29results about How to "Reduce delays" patented technology

The invention relates to the technical field of fault diagnosis models, and discloses an adaptive AI fault diagnosis model fusing reinforcement learning and an industrial equipment predictive maintenance system. Comprising an industrial equipment data acquisition and preprocessing module, a reinforcement learning-fused adaptive fault diagnosis model module, an equipment health degree evaluation and residual life prediction module, a predictive maintenance decision and execution module and a system management and integration module, and the output end of the industrial equipment data acquisition and preprocessing module is electrically connected with the input end of a self-adaptive fault diagnosis model module fusing reinforcement learning. According to the system, data during equipment operation are effectively collected, the equipment health degree evaluation module is matched to convert the feature deviation degree into 0-100 visual scores, the health trend visualization unit displays changes through a broken line diagram and a thermodynamic diagram, fault risk points are marked, historical backtracking and multi-equipment comparison are supported, non-professional personnel can also rapidly judge the equipment state, and the equipment health degree evaluation module is matched to convert the feature deviation degree into 0-100 visual scores. And the equipment management efficiency is improved by 30%.

This application provides a vehicle battery charging control system, method, and vehicle, relating to the field of new energy technology. The charging control system is applied to vehicles equipped with power batteries. The charging control system includes: an on-board communication module, a body control module, and a power domain controller. The on-board communication module is communicatively connected to the body control module, and the body control module is hard-wired connected to the power domain controller. The on-board communication module detects the battery voltage of the vehicle battery based on charging configuration information and sends a charging request to the body control module when the battery voltage meets preset charging conditions. The body control module responds to the charging request and wakes up the power domain controller via hard-wired connection. The power domain controller triggers the charging system to charge the vehicle battery. This charging control system reduces the possibility of over-discharge of the vehicle battery.

This application provides a user mining method and apparatus, electronic device, program product, and storage medium. The method includes: acquiring user data through public API interfaces of multiple social networking platforms and storing the acquired user data in an intelligent computing performance-enhanced file storage service deployed on a cloud computing platform; using a data processing cluster on the cloud computing platform to read the user data from the intelligent computing performance-enhanced file storage service, reconstructing features based on the user data, and storing the reconstructed feature data back to the intelligent computing performance-enhanced file storage service; constructing and training a reinforcement learning model on the cloud computing platform based on the reconstructed feature data; storing historical experience data of the agent in the intelligent computing performance-enhanced file storage service during the training process; deploying the trained reinforcement learning model to a production environment and using the reinforcement learning model to perform user mining based on real-time acquired user data.

The invention discloses an online computing unloading method and system for edge computing equipment of a graph neural network, and the method comprises the following steps: preprocessing input data, and constructing a node feature matrix of the edge computing equipment; constructing a graph convolution network model, and performing convolution processing on the graph matrix feature data to generate an unloading probability; generating a plurality of candidate adaptive unloading decisions according to the unloading probability, evaluating a weighted calculation rate of each strategy, and selecting an optimal unloading decision; and according to the generated optimal unloading decision, training and updating the model through experience playback. According to the method, the computing task is unloaded to the network edge terminal, the delay and energy consumption problems caused by cloud processing are effectively relieved, and efficient utilization of various edge computing devices in the power grid field is achieved.

The application provides a signal timing and networked automatic driving vehicle speed hierarchical collaborative control method based on reinforcement learning, and relates to the technical field of intelligent traffic control and vehicle-road collaborative control, and comprises the following steps: obtaining the stop-and-drive vehicle state information on the import lane served by each phase at each signal cycle boundary, and generating a reference timing scheme of the next cycle by a cycle timing intelligent agent based on the stop-and-drive vehicle state information; obtaining a refined discharge window; and outputting a continuous longitudinal acceleration control instruction to a target networked automatic driving vehicle by a networked automatic driving vehicle speed intelligent agent according to the refined discharge window, so that the arrival time of the target networked automatic driving vehicle matches the discharge window, and the collaborative control of traffic signal timing and vehicle speed is realized.

The application relates to a signal modulation method, a demodulation method and a LoRa communication system. A total bit stream is divided into a plurality of parts, and is mapped into a frequency index set through G frequency index values. G LoRa signals are generated according to corresponding spreading factors, and are superimposed to obtain a single multiplexing signal of the spreading factor. The single multiplexing signals of all the spreading factors are superimposed to obtain a transmission signal, and signal modulation is completed. The transmission signal is received. For each single multiplexing signal, a de-chirp operation and a discrete Fourier transform are sequentially performed to obtain a frequency domain signal of each spreading factor. G peak values with the largest amplitudes in each frequency domain signal are detected, and corresponding frequency index values are recorded. The frequency index set corresponding to each spreading factor is estimated according to the frequency index values, and the bit stream in the frequency index set is recovered through reverse mapping. All the recovered bit streams are combined to obtain a total bit stream, and signal demodulation is completed. The application maximizes the transmission rate while ensuring the reliability of LoRa communication.

The application belongs to the technical field of emergency help equipment, and provides an emergency help signal panoramic projection device, method and terminal, the device can continuously detect the spatial posture and rotating motion state of itself through the environment positioning module, generate corresponding posture data and direction data, the control module can obtain the data and determine the real-time projection pointing angle of the device, and then when the device rotates, the projection control instruction corresponding to the real-time projection pointing angle is generated and sent to the image projection module, so that the image generation unit in the image projection module drives the corresponding rotating transformation of the help image, and the light source in the image projection module irradiates the help image to project the visual help signal which is right and complete. The scheme realizes omnidirectional controllable intelligent projection, can make the help signal cover all directions, at the same time ensures that the projected image can be clearly observed, is more easily identified by the rescuer, reduces the delay caused by signal blur, and improves the probability of being rescued by the help seeker.

The utility model relates to the technical field of medicine management, and discloses a medical examination sampling device which comprises a supporting rod, the left side of the top end of the supporting rod is fixedly connected with a containing pipe, the right side of the exterior of the containing pipe is fixedly connected with a mounting seat, and the mounting seat is connected with a sealing cover through a locking assembly. A sliding rod is slidably connected to the interior of the sealing cover, a limiting rod is fixedly connected to the interior of the right side of the sliding rod, a rotating rod is fixedly connected to the interior of the right side of the supporting rod, a button is rotatably connected to the exterior of the rotating rod, and two limiting grooves are formed in the interior of the top end of the button; the left side of the outer portion of the sliding rod is sleeved with a strong spring, the left side of the sliding rod is fixedly connected with a piston head, and the left side of the bottom end of the supporting rod is fixedly connected with a connecting ring. According to the utility model, sampling operation can be carried out more quickly and smoothly, pause and delay caused by hand fatigue or tedious operation are reduced, and the overall working efficiency is improved.

This invention provides a method and system for dynamic correction of a hoist under speed loop control, relating to the field of data processing. The method includes: acquiring the hoist car's descent information during cargo entry into the hoist car, and performing dynamic correction of the hoist's descent based on this information; collecting the real-time motor speed of the hoist during cargo transportation, and adjusting the phase current of the hoist motor based on the real-time motor speed and a preset motor speed; and acquiring the hoist car's lifting information during cargo exit from the hoist car, and performing dynamic correction of the hoist's lifting based on this information. This method has the advantage of improving the real-time performance and accuracy of hoist correction.

This application provides a human-computer interaction method and system for emergency rescue scenarios. The method involves parsing and processing the multimedia information received from an emergency rescue requester to obtain an emergency rescue situation awareness model. Based on this model, a deep reinforcement learning algorithm is used to select possible emergency response strategies. Medical knowledge graph technology is then used to analyze the relationship between the requester's symptoms and possible causes to generate an emergency rescue guidance plan. Based on this guidance plan, a traffic flow prediction algorithm is used to optimize the rescue route in real time. Combined with road condition information collected using IoT technology, the route of the rescue vehicle is dynamically adjusted to generate an optimal rescue route and the latest rescue progress notification. Finally, the optimal rescue route and the latest progress notification are used to visualize the emergency rescue operation process, generating intuitive visual guidance. The technical solution provided by this application significantly improves the speed, accuracy, and effectiveness of emergency response.

This application provides a miniaturized phase shifter, and the phase shifter is applied to an antenna system and a communications device. The phase shifter includes: a radio frequency input transmission line, a radio frequency output transmission line, and a first branch and a second branch connected in parallel between the radio frequency input transmission line and the radio frequency output transmission line; the first branch includes at least one first switch component and at least one first transmission line, and the first switch component and the first transmission line are connected in series; and the second branch includes a second transmission line, a third transmission line, at least one second switch component, and one-to-one corresponding grounding components, one terminal of the second transmission line is connected to the radio frequency input transmission line, the other terminal is connected to one terminal of the third transmission line, and the other terminal of the third transmission line is connected to the radio frequency output transmission line; one terminal of each second switch component is connected to a connection node of the second transmission line and the third transmission line, and the other terminal is connected to the corresponding grounding component; and at least one first switch component and at least one second switch component are turned on or off at a same time.

The invention discloses a platform safety line grading early warning method and system based on deep learning, and belongs to the technical field of computer vision and public safety. The method comprises the following steps: initializing a system; acquiring video stream data of a platform monitoring area in real time and performing image preprocessing; identifying and positioning pedestrians in the video frame by using the target detection and segmentation model; based on the identified foot position information of the pedestrian, judging whether the pedestrian crosses a line or not; utilizing a multi-task attribute identification model to identify the attributes of the line-crossing pedestrians; calculating a minimum line-crossing distance of the line-crossing pedestrians and a target density factor of a line-crossing area, generating a comprehensive threat score, and generating a priority queue; and generating and executing a self-adaptive multi-mode safety reminder. According to the invention, the automatic, precise and graded management of the platform line crossing behavior is realized, the timeliness, accuracy and humanization level of early warning can be effectively improved, and the method is suitable for the safety protection of public traffic scenes such as high-speed railway stations and subway stations.

This invention relates to the field of clothing design technology, specifically to a computer-aided clothing design method and system. The method includes acquiring all operation information and original clothing parameters from user actions on the computer-aided design interface; identifying the clothing modification area and modification parameters operated by the user based on all operation information and original clothing parameters; performing regional structure analysis on the user-operated clothing modification area to obtain the modification impact area; performing a fitting simulation on the clothing modification area based on the modification impact area and modification parameters to obtain qualified simulated fitting parameters; and adjusting the user-operated clothing modification parameters based on the qualified simulated fitting parameters to obtain the clothing design parameters. The purpose of this invention is to solve the problems of inaccurate fabric simulation and system delays in existing design methods, which lead to extended clothing design time and limited designer creativity.

According to certain general aspects, the present embodiments relate generally to securing communication between ECUs. In some embodiments, this can be done by utilizing the excess space in the CAN protocols. According to certain other aspects, security features such as sender authentication and message originality can be implemented at the protocol level, reducing the delays associated with implementing security features at higher levels in the communication stack. Additionally, the complexity of the security configuration is minimized by implementing the security features in hardware.

The invention discloses a power grid power transmission line fault early warning method. The method comprises the steps of obtaining historical environmental factor data and historical power grid operation state data; extracting features related to power transmission line faults and effective features corresponding to different types of power transmission line faults based on the historical environmental factor data and the historical power grid operation state data; constructing a fault probability prediction model according to the characteristics related to the power transmission line fault; constructing a fault type prediction model according to the effective features corresponding to the different types of power transmission line faults; acquiring current environmental factor data and power grid operation state data in real time, and inputting the data into the fault probability prediction model and the fault type prediction model to predict the fault probability and the fault type of the power transmission line; and when the fault occurrence probability exceeds a preset probability threshold, starting an early warning program. According to the design, the fault probability and the fault type of the line can be judged, so that corresponding prevention measures can be taken in time, and the reliability and the safety of a power grid are improved.

The invention provides a ship construction data collaboration method and system based on a unified material code, and the method comprises the steps: carrying out the three-dimensional modeling of a general station, delivering the ordering information needed by the ordering of a general assembly factory, and forming an ordering data table; and then converting the material description of the general station into the material description of the general assembly factory by establishing a standardized mapping rule to obtain a general station and general assembly factory material information mapping relation table commonly used by both parties. And on the basis, the general assembly factory receives the order data table of the general place and issues an order list according to the order data table, and finally, the general assembly factory receives the general place design model and carries out collaborative revision and confirmation on the order data according to actual manufacturing requirements. According to the data collaboration method, the collaboration problem caused by inconsistent material codes between an overall design unit and a production and manufacturing unit is effectively solved, the efficiency and accuracy in the design and manufacturing process are improved, rework and delay are reduced, and the data collaboration method has important industrial application value.

The application provides a kind of full solid waste roadbed cushion material and preparation method thereof, by breaking and screening coal-fired slag, the first slag, the second slag and the third slag are divided according to different particle size range, then the mixed slurry is obtained by wet grinding after mixing phosphogypsum, steel slag powder and lead-zinc tailings powder according to the preset weight fraction; then the required quicklime, the first slag, the second slag and the third slag are weighed according to the preset weight fraction, and the water reducing agent, the mixed slurry and the water are fully stirred after pre-stirring, to obtain the full solid waste roadbed cushion material. Through the above-mentioned mode, the synergistic effect between the raw materials can be effectively utilized, the formation of hydration products is promoted, and the heavy metal ions are effectively solidified, so that the cheap and difficult-to-handle solid waste material is prepared into the full solid waste roadbed cushion material with good mechanical properties and high stability, which has high economic and environmental benefits.

The utility model relates to the technical field of PCB production, disclose a kind of PCB board production dust removal device, including device cavity;Suction mechanism is connected in the device cavity, to the gas of outside miscellaneous dust is inhaled into the device cavity;And filter mechanism, it is set in the device cavity, located above the suction mechanism, to filter the miscellaneous dust inhaled gas;Wherein, the filter mechanism includes the spiral filter assembly being set in the top or middle part of the device cavity, and the horizontal filter assembly being set below the spiral filter assembly.The utility model provides a kind of PCB board production dust removal device, not only can be filtered to polluted air multiple times, to effectively improve the quality of device whole to air filtration and purification, and filter mechanism whole is not prone to damage, effectively reduce the probability that device is delayed due to filter mechanism damage and the overall production efficiency of PCB production line.

The invention discloses an urban traffic flow prediction and signal optimization method based on a space-time diagram neural network. The method specifically comprises the following steps: S1, constructing an urban road network topological graph; s2, multi-source heterogeneous data fusion and feature extraction; s3, establishing a space-time diagram neural network prediction model; s4, a signal optimization algorithm based on a prediction result; s5, model training and online strategy updating; s6, integration and real-time reasoning optimization are carried out; according to the method, through constructing topological graph representation of an urban road network, fusing multi-source heterogeneous traffic data and adopting a deep learning architecture combining a graph convolutional network and a time sequence prediction model, accurate short-term prediction of traffic flows of intersections and road sections is realized, and a dynamic signal optimization algorithm is designed based on a prediction result, so that the traffic flow prediction efficiency is improved. The adaptive adjustment of traffic signal timing is realized, so that the traffic efficiency of a road network is improved, and the vehicle delay is reduced.

This application provides a gesture recognition method, and relates to the field of man-machine interaction technologies. The method includes: extracting M images from a first video segment in a video stream; performing gesture recognition on the M images by using a deep learning algorithm, to obtain a gesture recognition result corresponding to the first video segment; and performing result combination on gesture recognition results of N consecutive video segments including the first video segment, to obtain a combined gesture recognition result. In the foregoing recognition process, a gesture in the video stream does not need to be segmented or tracked, but phase actions are recognized by using a deep learning algorithm with a relatively fast calculation speed, and then the phase actions are combined, so as to improve a gesture recognition speed, and reduce a gesture recognition delay.

This invention discloses an intelligent perception and proactive early warning system for safety risks in multimodal transport of long and heavy cargo based on multi-source data fusion, relating to the fields of logistics transportation and intelligent transportation technology. It includes: a raw data input module; a business scenario intelligent perception module; a data collection and analysis module; a safety risk early warning module; an operation guidance module; and a network information receiving module. This invention also discloses a method for intelligent perception and proactive early warning of safety risks in multimodal transport of long and heavy cargo based on multi-source data fusion. This invention solves the complex safety challenges faced by long and heavy cargo in multimodal transport, improving the safety and efficiency of the transportation process.

This invention discloses an optimization method for lane layout and signal timing combination at T-junctions. First, it defines index variables related to the model, including intersection branch index, flow direction index, lane index, and saturation flow reduction coefficient index. Second, by introducing a double-ring standard phase, and considering the saturation flow reduction effect as well as the configuration of shared lanes and right-turn phases, a mixed-integer linear programming model is constructed with the signal cycle minimization as the objective function and lane layout, phase duration, saturation flow, traffic flow, flow ratio, and saturation degree as constraints. Finally, the branch-and-bound method is used to solve the model, obtaining the optimal lane layout and signal timing scheme for the T-junction, thereby improving vehicle throughput and reducing vehicle delays at the intersection.

The application relates to a right-turn vehicle separation method for a roundabout intersection under an intelligent network connection environment. The application aims to solve the problem that the existing roundabout intersection is significantly congested and diffused during a peak period, and the main entrance is oversaturated, leading to low traffic efficiency of the roundabout intersection during the peak period. The process is as follows: step one, relying on a roadside unit (RSU) to collect vehicle basic information, road state and other data in real time through V2X communication and video detection; presetting a roundabout density threshold value, a vehicle speed threshold value and a queue length threshold value; when the real-time collected roundabout density is greater than the roundabout density threshold value, or the collected vehicle speed is greater than the vehicle speed threshold value, or the collected queue length is greater than the queue length threshold value, step two is executed; step two, a right-turn vehicle separation strategy is executed; and the application is used in the field of roundabout traffic control.

This invention discloses an electrical error prevention proxy method and system. The method includes: detecting network status; if the network status changes from normal to abnormal, switching the entity exercising management authority from an edge proxy device to the error prevention industrial control host; the error prevention industrial control host detecting the status of the error prevention electrical terminal and issuing error prevention instructions; wherein, the management authority is the authority to manage the error prevention operation of the error prevention electrical terminal; if the network status is normal, the edge proxy device exercises the management authority, detects the status of the error prevention electrical terminal and issues error prevention instructions; the edge proxy device sends the status of the error prevention electrical terminal to the error prevention industrial control host. This invention can reduce network delays in electrical error prevention and avoid safety hazards caused by network delays in electrical error prevention.

The invention provides an industrial Internet of Things service information processing method and system based on identifier analysis, and relates to the technical field of industrial Internet of Things. After industrial equipment is powered on, identity information containing a unique industrial Internet identifier and a digital signature is generated through an integrated active identifier carrier, and an access request is sent to an edge intelligent gateway; the gateway queries an identifier analysis system based on the identifier, acquires corresponding equipment configuration information and automatically configures a data acquisition task; the core platform service layer performs associated storage management on the acquired data, and provides digital twinborn model information and a service interface for the application and twinborn layer according to a preset identification mapping relationship; the application and the twinborn layer dynamically construct and render a visual scene according to the active equipment identification list and related information, real-time synchronization and interaction of physical equipment and a digital twinborn model are achieved, and zero-configuration network access, virtual-real dynamic synchronization and full-process efficient and controllable service processing of industrial equipment can be achieved.

The utility model provides a battery cover plate dispensing device, which relates to the technical field of battery cover plate processing equipment, and comprises a base and a working mechanism, the upper end of the base is fixedly connected with the working mechanism, through the arrangement of the working mechanism, when the battery cover plate dispensing device is used, the automation and accurate control of the battery cover plate dispensing process are realized, and the working efficiency is improved. The air cylinder driving assembly reduces manual intervention through automatic operation, the production efficiency is improved, an adjusting plate and a fixing clamping ring of the adhesive dispensing assembly allow fine position and angle adjustment of an adhesive dispensing head, the adhesive dispensing accuracy is ensured, the equipment adaptability is improved through a lifting adjusting assembly, the device can adapt to battery cover plates with different thicknesses, and the product quality is improved. The fixing assembly clamps the battery cover plate through the fixing clamping strip and the connecting fixing plate, the stability in the dispensing process is guaranteed, the maintenance process is simplified through the automatic cleaning function of the cleaning assembly, the maintenance cost is reduced, and the operation convenience of a user is further improved through simple operation of the handheld rod.

The invention discloses a semiconductor scheduling method based on dynamic priority and reinforcement learning decision, relates to the technical field of semiconductor manufacturing scheduling, and aims to solve the problems of low utilization rate, delivery delay and poor dynamic adaptability of traditional scheduling equipment. The method comprises the following steps: fusing equipment space topology and process dependence, and generating a machine family low-dimensional coding vector; constructing a 13-dimensional dynamic feature vector, standardizing the 13-dimensional dynamic feature vector, inputting the standardized 13-dimensional dynamic feature vector into a strategy network consisting of a multi-head attention network and a feedforward network, and outputting a real-time priority; a reinforcement learning framework is constructed based on event driving, and the network is optimized through a three-level machine distribution rule, a negative penalty reward function and a natural evolution strategy. According to the invention, equipment load balancing and scheduling intelligent adaptation are realized, the equipment utilization rate is effectively improved, the wafer batch tardiness and completion time are reduced, and the method is suitable for a complex dynamic semiconductor manufacturing scene.

The utility model relates to the technical field of reinforcing steel bar sleeves, and discloses a double-layer reinforcing steel bar sleeve which comprises an outer cylinder serving as a connecting body, inner cylinders are arranged on the two sides of the inner wall of the outer cylinder, fixing boxes are fixedly connected to the opposite ends of the inner cylinders, and pull rods are connected to the inner walls of the fixing boxes in a sliding mode. The opposite ends of the pull rods are fixedly connected with sliding blocks, the opposite ends of the sliding blocks are rotationally connected with rotating blocks, and the opposite ends of the rotating blocks are rotationally connected with inserting blocks. The pull rod drives the sliding block, the sliding block drives the rotating block, the rotating block drives the inserting block, then the inner cylinder is inserted into the outer cylinder, when the pull rod is loosened, the sliding block is sprung by the spring, the sliding block drives the rotating block, and the rotating block drives the inserting block to be inserted into the fixing hole to be fixed, so that steel bar connection operation is smoother; and flow interruption and delay caused by inconvenient operation are reduced, and the continuity and efficiency of construction are improved.

The application relates to a driving task allocation method and device based on data-driven heavy-load combined train driving hierarchical evaluation and a computer device. The method comprises the following steps: acquiring train running state data of a heavy-load combined train driven by a to-be-analyzed driver in multiple driving sections; from each driving section, a target driving section meeting a preset static geographic condition is screened out, and target train running state data in the target driving section is screened out from the train running state data; according to the target train running state data and the train running state data, a target driving evaluation result of the to-be-analyzed driver is determined; from the to-be-analyzed drivers, a target driver whose target driving evaluation result matches a task type of a to-be-analyzed driving task is screened out, and the to-be-analyzed driving task is allocated to the target driver. By using the method, the running efficiency of the heavy-load combined train can be improved.