20 results about "Decay factor" patented technology

The invention relates to the technical field of computers, in particular to a component risk prediction method and system fusing a time decay factor and a storage medium, and the method comprises the steps: analyzing a component interaction relationship, building a node set and an edge set, and introducing the time decay factor to construct a time decay weighted component dependency relationship graph; performing dynamic graph structure feature analysis and time sequence modeling on the nodes and the associated edges thereof, and by extracting multilayer neighborhood features of the nodes, constructing a time sequence behavior model and calculating a structure lag coefficient to generate a graph feature set; introducing a service weight to carry out multi-layer dependency fusion and weight dynamic adjustment to obtain a graph feature set after weight fusion; and constructing a risk prediction input matrix, inputting the constructed risk prediction input matrix into a preset risk prediction model, and outputting a risk prediction score and a risk level of each component by the model. According to the method, the accuracy and timeliness of component risk prediction in a complex system can be remarkably improved.

The invention discloses a weighted exponential decay instantaneous phase inversion method based on automatic differentiation. Under the framework of automatic differential, the underground velocity model can be inversed by using phase and waveform information at the same time, and the advantages of constructing an initial model at the early stage of inversion and improving the resolution of the model at the later stage of inversion are combined. The method comprises the following steps: firstly, performing Hilbert transform on seismic wave data, extracting index instantaneous phase information of observation data and simulation data, and constructing a weighted instantaneous phase inversion objective function by introducing an attenuation coefficient and combining waveform information; secondly, carrying out back propagation on the loss of the objective function based on an automatic differential framework, and solving an update gradient; and finally, updating the speed model by using an SDG optimizer. According to the method, a layered multi-scale inversion effect can be realized by adjusting attenuation factor values, an objective function is constructed by weighting exponential attenuation phase information and waveform information, a better initial model is constructed before a conventional multi-scale full-waveform inversion process, the problem of cycle jump of full-waveform inversion is relieved, and the inversion efficiency is improved. And finally, a high-precision inversion result is obtained.

The invention discloses a self-adaptive task priority management method based on a dynamic entropy weight and a time decay factor, and relates to the technical field of computer task scheduling and resource allocation. The method comprises the following steps: acquiring a to-be-processed task set and constructing a multi-dimensional feature matrix, dividing features into benefit type indexes and cost type indexes, and respectively carrying out same-direction standardization processing on data by adopting forward and reverse normalization formulas; calculating entropy values and difference coefficients of the characteristic indexes in real time based on an information entropy theory, determining dynamic weights reflecting current task flow characteristic distribution, and calculating static basic scores of the tasks; introducing an exponential time decay model containing a natural constant and a time sensitive coefficient, performing nonlinear acceleration compensation on the static basic score according to the queue residence time of the task, and generating a final dynamic priority index; and screening the task corresponding to the maximum value according to the index to preferentially allocate computing resources. According to the method, the blindness of manual weighting can be eliminated, the hunger and deadlock problems of low-priority tasks are effectively solved through an exponential-level time compensation mechanism, and the method has the advantages of low algorithm complexity, high environment adaptability, high scheduling fairness and the like.

The application relates to a low-orbit satellite positioning satellite selection method, device, equipment and medium. The method comprises the following steps: acquiring adaptive parameters such as the population dimension of an alternative scheme and the number of satellite selection, and designing an alternative scheme matrix and a random step matrix. The alternative scheme matrix is resampled to generate an evolution matrix, and a decay factor is calculated according to the number of iterations and the number of fitness evaluations. A basis vector is obtained through the number of iterations, the number of evaluations and the alternative scheme matrix, and after the fitness of the evolution matrix is calculated, a dominant solution and a poor solution are generated. A new scheme matrix is calculated in combination with the basis vector, the decay factor and the like, and if the fitness of the new scheme matrix is better, the current optimal solution is replaced. The satellite weight in the historical evolution matrix is adjusted by using a dynamic weight distribution strategy, and the scheme matrix is further optimized. Iteration comparison is continuously carried out, the scheme matrix with the optimal fitness is taken as the current iteration result, and the maximum number of iterations is reached until the optimal satellite selection scheme is finally output. The method can improve the satellite selection efficiency and accuracy.

The embodiment of the invention provides a video recommendation method and device, equipment and a storage medium. The method specifically comprises the steps of obtaining interaction behavior data of a user for playing a video; under the condition that the interactive behavior data indicates that the negative feedback behavior occurs, performing feature extraction on the played video to obtain at least one disgust feature; determining the aversion score of the user for the aversion feature according to the basic aversion intensity, the continuous strike weight and the time attenuation factor corresponding to the aversion feature and the time interval between the occurrence time of the negative feedback behavior and the current time; and obtaining candidate videos in a pre-generated video recommendation list, and updating the candidate videos in the video recommendation list according to the aversion scores corresponding to the aversion features included in the candidate videos and a preset score threshold.

The invention relates to a table item popularity attenuation method and device, computer equipment, a readable storage medium and a program product. The method is applied to a storage system, the storage system comprises a plurality of table entries, and the method comprises the following steps: under the condition that an access request sent by a host successfully accesses data in the storage system, updating the access frequency of the table entries accessed by the access request; under the condition that the number of the target table items in the storage system meets a preset threshold value and / or the timing duration of a timer in the storage system is equal to a preset duration, determining a heat attenuation factor; the target table entries comprise table entries of which the access frequencies reach a frequency threshold value; and according to the popularity attenuation factor, attenuating the access frequency of each table item in the storage system. By adopting the method, the access frequency of each table item can be attenuated, so that the to-be-replaced table item is accurately screened from the cache based on the attenuated access frequency of each table item, and the utilization rate of the cache is further improved.

The application discloses a main bearing damage diagnosis method based on flow-heat coupling dynamic topology and a physical information graph autoencoder, and comprises the following steps: one-dimensional residual feature extraction is performed on multi-source SCADA time series data to obtain time series feature vectors of nodes; a Nusselt number is calculated, edge weights of a directed adjacency matrix are dynamically calculated based on the Nusselt number, a temperature gradient gating function and a thermal resistance parameter, and a dynamic flow-heat topology graph is constructed in combination with the time series feature vectors; a graph encoder is used to perform spatial aggregation on neighbor node features along a directed edge to obtain a predicted temperature of the main bearing at the next moment; a total loss function is calculated based on mean square errors of the predicted temperature and an actual temperature and physical consistency loss, and training is constrained to obtain a state reconstruction model; a residual error between the predicted temperature output by the model and the actual temperature, and a decay factor calculated based on a bearing fatigue life equation are used to dynamically adjust a warning threshold, and damage warning is triggered when the residual error exceeds the threshold.

The invention discloses a bus bunching prediction method and system based on a time-aware network. The method comprises the following steps: preprocessing bus arrival data, fusing multi-source spatial-temporal characteristics, introducing time period codes, and processing abnormal and missing values; targeted denoising is performed on data with different characteristics based on Fourier analysis, so that the data quality is improved; a time perception neural network is established, local features and long-term dependence are captured through a CNN-LSTM architecture, a time decay factor mechanism is introduced, dynamic modeling of a time sequence information importance decay process is realized, and a long-term rule of data and short-term disturbance after time calibration are effectively integrated; and a dual attention mechanism of time dimension and feature dimension is fused, key features are extracted and integrated, and multi-step prediction of bus bunching is realized. According to the method, the non-linear relation and the space-time dynamic rule of the bus data can be fully captured, and the problem of unequal intervals in a bunching time sequence is effectively solved.

The invention discloses a hybrid program load adaptive cache replacement method and system, and belongs to the technical field of electric digital data processing. The method comprises the following steps: constructing a cache management architecture; extracting a program counter value and a physical memory address of the current instruction in real time, forming a combined feature, inputting the combined feature into an online incremental learning model, and judging the category of memory access; updating the activeness weight of the micro-cluster according to the access category; an attenuation factor is introduced, the activeness weights of all the micro-clusters are updated regularly according to a preset time interval, and inactive micro-clusters are deleted automatically; the cache line hit rate of the reserved micro-clusters and the excess occupation amount of the reserved micro-clusters in the current period are calculated in real time, and sacrificial clusters in the reserved micro-clusters are found out; and expelling the cache line with the largest line age from the micro-cluster, and synchronously cleaning associated entries of the cache line. According to the invention, protection of high-value data and effective isolation of pollution sources are realized in a complex mixed program load scene.

This disclosure relates to the field of data processing technology, and in particular to a method, system, device, and medium for processing data popularity. The method involves: acquiring historical hot / cold trends and current access characteristics of the data; calculating the initial popularity based on the platform weight at the time of data addition, historical hot / cold trends, and a popularity decay factor; dynamically adjusting the time decay factor based on the interval between the data addition time and the target time; calculating the decay of the initial popularity based on the time decay factor to obtain the decayed popularity; and calculating the latest total popularity based on the decayed popularity and current access characteristics, and storing the updated data popularity in a database. This disclosure, by introducing a time decay factor and a dynamic parameter adjustment mechanism, solves the problems of popularity distortion caused by fixed parameter calculations, inaccurate reflection of new data popularity, and the Matthew effect in existing technologies. It achieves accurate and real-time calculation of data popularity, effectively improving database management efficiency and storage resource utilization.

The invention provides a diffuse reflection metasurface obtaining method based on an ivy algorithm and a related product. The method comprises the steps that a population is initialized according to the number of units of the diffuse reflection metasurface, the number of individuals in the population and the preset level quantity of phase discretization; at least adopting a multi-strategy-based ivy algorithm to perform multiple times of global exploration on the population according to a target function so as to optimize and update the population; and setting units of the diffuse reflection metasurface according to the optimized and updated individuals in the population. According to the technical scheme, in the process of optimizing and updating the population by adopting a multi-strategy-based ivy algorithm, a nonlinear attenuation factor and a flight-enhanced population spreading step length are introduced, so that the acoustic scattering uniformity of the diffuse reflection metasurface can be improved.

The application discloses a kind of complex multi-step network attack detection methods based on space-time fusion features, including model training stage and multi-step attack detection stage, the method of the present application is combined with graph neural network, time weight and LSTM, the space-time characteristics of complex multi-step attack are comprehensively analyzed, the complex correlation between and inside multi-step attack chain is effectively captured.Through five-tuple fingerprint construction traffic graph structure, node attribute fusion protocol type, timestamp and load characteristics, the designed edge generation algorithm can effectively represent the space-time correlation characteristics of multi-step attack.Design the adjacent matrix updating algorithm with time decay characteristics, dynamically adjust the information propagation intensity between nodes through the configurable decay factor, effectively suppress the interference of time sequence confusion noise on attack chain division.Design a three-layer composite network structure, form a progressive feature learning path of "space-global-time", solve the division problem of space-time overlapping multi-step attack chain.

According to the distortion compensation method and system of the coal and gas outburst physical simulation similarity criterion, a gas-solid coupling control equation of coal and gas outburst physical simulation is selected, and a similarity ratio relational expression of key parameters is established; processing the similarity ratio relational expression to obtain a first similarity criterion; based on an effective stress theory and a thermodynamic theory, constructing a mechanical strength attenuation function of the porous medium material under the action of gas pressure; carrying out a mechanical test under different gas pressure conditions to measure material strength parameters of the prototype and the model; according to the material strength parameters, strength attenuation factors under different gas pressure conditions are calculated, and corresponding strength attenuation functions are obtained through fitting; combining the intensity attenuation function with a first similarity criterion, deducing to obtain a distortion compensation equation of the similarity criterion, and constructing a distortion compensation similarity criterion; and finally determining a distortion compensation value of each physical property parameter. According to the method, the correction and compensation of the physical simulation similarity relation under the action of the gas pressure can be realized, and the accuracy of the physical simulation result is improved.

The application provides a configuration item credibility dynamic processing method and device, electronic equipment, medium and product. It relates to the field of computer operation and maintenance. The method comprises the following steps: obtaining configuration item data and metadata thereof, the metadata comprising an update time and a data type; calculating a time decay factor based on the update time and the data type, the time decay factor adopting an exponential function form, and the difference of the data type affecting the decay rate of the exponential function; and generating a dynamic score according to the time decay factor and a preset weight parameter, the dynamic score being used to represent the credibility of the configuration item data. Based on the configuration item credibility dynamic processing method provided by the application, the accuracy and operation and maintenance efficiency of CMDB data can be improved.

This invention relates to the field of data processing technology, and provides a method and system for creating, packaging, and one-click replicating digital olfactory recipes. The method includes: creating an olfactory time segment; parametrically editing the olfactory time segment, including an intensity envelope, to complete the creation of the digital olfactory recipe; wherein, firstly, the initial intensity and the average deviation of historical intensities of the same user in similar scenarios are obtained; then, an intensity decay factor is obtained based on the initial intensity, and the intensity exponential decay factor is compensated based on the average deviation, defining an intensity envelope; the digital olfactory recipe is packaged into a standardized olfactory script; the user initiates a one-click replica request to the server, the server verifies the user's identity, and if the user's identity verification is successful, a download link for the standardized olfactory script is returned. This invention can adaptively compensate according to individual user preferences, realizing personalized olfactory recipe creation, which is beneficial to improving the quality of the user's olfactory experience.

The application belongs to the technical field of conductance drift, and particularly relates to a reasoning mapping scheme for optimizing conductance drift, which comprises the following steps: S1: according to a conductance drift model, fitting actual device conductance drift data to obtain model parameters; S2: scaling input data and weights respectively as a weight input voltage V0 and a target mapping conductance C; S3: processing the input signal through the S2 step into the weight input voltage V0, multiplying the weight input voltage V0 by a decay factor to obtain a final input voltage V, and then mapping the final input voltage V to a memristor array through a peripheral board card circuit. Compared with a traditional method, the scheme provided by the patent solves the conductance drift problem, optimizes the performance of the system through a unique mechanism of multiplying the input signal by a compensation factor, and directly provides the compensation of the conductance drift by the input signal, thereby reducing the difficulty of hardware network implementation. The system using the scheme can reach an accuracy of 96% after being trained for 20 times, and the accuracy is improved by 10% compared with the traditional scheme.

The application provides a quantitative evaluation method for a knowledge system, which comprises evaluation of a primary index and a secondary index, the secondary index comprising a representation category, a knowledge granularity, a knowledge connotation, a knowledge entropy before fusion, a knowledge entropy after fusion, a decay factor, a system throughput, a reasoning real-time performance, a business diversity and a business effectiveness; the primary index comprising a representation capability, an understandability, an easy operability and a business support capability, the representation capability being calculated based on the representation category, the knowledge granularity and the knowledge connotation, and providing a selection basis for a knowledge representation method of a representation layer of the knowledge system, the understandability being calculated based on the knowledge entropy before fusion and the knowledge entropy after fusion, and providing a selection basis for a knowledge density of the representation layer of the knowledge system, the easy operability being calculated based on the decay factor and the system throughput, and providing a parameter configuration basis for a management layer of the knowledge system, and the business support capability being calculated based on the reasoning real-time performance, the business diversity and the business effectiveness, and providing a parameter configuration basis for an application layer of the knowledge system.

The invention belongs to the technical field of geotechnical engineering, and discloses a rock mass equivalent GSI parameter valuing method based on three-dimensional strength and strain softening, and the method comprises the steps: determining rock mass mechanical parameters and tunnel excavation parameters; calculating the plastic zone radiuses Rp and epp of the model by adopting the peak value GSIp and the ideal elastic-plastic model; calculating an attenuation factor k according to the determined parameters; determining a proportionality coefficient beta through an exponential decay relation based on the decay factor k and the model plastic zone radii Rp and epp, and further obtaining an equivalent parameter GSIeq; and finally, calculating the mechanical response of the surrounding rock by adopting the equivalent parameter GSIeq and the ideal elastic-plastic model. According to the method, the equivalent GSI parameters are introduced, the complex strain softening problem is converted into the ideal elastoplasticity problem of efficient calculation, the calculation efficiency is remarkably improved while the calculation precision is guaranteed, the influence of the three-dimensional stress state is fully considered, and a practical and reliable technical means is provided for stability analysis of tunnel engineering.

This invention discloses a method and system for improving the success rate of wide-lane ambiguity fixation in BeiDou dual-frequency positioning, belonging to the field of high-precision positioning technology. It solves the problems of existing methods being unable to adapt to dynamic changes in observation quality, suffering from old information contamination, and difficulties in setting the smoothing window length. The method includes: constructing a MW combined observation model for BeiDou dual-frequency positioning; establishing a time-decay-based weight allocation function to recursively smooth the wide-lane ambiguity floating-point solution sequence; performing integer fixing and reliability verification based on the smoothed wide-lane ambiguity floating-point solution estimate; and dynamically adjusting the exponential weighting factor according to the measurement residual to achieve adaptive smoothing. This invention introduces a time-decaying weight function to give higher weight to recent observations and establishes a recursive update model. Simultaneously, it introduces a residual-driven dynamic decay factor adjustment mechanism, effectively improving the reliability and timeliness of wide-lane ambiguity fixation in BeiDou dual-frequency positioning.

The invention discloses a multi-agent collaborative decision-making method based on an improved MADDPG, and belongs to the technical field of multi-agent multi-task allocation, and the method comprises the steps: carrying out the design of a reward function, and enabling the agents to form a surrounding formation; a layered multi-agent collaborative framework is designed; and finally, a priority experience playback mechanism is added to the MADDPG algorithm, an attenuation factor is added under OU noise noise to optimize an exploration mechanism, experience pools in the priority experience playback mechanism are divided into a common experience pool and a high reward round experience pool, and experience sampling times are introduced into an experience importance weight calculation formula. The method can effectively solve the problem of multi-agent and multi-task allocation.