276 results about "Regular distribution" patented technology

A method for converging on a route through a radio frequency (RF) communications network, that includes multiple nodes, includes (a) identifying all RF links between each node in the RF communications network; (b) determining a connectivity confidence interval for each RF link by: (aa) developing a calculated signal to noise ratio using a radio frequency communication link propagation loss model; (bb) determining a threshold signal to noise ratio based on a predetermined RF packet completion rate; (cc) determining a standard deviation value based on a signal strength and a noise level at a signal receiving node; (dd) calculating a Z number, associated with a normal distribution table which is based on the threshold signal to noise ratio minus the calculated signal to noise ratio with that result being divided by the standard deviation value; and (ee) assigning a CCI probability value based on said Z number. The method further comprises (c) setting a predetermined connectivity confidence interval minimum; (d) comparing each connectivity confidence interval for each RF link to the predetermined connectivity confidence interval minimum to determine which are greater than or equal to the predetermined connectivity confidence interval; (e) assembling each RF link corresponding to each connectivity confidence interval that is greater than or equal to the predetermined connectivity confidence interval together to identify various routes through the RF network; (f) calculating a resultant connectivity confidence interval for each route by multiplying together each connectivity confidence interval for each RF link of each particular route; and (g) comparing each resultant connectivity confidence interval to identify the route with the greatest resultant connectivity confidence interval. An apparatus for carrying out such a method is also presented.

The invention discloses an improved RRT<*> obstacle avoidance motion planning method based on a multi-degree-of-freedom mechanical arm, and belongs to the field of mechanical arm motion planning. A six-degree-of-freedom mechanical arm model with seven connecting rods and six rotary joints is built; parameters in a to-be-searched space are determined; if the distance is shorter than the distance of a path with lowest cost, the distances between a near node in a set to an initial point and the distance between the node to a random point are temporarily determined as the minimum path; a newly generated sigma is subjected to collision detection, and the node and the path are added if the newly generated path does not collide an obstacle interval; the steps are repeated until the optimal path is found; and the generated path is added into a path planning device. Compared with the prior art, the method has the following advantages that the random search characteristic is changed in a mode of adding normal distribution, the algorithm convergence rate can be increased through the heuristic search, the RRT<*> algorithm has the evolutionary optimization path, and a large number of calculations is not needed; and after Gaussian distribution of an inspiration point near a target point is added, the convergence rate is increased, and the search time is shortened.

The invention provides a sensor for nano gold particles and a preparation method thereof. The end surface of a multi-core optical fiber is of a conical-platform structure; a total-reflection film is plated on the surface of the conical platform; the nano gold particles which are distributed regularly are fixed on the end surface of the optical fiber plated with the total-reflection film; exciting light is injected into one fiber core of the multi-core optical fiber, is reflected to the end surface of the optical fiber at the film-plated position of the conical platform and generates total internal reflection on the end surface of the optical fiber, and a generated evanescent field excites a localized surface plasmon resonance effect of the nano gold particles; the reflected light is collected by the fiber core symmetrical to the fiber core injected with the exciting light, and the change of the physical quantity of external substances is sensed by the spectrum of the reflecting light. The sensor and the preparation method have the advantages that the multi-core optical fiber, a self-assembly technology of a near-field optical tweezer and the localized surface plasmon resonance effect of the nano gold particles are combined, and the near-field optical tweezer of the multi-core optical fiber can be utilized for capturing the nano gold particles, so that the optical self-assembly and regular distribution is carried out on the nano gold particles according to the distribution rule of the capturing areas; the structure is simple, the volume is smaller and the repeatability is high.

The embodiment of the invention provides a training sample data expansion method and device based on a variational auto-encoder, and relates to the technical field of big data. The method comprises the steps of obtaining an original sample; inputting the original sample into the encoder of the variational autoencoder, wherein the encoder of the variational autoencoder comprises two neural networks, the two neural networks output Mu and sigma respectively, and Mu and sigma are both functions of the original sample; according to the square of the Mu and sigma, namely sigma 2, generating a randomnumber of corresponding Gaussian distribution; randomly sampling the standard normal distribution to obtain a sampling value epsilon, and determining a sampling variable Z according to the sampling value epsilon and the random number of the Gaussian distribution; and inputting the sampling variable Z into a decoder of the variational autoencoder, decoding the sampling variable Z by the decoder ofthe variational autoencoder, and then outputting similar samples of the original samples, and taking the similar samples as extended samples. Therefore, the technical scheme provided by the embodiment of the invention can solve the problems that the time and labor are wasted and the efficiency is low when sample data is manually expanded in the prior art.

The invention discloses a process monitoring method which is based on non-Gaussian component extraction and support vector description. The method comprises the following steps: read-in of training data and data to be diagnosed, data preprocessing, establishment of a principal component analysis model, particle swarm optimization algorithm, non-Gaussian projection calculation, support vector data description, residual analysis, principal component estimation, fault detection and the model updating. By the method, the non-Gaussian components can be automatically extracted from operating data of an industrial process, thus avoiding the disadvantage that the conventional statistical process monitoring method assumes that data is subject to normal distribution, and the non-Gaussian projection algorithm based on the particle swarm optimization algorithm ensures the maximization of the non-Gaussian properties of the extracted independent components, and avoids the problem that the independent component analysis method is easy to be involved in the locally optimal solution. Compared with the conventional statistical process monitoring method, the method can find abnormity in time, effectively reduce the rate of false alarm, and obtain better monitoring effect.

The invention discloses a method for evaluating a reservoir of a fractured well. The method comprises the following steps: S1, establishing an evaluation database M of the reservoir, wherein M is composed of a parameter set X, a sample set U and an index set Y; S2, dividing all the parameters into three kinds, and establishing a three-stage evaluation system including an index layer, an element layer and a target layer; S3, calculating the correlation between each parameter and open-flow capacity in the sample set U, and establishing a single-factor evaluation criterion set V; S4, calculating the influence weight W of each parameter to fracture improvement effect and the influence weight K of each kind of parameters to the fracture improvement effect in the sample set U by using a grey correlation method; S5, calculating and evaluating the membership degree R of each parameter in the object X in four different stages in the single-factor evaluation criterion set V by using normal-distribution membership functions; and S6, carrying out blurring operation on the weight W and the membership degree R, and grading quantifiably and comprehensively. The method for evaluating the reservoir of the fractured well, which is disclosed by the invention, has high objectivity and accuracy; a novel decision method is provided for selecting a fractured improvement well.

The invention discloses an adaptive Kalman filtering method oriented for WiFi/PDR indoor integrated positioning. The method comprises: first, holding by a user a terminal device in a targeted area to receive the RSSI from each AP; in a lognormal distribution path loss model, using the weighted least squares method to obtain the location of the user and at the same time, through the estimation model in the PDR algorithm, obtaining the location of the user; then using the adaptive Kalman filtering to integrate the positioning information based on the propagation model with the positioning information from PDR for multiple times to obtain the best location of the user, wherein the adaptive Kalman filtering is reflected in the feedback mechanism, that is, each integration positioning result dynamically corrects the path loss index and other parameters of the weighted least squares method so as to ultimately make the propagation model more in line with the indoor environment. The method solves the problems of low positioning accuracy of only using the WiFi and of the cumulative errors existing in PDR under an indoor environment, and the method also can track the path loss index of the propagation model in real time and enhances the stability of positioning performance.

The invention relates to an integrating method of three-dimensional spatial distribution vortex arrays, and belongs to the field of diffraction optics. A pure-phase code of a two-dimensional uniform-strength vortex optical grating and a pure-phase code of a spiral Dammam wave zone plate are obtained through analytic operation and an optimization algorithm respectively, phase information of the pure-phase codes is overlaid, a lens factor is introduced, discretized phase values are loaded to a spatial light modulator according to the pixel number and the pixel size of the spatial light modulator, and then the corresponding three-dimensional vortex arrays can be obtained. According to the method, the equal-strength and regular-distribution three-dimensional vortex arrays can be obtained, and each vortex beam in the arrays has the specific topological charge number.

An identical object determination method of determining whether or not objects are detected by a plurality of sensors identical has a determining a relative position and speed thereof, a determining a position error thereof based on the determined positions and calculating a first probability that the detected objects are identical with respect to the position from the determined position error based on a normal distribution concerning the position error, a determining a speed error thereof based on the determined speed and calculating a second probability that the detected objects are identical with respect to the speed from the determined speed error based on a normal distribution concerning the speed error, a calculating a third probability that the detected objects are identical based on the calculated first probability and the calculated second probability, and a determining that the detected objects are identical if the third probability exceeds a first determination value.

The invention discloses a method of calculating exponential life-limited spares demand; the method mainly comprises the steps of simulating life of a life-limited spare to produce life random numbers ti of not less than 1000 life-limited spares, wherein the life random numbers ti comply with exponential distribution exp(Mu0); simulating service lives of life-limited spares; traversing and correcting the random numbers ti to obtain ti; subjecting the random numbers ti of step 2 to normal distribution fitting calculation, with the calculation result marked as N(Mu, Sigma2); calculating spares support probability P within support task time when the number of the life-limited spares is j; setting a threshold of the support probability so that a number S of the life-limited spares, satisfied when the support probability is greater than or equal to the above threshold, is the calculated demand for life-limited spares. The method of calculating exponential life-limited spares demand implemented according to the invention allows the demand for exponential life-limited spares to be calculated accurately.

The invention provides a numerical data compression method for a column storage database, which includes: extracting one column of data from a data table, dividing each column of data into multiple subsidiary columns, counting characteristic values of data distribution laws in each subsidiary column, and compressing by selecting the corresponding lightweight compression algorithm according to the characteristic values of the data distribution laws, wherein different data in each subsidiary column occupies identical space. The numerical data compression method is simple in operation and applicable to compression of data of regular distribution and irregular distribution.

The invention discloses a particle-swarm-based rapid optimization deployment method for wireless network sensors. The method is capable of improving a global optimization effect of a particle swarm optimization algorithm (PSO) and reducing particle resource use and greatly speeding up a solution speed. The method is as follows: using a feasible solution of each kind of deployment of the wireless network sensors as a particle to establish a particle swarm and initiating the speed and position of each particle and setting an iteration number and using the PSO to perform iterative search on the particles, wherein a disturbance term is added when the speed of the particles is updated; after the iterative search is performed for the set iteration times, obtaining a final optimization result so as to realize particle swarm optimization; the PSO using an effective coverage rate of a wireless network as a fitness; and the disturbance being the product of a disturbance amplitude and a random number selected through standard normal distribution and increasing or decreasing the proportion of the disturbance term through changing the disturbance amplitude. The particle-swarm-based rapid optimization deployment method for the wireless network sensors is suitable for the PSO to obtain a comparatively satisfactory solution under a condition that the number of particles is significantly small.

The invention relates to a method for inspecting abnormal data and an inspection device applying the method. The method comprises the following steps: performing sequencing on unfiltered data samples according to the size; selecting a proper percentile in the data samples, and calculating a percentile value corresponding to the percentile; and taking the calculation result as a numerical screening limit value of the data samples, and taking data, of which the value is more than the upper limit value of numerical screening or less than the lower limit value of numerical screening, in the data samples as abnormal data points. The invention improves the inspection precision by improving the method for inspecting the abnormal data, is favorable for improving the yield of products and reducing waste plates and the production cost, realizes the inspection on abnormally distributed type data which cannot be realized in the prior art, and improves the production efficiency.

The present invention relates to a system and method for tuning demand coefficients. Transaction data for product categories is received from a store(s). Price elasticity and uncertainty values are selected for the product categories. This transaction data may be seeded with generic price elasticity and uncertainty values. Product categories where the transaction history is not sufficient enough to generate accurate demand coefficients may be identified. Tuning parameters for a product category are estimated using price elasticity and uncertainty values. The tuning parameters include price elasticity mean and price elasticity standard deviation. A modified likelihood function is generated by applying a normally distributed price elasticity term. The modified likelihood function may then be solved for its maxima, thereby generating tuned demand coefficients which may be output to a pricing optimization system for product price setting, and/or may be stored for later product categories. New sales data may be received from the store(s). This data may be used to retrain the tuned demand coefficients.

The invention belongs to the technical field of the design, manufacturing and processing of strengthened and toughened mechanical parts, relates to a method for controllably improving the strength and toughness of parts and specifically relates to a method for the bionic coupling of strengthened and toughened mechanical parts. The method is characterized in that bionic coupling unit bodies formed by coupling bionic morphology coupling elements with strengthened structure coupling elements are distributed on the surfaces of the parts; and due to the regular distribution of the bionic coupling unit bodies and the strengthened structure superior to the raw material, the strength and toughness of the parts can be greatly and controllably adjusted after the parts are treated by using the method, thereby effectively improving the properties of the material, and greatly prolonging the service life of the parts. Meanwhile, the invention breaks through the traditional idea of modifying the whole material or whole surface, and utilizes laser, electric pulses and other high-energy beams as processing means to carry out local treatment on parts based on bionic principles, thereby having the great advantages of low preparation cost, high cost performance, no pollution, environmental protection and the like.

The invention relates to a method and a system for modeling I-UWB (Impulse-Ultra Wide Band) signal multipath transmission characteristics in an indoor visual range environment. The method comprises the following steps: dividing the I-UWB multipath signals into two confirmed clusters, each of which is provided with random arrival multipath rays; inputting the confirmed parameter which includes a multipath gain Alpha 0,0 of the initial ray of the first cluster, a multipath gain Alpha 0,1 of the initial ray of the second cluster, and the relative time delay Tm of the first and the second clusters; inputting the statistical parameters which includes the multipath arrival factor Lambada, the damped exponential Gamma of the multipath mean power and the logarithmic normal distribution standard deviation Sigma of the multipath gain of all the multipath rays which arrive subsequently in each cluster; initializing the functions of logarithmic normal distribution and exponential distribution; generating the first cluster and the second cluster; sorting the arrived multipath rays by the time sequence; and outputting the channel impulse response. The invention realizes the simplified I-UWB signal multipath transmission model, which has higher forecasting accuracy and is suitable for the indoor visual range environment without a plurality of environment topological data.

The invention discloses a video transmission-oriented network bandwidth detection and prediction method and a video transmission-oriented network bandwidth detection and prediction system. The network bandwidth detection and prediction method comprises the following steps of calculating a current available network bandwidth according to a media stream length and receiving time; and predicting an available bandwidth at the next moment according to network bandwidths in the first N time intervals by utilizing a normal-distribution-based bandwidth prediction model. According to the method and the system, the bandwidth is measured in a way of adding information to a media stream, so that the shortcoming of requirements on the transmission of a great number of test data packets in the conventional method is overcome, the available bandwidth of a network can be quickly acquired under the condition of no influence on multimedia transmission, and the method and the system are high in practicability; and the normal-distribution-based bandwidth prediction model is also used for quickly and accurately predicting the available bandwidth of the network at the next moment according to measured data, and quick regulation and accurate prediction can be realized even under the condition of greater bandwidth changes.

The invention relates to a method for fully utilizing wind and solar energy and substantially improving the efficiency of solar salt, pertaining to the technical field of ecological engineering. The invention comprises a water evaporation system, a suspension device of the evaporation system, a solution regular transmission system and a solution regular distribution system. By adopting the invention, full and sustainable use of the wind energy and the solar energy can be realized so as to speed up water evaporation, reduce energy consumption, shorten the cycle of solar salt and greatly improve salt production and the original production and design capacity on the basis of not increasing the area of a saltern.

The invention relates to a fruit image classification method and device based on a neural network and transfer learning. The method comprises the following steps: preprocessing the image to be classified, preprocessing the image, and then combining the SMOTE algorithm for data enhancement; Carrying out BN batch normalization operation on pixel points of the data enhanced image to make the pixel points conform to a normal distribution; The pixel points of the normal distribution are inputted into a fruit image classification model based on neural network and transfer learning, and the classification result of the image is outputted. This method can be used to classify the input images of arbitrary size. The method relies on the classification model of fruit image based on neural network andtransfer learning, which improves the efficiency of recognition and classification, reduces a lot of time cost and has high reliability.

The invention relates to an online virtual measuring method of distribution circuit impedance. The method comprises the following steps of: (1) numbering network topology of a distribution network, power circuit sections and nodes; (2) establishing a circuit section voltage dropout linear equivalent calculation model; (3) configuring measuring devices at the head end and the tail end of the circuit section and collecting M groups of data of voltage and current at the head and the tail ends as sample data; (4) analyzing the sample data and calculating impedance of the circuit section; and (5) judging whether the circuit section impedance residual error is subjected to normal distribution with a mean value of zero and a variance of Sigma2. The equivalent impedance is introduced in the online virtual measuring method disclosed by the invention; the circuit section voltage dropout linear equivalent calculation model is established; the impedance of the circuit section is analyzed and calculated by using mathematical methods such as regression analysis methods or mean value equation solving methods; and the online identification of the circuit parameters is realized. The method has the advantages of simplicity in calculation, high calculation speed and high calculation precision, makes up the shortcomings of the traditional method, is suitable for analysis and calculation of the intelligent distribution network and can guarantee precision and real-time performance of calculation.

The invention provides an advertisement click rate prediction method and device. The method comprises the steps that the number of times of presentation and the number of times of being clicked of an advertisement are counted; according to the number of times of presentation and the number of times of being clicked, the confidence interval of the advertisement click rate is calculated; and according to normal distribution, a value is sampled from the confidence interval and is used as a prediction value of the advertisement click rate. According to the invention, an adverse effect caused by the prediction deviation of the click rate can be solved.

The invention relates to a statistical process control method and a statistical process control device applying the method. The method comprises the following steps: acquiring initial data to be analyzed during production, dividing the initial data into non-continuous distribution data, autocorrelation data, nonnormal distribution data and normal distribution data, determining the control limit according to different data types respectively, and monitoring and/or analyzing the production process according to the determined control limit. The statistical process control method simplifies classification of control charts through data classification, solves the problem of calculating the control limit of limited data points subjected to normal distribution and nonnormal distribution, realizes monitoring of a complex system with a large quantity of parameters which can not be realized in the prior art, saves the data processing time, and improves the work efficiency.