113results about How to "Eliminate dependencies" patented technology

A batch process model-free output feedback control method with non-repetitive interference belongs to the technical field of industrial process control, and specifically comprises the following steps: step 1, giving a mathematical description for an optimal tracking control problem of a non-repetitive batch process; 2, designing an optimal control law based on output feedback; 3, a game Q function construction method based on two-dimensional output feedback in the intermittent process; 4, solving a two-dimensional output feedback problem of batch operation by utilizing reinforcement learning; 5, solving an optimal control strategy based on output feedback through iteration to enable the optimal control strategy to be finally converged to an ideal optimal value; according to the method, the problem that a traditional method for processing the complex characteristics of a two-dimensional system is too tedious is solved, through the output feedback control method based on reinforcement learning, the method does not depend on a system model, the influence on the control performance is reduced, and the control effect is greatly improved.

This invention discloses a diagnostic method for abnormal fluctuations in quality characteristics based on heterogeneous data optimization, applied to the field of abnormal fluctuation analysis of automotive crankshafts. Addressing the problem that existing technologies are ineffective in handling the ambiguity of multi-layered, multi-dimensional, and multi-source data, this invention constructs a dimensionality reduction and optimization method for multi-dimensional, multi-source quality characteristic data. This method eliminates the correlation between variables and sampling dimensions, extracts more principal features, and optimizes the dimensionality-reduced data. Then, a diagnostic model for abnormal fluctuations in quality characteristics based on the principal feature dimensionality reduction and optimization method is constructed, thereby obtaining the location results of abnormal quality in automotive crankshafts.

The invention discloses an industrial control private protocol vulnerability mining method based on protocol analysis. The method comprises the steps of 1, preprocessing industrial control protocol flow data; and 2, analyzing the target protocol by using the byte change characteristics and the heuristic rule. And 3, generating a test case of the target protocol by using the analysis specification. And 4, reconstructing a protocol state machine by using the state conversion track of the flow, and guiding the protocol to reach a target state for testing. And 5, tracking the state of the detected target and monitoring and positioning the triggered vulnerability through a network. According to the method, the analysis accuracy of the private protocol specification can be effectively improved, the test case acceptance rate, the test target anomaly rate and the state coverage rate in the fuzzy test process are remarkably improved, and meanwhile, more protocol states can be efficiently tested under the condition of the same number of test cases so as to mine more potential vulnerabilities in the protocol.

The application relates to the field of flight control technology, in particular to a hovering control method and device for a short take-off and vertical landing aircraft of a tilt-rotor type, wherein the method comprises the following steps: constructing a hovering nonlinear dynamics model of the short take-off and vertical landing aircraft containing a tilt-rotor power system; converting the hovering nonlinear dynamics model of the short take-off and vertical landing aircraft into a high-order full-drive system model to design a state tracking controller; designing an adaptive law based on a neural network to estimate nonlinear terms in the hovering nonlinear dynamics model of the short take-off and vertical landing aircraft; and feeding back the nonlinear terms to the state tracking controller to perform hovering mode stable control. Thus, the problems that the hovering control method based on dynamic inversion needs a large amount of linearization processing, the actuator dynamic characteristics are significantly different, the controller needs to verify the closed-loop performance through repeated simulation and experiment, and the multi-actuator decoupling efficiency of the aircraft is low and the robust control performance is poor are solved.

The application provides a polarization / inertial navigation combined navigation method based on a deep Kalman filtering network, belongs to the field of bionic polarization navigation, and comprises the following steps: taking intermediate characteristics in Kalman filtering, a polarization sensor and historical data of inertial navigation as inputs of the network, building a priori prediction error covariance matrix estimation network and a measurement residual covariance matrix estimation network in polarization / inertial filtering, simultaneously training the priori prediction error covariance matrix estimation network and the measurement residual covariance matrix estimation network by adopting a two-stage training strategy, predicting states and measurements, calculating optimal gain in polarization / inertial filtering, and fusing the prediction of states and new information according to the calculated optimal gain to obtain estimation of the states at the current time. The application realizes high-precision navigation of the polarization / inertial navigation combined navigation system and improves the environmental adaptability of the system.

The present application relates to the technical field of computer vision, deep learning and graphics rendering, in particular to a 3D portrait reconstruction method and system based on cascade diffusion prior and UV space physical constraint, which adopts a parameter efficient adaptation paradigm of freezing a unified DiT backbone and a cascade LoRA adapter, and sequentially completes UV space texture completion, illumination homogenization and intrinsic material decomposition; introduces a cross-intrinsic attention mechanism to realize multi-material pixel-level spatial alignment, and executes differential physical rendering supervision to constrain material physical consistency in the UV domain. The present application realizes 100,000-level field image generalization under less than 100 3D scanning samples, and the generated 4K PBR material assets have no baking light and shadow residues, and the physical behavior is consistent under re-illumination, which is suitable for high-fidelity 3D portrait asset production in the fields of digital human creation, film special effects and virtual reality.

The invention discloses a vital sign monitoring system based on a WiFi signal, which comprises two WiFi CSI data acquisition platforms, a transmitting end platform and a receiving end platform, the transmitting end platform and the receiving end platform are respectively connected with a transmitting end antenna and a receiving end antenna through radio frequency lines, the transmitting end antenna and the receiving end antenna are directional antennas, and the transmitting end antenna and the receiving end antenna are both directional antennas. The beam of the directional antenna points to the chest and abdomen of the human body, the height of the directional antenna is flush with the chest position of the human body, and wave-absorbing materials are arranged on the sight distance paths of the transmitting end antenna and the receiving end antenna. According to the invention, vital signs can be reliably extracted, and separation of breathing and heartbeat signals can be realized. The invention further discloses a vital sign monitoring method based on the WiFi signal.

This invention discloses a method for autonomously constructing an evolvable knowledge graph by integrating a multimodal large-scale model, relating to the fields of artificial intelligence and knowledge graph technology. The method includes: collecting raw multimodal data from text, images, audio, and structured tables; using a pre-trained multimodal understanding large-scale model to perform unified semantic encoding on the raw multimodal data, generating modality-independent deep feature vectors to form an initial multimodal feature pool; performing cross-modal clustering analysis on the vectors in the feature pool, grouping semantically similar vectors into the same feature cluster, with each feature cluster defined as a candidate knowledge concept node; constructing an initial concept relationship network based on the co-occurrence relationship and feature similarity between candidate concept nodes; iteratively optimizing and evolving the network according to preset graph quality evaluation indicators, outputting the final evolvable knowledge graph. This invention achieves the autonomous and unified construction and evolution of a knowledge graph from multimodal data.

The embodiment of the invention provides a Guest OS starting exception detection method and system, and the method comprises the steps: recording the starting moment of a virtual machine instance in response to a detected starting event of the virtual machine instance; under the condition that the virtual machine instance is in the running state, determining whether the virtual machine instance successfully responds to a detection request sent by a host machine where the virtual machine instance is located or not; if it is confirmed that the predefined starting exception judgment condition is met, it is judged that starting of a guest operating system Guest OS on the virtual machine instance is abnormal; wherein the startup exception judgment condition comprises that the virtual machine instance does not successfully respond to the detection request within a preset duration after the startup moment.

The invention relates to the technical field of mobile robot navigation, and discloses a robot adaptive motion control method and system, and the system comprises a step test module which is used for recording odometer data and control quantity data with a timestamp; the parameter fitting module is used for analyzing the odometer data and the control quantity data and establishing a mapping function of the control quantity and the movement speed; the delay compensation module is used for maintaining a historical control quantity queue and predicting the expected pose of the robot when the control instruction sent at the moment t is executed based on the control delay tau; the multi-objective optimization module is used for solving the optimal control quantity; and the online updating module is used for continuously collecting motion data in the motion process of the robot and dynamically updating the mapping function and the key parameters. According to the method, actual motion parameters are automatically obtained through step testing, different robot individuals and performance attenuation scenes are adapted, the problem of model mismatching is solved, and time delay compensation is controlled to reduce positioning deviation.

This utility model relates to the field of pipeline inspection technology, specifically to a drone inspection device for oil and gas pipelines. It includes an inspection instrument with an infrared camera fixedly mounted on one side. A removable battery is detachably mounted on the adjacent side via a clamping structure, and the removable battery has a charging slot for independent charging and maintenance. A sliding frame is fixedly mounted at the bottom of the inspection instrument, with hooks slidably embedded within it, allowing the device to be hung on the operator's clothing or mounted as a whole on a drone bracket. A display screen is embedded on the side of the inspection instrument opposite the removable battery. A microcontroller fixedly mounted inside connects to the removable battery via preset contacts to obtain power, and simultaneously connects to the infrared camera via flexible wires to transmit thermal signals and to the display screen to output detection results. This utility model solves the problems of shortened battery life due to reliance on drone power supply and the inconvenience of battery removal affecting continuous operation, achieving seamless switching between autonomous power supply and manual / drone dual-mode operation.

The utility model relates to the technical field of medical instruments, in particular to a hanging frame and a therapeutic apparatus, the hanging frame comprises a handle, a rod frame and a locking mechanism, the handle is provided with a fixed channel, and the side wall of the fixed channel is provided with a first limiting hole; the rod frame is provided with a second limiting hole, and a part of the structure of the rod frame is inserted into the fixing channel. The locking mechanism comprises a driving piece, a transmission piece and a locking piece, the locking piece is arranged in the rod frame in a sliding mode, the driving piece and the transmission piece are both movably arranged on the rod frame, the driving piece is movably matched with the transmission piece, the transmission piece abuts against the locking piece, and when the transmission piece moves relative to the driving piece, the locking piece is locked by the locking piece. And the locking piece can move towards and extend into the first limiting hole or be far away from and separated from the first limiting hole, so that the rod frame is detachably connected with the handle. The utility model mainly aims to provide a hanger, and aims to improve the disassembly and assembly convenience of the hanger.

The application relates to the field of mechanical manufacturing and automatic assembly, in particular to a fan support bearing head device and a method thereof; the device comprises a rack base, a composite spindle, a micro boring cutter, a glue injection head and a control and detection unit; the device receives detection signals through the control unit and drives each component, so that the outer pipe of the composite spindle drives the micro boring cutter to cut grooves under the driving of a fixed cam, and the inner pipe synchronously cooperates with the glue injection head to inject glue; the core is that the cooperative operation of material removal processing and fluid filling is realized in the same device, so that the solidified glue and the micro groove form mechanical interlocking; the device replaces the traditional physical extrusion mode, eliminates the assembly stress and bearing deformation caused by radial extrusion from the physical mechanism, effectively improves the fixing reliability of the axial thrust, and effectively captures the debris by using the negative pressure cavity to suck air flow, and completely solves the pollution problem of the assembly surface.

This invention belongs to the fields of medical image processing and deep learning technology. It discloses a pathological image classification method and system combining stable learning and hybrid enhancement. The method involves acquiring a pathological image dataset, dividing it into a training set, a validation set, a test set, and an external validation set, and preprocessing the dataset. A deep learning network combining stable learning and hybrid enhancement is constructed and trained using the training set. The optimal deep learning network model is obtained using the validation set. The test set and the external validation set are input into the optimal deep learning network model to output the pathological image classification results. This invention utilizes a well-fitting pathological image classification model to effectively improve the overfitting problem and weak recognition ability of traditional models for domain-biased data, enhancing the recognition accuracy of independent and identically distributed data, improving the robustness and generalization ability of the pathological image classification model, and increasing the diagnostic accuracy of pathological images.

The application discloses a kind of multi-agent system output compensation control methods under non-periodic DoS attack, first construct follower fuzzy approximation system model and non-periodic DoS attack model;When attack activation causes output loss, trigger attack compensation mechanism to keep last output to provide continuous compensation signal;Switch fuzzy state observer is constructed to estimate agent state, and adaptive law in attack activation and dormancy stage is designed using Lyapunov function;Global preset performance function is constructed to establish dynamic performance boundary, and consistency tracking error and normalized conversion error are calculated to construct barrier function;Distributed adaptive security control law is obtained by combining backstepping method.The application overcomes the defect that observation performance decreases due to output loss in attack stage and traditional performance function depends on error initial value, guarantees global preset performance and high-precision state estimation, solves the cooperative control problem under non-periodic DoS attack.

An upturning type door body hinge with a random stop function comprises a driving arm fixedly connected to a door body and a fixing frame fixedly installed on a box body, one end of the fixing frame is rotationally provided with an output shaft, the outer cylindrical surfaces of the two ends of the output shaft are sunken to form an outer connecting table and an inner connecting table respectively, and the outer connecting table and the inner connecting table are connected with the driving arm. One end of the connecting rod is clamped and mounted on the inner connecting table, and the driving arm is clamped and connected to the outer connecting table; the other end of the connecting rod is hinged to a pull arm, the other end of the pull arm is connected with a sliding rod, and the sliding rod is installed in a movable groove formed in the fixing frame in a sliding mode. A spring is arranged between the sliding rod and the fixed frame, when the door body is turned upwards, the connecting rod is driven to swing through the output shaft, the pull arm stretches the spring, and the tension of the spring is balanced with the gravity of the door body, so that the door body can be paused at any position.

The utility model provides a kind of machining center contact type mechanical tool changing structure, it is related to mechanical tool changing structure technical field, including stand mechanism, the front of the stand mechanism is equipped with main shaft box, the front of the stand mechanism is close to top end place, and punch arm mechanism is installed in the inside of main shaft box, the stand mechanism includes stand main body, the top end of the stand main body is equipped with motor with output end towards bottom, the front of the stand main body is close to top end place and is equipped with punch block, the inside of main shaft box is close to the front place and is equipped with main shaft body, the top end of the main shaft box is close to left side place and is screw-connected with first fastening screw, the periphery of the first fastening screw is equipped with tension spring, the top end of the main shaft box is close to the front place and is screw-connected with second fastening screw, mechanical quick tool changing can be realized, additional transmission device is not needed, with the advantages, such as quick tool changing efficiency, small size, cost saving.

The invention discloses a construction method and system of a medical dialogue thinking chain data set. The method comprises the following steps: firstly, preprocessing original medical session data, and screening a target session round to obtain structured data; then, in combination with downstream tasks such as medical dialogue generation and disease diagnosis assistance, a forward reasoning Prompt template is constructed, and the problems that an existing method depends on the response of a real doctor, and a thinking chain lacks clinical authenticity and depth are solved. And through a forward reasoning-review-reflection mechanism, on the basis of structured data and a Prompt template, generating a thinking chain and reaction data which conform to real clinical reasoning logic, and finally forming a medical dialogue thinking chain data set. The data set is consistent with clinical diagnosis logic, and does not need to depend on the reaction of an original doctor. The diagnosis accuracy, the response pertinence and the interpretability of the medical dialogue system can be effectively improved, and powerful support is provided for constructing the reliable and interpretable medical dialogue system.

This invention discloses an intelligent adaptive dose verification calculation method and system. The intelligent adaptive dose verification calculation method includes: obtaining preliminary dose distribution data for a patient based on a deep learning-based dose prediction model; dynamically assigning variance reduction parameters to particles in a Monte Carlo simulation based on the high and low dose distribution information in the preliminary dose distribution data, wherein the variance reduction parameters include simulation weights, with simulation weights assigned to particles predicted to be in high-dose regions being less than those assigned to particles predicted to be in low-dose regions; and using a Monte Carlo dose calculation engine, simulating the planned beam based on the dynamically assigned variance reduction parameters to obtain high-precision verification dose distribution data. This invention's intelligent adaptive dose verification calculation method ensures the physical authenticity of the final verification dose distribution and the credibility of the "gold standard."

This invention relates to an automatic attitude calibration method, apparatus, device, and medium for a TMR current sensor. The method includes the following steps: acquiring reference attitude information detected by the attitude sensor in the TMR current sensor during initial installation; acquiring the magnetic signal sensed and output by the TMR sensing element in the TMR current sensor in real time during use; detecting whether the TMR current sensor undergoes positional change during use; when a positional change is detected, acquiring the current attitude information detected by the attitude sensor at the current position in real time, calculating and outputting the measured current after the position change based on the magnetic signal, the current attitude information, and the reference attitude information; when no positional change is detected, calculating and outputting the measured current based on the magnetic signal and the reference attitude information. Compared with the prior art, this invention has the advantages of improving the measurement accuracy of the current sensor and having low deployment cost.

The invention discloses a wave beam regulation and control unit, a wave beam regulator and an antenna system, and the wave beam regulation and control unit comprises a first metal layer which comprises a first pedestal, the first pedestal is provided with four first spiral inductors, any two adjacent first spiral inductors are mutually symmetrical, and two adjacent first spiral inductors at the tail ends are connected; the second metal layer comprises a second base, four second spiral inductors are arranged on the second base, the four second spiral inductors are arranged in a central symmetry mode, and the tail ends of the second spiral inductors are connected at the central point of the second base; the third metal layer comprises a third base, four third spiral inductors are arranged on the third base, any two adjacent third spiral inductors are symmetrical to each other, and two adjacent third spiral inductors at the tail ends are connected. The deployment cost of the antenna system can be effectively reduced, and the aperture efficiency of the antenna system is improved.

The application relates to the field of image recognition and discloses a visual detection method for road crowd counting analysis, which comprises the following steps: acquiring a video stream and calibrating a reference pixel line, synthesizing an original two-dimensional space-time slice graph; calculating the optical flow divergence eigenvalue between adjacent pixel points and identifying a motion vector divergence boundary; comparing a divergence discrimination threshold, inserting a zero-value pixel isolation array at the divergence boundary position to construct a reconstructed space-time slice graph with a pixel-level connectivity blocking boundary; and determining a road crowd counting result according to the distribution number of target connected domains in the reconstructed space-time slice graph. The application introduces a physical isolation mechanism based on motion vector divergence, blocks semantic adhesion features at the data bottom layer, eliminates the dependence of the identification process on individual spatial contour integrity, and effectively improves the crowd counting precision and data processing efficiency under a high-density shielding environment.

The invention discloses a delay search method and system for a delay phase-locked loop, a computer and a medium, and the method comprises the following steps: enabling a signal to activate a finite-state machine, and jumping to a search state; based on the preset configuration of the finite-state machine, linear step length search is carried out in the guard band range to locate a target phase, if the target phase is not found in the guard band range, linear step length search is carried out again after the guard band range is expanded, and when the real-time phase is equal to the target phase, a symbol state is entered; in the symbol state, the slope of real-time phase change is calculated, and symbol matching is verified; and if the symbols are matched, entering a phase-locked state, fixing a real-time delay change value in the phase-locked state, and pulling up a phase-locked loop activation signal. According to the invention, the digital code which is output by the phase discriminator and represents the phase relationship between the reference clock and the input clock can be received, the delay is searched in a linear search mode, the control signal is sent to the delay line, and the clock delay is changed until the clock meeting the requirement is generated.

The invention discloses an optical bypass protection system supporting SRv6, and the system comprises a protection control module which is used for monitoring the operation state of a bypass protection system; the switching module is used for switching between a main channel and a bypass channel under the control of the protection control module, the main channel flows through the protection system, and the bypass channel is located in the optical bypass protection system; the optical bypass side SRv6 information adding module is deployed on the bypass channel and is used for carrying out segment routing packaging on the received message; wherein when the switching module is switched to the bypass channel, the optical bypass side SRv6 information adding module reconstructs a destination address of an SRv6 TE message into a segment identification address by constructing the SRv6 TE message. According to the scheme of the invention, the destination address of the SRv6 TE message is reconstructed into the segment identification address, so that the message reinjected to the main network is logically decoupled from the original destination IP address.

The invention provides a vehicle identity information offline authorization method and system, electronic equipment and a storage medium, and belongs to the technical field of vehicle information security. The method comprises the following steps: receiving an offline authorization code which is sent by a third-party application, is signed and issued by a vehicle factory cloud and comprises application salt and a digital signature; verifying the validity of the authorization code in a hardware security environment of the vehicle-mounted terminal; after the verification is passed, based on the application salt extracted from the authorization code and the safely stored vehicle identification code and equipment salt, safety calculation is executed to generate an identity derivative code; and outputting the identity derivative code for a third-party application to perform vehicle identity verification. According to the invention, the safety and controllability of the vehicle identity information in the off-line environment are improved.

This invention discloses a distributed data sharing system based on the DPoS blockchain. The system includes modules for data acquisition, identity authentication, blockchain notarization, distributed storage, and access control. The data acquisition module encrypts files using the SM4 bit slicing algorithm to generate ciphertext and an SM4 session key. The key is then encapsulated using an ECC public-private key system to generate a file fingerprint. The distributed storage module stores the ciphertext based on IPFS and returns a CID. The blockchain notarization module records file fingerprints, CIDs, and other information using the DPoS consensus mechanism, forming an immutable registration record. The identity authentication module uses an on-chain ECC public key to complete identity authentication and key distribution. The access control module verifies the user's private key signature and distributes a key for the user to decrypt the ciphertext. This system can achieve encrypted data storage, trusted identity authentication, and access control, ensuring the security and traceability of data sharing.

The utility model discloses a kind of high-efficiency pier concrete health-cultivation devices, including the water storage tank being arranged at the top of pier and the health-cultivation coating being wrapped in the outside of pier, the water storage tank bottom is equipped with drip irrigation valve;The health-cultivation coating sequentially includes capillary wicking layer and waterproof water-locking film layer from inside to outside;The waterproof water-locking film layer is a cloth and film structure;The capillary wicking layer is composed of geotextile nonwoven fabric layer and superabsorbent resin particle layer, wherein superabsorbent resin particle is evenly distributed on the surface of geotextile nonwoven fabric layer side and is fixed on geotextile nonwoven fabric layer by adhesive, the geotextile nonwoven fabric layer periphery is fixed with the cloth layer periphery corresponding to waterproof water-locking film layer, and the superabsorbent resin particle layer is arranged towards waterproof water-locking film layer;Water distribution mechanism is connected with drip irrigation valve, and water is drained to the periphery of the top of pier. The utility model does not need external power, and only relies on gravity and capillary effect to realize uniform, long-acting, water-saving automatic health-cultivation.

The invention provides a block compression method for marine survey data, electronic equipment and a storage medium. The method comprises the following steps: acquiring marine survey data, wherein the marine survey data comprises text data and image data; according to the data type of the marine survey data, partitioning the marine survey data to obtain a plurality of original data blocks, and partitioning text data in the marine survey data according to a first partitioning size; partitioning image data in the marine survey data according to a second partitioning size; respectively carrying out data compression on each data block in the plurality of original data blocks to obtain a plurality of compressed data blocks, and adopting different compression strategies on repeated character strings and non-repeated character strings in each data block in the plurality of original data blocks; and performing global assembly on the plurality of compressed data blocks according to the sequence of the plurality of original data blocks to obtain target compressed data. The method can give consideration to high compression rate, high compression speed and low resource consumption.

The invention belongs to the technical field of airborne embedded computers, and provides a processor starting method based on a dual-redundancy Flash storage architecture, and the method comprises the following steps: configuring a main Flash and a standby Flash which are mutually backed up for a processor, and storing the same bootstrap program for the main Flash and the standby Flash; in the starting process of the processor, sensing the starting state of the processor in real time, and judging whether the bootstrap program of the main Flash is abnormal or not according to the starting state; if the SPI chip selection signal is abnormal, dynamically switching the SPI chip selection signal of the processor to the standby Flash by the FPGA, triggering the processor to reset and restarting from the standby Flash; and after the processor is successfully started from the standby Flash, the bootstrap program of the main Flash is automatically repaired. According to the method, an NVRAM is not needed, all SPI interface processors are adapted, the false alarm rate is lower than 0.1%, the system recovery time is smaller than or equal to 2 seconds, and the starting reliability of the airborne system in the severe environment is remarkably improved.

The invention relates to the technical field of data processing of multi-source data fusion, and discloses an investigation data integration analysis method and system based on multi-source data fusion. According to the scheme, a multi-source sensor is deployed, original voltage is collected, and standardization is carried out in each physical quantity range; constructing temperature, pressure and vibration three-dimensional observation and states based on a unified time index, and setting initial values; recursive prediction is carried out by using a constant preserving model; counting a multi-measuring-point variance in a static window, determining observation noise in combination with the minimum quantization step size, and calculating fusion gain; linearly updating and correcting covariance, and then carrying out dimension-by-dimension amplitude limiting according to an abnormal upper limit; carrying out space averaging on the three-dimensional results of the multiple measuring points at each sampling moment, and carrying out rollback if necessary; and a fusion sequence and a statistical report are formed in the whole cycle. According to the method, the data scale is unified, subjective parameters and computing power burden are reduced, adaptive noise modeling along with the environment is achieved, stable noise suppression and fault resistance are achieved, and continuous output and comparable analysis are guaranteed.