62results about How to "Improve scheduling accuracy" patented technology

Disclosed are a self-driving car scheduling method, a car scheduling server and a self-driving car, which belong to the technical field of computers. The method comprises: receiving a request for taking a car; according to the request for taking a car and travel information about various self-driving cars within the range of management, determining at least one first candidate car in a plurality of self-driving cars; according to current position information, current road condition information and planned route information about each first candidate car, calculating a first time required by each first candidate car to reach a destination for taking a car; according to the first time corresponding to each first candidate car, determining the final candidate car in the at least one first candidate car, the final candidate car at least satisfying the expected time of reaching the destination; and issuing the request for taking a car to the final candidate car. By means of the present invention, at least according to the request for taking a car and the current position information, current road condition information and planned route information about various self-driving cars, the cars are scheduled, thereby having a high scheduling accuracy.

Systems and methods for providing remote access to interactive media guidance applications are provided. In particular, the systems and methods for providing remote access to interactive media guidance applications provided herein generally relate to enhancing remote access capabilities by providing algorithms and techniques for managing and correcting disparities that may be found between program listings data used by the interactive media guidance application and the remote access server.

Systems and methods for providing remote access to interactive media guidance applications are provided. In particular, the systems and methods for providing remote access to interactive media guidance applications provided herein generally relate to enhancing remote access capabilities by providing algorithms and techniques for managing and correcting disparities that may be found between program listings data used by the interactive media guidance application and the remote access server.

A pipe network node flow measuring and dispatching method based on pressure monitoring comprises the steps that node pressure of a predetermined node in a pipe network is measured actually, and according to the measured node pressure, based on the discrete point spatial interpolation computing method, node water heads of all nodes are worked out; the pipe section flow of a predetermined pipe section is measured actually, coefficients of friction resistance of all flow measuring pipe sections are determined according to the actually measured pipe section flow and the worked-out node water heads, pipelines are grouped through the spatial clustering algorithm according to the coefficients of friction resistance of the pipe sections, and coefficients of friction resistance of all pipe sections are determined; according to the worked-out node water heads, the coefficients of friction resistance of the pipe sections, pipe lengths and pipe diameters, the pipe section flows of all the pipe sections are worked out; according to the material equilibrium principle, the node flows of all the nodes are worked out, and therefore pipe network model updating and checking, pipe network dispatching, leakage quantity analysis, leaking point positioning and other pipeline maintenance and operation management are conducted conveniently. By applying the method, energy consumption and loss due to leakage can be reduced substantially, and the pipeline management level is improved.

The invention discloses a device for scheduling a push flow server. The device comprises a memory and a processor, the memory stores a scheduling program capable of running on the push flow server, and when the program is executed by the processor, the following steps are implemented: when a push request sent by a push flow client is received, determining a geographical area where the push flow client is located according to the push flow request; obtaining an IP address of a first level node server corresponding to the geographical area in a server group according to preset server level distribution; sending the obtained IP address to the push flow client, so that the push flow client pushes a live broadcast streaming media file to the first level node server corresponding to the IP address according to the IP address. The invention further provides a method for scheduling the push flow server, and a computer readable storage medium. By adoption of the device and method disclosed by the invention, the accuracy of scheduling the push flow server is improved, thus improving the stability of the uploading speed of video streams.

The invention relates to a vehicle leasing method and apparatus, computer equipment and a storage medium. The vehicle leasing method includes the steps: acquiring a map of a monitoring area, and determining a plurality of rental stations in the monitoring area according to the map; collecting shared vehicle access data of multiple rental stations in a plurality of monitoring time frames in a historical monitoring period; inputting the shared vehicle access data into a preset vehicle leasing model, and determining the conventional quantity demanded for the shared vehicles for the plurality of rental stations; according to the conventional quantity demanded, generating a leasing strategy of the monitoring area in the plurality of monitoring time frames of the current monitoring period; and identifying whether the shared vehicles arrive at the rental stations, and performing resource transfer on users of the shared vehicles which arrive at the rental stations based on the leasing policy.The vehicle leasing method can reduce the leasing cost of shared vehicles, and can improve the leasing efficiency of the shared vehicles.

The invention discloses a frequency band resource allocation instruction method in a frequency division multiple access sending mode. The method comprises the steps of carrying out sub-channel division on transmission bandwidth, and instructing a method for scheduled frequency band location through two parts of bitmap instruction fields; generating a resource allocation instruction signal for each terminal after a sending node determines frequency band locations allocated to each piece of terminal data; and determining the frequency band locations for bearing the data by the terminal through reading of the resource allocation instruction signals, and obtaining data to carry out follow-up related operations. According to the method, all bits of the resource allocation signals are used for bitmap instructions, the resource allocation flexibility is greatly improved, the scheduling precision is improved, the signal overhead is reduced, the signal structure is simple, and the calculation complexity is low.

The invention discloses a dynamic resource allocation method and device under a heterogeneous wireless network. The dynamic resource allocation method comprises the steps of: periodically obtaining network state information of each access network in the heterogeneous wireless network; receiving a data request sent by each access network in the heterogeneous wireless network; determining a wireless network adapter factor between each data packet to be allocated and each access network according to the obtained network state information; and aiming at a plurality of data packets to be allocated in a service datum, determining an access network to which a (k-1)th datum is allocated according to an access network to which a kth datum is allocated when the value of an effect function of all former k+1 data is maximum, wherein k is a natural number. According to the invention, the performance difference of different wireless network systems and the dynamic load conditions are comprehensively considered, and the dynamic joint scheduling of wireless resources among the access networks is realized, thus the scheduling precision can be improved and the utilization rate of the wireless resources is increased.

The embodiment of the invention discloses a cloud computing resource scheduling method. The method comprises steps of a machine learning algorithm being adopted to train a resource scheduling model according to a test data set; wherein the test data set comprises feature information representing that the current resource of the application program meets an appropriate resource usage amount; inputting the feature information required by the current application service into the trained resource scheduling model, and obtaining an appropriate resource usage amount required by the application service; and scheduling according to the appropriate resource usage amount and the current resource usage amount of the current application service. According to the cloud computing resource scheduling method provided by the embodiment of the invention, the cloud computing resources can be automatically and reasonably allocated, and compared with a mode that manual scheduling or arrangement is needed in the prior art, the requirement on technicians is low, the time is saved, and scheduling accuracy is high.

The invention discloses a request task scheduling method and a control centre server, wherein the method comprises the steps of: receiving node information reported by a CDN node to be scheduled, andconstructing multiple training samples based on the node information; creating a support vector machine model, wherein the support vector machine model comprises the specified number of binary classifiers; and furthermore, the specified number is determined based on the total number of CDN nodes to be scheduled; training the support vector machine model for several times by utilizing the constructed multiple training samples, wherein a corresponding weak classifier is generated after training for each time; and furthermore, the weak classifier has a weight value; and, on the basis of the weight values of various weak classifiers, combining the various weak classifiers into the final classifier, and scheduling received new request tasks in the CDN node to be scheduled through the final classifier. By means of the technical scheme provided by the invention, the request task scheduling precision can be increased.

The invention relates to a resource data scheduling system and method. The method includes: building a data interface in data transmission with a resource platform, and acquiring multiple monitoring data from a resource manager of the resource platform through the data interface, wherein the monitoring data include resource using rate; calculating average value of the resource using rate in a preset time period; comparing the average value with a preset resource threshold value, if the average value is greater than or equal to the resource threshold value, detecting whether hardware resources and software resources in the resource platform are usable or not, if yes, executing mainframe migration scheduling and business application expanding on the resource platform, and if not, giving an alarm for prompting. The system can automatically trigger resource scheduling and managing according to resource using rate, execute mainframe migration scheduling and business application expanding, solve the problem of elastic expansion and contraction of business application more efficiently and quickly, distribute resources more reasonably and guarantee efficient and stable running of the business application in the resource platform.

The application relates to the technical field of storage and discloses a storage controller. The storage controller comprises a distribution core, a plurality of ranking cores and a request processing core; the three types of cores are used for distributing IO requests respectively to the different ranking cores, generating a processing ranking index for each IO request, and processing the IO requests according to magnitudes of processing ranking indexes of the IO requests so as to distribute the IO requests received by the storage controller flexibly.

Systems and methods for providing remote access to interactive media guidance applications are provided. In particular, the systems and methods for providing remote access to interactive media guidance applications provided herein generally relate to enhancing remote access capabilities by providing algorithms and techniques for managing and correcting disparities that may be found between program listings data used by the interactive media guidance application and the remote access server.

The invention provides an online car-hailing scheduling method and device, electronic equipment and a storage medium. The method comprises the steps of determining supply and demand actual data of each grid in an online car-hailing service area at a current scheduling moment; based on the supply and demand actual data of each grid at the current scheduling moment, determining a local area supply and demand state of each empty vehicle in each grid at the current scheduling moment; determining the scheduling action of each empty vehicle in each grid at the current scheduling moment based on the local area supply and demand state of each empty vehicle in each grid at the current scheduling moment; and determining an online car-hailing scheduling result of the online car-hailing service area based on the scheduling action of each empty car in each grid at the current scheduling moment. According to the method, the device, the electronic equipment and the storage medium provided by the invention, the scheduling accuracy is high, and the user travel efficiency and the user travel experience are improved.

The invention relates to an electric power system scheduling method based on homogenization of multiple types of power generation resources, and belongs to the technical field of power system operation optimization. According to the electric power system scheduling method based on homogenization of multiple types of power generation resources, homogeneous modeling is adopted, the controllability of various types of generator sets is described, the calculation complexity is reduced by adopting aggregation technology, and an electric power system scheduling model based on multi-type power generation resource homogenization is constructed by adopting an existing linear programming calculation method. On one hand, the controllability of the active power of various power generation resources isimproved to realize the optimal scheduling of the controllability of the active power of various different power generation resources; besides, by means of polymerization technology, the scheduling complexity of an electric power system comprising multiple types of power generation resources is greatly reduced, and meanwhile high scheduling precision is kept. By applying the method disclosed in the invention, the controllability of various power generation resources including renewable energy sources can be fully developed, the uncertainty of multiple types of power generation resources in power system scheduling is reduced, and the flexible controllability of the power generation resources is improved.

The invention provides a parallel simulation-based urban traffic and transportation support system. A simulation system running parallel to a traffic and transportation guarantee system is built; through interconnection and information interaction with the actual system, the simulation system continuously obtains the latest transportation condition from the traffic and transportation guarantee system, builds entity simulation models, performs transportation condition analysis and scheme optimization through super-real-time simulation operation of the models and feeds results back to the urban traffic and transportation guarantee system; guarantee schemes can be generated continuously and circularly and expected effects of the schemes can be deduced and estimated. The system greatly shortens transportation dispatching time, greatly improves the dispatching accuracy and provides a new thought for building a parallel simulation system for decision support.

A resource management device of embedded real time operation system consists of core host, internal memory supervisor, file supervisor, device supervisor, network communication unit and application module. It is featured as using message mode to call system function between each unit and supervisor to carry out distribution and management on information resource i.e. confirming task property by initializing task and carrying out management and distribution treatment separately according to custom task or no custom task then carrying out dispatch operation.

The invention relates to a distributed photovoltaic power generation management and settlement method based on an electric power spot environment, and the method is used for the execution of dispatching equipment assisting the power dispatching of a power grid in the electric power spot environment, and comprises the steps: building a power dispatching function of the power grid according to the power dispatching parameters of the power grid and the power generation information of a photovoltaic power generation system; solving the power dispatching function to obtain an electric quantity dispatching optimal solution of the photovoltaic power generation system; and acquiring an electric quantity dispatching optimal scheme of the power grid according to the electric quantity dispatching optimal solution, sending the electric quantity dispatching optimal solution to a dispatching control system, and performing dispatching control on power generation and energy storage work of the photovoltaic power generation system and electric quantity clearing work of a clearing system of the power grid by the dispatching control system according to the electric quantity dispatching optimal scheme. According to the invention, the electric quantity of the distributed photovoltaic power generation system can be accurately scheduled and stored, and the lacked electric quantity can be dispatchingfrom the power grid when the power generation quantity and the energy storage quantity of the photovoltaic power generation system are less than the demand quantity, so that the actual application demand is met.

The embodiment of the invention provides a container creation scheduling method and device, a computer readable medium and electronic equipment. The container creation scheduling method comprises thesteps of obtaining information of a target mirror image layer of a to-be-created container in a cluster; determining statistical information of each node in the cluster about the target mirror image layer according to the information of the target mirror image layer; and determining a target node according to the statistical information of each node in the cluster about the target mirror image layer, wherein the target node is used for creating the to-be-created container. According to the technical scheme of the embodiment of the invention, the scheduling accuracy and the reasonable utilization of cluster resources are improved.

The invention belongs to the field of power distribution networks, and provides a multi-time-scale active and reactive coordinated optimization scheduling method and system for an active power distribution network. The method comprises the following steps: in a pre-stage of a set time interval, describing the uncertainty of renewable energy sources based on a typical scene, and optimizing the start-stop state and the output state of each adjustment resource in the active power distribution network; in a stage within a set time interval, according to an optimization result of each adjustment resource in the active power distribution network before the set time interval and an updated power prediction value within the set time interval, using a multi-correlation model to estimate a power prediction error so as to correct the power prediction value within the set time interval, and coordinating fast response resources in the active power distribution network based on the corrected power predicted value within the set time interval; and in the real-time stage, performing rolling optimization on output of quick response resources in the active power distribution network by adopting model prediction control. The adaptability of the active power distribution network to renewable energy sources is improved.

An embodiment of the invention discloses a method and a device for multiplying a matrix and a vector, the method comprises the following steps of: obtaining first indication information of a matrix element; reading a matrix element value of a non-zero element from a preset matrix according to the first indication information, and determining a first positional mark of the read matrix element value; acquiring second indication information of a vector element; reading a vector element value of a second positional mark corresponding to the first positional mark from input vector data according tothe second indication information; and obtaining a multiplication value of the matrix element value and the vector element value. By adopting the embodiment of the invention, the complexity of data processing can be reduced, the power consumption of data processing can be reduced, and the data processing efficiency can be improved.

The invention provides a power distribution method and system for a wind power cluster. The method comprises the following steps that: based on the grid-level dispatching instruction, the optimal windpower output of the wind farm cluster layer is determined by utilizing the preset active power coordinated distribution strategy; Based on the optimal wind power output of wind farm cluster layer, the optimal wind power output of sub-field layer is determined by using the pre-determined active power coordinated distribution strategy. Based on the optimal wind power output of the sub-field layer,the operation of the wind turbine control terminal in the unit layer is controlled. The active power output hierarchical distribution method of the invention combines the hierarchical thought of the large-scale system and considers the coordination and cooperation between different wind farms in the same area at the level of the wind power cluster, further realizes the relative minimization of thesystem network loss and improves the wind power consumption capacity at the same time.

The invention discloses an automatic driving vehicle scheduling method, a vehicle scheduling server and an automatic driving vehicle, and belongs to the technical field of computers. The method includes: receiving a ride request; determining at least one first candidate vehicle among multiple autonomous vehicles according to the ride request and the driving information of each autonomous vehicle within the management range; and determining at least one first candidate vehicle according to the current position information of each first candidate vehicle , current road condition information and planned route information, calculate the first time required for each first candidate vehicle to reach the destination; according to the first time corresponding to each first candidate vehicle, among at least one first candidate vehicle , determine the final candidate vehicle, and the final candidate vehicle at least meets the expected arrival time at the destination; send the ride request to the final candidate vehicle. The present invention at least performs vehicle scheduling according to the ride request and the current location information, current road condition information, and planned route information of each self-driving vehicle, and the scheduling accuracy rate is high.

The invention discloses a power headroom reporting method, which is used for terminal equipment and is characterized in that N reported PUSCHs (Physical Uplink Shared Channel) are included in a time unit, and N is greater than or equal to 2; wherein each reported PUSCH is a PUSCH used for sending a power headroom report; wherein the power headroom report comprises power headroom values of M carriers, the time unit is a time unit of a timer triggering power headroom report, N is greater than or equal to 2, and M is greater than or equal to 2. The invention further comprises terminal equipment using the method. The transmission efficiency of PHR can be improved, and the scheduling accuracy of the network equipment is improved.

The invention discloses an online second-hand vehicle source elastic scheduling method and system and a scheduling server. According to the embodiment of the invention, according to the SKU vehicle condition, the SKU display board price and the local SKU vehicle condition of the same type, the price of the local SKU display board of the same type, the ranking sequence of the SKUs is determined, the positions of the SKU units scheduled on a user client page are determined jointly through various data, the scheduling accuracy is greatly improved, the SKU units with high cost performance are fully displayed, and a second-hand vehicle source database is effectively utilized. Meanwhile, the price of the SKU display board is calculated; the SKU display board price can guide a merchant to provideSKU display board price debugging for the merchant, the method can effectively guide the merchant to adjust the SKU display board price to a proper price, and finally play a role in scheduling the SKU unit at the page position of the client, and the second-hand vehicle source database is further effectively utilized.

The invention provides a method and a device for realizing multi-core scheduling of a virtual machine. The method comprises the following steps: correspondingly configuring a main virtual machine of the virtual machine and N1 shadow virtual machines to N CPU cores; scheduling the main virtual machine and the shadow virtual machine to run on the corresponding CPU cores in the same time window through a time scheduling table, wherein the time scheduling table is configured with a time window for scheduling the operation of the main virtual machine and the shadow virtual machine. Based on the method, the CPU resource utilization rate and the real-time performance of the virtual machine can be better coordinated.

The invention discloses a service elastic expansion and contraction method based on emergency detection. The method comprises the following steps: taking an obtained service state as a work cycle; obtaining work load data before the service, and inputting the work load data into the long-term prediction model to carry out multi-step probability prediction to obtain prediction results of a plurality of time points in the future; calculating a multi-step deviation vector; updating a confidence interval upper bound value of the multi-step probability prediction result; if at least two values in the multi-step deviation vector are greater than a preset threshold value, determining that the current state is an emergency; otherwise, the emergency situation is not present; generating a trend prediction result; calculating the number of instances required by the service at the moment according to the preset workload upper limit of a single instance; obtaining the workload data of the previous service from the time sequence database, and carrying out single-step prediction to obtain a single-step prediction result; obtaining a future instance number of the service based on a base point offset method; and scheduling is carried out, so that a work cycle is completed.

The invention provides a batch processing machine scheduling method of dynamically-changed weight. The method comprises steps of generating a trigger event, determining a scheduled object, calculating the dynamically-changed weight, obtaining a product family of the top priority, distributing work and loads to an idle available batch processing machine, and determining a scheduling termination condition. For the scheduling object of minimized total weighted completion time, to-be-processed products in a buffer of the idle available batch processing machine are taken as research objects. The dynamically-changed weight is calculated according to the dynamic change condition of the number of the to-be-processed products of different product families, and the like. The product family of the top priority under the object of the minimized total weighted completion time is determined according to the dynamically-changed weight. Batching is performed according to the "at full batch principle". A first batch which is ordered according to the "first come first serve principle" is allocated with work and loads. Rolling cyclic execution is performed until the scheduling termination condition is met. The method is convenient to achieve, high in scheduling precision, and beneficial for popularization and application of industries such as a semiconductor industry adopting batch processing machines.

In an embodiment, a processor may be configured to dynamically infer one or more attributes of input and / or output registers of an instruction, given the attributes corresponding to at least one input registers. The inference may be made at the issue circuit / stage of the processor, for those registers that do not have attribute information at the issue circuit / stage. In an embodiment, the processor may also include a register attribute tracker configured to track attributes of registers prior to the issue stage of the processor pipeline. The processor may feed back, to the register attribute tracker, inferred attributes and the register addresses of the registers to which the inferred attributes apply. The register attribute tracker may be configured to may associate the inferred attribute with the identified register attribute tracker may also be configured to infer input register attributes from other input register attributes.

The invention provides a service scheduling method, device and system, and a quality monitoring system, and relates to the technical field of CDN scheduling, and the method comprises the steps: analyzing the service quality of a terminal, so as to determine whether the service quality of the terminal is lower than the preset service quality or not; under the condition that the service quality of the terminal is lower than the preset service quality, determining a media server for providing service for the terminal; determining whether the media server fails or not according to the service quality of a plurality of terminals provided with services by the media server; and under the condition that the media server fails, sending a first notification message carrying the identifier of the media server to a CDN scheduling server, so that the CDN scheduling server does not distribute a newly initiated access request to the media server.