30 results about "Trend extraction" patented technology

A method for two time-scales video stream transmission control is proposed, which mainly includes the following three steps. Firstly, a model-based long time-scale bandwidth trend extraction step is proposed to calculate a current bandwidth through a Transmission Control Protocol (TCP) throughput model, and predict a trend of network bandwidth by using an Exponential Smooth Model (ESM) according to the calculated current bandwidth. Secondly, a short time-scale bandwidth fluctuation detection step is proposed to divide a network status into three categories and according to different network statuses, additively increase or multiplicatively decrease the estimated bandwidth. Thirdly, a target bit rate adjustment step based on two asymmetrical thresholds is proposed to set an up-threshold and a down-threshold of bandwidth to avoid frequently adjusting a target bit rate of an encoder. The method can satisfy the requirements of TCP-friendliness, real-time, and smoothness for the video transmission.

The invention relates to an adaptive online monitoring data trend extraction method. The method combines two methods of combined morphological filter and empirical mode decomposition. First, an adaptive morphological filter structural element is dynamically constructed, and the morphological filter is used to The structural elements are filtered, and then the empirical mode decomposition is performed, and then the trend item is constructed according to the characteristics of the processed data, and finally the trend warning is carried out according to the trend item. The method has high computing performance, strong adaptability and good application effect.

The present invention relates to a cloud-based smart walnut motion pattern recognition method and system, the method comprising the following steps: S601, using the smart walnut internal sensor to collect user motion data in real time; S602, smoothing and preprocessing the constructed time series, and The processed time series extracts the movement trend according to a certain interval; S603, respectively calculates the acceleration and angular acceleration for the three coordinate axis directions, and performs the correlation analysis of the movement trend of the smart walnut; S604, the movement trend correlation of the smart walnut calculated in step S603 Construct the correlation coefficient feature of three coordinate axis direction dimensions according to the result; S605, use the correlation coefficient feature of the three coordinate axis direction dimensions constructed in step S604 to train the action recognition model; S606, perform intelligent walnut action recognition on the newly collected user motion data, Feedback the recognition result to the user. The beneficial effect is to effectively evaluate the exercise time and state of the smart walnut user and remind the user to exercise regularly.

The invention relates to a method for extracting and predicting the degradation trend of equipment performance. Real-time online data to obtain the current performance degradation time series of hydropower units; (2) use ITD to decompose the performance degradation time series of hydropower units into the sum of several stationary PRC time series and a trend component; (3) for all the obtained components Identify the complexity characteristics of the time series, and reconstruct the components to obtain the performance degradation trend according to the preset complexity characteristic threshold; (4) evaluate, judge and predict the performance of the hydroelectric unit equipment according to the obtained performance degradation trend, and use Detect equipment abnormalities in time. The invention can be widely used in the field of predicting the performance degradation trend of hydroelectric unit equipment.

The invention discloses a pressure transformer with a trend extracting function. The pressure transformer comprises a signal conditioning unit, a casing and a display module, wherein the signal conditioning unit is connected and packed in the casing, and the display module is arranged on the outer side of the casing, further connected with the signal conditioning unit and used for displaying parameters. The signal conditioning unit comprises an information storing module and a trend extracting module, wherein the information storing module is used for storing signals and system parameters, the trend extracting module is respectively connected with the information storing module and the display module, the trend extracting module can extract pressure information stored in the information storing module, the trend extracting module can acquire trend pressure signals in a set time quantum through operation, and the display module can display the extracted trend signals. The pressure transformer uses the trend extracting module to extract and operate pressure signals in a period in the information storing module, so that data change trend in the period can be extracted, and analysis on pressure signals is facilitated.

The invention discloses a turnout fault diagnosis method based on qualitative trend analysis. According to the method, first, Gaussian filtering and interval half trend extraction are performed on typical current running signals in the five states of turnout normality, unlocking difficulty, step conversion resistance fault, sawtooth conversion resistance fault and locking difficulty fault, fault trend rules are summarized, and a fault diagnosis knowledge base is established; second, Gaussian filtering and interval half trend extraction are performed on a to-be-diagnosed current running signal, obtained fault trends are matched with the fault trend rules in the fault diagnosis knowledge base, and corresponding matching degrees are obtained; if all the matching degrees are smaller than 0.5, fault reasons are determined through manual analysis; if not all the matching degrees are smaller than 0.5, the fault trend with the maximum matching degree is the turnout state; and finally a fault diagnosis result is output. The method is low in cost, simple in structure and convenient to operate, and the diagnosis result has high credibility.

The invention discloses a traffic flow change trend extraction method based on floating car data and belongs to the technical field of intelligent transportation. The traffic flow change trend extraction method based on the floating car data comprises obtaining historical floating car data of at least three mouths of a road chain and classifying the data according to feature days; performing denoising and smoothing on the historical floating car data; dividing the historical floating car data of every feature days into at least one time frame according to the change trend of a morning peak and an evening peak; performing preliminary clustering on the classified historical floating car data through a K-means clustering method; further clustering the historical floating car data through a controlled interval method or a two-value method according to a coarse granularity expression of traffic information. The traffic flow change trend extraction method based on the floating car data combines the coarse granularity expression of the traffic information, further merges the traffic flow trends on the basis of a K-means clustering method, enables the traffic flow change trend to be more salient, thereby providing more crystal clear references for traffic flow prediction, route planning and induction, route planning and the like.

The invention discloses a method for identifying early overflow layers while drilling in the process of oil and gas drilling, comprising the following steps: S1, selecting the real-time overflow monitoring parameters of mud logging while drilling that can be obtained in advance and can reflect the early stage of overflow; S2, respectively Use trend algorithm and data standardization method to process the selected overflow monitoring parameters; S3. Input the processed overflow monitoring data into the trained factor analysis model for overflow layer identification, and use the numerical value of the overflow while drilling data Qualitative evaluation in the form of the overflow layer identification curve; S4, calculate the dynamic difference and trend feature quantity of the total pool volume data corresponding to the late well depth, and extract the trend of the data. If the calculated dynamic difference of the total pool volume and If the trend characteristic quantity exceeds the set threshold, it is judged that the result of S3 is an overflow layer; S5, repeat steps S2 to S4 until all parameters are processed. The invention can detect and accurately predict the depth of the overflow layer in advance.

The invention discloses a heart rate turbulence tendency extraction method based on a cloud model and a scatter diagram. The heart rate turbulence tendency extraction method based on the cloud model and the scatter diagram comprises the specific steps of (1) collecting HRT samples, (2) drawing the Poincare scatter diagram of RR intervals, dividing coupling points into two types of coupling points, matching the two types of the coupling points with a decelerating cloud and an accelerating cloud, and determining two ortho-state cloud curvilinear equations through a reverse-cloud generator, (3) finding critical points of two types of the coupling points, calculating linear regression slopes section by section, and storing slope vectors, (4) calculating tj_mean when changes of types continuously happen on adjacent k coupling points, and substituting the tj_mean into the two ortho-state cloud curvilinear equations to calculate the degree of membership, (5) comparing a sign yun with a subsection slope before k coupling points, comparing the sign yun and a subsection slope after the k coupling points, and classifying the k coupling points into a front subsection or a later subsection, (6) updating subsection points, calculating the linear regression slopes section by section, and updating the slope vectors, (7) mixing adjacent subsections together when the adjacent subsections are the same in sign of slopes, and repeatedly carrying out the step (6) until adjacent subsections needing to be mixed together do not exist.