32 results about "Workload prediction" patented technology

The disclosed embodiments provide a system that configures a graphics-processing unit (GPU) in a computer system. During operation, the system predicts an incoming workload to the GPU. Next, the system identifies an operational floor for the GPU based on the incoming workload. Finally, the system uses the operational floor to configure the subsequent execution of the GPU, wherein the operational floor facilitates processing of the incoming workload by the GPU.

The disclosed embodiments disclose techniques for optimizing emergency department resource management. During operation, a system receives a set of parameters that is associated with a patient entering an emergency department. The system analyzes the set of parameters in a machine learning module to determine (1) a calculated acuity score that indicates an estimated severity of illness for the patient and (2) a set of workload predictions that predict a set of resources that will be needed to treat the patient in the emergency department. The system then uses the acuity score and the workload predictions to assign a set of predicted tasks that are associated with treating the patient into the work queues of the emergency department.

The invention discloses a method for predicting the software workload of a newly-added software project, belonging to the technical field of development of computer software. The method comprises the following steps of: discretizing the workload of history projects and dividing the history projects into project classes in designated number; calculating the condition probability and the priori probability of each project attribution in the classes of the project workload by using workload attribute data of the history projects; establishing a Bayes classification model and predicting the workload class of the newly-added project; adding the newly-added project subjected to workload classification prediction into history project data, repairing missing date, recalculating the condition probability of the project attribute on the project workload class and recalculating the priori probability of the project workload class, and repeatedly iterating until all probability distributions are converged; and finally predicting the workload of the newly-added software project by using the converged posterior probability distribution. Compared with the prior art, by using the method, the capability of a mode for predicting the workload of the software project is greatly improved.

The invention discloses a controller workload prediction method and system. The controller workload prediction method comprises the steps of building a multiple regression model according to controller workload prediction related indexes and controller workload prediction sample data; importing air traffic flow situation index prediction data in a controlled sector into the multiple regression model, and obtaining a controller workload prediction result. According to the controller workload prediction method and system, the air traffic flow situation multidimensional indexes which influence controller workload are considered completely and comprehensively, and thereby effective prediction for controller workload is achieved. The designed controller workload prediction system can be applied to an engineering unit and has great operability.

The invention relates to an HEVC-based DVFS control method and system, a processor and a storage device. The method comprises the following steps: performing decoding in cooperation with HEVC based on a CPU and a GPU in the manner of a streamline, performing entropy decoding on a binary bit stream input by a decoder, assigning decoder parameters, estimating the number of TU of a video frame i, analyzing an inverse transformation coefficient matrix, and inputting the inverse transformation coefficient matrix to a workload prediction module; on each synchronous node of the CPU and the GPU, predicting the working frequency of the CPU and the working frequency of the GPU by the workload prediction module based on the number of TU of the video frame i estimated by the entropy decoding, and inputting the working frequency of the CPU and the working frequency of the GPU to a frequency manner; adjusting the set working frequency of the CPU and the set working frequency of the GPU according to the number of the decoded video frames in a frame cache, and inputting the set working frequency of the CPU and the set working frequency of the GPU to a regulator; and setting the frequency of the CPU according to the final working frequency of the CPU by the regulator, and setting the frequency of the GPU according to the final working frequency of the GPU, and continuing to decode by the CPU and the GPU at the set frequency.

The invention discloses a controller workload prediction method and system. The method comprises a first step of determining a back propagation (BP) neutral network topology structure, collecting sector traffic flow state index samples of different times, and building a sample set; a second step of optimizing a weight and a threshold of a BP neutral network by use of a genetic algorithm, performing network training, and outputting an optimized BP neutral network; a third step of predicting the index of the controller workload through the optimized BP neutral network of the second step according to the input real-time data of the sector traffic flow state index; and a fourth step of responding to a warning if a result of predicted index of the controller workload meets a preset condition. Through adoption of the method, the reliability of the prediction result of the controller workload can be improved, actual demands of an air traffic control unit on real-time prediction and warning for the controller workload can be met, and the method can provide data support for improvement of operation management level and optimization of the controlled airspace structure.

Methods, systems, and apparatuses for automated driving or for assisting a driver and for determining driver workload prediction as a function of a driving route or path are disclosed. A system includes a predicted workload component, a route component, and a notification component. The predicted workload component is configured to determine that at least one section of a current route comprises a high driver workload. The route component is configured to modify the current route to generate an alternate route, wherein the alternate route avoids the at least one section that comprises a high driver workload. The notification component is configured to provide the alternate route to a driver or automated driving system.

The invention provides a line production line workload prediction method based on a production cycle. The method comprises the steps of forming an in-process product list information set; forming a virtual workpiece list information set; calculating a production periodic table according to the historical workpiece information; integrating the in-process product list information and the virtual workpiece list information as a set of elements to obtain a simulated workpiece list information set; obtaining a complete process flow corresponding to each piece of simulated workpiece list informationaccording to the production periodic table; calculating the estimated arrival time, the estimated start time and the estimated end time of each step of each simulation workpiece; and obtaining a prediction set of the prediction information of each simulation workpiece to form a statistical report. According to the characteristics and known data of the flow production line, the prediction method is utilized to quickly realize workload prediction of the whole production line in a period of time in the future, detailed prediction data are reserved, analysis of each dimension is performed on thedata, and the regulation and control purposes of production regulation and control, productivity early warning and the like are achieved in combination with the condition of the actual production line.

The invention discloses a working platform workload prediction method and system based on load prediction, and relates to the technical field of computer software. The method comprises the following steps: initializing a known data source calculation topology data set to train an optimized SOM network; obtaining to-be-allocated tasks in the task pool and grouping the to-be-allocated tasks according to the workload; enabling the SOM network model to acquire a training neural network model of each cluster by using the workload characteristics of each cluster in the neural network learning task pool; when a user publishes a new task, enabling the SOM network model to firstly obtain an initial workload of the user and determine a cluster to which the user belongs according to the initial workload of the user; and predicting the workload of the new task by utilizing the training neural network model of the cluster to which the new task belongs. The SOM network is trained and optimized through the known data, the training neural network model of each cluster is obtained according to the load characteristics of the clusters to complete prediction of the workload of the new task, pricing of the task is facilitated, the task is reasonably allocated, and the working efficiency of employees is improved.

The invention discloses a new-drilled well workload prediction method based on ensemble learning. The method is characterized in that a new-drilled well workload prediction model based on a random forest is built, a key hyper-parameter combination of the new-drilled well workload prediction model is optimized through a particle swarm optimization method, and a weighted voting mechanism is added ina decision stage; by adjusting the weight value of the weak classification decision tree, the generalization error of new-drilled well workload prediction is reduced, and the precision of the new-drilled well workload prediction model is improved. The key hyper-parameter combination in the new-drilled well workload prediction model is optimized by adopting the particle swarm optimization method,the influence of noise data in oil reservoir development historical data is reduced, and the stability and the operation speed of the random forest method are improved; a weighted voting mechanism isadded in the decision stage, wherein the weight proportion of a high-score decision tree is increased and meawhile the negative influence of a low-score decision tree on a prediction result is reduced, and thereby the precision of a new-drilled well workload prediction model is improved.

The invention discloses a working platform task workload assessment method based on big data, and relates to the technical field of working platform management. According to the method, a working platform decomposes a task, searches a matched packet receiving party from a talent pool of the working platform according to skill requirements of each sub-task, and evaluates the working content of the task received by the packet receiving party, and the evaluation is classified into content evaluation, complexity evaluation and similarity evaluation. The method is convenient to operate and use and simple to operate, the working efficiency of working platform management can be effectively improved, the original data stored in the working platform is compared with the evaluated working content, the working content influenced by various factors is calculated through comparison, and the workload is reanalyzed. The accuracy of workload evaluation of a working platform is improved, the work task is decomposed into a plurality of sub-tasks, it is avoided that due to too little classification, task classification is inaccurate, and later workload prediction is affected, and the prediction accuracy is improved.

The invention provides an oil and gas field asset evaluation method and device, and relates to the technical field of oil and gas exploration and development, and the method comprises the steps: obtaining oil and gas field basic parameters and bid invitation contract evaluation parameters; according to the recovery ratio data, the recovery degree data, the storage and production ratio data and the geological reserve data, determining a peak period yield and peak period stable yield period prediction model; determining a yield and workload prediction model according to the storage and production ratio data, the oil production speed data, the decline stage yield data, the yield increase yield data and the single well productivity data; determining a well group optimization prediction model according to the recovery ratio data, the reservoir effective thickness data, the well spacing data and the permeability data; and generating an oil and gas field asset evaluation result according to the peak period yield and peak period stable production period prediction model, the yield and workload prediction model and the well group optimization prediction model. According to the method, the prediction speed is optimized, and a more accurate prediction result is obtained based on limited parameters.

The invention discloses an inspection robot work adjustment method based on a regression analysis algorithm, and the method comprises the following steps: S1, obtaining an inspection route of an inspection robot, and recording the positions of charging piles and robot shutdown rooms distributed on the inspection route; S2, establishing a battery loss degree mapping table according to charging of the charging pile and charging of the shutdown room; s3, obtaining the working state of the inspection robot and the historical data of the corresponding battery power use characteristics, and establishing a workload prediction model by adopting a regression analysis prediction algorithm; s4, acquiring a current inspection task of the inspection robot, judging whether the battery capacity of the inspection robot meets the requirement of the current inspection task or not according to the workload prediction model, if so, entering step S6, and otherwise, entering step S5. According to the inspection robot, in the inspection process, the electric quantity can fully meet the inspection requirement, meanwhile, charging is reasonably conducted, the battery is protected, damage to the battery is reduced, and the service life of the inspection robot is prolonged.

The invention relates to a staged workload prediction method and system based on composite variable characteristics, and the method comprises the steps: building a test workload prediction model and a staged workload prediction model, and firstly, obtaining the total prediction workload through the test workload prediction model; the total prediction workload is used as input, a staged workload prediction model is used for calculation to obtain the predicted staged workload, and aiming at the problem of how to evaluate the project test workload, on the basis of fully considering various influence factors, a linear regression algorithm and a variational self-encoding VAE algorithm are used for evaluating the project test workload. And predicting the overall test workload of the project and the workload of each stage of the project test work so as to achieve the purpose of accurately predicting the project test workload.