocessing of the collected data, and constructing a state feature vector; state analysis and dynamic weighted compensation are carried out based on the state feature vector, and a compensation torque considering the response speed and the suppression capability is generated by constructing an effective adjustment intensity coefficient and a disturbance separation factor; the compensation torque is superimposed to the speed loop output to generate a final motor torque command. The control logic conflict caused by the fact that a traditional observer cannot distinguish an expected load from unknown disturbance is solved, and high dynamic response and high stability are both considered.

The application provides an anti-interference emergency communication signal enhancement device, which comprises a multi-band antenna array, the multi-band antenna array is connected with a preprocessing circuit, the preprocessing circuit is connected with a machine learning processing chip and a power amplifier, the machine learning processing chip is connected with an environment sensor group, the preprocessing circuit, the machine learning processing chip and the environment sensor group are powered by a dual power supply system, the multi-band antenna array, the preprocessing circuit and the power amplifier are isolated in a high-frequency chamber by a split shielding cavity, and the machine learning processing chip, the dual power supply system and the environment sensor group are isolated in a low-frequency chamber by a split shielding cavity. The machine learning processing chip is used to realize real-time interference filtering parameter adjustment, improve dynamic response capability and anti-interference capability, meanwhile, the split shielding cavity is used to ensure that the high-frequency circuit and the low-frequency circuit work without interference, improve electromagnetic compatibility capability and maintain device operation stability.

The invention belongs to the technical field of permanent magnet synchronous motor control, and discloses a sensorless control method and device for a permanent magnet synchronous motor, and the method comprises the steps: building a super-spiral sliding-mode observer, introducing an observation error linear term in the super-spiral sliding-mode observer, and constructing a second-order sliding-mode model; constructing a back electromotive force adaptive optimization module in the super-spiral sliding mode observer by adopting a complex coefficient filter; constructing a phase-locked loop by adopting a fourth-order active-disturbance-rejection algorithm, and estimating a rotor position error in real time by adopting a state expansion observer; a primary linear expansion state observer and a secondary linear expansion state observer are connected in series to form an improved linear active disturbance rejection controller. According to the method, buffeting of the sliding-mode observer can be suppressed, and the counter electromotive force estimation precision is improved; the dynamic response capability of the system under the working conditions of medium-high speed and sudden loading is enhanced, the method is more suitable for complex working conditions of high speed, variable load and the like, and the robustness and the control precision of the system are remarkably improved.

This invention belongs to the field of medical temporal signal processing technology, specifically a method for predicting respiratory-related displacement based on attention-embedded dilated convolution. The method involves acquiring respiratory-related temporal data of the target object over continuous time; inputting this data into a trained respiratory-related displacement temporal prediction network model, which outputs the predicted respiratory displacement. The network model comprises multiple cascaded dilated convolutional residual blocks and a prediction layer. Each dilated convolutional residual block includes a causal dilated convolutional unit, a self-attention mechanism unit, and a residual connection unit. The output of the last dilated convolutional residual block is input into the prediction layer, which outputs the respiratory-related displacement of the target object at future time points. This invention dynamically weights key time steps during each layer's temporal feature extraction process, thereby improving the predictive ability for respiratory-related displacement and tumor motion.