73 results about "Performance results" patented technology

The invention relates to the technical field of smart factory decision-making, and discloses a digital twin-driven smart factory AI intelligent decision-making system, which comprises a data acquisition and preprocessing module used for acquiring and preprocessing real-time and historical data of a production line; the twinborn modeling identification module is used for constructing a twinborn body and carrying out parameter identification and uncertainty quantification; the alignment evaluation credible module is used for comparing twin prediction data with real data and generating availability marks and trust scores; the strategy simulation optimization module is used for generating candidate strategies and risk evidences based on the constraints and the key performance indicators; the grayscale online publishing module is used for screening and grayscale publishing a strategy according to the trust score and the performance result; the operation evaluation auditing module is used for collecting operation data and generating an auditing packet; and the drifting detection updating module is used for detecting distribution drifting and recalibrating and updating the twinborn body and the strategy library. According to the invention, the reliability, robustness and continuous optimization of the production decision of the smart factory are realized.

The invention discloses an intelligent production and quality monitoring system and method for premixed concrete, and particularly relates to the technical field of concrete production quality monitoring, and the method comprises the following steps: constructing an initial control parameter set; environment data is collected and acts on the initial control parameter set to generate an applicable control parameter set; performing mixing production based on the applicable control parameter set, collecting process behavior data, and inputting the process behavior data into a machine learning model to predict compressive strength and setting time; comparing the prediction result with a performance target, if the tolerance is exceeded, determining that the quality is low and correcting control parameters, and if the tolerance is not exceeded, determining that the quality is high and allowing discharging; according to the method, the initial control parameter set is constructed by using the historical weighing deviation and the strength actual measurement result, the applicable control parameter set is dynamically generated in combination with the real-time environment data, the performance result is predicted through the model and compared with the preset target, and a quality judgment and parameter correction closed loop is formed. And the parameter rationality, the environmental adaptability and the quality control intelligent level of concrete production are improved.

The invention discloses a task scheduling method and device, a storage medium and electronic equipment, and relates to the field of cloud computing. The method comprises the steps that a task model is constructed according to a target training task, and the task model is used for representing attributes and resource requirements of the target training task; on the basis of the task model, predicting performance results when the target training task is executed on the N computational nodes to obtain N performance results, N being an integer greater than 1, and the performance results comprising execution efficiency and resource utilization rate of the target training task on the computational nodes; filtering the N computing nodes according to the N performance results and the network environment information to obtain S first computing nodes, and determining a target node from the S first computing nodes; and executing the target training task through the target node. The technical problem of low deep learning training task scheduling efficiency in the prior art is solved.

The invention relates to the technical field of data center resource scheduling, and discloses an intelligent data center optimization method based on distributed GPU computing power scheduling. According to the method, historical scheduling records are quantitatively evaluated, historical strategies are distributed to experience groups of different levels according to scheduling effect indexes and strategy trust levels, and a structured strategy library is formed. And for a target performance level defined in the dynamic behavior model of the data center, calculating a serialized scheduling scheme for achieving a target, and selecting candidate strategies from the structured strategy library. And carrying out adaptive correction on the serialized scheduling scheme and the candidate strategy by adopting a fusion decision-making method to generate a final scheduling strategy. And executing the final scheduling strategy, generating a control command in combination with the real-time resource data to implement resource adjustment, and recording an actual performance result. By comparing the deviation between the actual performance result and the target performance level, the deviation source is diagnosed, the experience grouping content is revised, and the structured strategy library is optimized.

An axial flow compressor rapid prediction method based on self-adaptive correction comprises the following steps that a one-dimensional performance rapid prediction model of an axial flow compressor is constructed based on compressor geometric parameters and a one-dimensional average pitch diameter theory; based on a one-dimensional performance rapid prediction model, taking a relative error between a calculation result and experimental performance as an optimization target, introducing a correction factor into a lag angle and total pressure loss empirical model as an optimization parameter, and carrying out optimization of the correction factor; establishing a mapping relation among the compressor inlet speed coefficient, the inlet relative Mach number and the optimal correction factor under different working conditions; based on an existing compressor experimental performance result, a correction factor optimization method is adopted, a mapping relation between the multiple compressor inlet working condition parameters and the optimal correction factor is constructed, and a data set is formed; a BP neural network is utilized to train a data set, so that the data set has generalization ability; and calling the trained BP neural network in the one-dimensional performance rapid prediction model to realize adaptive correction of the empirical model and rapid prediction of the performance.

The invention discloses a pavement anti-slide performance prediction method and system based on hyperspectral image texture features. The method comprises the following steps: acquiring an airborne hyperspectral image of a pavement; performing wavelet transformation on each wave band of the hyperspectral image to obtain a high-frequency component, calculating Shannon entropy and energy on the high-frequency component, and further constructing a wave band score by using the Shannon entropy and the energy to screen out a spectral characteristic wave band reflecting road surface characteristics; screening texture parameters based on the texture features of the spectral feature wavebands to form a texture parameter data set; and finally, inputting the texture parameter data set into the CGS-XGBoost model for training, and predicting anti-skid performance results of other unknown pavements. According to the technical scheme, the limitation of existing evaluation indexes on road surface texture characterization is expanded, the anti-skid performance is predicted from the perspective of macroscopic image texture, the defect that the measurement area of a traditional method is small is overcome, and the accuracy and universality of road surface anti-skid performance prediction are remarkably improved.

The invention relates to a talent data analysis method and device based on multi-dimensional data fusion and electronic equipment, and the method comprises the steps: carrying out the clustering of multi-dimensional talent data, obtaining a data set, extracting target data from the data set, and obtaining the performance result of the target data; generating an internal driving force result of the target data, performing first-order motivation correlation analysis based on the performance result and the internal driving force result to obtain a first-order motivation analysis result, and generating a second-order motivation analysis result corresponding to the target data according to the performance result and the performance level average score; and generating an evaluation result of the multi-dimensional talent data based on the first-order motivation analysis result and the second-order motivation analysis result. Quantitative analysis is carried out through value output withdrawn from a first-order motivation analysis result and value change withdrawn from a second-order motivation analysis result, a complete quantitative evaluation model is provided, multi-dimensional and multi-level analysis can be carried out on talent data on the whole, and the limitation of talent data analysis in the prior art is avoided.

To provide a gaming machine that can enhance the excitement of pre-announcement sequences. [Solution] A gaming machine that allows a player to select an option from a set of selectable options, and in which case a variation performance can be executed in a manner corresponding to the selected option, wherein the options include a pre-notification performance that notifies the performance content to be executed before the performance result of the variation performance is displayed, and if the pre-notification performance is selected, a performance content pre-notification performance can be executed that notifies one or more performance contents from the performance content to be executed.

To provide a game machine capable of improving interest in a game.SOLUTION: A plurality of kinds of expectation degree performances suggesting an execution expectation degree of a special game can be executed during a variation performance, an emphasis performance emphasizing the expectation degree performance can be executed, and the number of kinds of the emphasis performances executed during the variation performance resulting in a winning result in a first game state tends to be larger than the number of kinds of the emphasis performances executed during the variation performance resulting in a losing result in the first game state.SELECTED DRAWING: Figure 132

Provided are a logistics robot simulating system and method, the system comprising: a performance unit for performing product production simulation using logistics robots; and an operation control unit, which determines a reference operation quantity of logistics robots on the basis of performance results of the product production simulation performed by the performance unit on the basis of a preset initial operation quantity of logistics robots, changes the reference operation quantity according to a preset change rule, and determines an optimal operation quantity of logistics robots on the basis of additional performance results of the product production simulation additionally performed by the performance unit on the basis of the changed reference operation quantity.

The invention discloses a load testing method and device, equipment, a storage medium and a product. A hardware channel of a PCIe dummy load card is tested to obtain a hardware test result, and the PCIe dummy load card refers to a card for simulating heating values under different loads during a heat dissipation capability test of a liquid cooling system based on a PCIe interface; testing the control instruction of the PCIe dummy load card to obtain an instruction test result; testing the heating parameter of the PCIe dummy load card to obtain a heating test result; and determining whether the performance of the PCIe dummy load card reaches the standard or not according to the hardware test result, the instruction test result and the heating test result, and obtaining a performance result. According to the method, the PCIe dummy load card is adopted to simulate the PCIe card to generate heat to complete the performance test, and a more accurate dummy load card performance result is obtained through the three tests, so that more accurate and effective PCIe heating data is obtained, and the development speed of the heat dissipation capability of the liquid cooling system is increased.

The invention discloses a slurry formula performance prediction and grade optimization method for horizontal directional drilling. The method comprises the following steps: S1, slurry experiment and data acquisition based on an HDD process; s2, establishing an HDD slurry viscosity nonlinear regression model; s3, defining a slurry performance grade for HDD construction; s4, performing Monte Carlo simulation and formula space global performance scanning; s5, visualizing the result of the step S4; S6, performing HDD field formula optimization and decision making; the method provides a standardized process and decision basis for slurry preparation. The initiative'performance grade distribution cloud picture 'enables an engineer to see all possible performance results of the formula clearly at one eye, and blind trial and error are thoroughly avoided. The lowest-cost formula meeting the requirement can be quickly locked, and the material cost is saved; meanwhile, field debugging time is shortened, and the project progress is accelerated.

The invention relates to the technical field of transmission gear shifting testing, and discloses a transmission static gear selecting and shifting performance testing device and a testing method thereof.The transmission static gear selecting and shifting performance testing device comprises a gear selecting assembly, a gear shifting assembly, a detecting assembly and an adjusting assembly.The gear selecting assembly comprises a gear selecting mechanical arm and a first motor, and the gear selecting mechanical arm is fixedly connected with a rotating shaft of the first motor in a sleeving mode; the gear shifting assembly comprises a gear shifting mechanical arm and a second motor, the gear shifting mechanical arm comprises two connecting ends, one end of the gear shifting mechanical arm is fixedly connected with a rotating shaft of the second motor in a sleeved mode, and the other end of the gear shifting mechanical arm is fixedly connected with a gear shifting shaft in the transmission in a sleeved mode. The detection assembly comprises a first encoder, a second encoder, a first force sensor and a second force sensor, the first force sensor and the second force sensor are connected with the gear selecting mechanical arm and the gear shifting mechanical arm respectively, and the adjusting assembly comprises two adjusting seats which are connected with the gear selecting mechanical arm and the gear shifting mechanical arm respectively. The static gear selecting and shifting performance testing device is simple in structure, high in adaptability and low in cost, and comprehensive and objective static gear selecting and shifting performance results of the transmission can be obtained.

The invention discloses a performance accounting system and method based on remote office, and belongs to the technical field of remote office systems. The system comprises five core modules: an intelligent work baseline setting module, a multi-dimensional output tracking module, a real-time cooperation value accounting module, an elastic performance calibration module, and a visual performance board and growth module. The functions of the modules are as follows: the intelligent working baseline setting module generates a dynamic performance baseline and provides an accounting basic target; the multi-dimensional output tracking module collects and quantifies employee output data and calculates an output efficiency index; the real-time collaboration value accounting module splits the collaboration chain value and calculates individual collaboration quality indexes; the elastic performance calibration module dynamically adjusts a calibration rule and outputs an environment correction coefficient; and the visual performance board and growth module records loop data, and outputs a final performance result and attribution analysis. The method has the advantages that closed-loop linkage of modules can be realized, dynamic threshold judgment is realized, and remote office scenes can be accurately adapted.

Investment portfolios are evaluated based on thematic concepts. Natural language processing may be employed to determine thematic mentions of concepts in publications. Thematic scores associated with the organizations may be generated based on the thematic mentions and publication weights. Thematic models may be determined based on the organizations and performance outcomes such that each thematic model may predict a performance outcome based on the thematic scores associated with the organizations. Thematic models may be employed to predict the predicted outcomes for the organizations based on the thematic scores such that the predicted outcomes may be displayed in a report.

The application discloses a co-entropy ceramic protective coating sintering temperature and aerodynamic thermal load cooperative control system and relates to the technical field of sintering temperature control. The system comprises an information acquisition module: acquiring previous sintering information, wherein the previous sintering information comprises sintering temperature and performance results, and the performance results are the performance of aerodynamic thermal load. The application considers coating material and structure information and environmental information to formulate a reference directory under different basic information in the whole process, and the sintering temperature is regulated according to the aerodynamic thermal load of the coating material structure and environmental manufacturing demand, and then the density, interface bonding and residual stress distribution of the coating are realized through the controllable sintering temperature curve at the preparation end, so that the aerodynamic thermal load of the coating in use is ensured, the sintering temperature is designed and optimized, the response characteristics of the coating to the aerodynamic thermal load in use are within the user demand, and the closed-loop design of optimizing the response of the service end by controlling the manufacturing end is embodied.

The application discloses a kind of performance result optimization method and system of deep learning model training, method includes: in response to the end of deep learning model training, the performance result index of deep learning model training is acquired and the debugging parameter of deep learning model training is acquired;Search to debugging parameter and from the performance result index corresponding to debugging parameter in search result, the performance result index corresponding to current search is compared with the performance result index corresponding to last search, wherein the performance result index corresponding to last search includes the performance result index of deep learning model training;In response to the comparison result meets preset condition, stop searching to debugging parameter and determine the final parameter of debugging parameter;Based on each final parameter determined, deep learning model training is carried out again to obtain the performance result after optimization.Through the scheme of the application, performance evaluation cycle is shortened, and performance evaluation efficiency is improved.

The present disclosure provides a performance monitoring method and device, electronic equipment and storage medium, and relates to the technical field of information security. The method comprises: adding performance monitoring annotations and scene annotations to a business interface in a to-be-monitored scene; obtaining object granularity performance data of the business interface for each business initiation object based on the performance monitoring annotations during processing of a business request by the business interface; and performing statistics on the object granularity performance data of the business interface based on the scene annotations to obtain scene performance result values for each business initiation object in the to-be-monitored scene. The method can distinguish and monitor the performance of the business interface in processing the business request according to the granularity of the business initiation object, and can also automatically statistically integrate the performance data of all business interfaces in a to-be-monitored scene, thereby realizing a flexible scene monitoring scheme.

This application discloses a task scheduling method, apparatus, storage medium, and electronic device, relating to the field of cloud computing. The method includes: constructing a task model based on a target training task, wherein the task model characterizes the attributes and resource requirements of the target training task; predicting the performance results of the target training task executed on N computing nodes based on the task model, obtaining N performance results, where N is an integer greater than 1, and the performance results include the execution efficiency and resource utilization of the target training task on the computing nodes; filtering the N computing nodes according to the N performance results and network environment information to obtain S first computing nodes, and determining a target node from the S first computing nodes; and executing the target training task through the target node. This application solves the technical problem of low scheduling efficiency for deep learning training tasks in the prior art.

Provided are a method and apparatus for monitoring a model in beam management by using artificial intelligence and machine learning. The method may include: in relation to a reference signal configured for a terminal, receiving second reference signal resource set configuration information of the reference signal for monitoring an AI / ML model; on the basis of the second reference signal resource set configuration information, measuring signal strength or signal quality for the reference signal; and reporting the performance result of the AI / ML model by comparing a measured value of the reference signal with a predicted value of the reference signal inferred via the AI / ML model.

The application discloses a chip performance analysis method, device and equipment and a storage medium. The method comprises the following steps: obtaining input data of a chip to be analyzed through a parameter analysis layer, analyzing the input data, and obtaining analysis parameters and timing data; performing performance index calculation according to the analysis parameters and the timing data through an operation layer, and obtaining a calculation result; and generating a performance analysis report of the chip to be analyzed according to the analysis parameters and the calculation result through a report layer. The input data is collected and analyzed through the parameter analysis layer, the output analysis parameters and the timing data are standardized, and standardized and reliable original data support is provided for performance calculation. The operation layer combines the configuration parameters and the timing data, outputs two types of performance results of a full window and a slice, considers the overall and segmented performance analysis of the chip, and accurately locates performance fluctuation. The report layer integrates various parameters and calculation data, automatically generates a complete performance analysis report, realizes regular presentation of analysis results, and facilitates reference for checking, tracing and optimization.

The application relates to the technical field of intelligent factory decision-making, and discloses a digital-twin-driven intelligent factory AI intelligent decision-making system, which comprises a data acquisition and preprocessing module, a twin modeling and identification module, an alignment evaluation and credibility module, a strategy simulation and optimization module, a gray online release module and a running evaluation and audit module.The data acquisition and preprocessing module is used for acquiring real-time and historical data of a production line and performing preprocessing; the twin modeling and identification module is used for constructing a twin body and performing parameter identification and uncertainty quantification; the alignment evaluation and credibility module is used for comparing twin body prediction data with real data, generating availability labels and trust scores; the strategy simulation and optimization module is used for generating candidate strategies and risk evidence based on constraints and key performance indicators; the gray online release module is used for screening and releasing strategies in gray according to the trust scores and performance results; the running evaluation and audit module is used for collecting running data and generating an audit package; and the drift detection and update module is used for detecting distribution drift and recalibrating and updating the twin body and the strategy library.The application realizes the reliability, robustness and continuous optimization of intelligent factory production decision-making.

The invention relates to an electric power large model pruning method and system based on side computing power equipment. The method comprises the following steps: acquiring multi-source operation data and hardware parameter data of side equipment acquired by the side equipment; performing time domain and frequency domain feature extraction on the multi-source operation data to obtain a time domain mutation feature and a spatial correlation feature, and determining a pruning threshold according to the time domain mutation feature and the spatial correlation feature; performing performance analysis on the hardware parameter data to obtain a parallel performance result and a data processing result; performing pruning optimization on the large electric power model based on the pruning threshold to obtain a pruned preliminary model, and performing structure adjustment on the preliminary model according to the parallel performance result and the data processing result to obtain an adjusted model; and performing verification adjustment on the adjustment model based on a side computing power device to obtain a target model structure after verification adjustment. The prediction accuracy of the model in the transmission, transformation and distribution scene is improved, the hardware parallel computing efficiency is improved, and the computing delay is reduced.

A website building system (WBS) hook code optimization system includes a website session repository, an event receiver, a hook code bundler, an optimization determiner, a code creator, and a hook code executor. The website session repository stores historical performance data from a plurality of hook code executions. The event receiver receives an event associated with hook code, and the hook code bundler creates a hook code bundle based on that event. The optimization determiner analyzes the historical data to generate optimization instructions for the bundle. Following these instructions, the code creator generates execution code for an optimized hook code bundle. The hook code executor runs the optimized bundle, generating current performance results for the event. Finally, the executor logs these results back into the repository, updating the historical data to enable subsequent optimizations in a continuous feedback loop.

The present invention relates to an apparatus and a method for diagnosing Parkinson's disease. The method for diagnosing Parkinson's disease according to the present embodiment may comprise: a step for presenting one or more athletic ability evaluation missions and one or more non-athletic ability evaluation missions to a user, and collecting, from the user, performance results for the one or more athletic ability evaluation missions and performance results for the one or more non-athletic ability evaluation missions; a step for extracting one or more athletic ability features and one or more non-athletic ability features from the performance results for the one or more athletic ability evaluation missions and the performance results for the one or more non-athletic ability evaluation missions; and a step for diagnosing the presence or absence of Parkinson's disease in the user on the basis of at least one of the results of extracting the one or more athletic ability features and the one or more non-athletic ability features.

The invention discloses a simulation method for a distributed cooling system, and the method comprises the following steps: building a topological relation graph and a three-dimensional digital-analog graph of the distributed cooling system, building a tail radiator module, an air inlet channel module and a motor power module according to the topological relation graph and the three-dimensional digital-analog graph, and obtaining the characteristic parameters of each assembly of the distributed cooling system; establishing a one-dimensional model of the internal liquid side of the distributed cooling system by utilizing the topological relation graph and the characteristic parameters; according to the spatial position relation and the size data of the radiator module and the air inlet channel module, a distributed cooling system air inlet side three-dimensional calculation model is established; and establishing a fluid-solid-thermal coupling calculation model of the distributed cooling system by combining the liquid side one-dimensional model and the air inlet side three-dimensional model, continuously iterating in the model solving process, and determining a performance result of the distributed cooling system in the quasi-steady-state process. According to the simulation method provided by the invention, a relatively accurate performance result of the quasi-steady-state cooling system is obtained, and the design efficiency of the cooling system is remarkably improved.

Various techniques are provided for training a neural network to classify images. A convolutional neural network (CNN) is trained using training dataset comprising a plurality of synthetic images. The CNN training process tracks image-related metrics and other informative metrics as the training dataset is processed. The trained inference CNN may then be tested using a validation dataset of real images to generate performance results (e.g., whether a training image was properly or improperly labeled by the trained inference CNN). In one or more embodiments, a training dataset and analysis engine extracts and analyzes the informative metrics and performance results, generates parameters for a modified training dataset to improve CNN performance, and generates corresponding instructions to a synthetic image generator to generate a new training dataset. The process repeats in an iterative fashion to build a final training dataset for use in training an inference CNN.

A website building system (WBS) hook code optimization system includes a website session repository, an event receiver, a hook code bundler, an optimization determiner, a code creator, and a hook code executor. The website session repository stores historical performance data from a plurality of hook code executions. The event receiver receives an event associated with hook code, and the hook code bundler creates a hook code bundle based on that event. The optimization determiner analyzes the historical data to generate optimization instructions for the bundle. Following these instructions, the code creator generates execution code for an optimized hook code bundle. The hook code executor runs the optimized bundle, generating current performance results for the event. Finally, the executor logs these results back into the repository, updating the historical data to enable subsequent optimizations in a continuous feedback loop.

The invention belongs to the technical field of resource efficiency evaluation, and discloses a big data-oriented resource information efficiency evaluation management system and method, and the system comprises a resource panoramic modeling module which carries out the factor modeling of a to-be-evaluated resource, enables the resource information to be divided into a function factor and a constraint factor, and forms a resource panoramic efficiency vector; calculating single resource efficiency based on the resource panoramic efficiency vector; the resource interaction association module is used for constructing a resource interaction association network among the resources based on the single resource efficiency result, dividing edges among the resources into a gain edge and a retardation edge, and propagating the single resource efficiency in the resource interaction association network to obtain the overall system efficiency; the efficiency balance baseline module is used for dynamically monitoring the efficiency of the whole system, counting the average value and the fluctuation range of the overall efficiency according to a sliding time window, and establishing a self-adaptive efficiency baseline; judging whether the current efficiency state of the system needs to be optimized or not through the baseline deviation amplitude; and the scientificity and the dynamism of resource information efficiency evaluation are improved.