30 results about "Temporal discretization" patented technology

The invention provides a network dithering algorithm based network congestion reflection method. The method includes the steps: step 10, taking a certain time interval as a unit for time discretization of continuous time, taking Di as transmission delay of data packets to calculate an average value of transmission delay from Ti-1 to Ti and an average value of transmission delay from Ti to Ti+1, and obtaining a change rate DRi of transmission delay according to the two average values; step 20, adjusting a reference value per unit time: E=E+DRi*delta t, wherein E refers to a value of expectation, delta t refers to time difference between two frames; step 30, judging whether the reference value needs to be reset or not, and if yes, resetting the reference value according to an equation: E=Di+DRi*(Ti+1-Ti) / 2; step 40, calculating a dithering value Ji according to an equation: Ji=(15*Ji-1+|Di-E|) / 16, and using the dithering value to represent network congestion conditions, wherein the larger the dithering value, the more serious the impending network congestion. By the network dithering algorithm based network congestion reflection method, network congestion conditions can be reflected timely and accurately, and sensitivity to network congestion is improved to lower packet loss rate.

A method for assisting the navigation of an aircraft comprises: assembly by families of data by a processing unit of predetermined data and acquired data, including meteorological data, the families being predefined, each data value being associated with a time window of validity; formatting the data of the families to associate with each data value a type of graphical representation as text or a scalar, vector, surface, or volume; selection of families of data to be displayed; choice of a display time window for each family of data to be displayed; spatio-temporal discretization of the trajectory; spatio-temporal correlation of the discretized trajectory with each family of data as a function of the time windows of validity, in the display time windows of the family, to extract a sub-assembly from each family of data, by the processing unit; display of the sub-assemblies in a single representation on the same display screen.

The invention provides a relevant failure prediction method and system. The relevant failure prediction method comprises the steps of acquiring historical failure information of a distributive computing system and acquiring a failure information sample database according to the historical failure information; acquiring effective failure information in the failure information sample database and performing time discretization processing on the effective failure information to obtain a sample data matrix; acquiring a current probability shared risk group structure of the distributive computing system according to the sample data matrix; acquiring current failure sample information of the distributive computing system that is collected in real time; performing failure prediction on the distributive computing system according to the current probability shared risk group structure and the current failure sample information. The method and the system can describe the diversity and transmissibility characteristics of relevant failures comprehensively and accurately, improve prediction precision, effectively increase the prediction method execution efficiency and facilitate online failure deployment and implementation, and are of great importance for improving the availability and the reliability of the distributive computing system.

The invention discloses a method for predicting a user participation behavior of a hot topic by an improved RBF (Radical Basis Function) neural network, and belongs to the field of computer network information technology analysis. Firstly, the situation that a neural network can take a good fitting effect for a complex nonlinear relationship among user behaviors is considered, and further a user participation behavior prediction model is built by adopting the RBF neural network; secondly, a mapping relationship between a user attribute and the participation behavior has uncertainty, and a cloud theory is introduced for optimizing an activation function of a hidden layer in an RBF; and finally, topic popularity is subjected to exponential function model-based parameter fitting by utilizing time discretization and time slicing methods for a characteristic that the participation behavior of a user is changed with time, so that a topic popularity change trend is obtained.

The invention discloses an SCR denitration catalyst life prediction method based on mass running data; the SCR catalyst life is obtained through analysis of coal-fired power plant denitration system industrial process data, and the SCR denitration catalyst life prediction method specifically includes the following steps: firstly, obtaining history running data from a DCS data source system, finishing original data into a sample set according to time labels, filtering data with unchanged numerical value, wild point data and ultralimit data, and filtering the non-stationary state of the data; then, filtering influence of working condition hybridity by using a working condition cutting and data loading method; next, discretizing the sample set according to the time, and extracting and calculating the relationship of the ammonia nitrogen molar ratio and the denitration efficiency; and finally, excavating a degradation law of the catalyst performance along with time series by using an analysis model, and predicting the residual life of the catalyst according to a reloading criterion and field actual flue gas conditions. Through data integration and analysis of the actual running performance of the catalyst with the same type units and similar running conditions, the catalyst life management characteristic function can be established in a larger data set.

The invention discloses a cellular network mobile device energy consumption optimization method with a controllable dispatch sequence. Time discretization is performed, and therefore time slots are formed. Under the condition that the demand for the transmission performance of mobile applications is met, dispatch sending of flow data of the mobile applications is performed according to the optimal dispatch sending time slot which is calculated in advance, and when multiple applications of a mobile device need to send a certain amount of flow data, flow of the mobile applications is divided into multiple task units; based on the given task unit dispatch sequence, the given arrival time slots, the given deadline and the given user performance demand limitation, under the condition that a transmission energy consumption function, a tail power consumption function and a performance expenditure function are known and according to an energy consumption optimization dynamic programming model, the iterative algorithm is used for solving the transmission dispatch time slot of each task unit, and transmission dispatch is performed on flow units according to the transmission dispatch time slots. According to the cellular network mobile device energy consumption optimization method with the controllable dispatch sequence, under the condition that the demand for the performance of the mobile applications of the user is met, comprehensive optimization is performed on transmission energy consumption and tail energy consumption. The cellular network mobile device energy consumption optimization method with the controllable dispatch sequence has the advantage that the transmission dispatch sequence of the task units is controllable, and the cellular network mobile device energy consumption optimization method can be achieved on mobile terminal devices.

The invention discloses a space flexible mechanism dynamic reliability analysis method based on PSO. The space flexible mechanism dynamic reliability analysis method includes the specific steps of (1) building a time-variation reliability model of an in-orbit space flexible mechanism according to the reliability modeling theory, (2) carrying out time discretization on the time-variation reliability model, and dividing the time-variation reliability model into a plurality of time points t1, t2, ..., tn, (3) building performance functions corresponding to the time points, and then building an association coefficient rou(t, t+deltat) between t and t+deltat by deriving normal vectors of the performance functions, (4) obtaining reliability coefficients beta1, beta2, ..., betan with the PSO, (5) calculating the up-crossing rate in cooperation with the association coefficient rou(t, t+deltat) between t and t+deltat, and (6) calculating the failure probability with the up-crossing rate method. By means of the space flexible mechanism dynamic reliability analysis method, the accuracy of a high-dimensional nonlinear flexible time-variation reliability analysis result is improved.

The invention discloses a special effect design method of dynamic crowd tracking based on Kalman filtering. The special effect design method includes the following steps: (1) a user inputs the number and the property of individuals of a crowd and the overall control parameters of the crowd so as to generate the crowd animation which is required by the user in real time; (2) a virtual leading point is designated for every individual of the crowd, the process of the simulation of crowd motion is discretized according to time, each time step length is sigma, the initial time of the motion is t0, a Kalman filtering method is utilized to calculate the optimum leading force, namely, F, of the individual in a random step n at the time that t0 plus sigma multiplied by n, and the motion parameter of every individual is readjusted so that the leading function of the leading point to the individual is realized; the individual is led to move towards the destination of the next time step length, the formation of the overall crowd moves to the ultimate formation which is designated by the user in the process, and the general formation and the moving process of the dynamic crowd change are guaranteed to be smooth and natural. The special effect design method is an economical and effective method for generating the beautiful crowd animation effect of three-dimensional deformation.

The invention discloses a yield prediction method for a volume fracturing horizontal well of a low-permeability heterogeneous stress-sensitive reservoir stratum. The yield prediction method mainly comprises the following steps that basic parameters of the reservoir stratum, fluid and a horizontal shaft are collected; a heterogeneous gas reservoir is partitioned into at least two vadose zones in the length direction of the horizontal shaft, and the vadose zones include artificial fractures; a vadose unit simulation model of each vadose zone is established; vadose simulation models of at least two vadose units are established; the vadose simulation models of all the vadose units are coupled; time discretization is carried out, the previous five steps are repeated, and the unstable-state yield of the volume fracturing horizontal well of the ultra-low-permeability heterogeneous stress-sensitive reservoir stratum is predicted. According to the technical scheme, the unstable-state yield of the volume fracturing horizontal well of the ultra-low-permeability heterogeneous stress-sensitive reservoir stratum can be predicted, the defects that in the prior art, only homogeneity gas reservoirs can be predicted without consideration of the starting pressure and stress sensitive effects are overcome, and thus the effectiveness and the effect of fracturing transformation of the tight gas reservoir horizontal well are improved.

The invention belongs to the field of time domain calculation electromagnetism, and relates to an electromagnetism multi-scale calculation method based on a hybrid grid and a time step length. According to the method, a multi-scale time domain electromagnetic problem is decomposed into a macro scale part and a micro scale part, scale separation is achieved, the solving scale of an original problemis effectively reduced, a mixed grid suitable for the characteristics of a micro unit is adopted in the micro unit for discretization, and the solving freedom degree of the micro unit is further reduced. And in the aspect of time iteration, independent time step lengths are adopted for different microcosmic units for time discretization, so that the defect of too many times of time iteration caused by too small local microgrid size is effectively avoided. After the microscopic information is solved through the microscopic problem, coupling between the macroscopic problem and the microscopic problem is achieved, the macroscopic problem is substituted into the macroscopic problem, iteration solving is conducted through the macroscopic time step length, and finally the technical effect of efficiently solving the multi-scale time domain electromagnetic problem is achieved.

The disclosure discloses a method for obtaining formation parameters of a gas hydrate reservoir through well testing interpretation, which comprises: (1) establishing a physical model for well testing interpretation of the gas hydrate reservoir according to multiphase flow, hydrate dissociation, secondary hydrate formation and heat transfer exhibited in the well testing process of the gas hydrate reservoir; (2) establishing a mathematical model for well testing interpretation of the gas hydrate reservoir; (3) conducting spatial discretization and temporal discretization on the mathematical model, and adopting a finite volume method to obtain a numerical solution of bottom-hole pressure; (4) calculating a bottom-hole pressure variation curve and a production rate variation curve in the well testing process of the hydrate reservoir, and drawing theoretical curves of bottom-hole pressure difference and pressure difference derivative; and (5) matching a measured pressure curve with the theoretical curve to obtain the relevant formation parameters.

The invention discloses a method for controlling group mobility, which is based on the RBF and includes the steps: firstly, a user inputs each parameter of the group to form real-time group animation which is in line with the requirements of the user; secondly, the user specifies a series of control points and sets the corresponding velocity amplitude thereof, which represents the guide velocity to which an individual in the group moving to the point needs obey; thirdly, a RBF interpolation method acts on scattered control points input by the user to form a vector field which represents the velocity of the individual in the group and is used as a guide component for the mobility of the individual in the group; and fourthly, the simulated process of the mobility of the group is divided into a plurality of step length according to discrete time, the guide motion component is looked up for each step length individual in the group to calculate the voluntary roaming motion of the individual, and the individual moves towards the position of next time step length under the combined action of the guide motion component and the voluntary motion component. The method adopts the velocity vector field as the main controllable factor for the motion, has the advantage of fast formation and supports real-time interactive amendment.

The invention relates to the mobile telecommunication field, in particular to a method of uploading multiple network-element logs in batches, which is characterized in that an operation maintenance center gives log-uploading commands simultaneously to multiple network elements, the commands have standard FTP information, after the network elements receive the commands, a key data with a unique characteristic of each network element is calculated and sorted according to the promised algorithm among the network elements, and the uploading time and batch of each network element is confirmed. The method of uploading multiple network-element logs in batches of the invention is employed to discretize the uploading time of the multiple network elements, and simultaneously the multiple network elements are stored according to a certain strategy, thereby avoiding the impact caused by integrated report to the operation maintenance center. Further, the method of uploading multiple network-element logs in batches can integrally manage the multiple network elements, and gives the log uploading commands simultaneously, and thereby the operation is simple for users.

A computer implemented method for simulating a final pattern of a droplet of a fluid having a plurality of fluid properties is disclosed. The method includes using lubrication equations to represent solute flow, diffusion and evaporation of a droplet on a substrate. The method further includes solving the lubrication equations through temporal discretization and spatial discretization; and deriving the final pattern of the droplet from results of the solving. The final pattern is stored on a computer readable medium.

A computer implemented method for simulating a final pattern of a droplet of a fluid having a plurality of fluid properties is disclosed. The method includes using lubrication equations to represent solute flow, diffusion and evaporation of a droplet on a substrate. The method further includes solving the lubrication equations through temporal discretization and spatial discretization; and deriving the final pattern of the droplet from results of the solving. The final pattern is stored on a computer readable medium.

The invention discloses a critical path method for the starting number of portal cranes in a container yard, and relates to the technical field of quay management and yard management. The method of the invention comprises the following steps: decomposing ship handling tasks to parallel task operating lines, to a Bay bit sequence and then to a Bay bit container handling task sequence by utilizing a berthing plan and a cargo plan; according to the starting time of the bridge crane of a cabin handling task, calculating the starting time of lorries and the starting time of the portal cranes; after time discretization, inversely calculating the number of the parallel yard container handling tasks of any event; and after summing up to a Bay bit and continuous Bay bit cluster, deducing the number of the portal cranes in parallel operation at random time; and further calculating the starting number of the portal cranes by statistical criteria. The method does not relates to complicated models and algorithms, can adapt to predict the starting number of the portal cranes on a large quay, and is favorable for realizing the object of energy saving and full use of the bridge cranes and the portal cranes.

The invention discloses an unmanned aerial vehicle relay deployment method based on multi-detection unmanned aerial vehicle cooperative search. The method comprises the following steps: firstly, discretizing time into a time slot form, so that a detection unmanned aerial vehicle can execute a corresponding action in a specific time slot; randomly deploying a plurality of detection unmanned aerialvehicles in the target area, sequentially detecting the network capacity of the current position by each detection unmanned aerial vehicle, interacting the position information and the network capacity information among the detection unmanned aerial vehicles, and updating the positions of the detection unmanned aerial vehicles according to the obtained position information and the obtained networkcapacity information; and finally, performing network capacity detection, information interaction and detection position update iteration circularly until all detection unmanned aerial vehicles converge to the same position, wherein the position is the optimal deployment position. The unmanned aerial vehicle relay deployment can be carried out through real-time capacity detection, and the methodhas the advantages of being easy to operate and high in practicability.

The invention discloses a real time Kalman filtering recursion modulation method of parametric array sonar emission signal. The method comprises the following steps: take the signal to be modulated as an incoming signal after time discretization sampling; get a recursive envelope squared value of the parametric array emission signal through the Kalman filter; execute square foot over the obtained envelope squared value of the parametric array emission signal to get the envelope value of the emission signal; and then modulate the envelope and a carrier wave to get the parametric array emission signal. In the invention, the single in transient broadband unknown wave can be modulated distortionlessly; the computational complexity is small; and the noise in the incoming signal can be inhibited.

An apparatus for detecting audible signals from a sound source is disclosed which has a sampling apparatus for converting incoming continuous signals in a manner that discretizes with respect to time, wherein the apparatus has a delay line module which receives samples from the sampling apparatus and which is set up to simultaneously provide both a sample ri that is applied to its input and at least one delayed sample (I), and one or more arithmetic units equipped with filtering properties. The arithmetic units each have a switching unit which acts as a coefficient generator and which is set up to take a sampling clock i as a basis for controlling whether values ri and / or (I) provided by the delay line module are ignored or fractions thereof produced directly or by right-shifting are either added or subtracted to / from the previous computation result y I h D(i -1) or y Q h D(i -1), an adding / subtracting unit and a result memory that can be controlled by means of the sampling clock i. In addition, the ratio of the sampling frequency F to the filter frequency fh preferably follows the relationship: F = 4fh s (1; 2; 3; 4;...) ± 15%. Furthermore, an associated method is provided.

The invention discloses an SCR denitration catalyst life prediction method based on mass running data; the SCR catalyst life is obtained through analysis of coal-fired power plant denitration system industrial process data, and the SCR denitration catalyst life prediction method specifically includes the following steps: firstly, obtaining history running data from a DCS data source system, finishing original data into a sample set according to time labels, filtering data with unchanged numerical value, wild point data and ultralimit data, and filtering the non-stationary state of the data; then, filtering influence of working condition hybridity by using a working condition cutting and data loading method; next, discretizing the sample set according to the time, and extracting and calculating the relationship of the ammonia nitrogen molar ratio and the denitration efficiency; and finally, excavating a degradation law of the catalyst performance along with time series by using an analysis model, and predicting the residual life of the catalyst according to a reloading criterion and field actual flue gas conditions. Through data integration and analysis of the actual running performance of the catalyst with the same type units and similar running conditions, the catalyst life management characteristic function can be established in a larger data set.

The invention provides a dynamic space network construction method for traffic demand prediction, and relates to the field of space networks. The method is characterized in that the method comprises the steps: time discretization: dividing continuous time points into different time periods; node space relationship matching: calculating the stability index S of the node, and adopting different matching strategies according to the stability condition of each node space relationship; dynamic space network modeling: in each time division, carrying out spatial relationship matching on each node, and constructing a dynamic space network, wherein the dynamic space network is further based on a space network, and heterogeneous network modeling is adopted. According to the dynamic space network construction method for traffic demand prediction provided by the invention, the dependency relationship between regions (or stations) can be more accurately reflected, the resource scheduling efficiencyduring traffic prediction is high, and the traffic demand prediction result can be effectively improved.

The present invention relates to a method and system for predicting associated faults, wherein the predicting method for associated faults comprises the following steps: obtaining historical fault information of a distributed computing system, and obtaining a fault information sample database according to the historical fault information; The effective fault information of the distributed computing system is obtained by discretizing the effective fault information in time to obtain the sample data matrix; according to the sample data matrix, the current probability sharing risk group structure of the distributed computing system is obtained; the current data of the distributed computing system collected in real time is obtained Fault sample information: According to the current probability sharing risk group structure and current fault sample information, fault prediction is performed on the distributed computing system. The present invention can fully and accurately describe the diversity and propagation characteristics of associated faults, improve prediction accuracy, effectively improve the execution efficiency of prediction methods, facilitate the deployment and implementation of online faults, and have great significance for improving the availability and reliability of distributed computing systems .

The invention provides a network dithering algorithm based network congestion reflection method. The method includes the steps: step 10, taking a certain time interval as a unit for time discretization of continuous time, taking Di as transmission delay of data packets to calculate an average value of transmission delay from Ti-1 to Ti and an average value of transmission delay from Ti to Ti+1, and obtaining a change rate DRi of transmission delay according to the two average values; step 20, adjusting a reference value per unit time: E=E+DRi*delta t, wherein E refers to a value of expectation, delta t refers to time difference between two frames; step 30, judging whether the reference value needs to be reset or not, and if yes, resetting the reference value according to an equation: E=Di+DRi*(Ti+1-Ti) / 2; step 40, calculating a dithering value Ji according to an equation: Ji=(15*Ji-1+|Di-E|) / 16, and using the dithering value to represent network congestion conditions, wherein the larger the dithering value, the more serious the impending network congestion. By the network dithering algorithm based network congestion reflection method, network congestion conditions can be reflected timely and accurately, and sensitivity to network congestion is improved to lower packet loss rate.

The invention discloses a cellular network mobile device energy consumption optimization method with a controllable dispatch sequence. Time discretization is performed, and therefore time slots are formed. Under the condition that the demand for the transmission performance of mobile applications is met, dispatch sending of flow data of the mobile applications is performed according to the optimal dispatch sending time slot which is calculated in advance, and when multiple applications of a mobile device need to send a certain amount of flow data, flow of the mobile applications is divided into multiple task units; based on the given task unit dispatch sequence, the given arrival time slots, the given deadline and the given user performance demand limitation, under the condition that a transmission energy consumption function, a tail power consumption function and a performance expenditure function are known and according to an energy consumption optimization dynamic programming model, the iterative algorithm is used for solving the transmission dispatch time slot of each task unit, and transmission dispatch is performed on flow units according to the transmission dispatch time slots. According to the cellular network mobile device energy consumption optimization method with the controllable dispatch sequence, under the condition that the demand for the performance of the mobile applications of the user is met, comprehensive optimization is performed on transmission energy consumption and tail energy consumption. The cellular network mobile device energy consumption optimization method with the controllable dispatch sequence has the advantage that the transmission dispatch sequence of the task units is controllable, and the cellular network mobile device energy consumption optimization method can be achieved on mobile terminal devices.

Disclosed is a device for detecting an acoustic signal of a sound source, said device having a sampling device for time-discretized conversion of an incoming continuous signal, wherein said device has a delay line module and one or more Computing means, the delay line module receives sampled values ​​from the sampling device and is designed to simultaneously provide not only the sampled value ri applied to its input but also at least one delayed sampled value (I). The computing means each have a switching means acting as a coefficient generator, an adder / subtractor and a result memory controllable by means of the sampling clock i, the switching means being designed to be controlled according to the sampling clock i, whether the inversion is determined by The values ​​ri and / or (I) provided by the delay line module are either optionally added to the part of the calculated result yIhD(i-1) or yQhD(i-1) so far, directly or by right-shifting or Subtract the value from the portion. Furthermore, the relationship between the sampling frequency F and the filter frequency fh is preferably in accordance with the relationship: F=4·fh·(1; 2; 3; 4; . . . )±15%. Additionally, a method of belonging is provided.

The present invention relates to a method for determining a corrected rotational speed signal and to an electric motor arrangement. In one step of the method, a periodic rotational angle signal havinga rotational-angle-dependent, periodic rotational angle error is determined. The rotational angle signal is converted into a digital raw rotational speed signal (16) which, on account of the rotational angle error, has an interference ripple, the period duration of which corresponds to the period duration of the rotational angle error. The digital raw rotational speed signal (16) comprises a plurality of sampled values (07) which are spaced apart in terms of time by a sampling period duration. Half the period duration of the interference ripple of the raw rotational speed signal (16) is determined. An average is calculated for at least one pair of sampled values (07) of the raw rotational speed signal (16), the temporal spacing of said values being the determined half period duration of the interference ripple of the raw rotational speed signal (16), apart from a temporal discretization error. An average value is obtained from the average calculation. A correction value which is dependent on the sampling period duration and on the frequency of the interference ripple is formed in order to take the discretization error into account. The corrected rotational speed signal is formed from a difference between the average value and the correction value. The correction signal varies in a sawtooth-shaped manner on the basis of the deviation of the sampling period duration from the period duration of the interference ripple, and the maximum amplitude of the correction signal also increases with the frequency of the interference ripple.

The invention discloses a yield prediction method for a volume fracturing horizontal well of a low-permeability heterogeneous stress-sensitive reservoir stratum. The yield prediction method mainly comprises the following steps that basic parameters of the reservoir stratum, fluid and a horizontal shaft are collected; a heterogeneous gas reservoir is partitioned into at least two vadose zones in the length direction of the horizontal shaft, and the vadose zones include artificial fractures; a vadose unit simulation model of each vadose zone is established; vadose simulation models of at least two vadose units are established; the vadose simulation models of all the vadose units are coupled; time discretization is carried out, the previous five steps are repeated, and the unstable-state yield of the volume fracturing horizontal well of the ultra-low-permeability heterogeneous stress-sensitive reservoir stratum is predicted. According to the technical scheme, the unstable-state yield of the volume fracturing horizontal well of the ultra-low-permeability heterogeneous stress-sensitive reservoir stratum can be predicted, the defects that in the prior art, only homogeneity gas reservoirs can be predicted without consideration of the starting pressure and stress sensitive effects are overcome, and thus the effectiveness and the effect of fracturing transformation of the tight gas reservoir horizontal well are improved.