95 results about "Probability curve" patented technology

The invention relates to a high-voltage single conductor cable service life evaluating method and system. A high-voltage single conductor cable service life evaluating method comprises the following steps of acquiring high-voltage single conductor cable original operation zero-failure data, conducting estimation for shape and size parameters of a double-parameter Weibull distribution model built for evaluating high-voltage single conductor cable service life via a corrected maximum likelihood function to acquire shape parameter estimation values and size parameter estimation values, and acquiring a high-voltage single conductor cable service life distribution probability curve via the double-parameter Weibull distribution model according to the shape parameter estimation values and size parameter estimation values. Accurate fault analysis and fault prediction can be provided with fewer samples by the use of the Weibull distribution model; parameter estimation can be achieved even when zero-failure samples do not exist by the corrected maximum likelihood function; and accuracy for insulation cable service life estimation can be improved.

The invention relates to a tobacco manufacturing stability inspection. It solves the issue of failure to effectively identify the factors affecting the tobacco quality, resulting in the inability of reasonable quantitative and qualitative analysis with simple application, effectively getting rid of all kinds of abnormal status, ensuring the quality of tobacco. Using time sequence scuff, distributional statistical chart and accumulated probability curve, separating the process stability from the overall process, judging the conformity of its distribution and that of the normal distribution, effectively identifying the process variation factors, tracking the time sequence zone, analyzing the variation source, type and the affecting degree to the quality index.

The invention discloses a three-dimensional matching algorithm between an adaptive window and a weight based on a picture edge. The algorithm comprises the following steps of: firstly, rapidly and dynamically selecting a window supporting size by picture edge information; providing a weight model which accords with a probability curve according to the characteristic that a matching value is changed along the geometrical distance from a neighborhood point to the center of the window; and combining with the color similarity constraint, wherein the weighted color distance cumulative sum is the similarity quantity, and computing parallax errors by points to obtain a dense disparity map. After the algorithm is used, matching noise can be effectively reduced, the matching precision between a high-edge region and a low-texture region can be improved, and the rapid and high-efficiency three-dimensional matching can be realized.

In order to improve the condition that a certain deviation of a lightning trip-out rate exists between lightning current amplitude probability distribution functions recommended by regulations and the IEEE (Institute of Electrical and Electronic Engineers) and practical lightning current amplitude probability distribution, the invention discloses a piecewise fitting method of a lightning current amplitude probability distribution function. According to the method, a numerical analysis method is adopted to perform piecewise fitting on the lightning current amplitude probability distribution function, continuity correction is performed, and a curve of the lightning current amplitude probability distribution function, which is obtained in the way, is strongly relevant to a practical lightning current amplitude probability curve. As is well-known, the lightning current amplitude probability distribution function is an important calculating parameter for calculating the lightning trip-out rate of an electric transmission line, and the accuracy of a value directly influences the accuracy of a calculating result of the lightning trip-out rate. Therefore, by using the fitting method disclosed by the invention, the more accurate lightning trip-out rate can be obtained, and operation reference is supplied to the department of electric power operation for designating a lightning protection weak area of a regional power grid.

The invention discloses a method for calculating the farthest detection distance of a Gm-APD laser radar. The method comprises the following steps: 1, acquiring echo intensity by using a laser radar equation; 2, obtaining the triggering probability of each time interval of the Gm-APD laser radar; 3, acquiring simulation data under different statistical frame number conditions by using a Monte Carlo method; 4, echo parameters are obtained through a signal detection method according to the Gm-APD trigger probability curve obtained in the step 2 and the simulation data obtained in the step 3; and5, calculating echo detection probability and confirming optimal parameters of the system. The Gm-APD performs optimal argumentation on parameters such as laser parameters and optical aperture underthe condition that the index parameters are determined, so that the laser radar system realizes optimal power consumption and volume on the premise of meeting the indexes, and has a very important promotion effect on further engineering of the Gm-APD laser radar.

The invention provides a mine pressure big data real-time prediction system and method based on a fully mechanized coal mining face and relates to the technical field of mine pressure. The system comprises an incoming pressure analysis module, a statistical calculation module, a prediction module and a display module, the incoming pressure analysis module analyzes support working cycle, divides working face support areas, divides working face incoming pressure and calculates incoming pressure step or stable step of a top plate at a support according to mine pressure big data; the statistical calculation module is used for counting all stable step pitches or incoming pressure step pitches of the top plate at the bracket position and fitting an incoming pressure probability curve; the prediction module predicts the roof incoming pressure probability of the support according to the incoming pressure probability curve and predicts the incoming pressure strength of the roof; and the displaymodule is used for displaying a support and support area roof pressure prediction result and a production prediction report including a single-support analysis and prediction comprehensive graph, anarea analysis and prediction comprehensive graph, a mine pressure prediction table and the like, and accurately predicts mining pressure by utilizing working face support mine pressure monitoring bigdata so as to guide roof management.

The invention discloses a wind power storage energy power and capacity optimization and configuration method based on economic dispatching. A weibull probability curve of wind power is established; a random planning model of a wind power system energy storage apparatus is established based on economic dispatching; and the random planning model of the energy storage apparatus, established by a "2m+1" point estimation method, is solved by adopting a parallel branch and bound method to obtain an optimal power and capacity configuration scheme of the energy storage apparatus. Aiming to solve the problem in wind storage effective matching in uncertain wind power environment, the random planning model of the energy storage apparatus with optimal power and capacity configuration is proposed; and based on a comprehensive solving scheme with point estimation and parallel branch and bound strategies, the energy storage random planning model is rapidly and effectively solved, so as to determine the optimal power and capacity configuration scheme of the energy storage apparatus.

The invention discloses a prediction method for fatigue reliability of a cable support in a subway tunnel. The method comprises: considering nondeterminacy of model parameters, using a response surface method to generate parameter samples; according to the parameter samples, establishing a finite element model, converting a track irregular spectrum to an irregular spatial waveform, used to adjust node coordinates of steel rail nodes in the model; the steel rail node being provided with a vertical pressed non-pulled unit, used to simulate wheel-rail contact; establishing a train-rail model, a tunnel-soil mass model, and a support-cable model and solving, to obtain stress time-history of the node on the root of the support, amplitude of the node being maximum; using a rain-flow counting method to calculate fatigue life; according to the parameter samples and a fatigue life value to fit the response surface and calculating a reliability index; accordingly, a series of fatigue life target values can be assigned, calculating the corresponding reliability index, so as to draw an invalid exceedance probability curve of the support.

The invention relates to a method for determining parameters of excitation system property evaluation model. The excitation system property evaluation model is a tree structural model. The top level of the tree structural model is a to-be-evaluated excitation system property, the bottom level is a plurality of basic evaluation factors, and the middle level is a classification evaluation factor. Parameters herein include a grade evaluation threshold value of a basic evaluation factor and the weights of the basic evaluation factor and the classification evaluation factor. The method for determining parameters includes the following steps: 1. acquiring a sample data which includes basic evaluation data of the basic evaluation factor which corresponds to a plurality of different excitation systems; 2. acquiring the corresponding grade evaluation threshold of each basic evaluation factor by using the probability curve knitting separation method; and 3. separately acquiring the corresponding weight of the basic evaluation factor and the classification evaluation factor by using 1-9 scale method. Compared with prior art, according to the invention, the method is simple and scientific, and can indirectly increase the reliability of an evaluation model.

The invention provides a method for evaluating repetitive frequency laser damage resistance of an optical thin film. The method comprises the following steps: irradiating different test points of a thin film sample to be tested with a series of sequence laser pulse with a certain energy density, determining and recording a damage state of each test point, and calculating small damage probability at the energy density; repeatedly changing the energy density of the laser to irradiate different test points of the thin film sample to be tested, determining and recording the damage probability of each test point, and respectively calculating the small damage probability at each energy density; respectively taking the energy density and the small damage probability as a horizontal coordinate and a longitudinal coordinate to draw a change curve of the small damage probability with the laser energy density, and evaluating the repetitive frequency laser damage resistance of the optical thin film at the energy density corresponding to a damage probability peak value of the probability curve. The energy density is higher, the repetitive frequency laser damage resistance of the optical thin film is better. The method helps simplify and practice methods for evaluating the repetitive frequency laser damage resistance of optical films.

The invention relates to a peak-to-average ratio suppression method based on the optimal amplitude distribution of an OFDM signal, which solves the problems of low power amplifier efficiency and signal distortion of a transmitter in the prior art. The purpose of peak-to-average ratio suppression proposed by the peak-to-average ratio suppression method disclosed by the present invention is not to reduce the highest peak value, but rather improves the efficiency of amplifying the OFDM signal, as well as consider the maximum peak signal during implementation, also consider the probability of occurrence but not the largest amplitude of the sub-peak signal, the constraint of Euclidean distance of constellation point vector in frequency domain on the amplitude distribution curve of OFDM signal is explored, and the optimal amplitude distribution probability curve of OFDM signal satisfying the highest efficiency of power amplifier is obtained. Based on this criterion, the limiting function andACE algorithm of M iterations are trained. The test results show that, in QPSK modulation, 3.15 dB of IBO performance improvement is obtained compared with the original OFDM system after the final time domain transmission signal is optimized by the invention, and the IBO of the single carrier system is only 0.05 dB less than the original OFDM system under the same conditions.

A method and apparatus for estimating device availability in a pervasive peer to peer environment is presented. The method and apparatus include a linear estimation approach and a bin estimation approach, where the bin estimation approach may include enhanced bin estimation or adaptive bin estimation. A central node queries available contacted devices and builds a history of times and availability indications for the contacted device, and subsequently computes an availability probability curve based on assumptions regarding the availability indications and the time since the last received availability indication.

The invention discloses a user activity anomaly detection and traffic prediction method based on big data analysis, which comprises the following steps: 1) obtaining regional number and time period information of an undetermined abnormal region; 2) using historical big data to evaluate an empirical cumulative probability curve of the undetermined abnormal region at time period activeness; then, using the empirical cumulative probability curve to determine a traffic activeness abnormal value of each moment in a prediction region; 3) finding eight regions adjacent to the space of the predictionregion with the prediction region as the center; then, using measurement sim (i, j) of spatial similarity of two adjacent regions i, j to find a spatial similar region k of the prediction region; andfilling a vacancy value yi, t of the time sequence of the prediction region; 4) performing traffic prediction on the time sequence of the prediction region obtained in the step 3) and finishing user activity anomaly detection and traffic prediction based on big data analysis. The method can realize accurate traffic prediction of the prediction region.

The present invention provides a trend prediction analysis method, equipment and storage medium. The method includes: preset analysis target range, time interval and cycle; obtain preset target market data; count the trends of each cycle in the time interval The number of targets, calculate the trend probability of the corresponding period according to the number of targets of each trend in each cycle, the trend includes rising, flat, falling; generate the trend probability curve in the time interval, and predict the next cycle according to the trend probability of the current cycle Periodic trend; displays the trend probability curve over the time interval. The present invention predicts the trend of the next stage through the current trend probability, and can obtain the inflection point of trend transition at the first time, avoiding the delay of information; introducing a probability smoothing algorithm, smoothing the trend probability can reduce probability glitches, making the probability stable and continuous distribution, which can predict a period of future market more stably and provide accurate probability.

The invention discloses a landslide prediction method. The method comprises the steps of 1, collecting and calculating control factors; 2, collecting range and time information of a historical landslide area; 3, generating landslide curves according to the collected historical landslide area and control factors; 4, clustering the landslide curves to form one or more landslide models; 5, generatinglandslide curves for control factors of a prediction area, comparing the landslide curves with all the clustered landslide curves in the landslide models, and calculating the Euclidean distances between the centers of mass of the landslide curves of the prediction area and the center of mass of each clustered landslide curve; 6, obtaining a landslide probability-distance relationship curve according to the Euclidean distances; 7, obtaining a landslide probability curve according to the landslide probability-distance relationship curve. By means of the landslide prediction method, the technical problems of a prediction method with inner factor research as a main part for prediction and monitoring as a main part for forecasting in the prior art are solved.

The invention discloses an ecological risk assessment method of a polycyclic aromatic hydrocarbon compound and a mixture thereof in a marine water environment. The assessment method comprises the following steps of determining the type and concentration of the polycyclic aromatic hydrocarbon in the marine water environment; screening the representative biological species having the flora characteristics of the creatures of the China sea area; acquiring the chronic toxicity data of polycyclic aromatic hydrocarbon in the water body of a target sea area; using a statistical extrapolation method to calculate the predicted no effect concentration of polycyclic aromatic hydrocarbon; using a joint probability curve method to calculate the level of the ecological risk of single polycyclic aromatic hydrocarbon in the water body of the target sea area; using a toxicity equivalent factor method to calculate the joint ecological risk of the polycyclic aromatic hydrocarbon mixture in the water body of the target sea area; identifying the risk of the monomer and the mixture of the polycyclic aromatic hydrocarbon according to the above results. The ecological risk assessment method of the polycyclic aromatic hydrocarbon compound and the mixture thereof in the marine water environment provided by the invention realizes the ecological risk assessment of the polycyclic aromatic hydrocarbon monomer and the mixture thereof in the China marine water environment by using the toxicity equivalent and the probability risk assessment method based on the species sensitivity distribution theory, and provides a decision basis for the marine ecological risk management of the China organic pollutants.

An ordered charging scheduling method based on electric vehicle charging load prediction comprises the following steps: A) calculating an initial state-of-charge probability curve when an electric vehicle is charged, and obtaining a time length required by charging in combination with the charging power of the electric vehicle; B) obtaining a daily charging load curve of the electric vehicle; C) taking the minimum charging cost of the user side and the minimum load peak-valley difference of the power grid side as objective functions, and establishing an electric vehicle ordered charging optimization model in the district; D) calculating the optimal solution of the objective function of the orderly charging optimization model of the electric vehicles in the area under the constraint condition, and taking the optimal solution as a charging scheduling method. The invention has the substantive effect that the electric vehicle charging load prediction model is established on the basis of detailed research on the charging load characteristics of various electric vehicles. The minimum charging cost of the user side and the minimum load peak-valley difference of the power grid side are taken as an objective function, an electric vehicle ordered charging optimization model is established in the district, and the objective function is solved to obtain the optimal ordered charging scheduling method.

The invention provides a method and a device for recovering lost users, an electronic device and a computer-readable storage medium, belonging to the technical field of data processing. The method includes acquiring characteristic data of one or more indicators of a target user; Calculating a viscosity coefficient of the target user according to the feature data; Generating a backflow probabilitycurve of the target user through the viscosity coefficient, wherein the abscissa of the backflow probability curve is a drain time and the ordinate is a backflow probability; Finding a drain time interval corresponding to a preset expected recovery probability interval in the return probability curve, and determining the drain time interval as a target time interval for implementing a recovery measure; The remedial measures are implemented within a target time period within which the remedial measures are to be implemented. The present disclosure provides a method for analyzing the return probability of lost users and taking remedial measures based on the analysis results, which can effectively save the lost users and improve the vitality of a website or App.

The embodiment of the invention discloses a method for pushing program information. The method is applied to a server-side, and comprises the following steps of: obtaining a corresponding skip probability curve when each user watches each program; arbitrarily selecting two users who watch a first program, comparing whether the similarity of the skip probability curves is higher than a preset first threshold value or not when a first user and a second user watch the first program; if higher, sending program information which is watched by the first user and of the same kind with the first program to a client-side of the second user; and / or sending the program information which is watched by the second user and of the same kind with the first program to the client-side of the first user. By applying the embodiment of the invention, the precision for pushing the program information is improved.

The invention relates to a method and system for determining clustering parameters for generating map lane lines, and the method comprises the steps: 1, setting the value minPts of the minimum numberof core points in a radius region, and determining a distance array from each line segment in the lane lines for clustering to the nearest kth line segment; setp 2, converting the distribution of thedistance array from each line segment to the nearest kth line segment into a cumulative probability curve; and step 3, determining the value eps of the radius parameter of the density cluster as the distance corresponding to the inflection point of the cumulative probability curve or the median of the distance array. According to the method, the density clustering radius parameter is determined according to the distance between the lane line center line segments used for clustering, the determination process can be automatically determined through the system, automatic parameter adjustment inthe map lane line clustering process is achieved, and a good result can be obtained without manual intervention.

The invention belongs to the technical field of mathematics teaching and particularly relates to a teaching optimization method applying mathematical probability events. The method comprises steps that for preparation, two types of balls having the same sizes and materials and different colors are prepared; the balls are disposed into a demonstration device, the demonstration device is fixed to adamping platform, and the control wind speed is controlled and does not exceed 0.05 m / s; a pressure sensor is disposed on a push head of the demonstration device, the push head is pressed, when a pressure value reaches a predetermined value, the push head is released after 3s pause under the pressure force, after the push head is automatically rebounded, after 3s pause, the push head is pressed again; a camera is placed in a laboratory, the camera is connected with a computer, and the computer is connected with projectors of multiple classrooms; the quantity of balls falling into a first collection frame and a second collection frame is separately counted by setting a video counter, and respective probabilities are calculated by the computer; the quantity of balls is changed, the above steps are repeated, and the probability curve is drafted by the computer. The method is advantaged in that multiple sets of experiments can be simultaneously accomplished, teaching efficiency is high, and strong contrast property is achieved.

The invention discloses a voltage sag loss assessment method considering protection action time. The method comprises the following steps: assessing a protection measurement current of equipment; determining the action time of a protection device; establishing a sensitive load voltage tolerance curve model VTC containing action time and dividing the sensitive load voltage tolerance curve model VTCinto uncertain areas; establishing a corresponding random probability density function; acquiring voltage sag historical data of a line, preprocessing the voltage sag historical data, and performingcalculating to obtain the accumulated time length of direct shutdown of the line and the direct economic loss; establishing a cumulative probability curve CPSD of the line voltage sag duration; establishing an equipment voltage sag sensitivity model and an equipment trip probability evaluation model; establishing a production line voltage sag sensitivity model and a line trip probability evaluation model, and obtaining indirect economic loss; and performing calculating to obtain the total economic loss. According to the method, the action time of the protection device is considered, the influence of voltage sag on equipment shutdown is distinguished, historical voltage sag data analysis is combined, and the accuracy of loss assessment is improved.

The invention discloses an intelligent substation reactive power optimization configuration method based on an improved optimal covering method. The method includes the steps that firstly, an annual reactive load curve is generated according to historical reactive data of an intelligent substation, and a reactive load probability curve is acquired based on the annual reactive load curve and a probability distribution principle; secondly, a region formed by the reactive load probability curve and a coordinate axis is covered with a capacitor bank and a dynamic reactive power compensation device; thirdly, an objective function is acquired by taking minimum of a mismatch area function and an investment cost function into consideration; fourthly, an optimal configuration scheme is acquired through a genetic algorithm according to the objective function. The purpose of coordinated configuration of the dynamic reactive power compensation device and the capacitor bank under the new reactive power compensation situation of the intelligent substation is effectively achieved, reactive power configuration is optimized from the probability aspect, minimum control errors are accumulated in the probability, the equipment utilization probability is increased, and the method has theoretical and guiding significance in the new construction situation of an intelligent power grid.

The invention discloses withstand explosion detection device and method for a power capacitor. The detection device comprises a photoelectric isolation system, a data acquisition system, a data processing and analyzing system and a man-machine interaction display system. The photoelectric isolation system sends received data after electric isolation to the data acquisition system which is composed of an attenuation unit and an acquisition device, the data sent by the photoelectric isolation system is sent to the acquisition device through an attenuator by configuring coefficient of the attenuator, an output end of the data acquisition system is connected to an input end of the data processing and analyzing system serving as a core component to receive data of the data acquisition system and complete analytic calculation, and an output end of the data processing and analyzing system is connected to a man-machine interaction interface which is responsible for configuration, control and detection result output of a test task. The withstand explosion detection device is used for detecting the characteristics of withstand explosion of the capacitor, and a withstand explosion probability curve of a shell of the capacitor can be drawn conveniently.

The invention discloses an accelerated test and prediction method for the service life of a laser optical component so that a problem that the service life of the large-caliber optical component underrandom flux illumination can not be obtained in the prior art can be solved. The method comprises: fitting is carried out by using a transmitting number or time as an independent variable and a corresponding service life probability as a dependent variable to obtain a service life probability function of a large-caliber optical component under a specific laser parameter; and the irradiation fluxof illuminating laser with a set parameter is changed to obtain a service life probability curve of the large-caliber optical component under different irradiation fluxes and data fitting is carried out on the function to obtain a service life flux function under the specific service life probability. According to the invention, accelerated testing of the service life of the laser optical component is realized; the test time is shortened. With the test result, the service life of the large-caliber optical component under the specific flux can be obtained; and on the basis of the classical reliability theory, the reliability analysis and prediction are carried out on the laser large-caliber optical component in the laser system. The accelerated test and prediction method is suitable for theservice life testing field of the large-caliber optical component.

Appointment reminder method for customer no-show behavior: first determine the customer’s service reservation request rule; according to historical data, fit the customer’s no-appointment probability curve to determine the customer’s no-appointment probability; pre-determine a service appointment opening time window T; determine the appointment rule; Discretization processing of reservation opening time window; determine the state of the reservation system; analyze the state transition law of the reservation system and the expected service utilization rate of different states at different stages; reversely recursively deduce the expected service resource utilization rate of each stage and state; clarify Optimal reminder time; determine the optimal appointment opening time window. The present invention treats the appointment reminder time differently according to different appointment lead times, so as to achieve the purpose of refined management; adopts discretization to process the appointment time window, and combines the dynamic programming method to determine the optimal reminder time; and gives the best appointment opening time window The determination method to maximize the utilization of service resources.

The invention discloses a power quality disturbance identification method and system based on a variable recurrent neural network, and a medium. The method comprises the following steps: extracting power quality disturbance waveform data characteristics, inputting the characteristics into a pre-trained variable recurrent neural network, and generating an output probability curve corresponding to Ndisturbances through an activation function; wherein the variable recurrent neural network comprises a random layer used for introducing a potential random variable to the output layer, and finally determining the disturbance type at a certain moment and / or the start and stop moments of a certain disturbance according to the output probability curve. According to the invention, the invention canachieve the recognition and classification of power quality disturbance types at any time, can accurately recognize the start and stop time of disturbance, and can achieve the multi-node, quick and accurate power quality disturbance recognition after the algorithm is introduced into a common household electric meter.

The embodiment of the invention discloses an eye behavior detection method and device and a storage medium. The method includes: sliding a time window on a video frame of the target video, obtaining an image sequence of the target object, wherein the image sequence comprises a plurality of images with a time sequence relationship; carrying out feature extraction on the eye region of each image; obtaining initial feature information of a plurality of eye regions with a time sequence relationship; according to a time sequence relationship of the eye region, fusing the initial feature informationand the difference feature information of the eye region; obtaining a plurality of pieces of fusion depth feature information with a time sequence relationship; according to a plurality of pieces offusion depth feature information with a time sequence relationship, obtaining probability values of eye behaviors of each image as preset eye behaviors, wherein the g probability values have a time sequence relationship; generating a probability curve coordinate system according to the plurality of probability values with the time sequence relationship, and determining whether a target object hasthe preset eye behaviors or not according to the probability curve coordinate system. According to the invention, the accuracy of eye behavior detection can be improved.

The invention discloses a rotor and stator subsystem rub-impact reliability analysis method based on an agent model. The method comprises the steps of 1, establishing a finite element simulation model of rotor and stator rub-impact; 2, finding out uncertain parameters existing in a design stage, processing, assembling and working operation links, determining main uncertain sources influencing rotor and stator rub-impact and probability distribution types of the parameters through sensitivity analysis, and inputting the parameters; 3, adding carefully selected samples near the curved surface based on the limit state by utilizing a self-adaptive experimental design algorithm, and constructing an agent model to replace a finite element simulation model; 4, predicting the function response of the MCS sample point at a given initial rotating speed by using the trained agent model, and calculating the failure probability according to the function response result; and 5, iteratively increasing the rotating speed within the considered rotating speed range, repeating the step 4, sequentially calculating the failure probability at each rotating speed, and finally obtaining a failure probability curve.

The invention discloses a test evaluation method for an engine super knock safety margin. The test evaluation method for the engine super knock safety margin comprises the following steps that (1) detonation pressure data are collected by monitoring super knock in an engine bench test to determine a probability curve of the distribution of the super knock detonation pressure of an engine; (2) By monitoring the occurrence of super knock in an endurance test of a whole vehicle, the occurrence frequency of super knock is counted so as to determine the occurrence frequency of the super knock undera certain endurance mileage of the whole vehicle; and (3) the failure probability of parts due to over-explosion under the certain mileage of the whole vehicle is determined through detonation pressure prevention strength values of parts and the probability curve of the distribution of the over-explosion detonation pressure of a bench. Design evaluation is provided for part design by scientific methods, and part failures caused by the super knock can be greatly reduced.