30 results about "Trend extraction" patented technology

The invention is directed generally to providing methods and systems for trend extraction and analysis. Embodiments include methods and systems for trend extraction and analysis of information extracted from dynamically changing data included in computer systems and/or networks. Various exemplary embodiments are provided that may generate characteristic indicators for trend(s) and/or distribution(s) for one or more data sources by use of, for example, temporal indicators derived through analysis of the difference in contribution separate portions of the data to the whole data set being considered, contribution of individual sources, and/or the interaction of the separate portions of the data with one another. Some exemplary approaches may include the use of singular value decomposition (SVD) and higher-order singular value decomposition (HOSVD) data extraction and analysis techniques. One use of these techniques is in the analysis of the dynamic data contained in Weblogs and the blogosphere.

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 relates to a sea level change nonlinear trend extraction method, which comprises the following steps of decomposing a sea level change time sequence by utilizing an empirical mode decomposition method to obtain an intrinsic mode function and a trend term with different frequencies; eliminating a low-frequency false component by using a modal function method, and marking and eliminating a high-frequency noise component by using a frequency divergence method; sequentially carrying out frequency spectrum analysis by adopting a Fourier method to obtain frequencies corresponding to the peak values in each component frequency spectrogram, removing the invalid frequencies according to a Nyquist theorem, and converting the frequencies into periods; obtaining all possible embedded calculation windows, through the singular spectrum analysis, taking the trend term of empirical mode decomposition as a reference, and selecting the trend term with the minimum difference as the final nonlinear trend of the sea level change. According to the method, the automatic selection of the optimal window and the automatic extraction of the optimal sea level change trend can be realized, the adaptability is good, the efficiency is high, the trend extraction is stable, and the influence of the time sequence length is small.

The invention relates to an equipment performance degradation trend extraction and prediction method. The method comprises the following steps: (1) establishing a hydroelectric generating set standardhealth model which comprehensively considers active power and a working head coupling effect and is based on inverse distance weighting, and obtaining a performance degradation time sequence of a current hydroelectric generating set according to acquired real-time online data of the hydroelectric generating set; (2) decomposing the performance degradation time sequence of the hydroelectric generating set into a sum of a plurality of stable PRC time sequences and a trend component by utilizing ITD; (3) performing complexity characteristic identification on all the obtained component time sequences, and reconstructing the components according to a preset complexity characteristic threshold to obtain a performance degradation trend; and (4) evaluating, judging and predicting the performanceof the hydroelectric generating set equipment according to the obtained performance degradation trend, and finding out equipment abnormity in time. The method can be widely applied to the field of hydroelectric generating set equipment performance degradation trend prediction.

The invention discloses a time sequence trend extraction and prediction method based on compressed sensing, and belongs to the field of time sequence prediction methods. According to the method, random sub-sampling is carried out on an original time sequence to obtain an observation vector, and then trend extraction is carried out on the original time sequence through multiple times of reconstruction under the condition that the sparsity of a reconstructed signal is determined. An optimal trend sequence is found in the plurality of reconstruction trends by utilizing a similarity evaluation index based on the Euclidean distance sequence, and future trend development of the optimal trend is predicted by using a support vector regression prediction method. The trend information of the original time sequence can be well extracted only by utilizing the observation vector of random sub-sampling of the original time sequence, and the invention has a certain compression ratio and is more suitable for being used under the condition of big data mining processing. The sequence trend information is predicted, so that the problems of long-term prediction, noise interference and the like of thetime sequence trend can be effectively solved, and the result is more scientific and effective.

The invention provides a thermal power generating unit primary frequency modulation index calculation method and a thermal power generating unit primary frequency modulation index calculation system based on trend extraction. The method comprises the steps: searching the actual power and power grid frequency historical data of a thermal power generating unit according to the characteristic quantized value of a thermal power generating unit primary frequency modulation data segment, and obtaining the thermal power generating unit primary frequency modulation data segment; trend extraction is carried out on the obtained primary frequency modulation data segment of the thermal power generating unit; on the basis of the trend of each sub-data segment, determining the sub-data segment of whichthe trend change direction is opposite to the frequency change direction and the amplitude change is maximum as a primary frequency modulation action segment; and on the basis of the amplitude variation of the primary frequency modulation action section and the sample timestamp, obtaining parameter values required by calculation of the primary frequency modulation performance index of the thermalpower generating unit, and calculating the primary frequency modulation performance index of the thermal power generating unit. The method can effectively overcome the problems of a current thermal power generating unit primary frequency modulation performance index calculation method, and is of great significance in promoting the primary frequency modulation performance assessment of the thermalpower generating unit by the power grid, improving the stable operation level of the power grid and large-scale new energy consumption.

According to the mobile high-definition video intelligent bidirectional detection bandwidth control method provided by the invention, a target end frame-level time delay trend extraction method and an extended RTP/RTCP driven source end rapid adaptive method are fused, so that the accuracy of the target end frame-level time delay trend extraction method is reserved; according to the method, the coding rate can fully use the available video bandwidth of the link to achieve the best subjective effect, meanwhile, the characteristic of high speed of the RTP/RTCP is fused, the packet loss phenomenon before congestion is detected due to the fact that the link cache of mobile high-definition video communication is too small is avoided, the experience of the mobile high-definition video is improved. A video bandwidth control architecture is provided on the basis, so that the video bandwidth control architecture has a better video effect for a fluctuating mobile network, is high in feasibility in practical application, is simple, efficient and high in practicability, serves as the most important link for guaranteeing the quality of a mobile high-definition video, and plays an extremely important role in mobile high-definition video communication.

The invention aims to provide a trend automatic extraction method based on a data block. The method comprises the following steps: firstly, collecting axle temperature data of a certain train as original data, and then drawing an original data curve; searching for an optimal segmentation value of the original data, extracting a data envelope center line, and searching for a local extreme point ofthe data envelope center line; and finally, dividing the data into data blocks according to the obtained local extreme points, and performing trend extraction analysis in each data block to provide atheoretical basis for further predicting axle temperature change, judging train faults and the like. The method solves the problems that in the prior art, the trend is extracted after data processingmethods such as filtering and smoothing are adopted for the collected data, the processed data result directly influences the trend result, the data smoothness degree is difficult to determine, and data features are lost during excessive smoothing.

The invention provides a bearing fault early warning method and device, and the method comprises the steps: obtaining a plurality of first bearing temperatures of a bearing in a recent time period, and a first working state parameter combination, corresponding to each first bearing temperature, of rotating equipment, and enabling each first working state parameter combination to comprise at leastone first working parameter; Carrying out trend extraction on the plurality of first bearing temperatures to obtain a second bearing temperature, and carrying out trend extraction on each first working state parameter combination to obtain a second working state parameter combination, the second working state parameter combination comprising at least one second working parameter; Processing the second working state parameter combination by adopting a regression model of a convergence state, and outputting a third bearing temperature; And determining whether the bearing has a fault according tothe second bearing temperature and the third bearing temperature. According to the scheme, the fault symptom of the bearing can be early found, so that necessary guarantee measures can be taken in advance, and more serious faults are avoided.

The invention discloses a degradation characteristic trend extraction method of a brushless DC motor. The vibration signals in the operation process of the brushless DC motor are extracted by an acceleration sensor, the vibration acceleration signals act as the motor performance parameters, and the change trend of the vibration acceleration signals is analyzed by the full frequency domain energy mean so as to represent the degradation process of the brushless DC motor. The method comprises the following steps of 1) installation of the tested brushless DC motor and test equipment; 2) acquisition of the vibration signals; 3) EMD decomposition; 4) calculation of the marginal energy spectrum; 5) calculation of the full frequency domain energy spectrum mean; and 6) generation of the degradation characteristic trend.

The invention provides a data analysis method and apparatus, an electronic device and a readable storage medium, and relates to the technical field of information processing, the method comprises thesteps of obtaining access data, and preprocessing the access data; inputting the preprocessed access data into a trend extraction filter to obtain trend data; and analyzing the access data according to the trend data. The method comprises the following steps: preprocessing acquired access data to filter out invalid access data; and then extracting trend data which comprises trend components and noise components from the preprocessed access data based on time by utilizing a trend extraction filter, thereby ensuring that the access data can be accurately analyzed according to the trend data, andfurther ensuring that the access data is effectively evaluated.

The invention provides a hydropower station equipment real-time monitoring data monitoring system and monitoring method based on trend alarm, and belongs to the technical field of hydropower station monitoring. The method comprises the following steps: respectively acquiring real-time monitoring data and historical data; constructing a trend alarm model, and obtaining an alarm threshold according to the trend alarm model; inputting the real-time monitoring data into the trend alarm model to obtain a real-time output value; whether the real-time output value is larger than an alarm threshold value or not is judged, when the real-time output value is larger than the alarm threshold value, an alarm signal is output, and monitoring of the real-time monitoring data of the hydropower station equipment is completed. According to the method, early warning of the monitored hydroelectric equipment, evaluation of the state of the monitored hydroelectric equipment and decision support information can be effectively provided from trend extraction, the accuracy and reliability of real-time monitoring of a hydropower station are improved, and the technical effects of saving the labor cost are achieved.

The invention relates to a signal high-frequency oscillation characteristic processing method. The method comprises the following steps: step 1, storing an oscillation solution in a knowledge base; step 2, for a signal to be processed, when a matched oscillation solution is searched in the knowledge base, skipping to step 3, otherwise, skipping to step 4; step 3, processing the to-be-processed signal by using the searched oscillation solution, generating a processing report and skipping to step 6; 4, obtaining a standard waveform, and obtaining the waveform characteristics of the waveform to be processed according to the corresponding standard waveform; step 5, determining an oscillation signal processing method of the to-be-processed signal according to the determined waveform characteristics of the to-be-processed signal; and step 6, ending. According to the method, the authenticity, flexibility, leakproofness and reliability of the waveform analysis process can be improved. Meanwhile, the method adapts to unpredictable oscillation interference of the waveform, and practical and effective trend extraction can be continuously carried out on high-frequency oscillation of which the complexity degree is changed.