455results about How to "Reduce lag" patented technology

The invention discloses a live broadcast time shift method which is applied to a live broadcast network including a client and a server. The live broadcast time shift of a user on the client is divided into pause and play. The client acquires the time label of a first index file currently parsed by the client according to pause, and acquires the played time of a video clip corresponding to the first index file. The client determines the timestamp of pause in a live broadcast video according to the time label and the played time. When the client receives play, a live broadcast time shift request is sent to the server according to the time shift time and timestamp of play. The live broadcast video in the live broadcast network is parsed and played according to a second index file returned by the server. After a user who watches live broadcast carries out live broadcast time shift, discontinuous screen caused by play screen hysteresis and play delay is reduced, and the user experience is improved. The invention further discloses the client.

The invention discloses a system resource dynamic capacity expanding and shrinking method and device, a computer device and a storage medium which are applied to the technical field of system configuration, and are used for solving the problems that the timeliness of existing system resource scheduling is low and the system stability is reduced. The method provided by the invention comprises the following steps of capturing a system log of a target system in real time, and obtaining current service data of the target system from the system log; obtaining allocated system resources allocated tothe target system by the system resource pool; analyzing the current service data according to a preset service index rule to obtain a current service operation index of the target system; calculating a resource use condition index of the target system according to the service operation index and the distributed system resource; if the resource use condition index is smaller than a preset index lower limit value, performing capacity reduction processing on the target system; and if the resource use condition index is greater than the preset index upper limit value, carrying out capacity expansion processing on the target system, wherein the index upper limit value is greater than the index lower limit value.

Methods and systems for providing a media guidance application, which provides media listings during the credits accompanying a work. The media guidance application may automatically retrieve listings of related works (e.g., movies, television shows, webcasts, etc.) featuring the individual member of the cast or crew of the work and supplement or replace each cast or crew member's name with listings for the related works.

The invention provides an electric vehicle cooling system which comprises three circulating loops: a small circulating loop, a large circulating loop and a warm air loop, wherein the three loops have a shared cooling pipeline, and the shared cooling pipeline is provided with a heat radiator and an electric pump; the small circulating loop is that a cooling liquid sequentially passes through the electric pump, a motor controller and a motor through a cooling pipe, and does not pass through the heat radiator; the large circulating loop is that a cooling liquid sequentially passes through the electric pump, the heat radiator, the motor controller and the motor through a cooling pipe; the warm air loop is that a cooling liquid passes through the electric pump, a warm air heat exchanger, the motor controller and the motor through a cooling pipe; the cooling system also comprises an expansion box; the heat radiator and a pressure-limiting valve of a warm air loop heat exchanger are connected with an upper water port of the expansion box; the lower water port of the expansion box and a main cooling pipe are connected in parallel to a water inlet of the electric pump; temperature sensors are installed at the inlet part of the motor controller, the cooling liquid outlet part of the electric pump and the outlet part of the heat radiator; sensor signals above are all connected to a cooling system control unit; and the cooling system control unit controls the operation of the system according to the information obtained from a vehicle-mounted bus as well as sensor information. The invention also provides a control method for the electric vehicle cooling system, and the normal operation of the power part of the electric vehicle is ensured.

A movable optical element adjusting and positioning device relates to a photoetching apparatus. A movable lens set in the device is fixed on a main lens cone by a laminated spring plate and can move along an optical axis or carry out horizontal inclination adjustment under the action of a driving motor. The driving of the driving motor adopts a long handle of a level mechanism in a driving mechanism of a gapless flexible bearing, and a short handle thereof drives the movable lens set to move along the optical axis. The adjusting and positioning device achieves an axial movement accuracy which can be lower than 0.5 microns and a horizontal adjustment accuracy which can reach 0.5mrad. The adjusting and positioning device has compact structure, stable operation, no friction or lubrication, low hysteresis and good repeatability, and the movement stroke of the movable lens set can reach 0.2 mm and the linearity is good.

The invention relates to an MXene@CS@PDMS three-dimensional porous composite material and a preparation method and application thereof. A high-sensitivity and high-reliability piezoresistive sensor isprepared by using the material. Since the cube sugar micro-pore granularity is proper and the structure is loose, the prepared PDMS has a three-dimensional sponge-like porous structure by using the cube sugar as a template and has the high flexibility and deformability. With introduction of the MXen@CS@PDMS system, an MXene nanosheet is attached to the PDMS surface tightly after successful loading of the MXene, so that a reliable three-dimensional conductive material is formed. Moreover, reaction parameters in the process of preparing the sensor are optimized to obtain the MXene@CS@PDMS composite three-dimensional structure with good stress and strain is obtained.

A composite type composed of polyurethane surface layer and main body of rubber is prepared through treating the surface of main body, synthesizing and modifying the prepolymer of polyurethane, and casting the surface layer. Its advantages are high elasticity and antiwear nature, and low rolling resistance.

The invention provides a networking method and apparatus of a fog wireless access network based on artificial intelligence. The method comprises the following steps: a central computing logic module receives reported data, wherein the reported data comprise measurement report data reported by a user terminal in the wireless access network, wireless transmission data reported by a base station andoperation and maintenance data reported by the wireless access network; based on these reported data, the central computing logic module configures a wireless access network working mode matching a user behavior, a service attribute and the wireless access network performance through a machine learning algorithm; and based on the access network working mode, an edge computing logic module judges adeployment method for optimizing current configuration, wireless resources, computing resources and cache resources of an edge communication entity. The proposed networking method is based on the machine learning algorithm, can adapt to various service demands, the wireless access network is flexibly constructed, and different application scenes and performance target demands are met.

The invention belongs to the field of equipment fault diagnosis, and particularly relates to a sewage treatment equipment fault diagnosis system and method based on Lasso regression. The fault diagnosis system comprises a sewage treatment subsystem, a plurality of supervision clients and a plurality of data acquisition devices; the sewage treatment subsystem comprises a central console, a communication server, a total data storage server and a data acquisition server; the central console comprises a control module, a data interface module, a data classification module, a machine learning module and a data communication module; and by acquiring data on the data acquisition devices, the machine learning model carries out learning training on the data so as to construct a plurality of sub regression prediction models for predicting fault types of various types of sewage treatment equipment. The sewage treatment equipment fault diagnosis system and method disclosed by the invention providea certain basic diagnosis help for engineering supervision and treatment decisions of sewage treatment, and effectively meet requirements for real-time monitoring and diagnosis on the sewage treatment environment fault problem in the sewage treatment process.

Embodiments of the invention provide a speech translation method and device, and a device for speech translation. The method specifically includes the steps of obtaining a text that corresponds to a speech recognition result that has undergone punctuation adding processing; obtaining a target clause from the text; translating the target clause and outputting the obtained first translation result;and when a current pause corresponding to the speech recognition result is detected, performing a second translation on the text corresponding to the speech recognition result that has undergone punctuation adding processing and between a previous pause and the current pause, and outputting the obtained second translation result so as to replace the first translation result with the second translation result. In the embodiments of the invention, the lag of the translation result with respect to a speech signal can be effectively reduced by the first translation result, and the quality of the translation result finally provided to a user can be enhanced by the second translation result.

The invention is suitable for the technical field of audios and videos, and provides a method and a device for processing play abnormality of an audio stream, a computer device and a computer readablestorage medium. The method comprises the steps of obtaining audio frame data in a cache; parsing decoding parameters in the audio frame data to obtain related audio data and recording the time stampof the first frame of the audio frame data; decoding and rendering the parsed audio frame data; obtaining the actual audio rendering time based on the related audio data and the time stamp of the first frame of the audio frame data; obtaining a real-time code stream playback time drift based on the actual audio rendering time; judging whether the audio frame data are abnormal based on the real-time code stream playback time drift; and when the judgment result is affirmative, starting and resetting an audio processing operation. According to the embodiment, the problem of sound interruption andstop caused by factors such as code stream shaking, signal instability or code stream hop can be processed efficiently, the stability of audio play is improved, and the conditions of unsmooth audio play are reduced.

The embodiment of the invention relates to a resource management method and device, a mobile terminal and a computer readable storage medium. The above method comprises the following steps that: when a detection result shows that an application in a preset application list operates in a foreground, determining a corresponding foreground process; collecting the occupancy rate of each CPU (Central Processing Unit) in multi-core CPUs; selecting the CPU with the lowest occupancy rate, and distributing the foreground process to the CPU with the lowest occupancy rate to be operated; and when a background process operates on the CPU with the lowest occupancy rate, distributing the background process to other CPUs to be operated. By use of the above resource management method and device, the above mobile terminal and the above computer readable storage medium, a situation that the application which operates in the foreground slowly operates since resources are seized can be reduced.

The invention discloses a medical knowledge graph construction method which can make full use of a multi-network knowledge base to dynamically construct a knowledge graph and is good in integrity andhigh in accuracy. The medical knowledge graph construction method comprises the following steps: (10) medical text data acquisition: crawling medical text data including disease description, symptom description, questions and answers and comments in a plurality of data sources through a web crawler technology; (20) obtaining a medical knowledge graph: extracting medical entities and relationshipsbetween the medical entities from the medical text data through a dictionary and rule method to obtain the medical knowledge graph comprising a medical entity graph and a medical entity relationship graph; (30) knowledge graph storage: storing the medical knowledge graph in a Neo4j graph database; and (40) entity fusion: performing entity fusion by adopting an integrated entity alignment method.

The present invention discloses a method and an apparatus for pushing live stream media data. The method comprises the steps of determining a target server in a server cluster used as a live relay node; acquiring live source data from a preset server by the target server; and pushing the live stream media data to audience clients by the target server, wherein the live stream media data is stream media data obtained after the live source data is transcoded, and the live source data is multimedia data that is collected and uploaded to the preset server by an anchor client in live broadcast. According to the method, the server cluster used as the live relay node can share the service pressure of the preset server, so that stalling of the audience client is relieved, the live broadcast quality is improved, the method and apparatus are highly available, and the user experience is improved.

The invention discloses a deep learning-based social network rumor detection method. The method comprises the steps of collecting social network data as sample data; performing marking and word segmentation on the sample data, and constructing a dictionary and sample word vectors having fixed lengths and taking numbers of sample words in the dictionary as elements; limiting a quantity of sample words contained in sample sentences of the sample data to be a fixed value; constructing a sample sentence matrix by adopting a Word2Vec method, wherein a row vector group of the sample sentence matrix is sample word vectors of all the sample words in the sample sentences; training the sample sentence matrix by adopting a deep learning method LSTM, and building a multilevel training model; constructing a to-be-detected sentence matrix by adopting a method as same as the method for constructing the sample sentence matrix; and performing classification detection on the to-be-detected sentence matrix according to the multilevel training model to obtain a rumor detection result of the to-be-detected social network data. According to the method, social network rumors can be effectively detected.

The invention discloses a method and device for generating a special effect video, belonging to the technical field of computers. The method includes the following steps: receiving a special effect adding instruction, and recording special effect information corresponding to the special effect adding instruction; during video recording, when a recording completion instruction is received, ending the video recording, and displaying a video editing page; and when a special effect editing completion instruction triggered by the video editing page is received, performing special effect editing onthe recorded target video according to the special effect information to obtain a special effect video corresponding to the target video. By adopting the scheme of the invention, the occurrence of thelow-speed operation of a terminal can be reduced.

A control device and a control method for compensating torque adjusting delay of a wind generating set are provided. The control device comprises the wind generating set, a pitch control mechanism, a parameter detecting module, a main controller, an overvoltage protecting module and a frequency conversion controller. By the adoption of the control method for compensating the torque adjusting delay of the wind generating set, the influence of torque adjustment on a whole machine is reduced. On one hand, a feed forward link is added in the control device and the control method, thereby greatly removing the delay caused by torque control and the influence of a random wind speed on quality of system control, optimizing power output of the set and improving responding speed of a subsystem controlled by a fan; on the other hand, a damper is added so as to reduce influence of load impact on a set transmission chain. By means of the control device and the control method for compensating the torque adjusting delay of the wind generating set, responding delay of the wind generating set caused by the torque control and the impact on transmission chain loads can be reduced.

The invention discloses a method for dynamically updating knowledge map, which is used for solving the synchronization problem between an encyclopedia knowledge map and a data source thereof. The invention takes the hot content on the World Wide Web as a starting point. Named entities are extracted as seed entities from which updates are likely to occur. Other entities associated with the seed entity are then captured on the encyclopedia site as extension entities. Then, a certain number of entities are obtained from the encyclopedia website to carry out feature engineering, and the update information of entity feature representation is mined by machine learning algorithm, and the predictor is constructed. Entities with high update probability are selected from the extended entities by using the predictor. Finally, the dynamic updating of knowledge map is realized by using seed entities and extended entities with high updating probability as updating objects under the circumstance of limited access to data sources.

The invention discloses a transverse extension preview switching control method for an autonomous vehicle. The transverse extension preview switching control method employs upper and lower layer controllers. In the upper layer controller, a front road curvature value and a transverse position deviation of the vehicle and a road center line are selected as eigenvalues of an extension set, the extension set is divided, a correlation function is solved, and vehicle-road system states are divided into a classic domain, an extension domain and a non-domain. In the lower layer controller, the classical domain adopts a PD feedback controller based on the transverse position deviation and a course deviation, the extension domain adopts a PD feedforward-feedback controller based on the front road curvature value, and a vehicle-road system in the non-domain is in an out of control state, emergency braking is adopted, and the switching control of two control strategies in the classic domain and the extension domain is realized according to different states of the vehicle-road system. The transverse extension preview switching control method for autonomous vehicle successfully applies an extension control theory to the field of transverse control of the autonomous vehicles, and satisfies the requirements for transverse control precision of vehicles.

A clamping circuit including a clamping diode, a bias line, and a clamp line is incorporated into a pixel circuit of amorphous silicon sensor arrays. The clamp diode in each pixel prevents the voltage across the photodiode from dropping below a specific threshold. By keeping the photodiode under reverse bias even under conditions that may otherwise saturate the pixel, image lag is reduced. In full fill factor amorphous silicon sensor arrays, a clamping circuit includes a clamp TFT, a bias plane, a clamp line, and a drain line. The clamp TFT reduces lag and blooming by draining off excess current developed under overexposure conditions. A method to globally reset a sensor array and a method to test and repair a TFT matrix in full fill factor sensor arrays without damaging the overlying collection electrode and sensor layer are also provided.

The invention relates to a cruciform spring element as preferably mounted in measuring devices for measuring the mass flow rate of bulk materials according to the Coriolis principle. To this end, two bearing elements (1, 2) that can turn relative to one another are provided in the measuring branch and have at least four leaf spring elements (3, 4, 5, 6) that cross one another in an orthogonal manner and join both bearing elements (1, 2) to one another. The crossing leaf spring elements (3, 4; 5, 6) of one radial direction (8) are arranged at least in pairs and cross at the rotational axis (7). The ends of each leaf spring pair (3, 4; 5, 6) are respectively fastened, on one side, to different bearing elements (1, 2) whereby transferring radial pressure loads and tensile loads in each direction (8) at the same time.

The present invention relates to a method for quickly measuring speed with high precision, in particular to a method for quickly measuring the speed of a motor with high precision. By labeling a Hall signal of the detected motor, the rotation rate of the motor which is accurately acquired by using a small section of signal in the rotation process of the motor is realized. Under the condition of stable speed, and the magnitude relation of the lengths of three Hall signals which are generated during the process of one complete revolution of the detected motor is labeled for determining the corresponding relation of a Hall element and different poles during the actual operation. Meanwhile, the trace and the positioning of the corresponding poles of the Hall signals during the operation process of the motor are completed. The primary advantage of the method is to use the small section of the signal generated during one complete revolution of the motor to accurately acquire the rotating speed of the motor speed. The quick and highly-precise speed measurement is realized.

Use of the hydraulically driven device in a series configuration with a minimally restrictive turbocharger is defined which will allow a very responsive and powerful boosting system to reach boost levels of 4-5 pressure ratio (PR) to support and enable OEM engine downsizing trends. An electric supercharger is also considered. A hydraulic drive assists to increase the acceleration rate of a turbocharger impeller / turbine shaft assembly and provide a secondary means of driving the compressor impeller at lower engine speeds where exhaust gases alone does not generate adequate shaft speeds to create significant induction boost. The hydraulic circuit includes a dual displacement motor, which provides high torque for acceleration yet converts to a single motor for high-speed operation. When the exhaust driven turbine function allows compressor speeds, beyond which the hydraulic system can contribute, a slip clutch allows disengagement of the hydraulic drive. In an alternative embodiment, the hydraulic drive provides means of forced induction air alone.

The invention discloses a continuous type pad printing and coding device. The continuous type pad printing and coding device comprises a workbench, a PLC controller and a pad printing machine. The continuous type pad printing and coding device further comprises a laser coding machine, a conveying component and a drying component. The conveying component comprises a motor, a transmission case, a main chain wheel, a chain and an auxiliary chain wheel. The chain is fixedly provided with multiple fixing seats. Locating blocks are connected to the fixing seats. Brackets are connected to the locating blocks, and the brackets are provided with mold grooves. The side wall of the transmission case is provided with a pad printing locating piece and a coding locating piece. Each locating block is provided with a left groove and a right groove. One end of the transmission case is provided with an infrared sensor. A motor shaft of the motor is sleeved with a wheel disk. Three sensing columns are distributed on the wheel disk. The infrared sensor, the pad printing machine and the laser coding machine are electrically connected with the PLC controller. The continuous type pad printing and coding device conducts continuous pad printing and coding operations on printed objects on a large scale and greatly improves the production efficiency; and pad printing and coding are clearer, the precision is high, the manufactured products are free from fading after being used for a long time and high in recognition capacity.

The invention discloses a method and device for recommending content distribution network systems. The method comprises the steps of determining a plurality of content distribution network systems deployed by a multimedia website; in allusion to each content distribution network system, obtaining a real-time quality evaluation parameter set and a historical quality evaluation parameter set which reflect the content distribution network system; determining a service quality order of the plurality of content distribution network systems according to the respective real-time quality evaluation parameter sets and the historical quality evaluation parameter sets of the plurality of content distribution network systems; and determining a recommendation sequence of the content distribution network systems according to the service quality order of the plurality of content distribution network system so as to recommend the content distribution network systems based on the recommendation sequence. The scheme of the invention is conducive to improving the video playing fluency, reduce the circumstance that the video cannot be loaded for a user terminal side and improve the video watching experience of users.

The invention relates to the technical field of road tunnel ventilation, in particular to an environment prediction-based tunnel ventilation control method and system. The tunnel ventilation control method comprises the steps of setting target values of indexes related to a tunnel environment; collecting detection values of the indexes at a current moment; according to the detection values of theindexes and a prediction model built through deep learning, predicting prediction values of the indexes at a next moment; and according to the detection values, the prediction values and the target values of the indexes, adjusting ventilation volume of a tunnel. The tunnel ventilation control system comprises a fan set, a detection apparatus and a control apparatus, wherein the fan set is arrangedin the tunnel; the detection apparatus is arranged corresponding to the fan set; and the control apparatus is connected with the fan set and the detection apparatus. According to the environment prediction-based tunnel ventilation control method and system, the hysteresis of air volume feedback control can be reduced and the ventilation volume control accuracy can be improved.

The invention provides an FPGA-based flux gate micro signal detecting system and an FPGA-based flux gate micro signal detecting method. The FPGA-based flux gate micro signal detecting system comprises a detecting module, an output module and a processing module which is realized by FPGA, wherein the detecting module converts a flux gate signal to a digital signal and transmits the digital signal to the processing module. The processing module performs processing for obtaining a secondary harmonic signal and transmits the secondary harmonic signal to the output module. According to the FPGA-based flux gate micro signal detecting method, after power amplification is performed on a sine wave with frequency of f0 / 2, saturation activation is performed on the activation coil of a flux gate sensor for generating a flux gate signal with frequency of f0. After filtering and analog-to-digital conversion are performed on the flux gate signal, cross-correlation detection is performed on the obtained digital signal and two reference signals for obtaining a secondary harmonic signal, the amplitude and the phase of the secondary harmonic signal. Finally digital-to-analog conversion is performed on the secondary harmonic signal, and the obtained analog signal drives the load to operate. According to the FPGA-based flux gate micro signal detecting system and the FPGA-based flux gate micro signal detecting method, related operation is realized by means of the FPGA. The FPGA-based flux gate micro signal detecting system and the FPGA-based flux gate micro signal detecting method have advantages of effective hysteresis reduction in signal transmission, high noise inhibition capability, convenient signal processing, and high measurement accuracy. Furthermore the FPGA-based flux gate micro signal detecting system and the FPGA-based flux gate micro signal detecting method can detect nanovolt-grade weak signals.

Example embodiments disclose an image sensor capable of preventing or reducing image lag and a method of manufacturing the same. Example methods may include forming a gate insulating film and a gate conductive film doped with a first-conductive-type dopant on a semiconductor substrate; forming a transfer gate pattern by patterning the gate insulating film and the gate conductive film; and fabricating a transfer gate electrode by forming a first-conductive-type photodiode in the semiconductor substrate adjacent to one region of the transfer gate pattern, by forming a second-conductive-type photodiode on the first-conductive-type photodiode, and by forming a first-conductive-type floating diffusion region in the semiconductor substrate adjacent to the other region of the transfer gate pattern.

The invention discloses a construction method of a dynamic knowledge graph. The method comprises the steps of S1, obtaining original data information; s2, storing and processing the original data information; s3, establishing a knowledge graph primary framework according to the processed original data information; s4, performing knowledge updating on the primary framework of the knowledge graph; s5, performing multiple operations on the S4 to realize multiple updating of the primary framework of the knowledge graph; s6, completing the establishment of the dynamic knowledge graph; according tothe method, the construction and updating of the knowledge graph are efficiently realized, and the initial accuracy and integrity of the knowledge graph are improved.

The invention discloses a moving average and neural network-based virtual machine load prediction method and system. The method comprises the steps of S1, collecting divided-period historical load data of a to-be-predicted time period and continuous load data before the to-be-predicted time period; S2, obtaining the divided-period historical load data of the to-be-predicted time period, and after a double moving average method is used for performing predication calculation on load inertia of a virtual machine, performing prediction to obtain a first load inertia prediction value of the next time period; S3, obtaining the continuous load data before the to-be-predicted time period, and in combination with the first load inertia prediction value, performing prediction by adopting an RBF neural network to obtain a second load inertia prediction value; and S4, taking the second load inertia prediction value as a final virtual machine load inertia prediction value and performing outputting. According to the method and the system, the hysteresis of continuous load prediction can be reduced, the timeliness of load prediction can be improved, the prediction accuracy can be improved, and the adaptive flexibility of abnormal conditions is high; and the method and the system can be widely applied to the field of virtual machine load prediction.