68 results about "Linear trend" patented technology

The invention discloses an ARIMA-BP neutral network-based bridge monitoring data prediction method. The method comprises the following steps of 1) building an ARIMA(p,d,q) model of bridge monitoring data; 2) performing monitoring value estimation on a monitoring cycle of existent actual measurement data through the ARIMA(p,d,q) model to obtain an estimated value Ypn; performing monitoring value estimation on an expected prediction cycle of inexistent actual measurement data to obtain an estimated value Ynm; 3) obtaining a residual error of the monitoring data: performing subtraction on an actual value of the monitoring data and the estimated value Ypn of the ARIMA(p,d,q) to obtain a residual error Epn of a prediction result; 4) performing a process for predicting a residual error by a BP neutral network; 5) predicting a residual error in a prediction cycle through the established network to obtain an estimated value Enm; and 6) performing superposition on linear trend information predicted by the ARIMA(p,d,q) and a prediction result of the residual error by the BP network to obtain a combined prediction model result. According to the method, nonlinear information is effectively analyzed and a prediction error is relatively small.

The invention discloses a medium and long term demand forecasting method for tendency and periodicity commodities. The method is suitable for medium and long term forecasting of the commodities which have periodicity, are influenced by trend variation and are sensitive to seasons. On the basis of exponential smoothing, linear trend, seasonal variation and random variation time series are decomposed, full consideration is given to influences of the trend variation and period variation on commodity transaction data, commodity transaction data according with a forecast model are selected according to a commodity data selection principle, corresponding smoothing parameters are determined, various indexes in the forecast model are respectively calculated through a quantitative method, and a predicted value of the commodities in a certain period in the future is obtained. Therefore, scientific judgments are provided for production, manufacturing, storage, marketing and the like of various enterprises in commodity supply chains.

An instrument and method for variable depth tillage is provided. A soil engaging implement has a pair of load cells and at least one strain gauge set mounted thereon. The load cells are used to determine a linear trend of topsoil resistance pressure change with depth as the soil engaging implement is drawn through the soil. The strain gauges are used to measure torque on the soil engaging implement caused by the load transmitted through the at least two load cells as well as the load applied to the point of the soil engaging implement. The linear trend of topsoil resistance pressure change with depth and the torque on the soil engaging implement are then used to determine both measured and predicted mechanical soil resistance to penetration applied to the point and the difference between the two values serves as an input for tillage depth adjustment.

The invention discloses an aging early warning method of a power lithium battery. A power type MOSFET and a current limit resistor are connected with two ends of a battery pack, the MOSFET tube is conducted by 2ms under the control action of a controller, a single pulse current with width of 2ms is generated in the battery and the current limit resistor, and internal ohmic resistance of the lithium battery can be acquired according to voltage difference and a resistance value of the current limit resistor. Abnormal value elimination and data smooth processing for the power lithium battery ohminternal resistance data acquired in real time can be carried out, integral data is selected to carry out linear trend fitting of acquired gradient contrast to determine an aging inflection point, andearly warning of the battery health state is carried out. The method is advantaged in that internal ohmic resistance measurement of the power lithium battery can be simply and rapidly carried out, aging early warning of the power lithium battery is realized through utilizing the measured internal ohmic resistance data, and the method is of great importance to facilitating healthy development of the electric automobile industry.

An instrument and method for variable depth tillage is provided. A soil engaging implement has a pair of load cells and at least one strain gauge set mounted thereon. The load cells are used to determine a linear trend of topsoil resistance pressure change with depth as the soil engaging implement is drawn through the soil. The strain gauges are used to measure torque on the soil engaging implement caused by the load transmitted through the at least two load cells as well as the load applied to the point of the soil engaging implement. The linear trend of topsoil resistance pressure change with depth and the torque on the soil engaging implement are then used to determine both measured and predicted mechanical soil resistance to penetration applied to the point and the difference between the two values serves as an input for tillage depth adjustment.

This invention discloses a kind of generating method of real-time monitored brain potential trend chart by adopting computer technique and non-linear dynamics analysis method. It relates to a kind of generating method of brain potential non-linear trend chart. It comprises the following steps: first, collecting original EEG information from different position of scalp; A / D converting to digital EEG information; according to the non-linear dynamics theory and by adopting the trail of state space in time sequence reconstruction system, within one second, converting the conductive digital EEG information to non-linear index that relating to dimension, complexity and approximate entropy, which reflected the current non-linear dynamics characteristic; drawing the non-linear trend chart by using the non-linear index of time sequence, to reflect the non-linear dynamics characteristic variation condition of entire course. This invention can real-time monitor the anaesthesia effect, reflect the variation of anaesthesia effect accurately and promptly.

The invention discloses a strong reflection clutter elimination method based on a double-source IR-UWB (Ultra-Wideband) bioradar. Antennas of two different central frequencies are utilized to perform detection at the same time. Firstly, radar echo signals of two different frequencies are subjected to intra-channel pretreatment respectively, including distance accumulation in a space domain, and normalization, linear trend elimination and low pass filtering in a time domain, so as to achieve enhancement of weak life signals of human bodies; and then the radar echo signals of two different frequencies are subjected to inter-channel treatment after pretreatment, the self-adaption clutter elimination technique is adopted to eliminate interference of strong reflection clutter in a radar echo, thereby improving the radar detection accuracy.

The invention discloses a method for stability analysis of reference point for GNSS automatic deformation monitoring, which comprises the following steps: calculating the GNSS observation data to obtain the coordinate residual time series of the reference point; detecting and removing rough errors in the coordinate residual time series of the reference point; extracting and removing the linear trend and the cyclical term in the series; conducting interpolation to the time series if there exist missing data in the coordinate residual time series after the removal of the trend and the cyclical term; extracting the common-mode error of the residual time series and removing the common-mode error from each residual series; and finally, using a quantification calculating method to determine whether the remaining residual series follows a normal distribution or not. If it is, then the reference point is regarded as stable; otherwise, instable. According to the invention, it can effectively eliminate the common error of a station, improve the signal-to-noise ratio of the coordinate time series, and finally make the main error of the coordinate time series of the station limited to observed random noise to judge whether the station is stable or not.

The invention discloses a method for identifying a hydrologic time sequence nonlinear trend. The method comprises: computing the maximum wavelet decomposition level according to a sequence length, and determining a specific discrete wavelet transformation method; obtaining corresponding sub-sequences at different decomposition levels; computing the wavelet energy density value of each sub-sequence to obtain a wavelet energy density function of a to-be-analyzed hydrologic time sequence; decomposing a white noise sequence by use of a discrete wavelet transform method to obtain a sub-sequence; taking the average value of the wavelet energy density function of each white noise sequence as a standard wavelet energy density function, and obtaining a confidence interval of the standard wavelet energy density function; and comparing the positional relationship between the wavelet energy density value of a to-be-analyzed hydrologic time sequence sub-sequence on the maximum time scale and the confidence interval of the standard wavelet energy density function. By adopting the method for identifying the hydrologic time sequence nonlinear trend, the problems that a wavelet analysis method is short of a reliable hydrologic physical basis in the aspect of hydrologic time sequence trend identification and cannot effectively estimate the significance and uncertainty of the hydrologic sequence nonlinear trend are solved.

The invention discloses a long-acting slow-release pharmaceutical preparation. The pharmaceutical preparation contains the following components in percentage by weight: 25%-60% of a water-insoluble or sparingly-soluble drug and 40%-75% of a high-molecular polymer. After the pharmaceutical preparation is subjected to one time of intramuscular injection administration, the rate of a maximum plasma concentration to a minimum plasma concentration in a main release period is less than 5; the slope of a linear trend line of a cumulative release curve is less than 8 under an in-vitro simulated release condition; the daily release amount is less than 8.5%; and the simulated release condition is a buffer solution with the temperature of 37+ / -0.5 DEG C and the pH of 6.8-8.4. The prepared long-acting slow-release pharmaceutical preparation has the beneficial effects that an obvious release delay period or a burst release phenomenon are avoided after the administration, the steady state plasma concentration can be rapidly achieved, the plasma concentration with a relatively small fluctuation range can be maintained in several weeks or longer through once administration, and the pharmaceutical preparation can take effect rapidly and has good compliance with a patient.

The invention discloses a remote life micro-motion signal detecting method based on a terahertz Doppler radar, and particularly relates to the technical field of life micro-motion signal detection. The method comprises the steps of receiving four living body reflection echo signals in real time by the terahertz Doppler radar, and digitizing the four living body reflection echo signals by a multi-channel digitizer; suppressing static clutters in the four living body reflection echo signals; suppressing linear trend terms in the four static-clutter-free living body reflection echo signals; performing treatment such as complex signal demodulation on the four signals subjected to static clutter filtering and linear trend suppressing respectively, and processing results through set empirical mode decomposition to acquire a reconstructed vital sign signal; performing Fourier transform on the reconstructed vital sign signal to obtain a signal spectrum; suppressing a harmonic component, and taking peaks in the spectrum subjected to in-band harmonic suppression as estimated values of respiratory rate and heart rate respectively.

The invention relates to a method for inverting sea wave height and period based on water pressure, comprising: using a water-pressure-type wave height meter to measure underwater pressure intensity to obtain water pressure intensity time-series data; subjecting the water pressure intensity time-series data to Flourier transform to obtain a pressure spectrum; calculating pressure correction coefficient of sea wave by considering linear trend change of water pressure and attenuation of pressure wave in depth under water-layer filtering effect, and correcting the water pressure; inverting sea wave height and period of high temporal-spatial resolution according to spectral moment method or by subjecting the pressure spectrum to Fourier inverse transform and using zero-crossing separable wave method.

The method for making accurate restructure of difference measurement by using frequency domain method is characterized by that according to the difference value obtained by measuring two different separation distances or shearing quantities of measured object, making two groups of difference values undergo the process of cycle continuation and utilizing Fourier obverse and inverse transformation to obtain two initial evaluation curves, then calculating the difference value of ratio of zero modulation error and correspondent sensor separation distance or shearing quantity and combining two initial evaluation curves and eliminating the influence of zero modulation error to obtain the evaluation curve of measured parameter, then removing the linearity tendency in the evaluation curve to obtain one accurate evaluation curve having no theoretical error so as to implement the accurate restructure of measured parameter.

The invention provides a method for judging gas fleeing time. The method comprises the steps that a change curve of an oilfield produced gas-oil ratio along with injection time is built; according to the change rate of the gas-oil ratio, the change curve of the oil field produced gas-oil ratio along with the injection time is divided into a primary gas appearing stage, a gas fleeing channel primary forming stage and a complete gas fleeing stage which are marked as the stage A, the stage B and the stage C respectively; the linear trend lines of change curve bodies of the oilfield produced gas-oil ratio along with the injection time in the stage A and the stage C are respectively built, and the cross point of the two linear trend lines is taken as the gas fleeing time. The method for judging the gas fleeing time can accurately judge the gas fleeing time.

The invention relates to a modeling method for a quartz flexible accelerometer starting model, and belongs to the field of navigation and control research of delivery vehicles. The method comprises the following steps of: performing data sampling on a quartz flexible accelerometer starting process; de-noising the sampled quartz flexible accelerometer starting process data by using Daubechies wavelet to acquire time sequence data; extracting an exponential tread term and a linear trend term of the time sequence data; standardizing the time sequence data after the trend term is extracted; determining category of the time sequence model according to characteristics of an autocorrelation function and a partial autocorrelation function of the time sequence data and modeling by using a proper model; and estimating parameters of the time sequence data model and establishing the time sequence analysis-based quartz flexible accelerometer starting model. The invention provides the modeling method for the time sequence analysis-based quartz flexible accelerometer starting model. Research personnel are helped to more deeply analyze the accelerometer starting process.

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 discloses a diagnosis method for through-flow stage efficiency abnormality of a steam turbine high-pressure cylinder. The invention studies how to diagnose the cause of abnormal through-flow parameters of a high-pressure cylinder. According to the scheme, the method comprises the following steps: first, testing the influence of temperature change of inlet steam on the through-flow stage efficiency of a high-pressure cylinder, and judging whether there is a steam leakage phenomenon inside the high-pressure cylinder according to the linear trend of expansion of steam in a steam turbine; then, closing the electric doors of extraction steam in all sections of the high-pressure cylinder, measuring the pressure and temperature of inlet steam, exhaust steam and extraction steam in all sections, calculating the ratio of extraction steam pressure to after-stage flow under variable conditions, comparing the extraction steam pressure with the pressure after corresponding through-flow stage; and finally, determining the cause of through-flow stage efficiency abnormality of the high-pressure cylinder. The cause of through-flow stage efficiency abnormality of a steam turbine high-pressure cylinder is diagnosed according to boundary conditions by making use of the expansion characteristics of superheated steam in the steam turbine and the flow characteristics of superheated steam in the channel. The invention fills a technology gap in diagnosis of through-flow stage efficiency abnormality of a steam turbine high-pressure cylinder in our country.

The invention discloses a continuous feature automatic binning algorithm based on similarity combination. The algorithm comprises modeling data, decision tree initial binning, 100 equal frequency binning, linear trend judgment, trend + ChiMerge binning, IV, correlation and the like, and a final scoring card model is obtained. According to the algorithm, the whole process is realized by using python; the time consumption and the information loss in the process of obtaining the segmentation point conforming to monotonicity are greatly reduced; meanwhile, the intervention of analysts on the sub-boxes is reduced; The stability of the model is well tested, the subjective influence of an analyst on trend judgment is weakened through the algorithm that box merging is conducted according to the trend of the data, the default monotonicity presented by a box separation result is better supported by modeling data, the persuasion of box separation is enhanced, and the expression capacity of variables is improved.

Reliable and efficient search windows are provided by allowing the adaptation of the code search window to be dependent on inaccuracy measures of relations between a cellular frame time and a satellite reference time. This inaccuracy is calculated in a positioning node (21) of the cellular communications system (1), preferably by filtering of measurements received from user equipments. Linear trend Kalman filtering followed by post processing of estimation errors is presently preferred. In order to ensure non-ambiguous interpretation of the received time stamps of received satellite signals (55) provided by user equipments (10), a pseudo propagation delay is computed in both the user equipment (10) and the positioning node (21) based on GPS acquisition assistance data. The GPS time stamp is then defined referring to the determined pseudo propagation delay.; In a preferred embodiment, the pseudo propagation delay is assured to be situated within a pre-determined time interval.

The invention provides a practical internal substratum automatic interpretation method. The practical internal substratum automatic interpretation method comprises the first step of conducting single-well fine sand group division, the second step of conducting fine synthetic record calibration, the third step of calculating an index bed controlled residual value through the fine synthetic record calibration, the fourth step of conducting nonlinear trend surface fitting on the residual value of each single well through a support vector machine method to obtain the residual position of sand groups and the fifth step of conducting automatic interpretation of the sand groups. According to the practical internal substratum automatic interpretation method, the nonlinear trend surface fitting of a support vector machine controlled by an index bed is used for automatically tracking a substratum, by means of the substratum interpreted through the method, the result is more objective and accurate, the working efficiency is effectively improved, and the interpretation error is reduced.

The invention provides a target locating method for through-the-wall life detection radar. The target locating method comprises the following steps that step A: original signals are inputted and the original signals are arranged into a B-SCAN chart according to the channel number; step B: the static clutter and the linear trend term are removed for each path of original signal; step C: energy accumulation is performed on the life micro signal in the time domain and the peak in the energy spectrum is detected so as to determine the distance estimation value between the target and two receivingantennas; step D: FFT is performed on all the data of the slow time direction in the distance according to the distance estimation value given in the step C so as to obtain the echo frequency information of the distance and determine the distance between the target and the receiving antennas; step E: the time delay curve after wall delay correction is obtained according to the distance from the target to the two receiving antennas given in the step D; and step F: the intersection points of the time delay curve are solved to act as location estimation of the target.

A computer implemented method, computer program product and system for monitoring a plurality of hardware or software system resources for identifying hidden trends in the behavior of the system resources, includes: collecting metrics of at least one system resource indicative of the behavior of at least one system resource; for each of the at least one system resource, determining a spectrum representative of a time-based signal of the collected metrics; performing a wavelet transform on each of the at least one spectrum; and analyzing the result of the wavelet transform to identify possible linear trends in the behavior of the at least one system resource.

The invention discloses a distribution network rapid reconstruction method taking voltage stability into consideration. The method comprises the following steps: 1, introducing a rapid and convenient voltage stability factor calculation method reflecting the voltage stability, and constructing an integrated factor minimization optimization model comprising network loss and a voltage stability factor; 2, constructing a rapid and non-iterative linear trend calculation method for reducing enormous time consumed for trend calculation, and based on the trend calculation method, calculating elements needed for the optimization model; and 3, introducing and organically combining individual-dependent parameter setting, an individual-dependent mutation strategy and a successful-parent-selecting framework, for improving a differential evolution algorithm, and constructing a differential evolution algorithm with an individual-dependent mechanism and the successful-parent-selecting framework, for solving distribution network rapid reconstruction taking the voltage stability into consideration. According to the invention, change of the network loss and voltage level in reconstruction can be reflected objectively, through parameter adjustment, grid structures with different significance can be obtained, and guidance is provided for actual operation.

The invention discloses a load prediction method for multiple cloud data center. The method comprises the following steps of: acquiring a log record file for recording the resource use condition of a virtual machine at each time point, and extracting required characteristic quantity data and historical load data from the log record file; converting the characteristic quantity data and the historical load data into corresponding input characteristic sequences and historical load vectors; using a pre-constructed neural network model to calculate and obtain a nonlinear component of load prediction; calculating to obtain a linear component of load prediction by using a pre-constructed autoregression model; and integrating a nonlinear component and a linear component of the load prediction to obtain a final load prediction result. The method comprehensively considers the linear trend and nonlinear characteristics of a load sequence along with the time change in a multi-data center environment, combines the neural network model with a statistical learning method of the autoregression model, and can effectively improve the prediction precision of the future load.

A logarithmic analog to digital conversion method for an analog input signal includes a logarithmic amplification with progressive compression of the input signal delivering a sequence of several secondary analog signals. The trend of the values of at least some of the secondary signals is a function of the values of the analog input signal including regions corresponding to a linear trend of the secondary signals as a function of that of the input signal expressed in a logarithmic scale. The method also includes a comparison of at least some of the secondary signals of the sequence with a common reference signal whose value lies within each of regions, supplying a thermometric code information item, and a generation of a first digital word from the thermometric code information item.

The invention relates to an in-situ seismic oscillation prediction model and a construction method thereof, and the construction method comprises the steps: obtaining acceleration recording data of each strong vibration observation station, carrying out the preprocessing of the data, and screening out standard data; removing a linear trend from the screened event record data of each strong vibration observation station, and manually selecting the record data of the first arrival time of the P wave and the S wave; carrying out one-time integration and two-time integration on the vertical acceleration record of each strong vibration observation station in the recorded data so as to respectively obtain a speed time history and a displacement time history; carrying out continuous Butterworth band-pass filter filtering processing on the displacement time history, so as to obtain P wave amplitude early warning parameters Pvall and Paall; calculating a peak velocity PGV and a peak acceleration PGA according to an instrument intensity calculation standard requirement; carrying out correlation fitting on the P-wave amplitude early warning parameters Pvall and Paall, and obtaining a correlation model between the Pvalue and the PGV and a correlation model between the Paall and the PGA. The model improves the reliability of a prediction result.

The invention discloses an MEMS gyro denoising method based on a wavelet threshold, which comprises the following steps of sampling a MEMS gyro signal; performing the trend analysis on the MEMS gyro signal; determining the wavelet bases and decomposition layers based on the results of trend analysis, selecting the wavelet coefficient transform function based on the determined wavelet basis and decomposition layer number, and distinguishing the noise and detail of the signal based on the selected wavelet coefficient transform function, selecting threshold functions and thresholds to remove thenoise components from the detailed part of the signal. By analyzing the output signal of MEMS gyroscope, the drift of MEMS gyroscope has a linear trend, which is a slow-changing process. The analysisshows that the output signal of MEMS gyroscope is a stationary signal with low frequency. On the basis of analysis, the influence of random error of MEMS gyroscope is reduced, and the real-time outputprecision of MEMS gyroscope sensor is improved. Because only wavelet coefficient transform is used, the calculation is simple, is easy to realize and has the advantage of high precision.

A computer implemented method, computer program product and system for monitoring a plurality of hardware or software system resources for identifying hidden trends in the behavior of the system resources, includes: collecting metrics of at least one system resource indicative of the behavior of at least one system resource; for each of the at least one system resource, determining a spectrum representative of a time-based signal of the collected metrics; performing a wavelet transform on each of the at least one spectrum; and analyzing the result of the wavelet transform to identify possible linear trends in the behavior of the at least one system resource.

The invention belongs to the technical field of online information propagation prediction, and discloses a social media network event propagation key time prediction method, a system, and a medium. The method comprises the steps: carrying out the classification according to different time sequence features of social media network event online information; carrying out smoothing processing on the popularity time sequence with too strong volatility by adopting a Hort linear trend method; recognizing a time interval in which a key node propagated by the preprocessed social media network event occurs; carrying out time window division on the preprocessed time series data, and extracting time series, fluctuation and text emotion features based on online information data; constructing a prediction model training sample, and training a prediction model by adopting an XGBoost algorithm according to the training sample and the number of future time windows; and predicting the occurrence time ofthe key nodes in the event propagation process of the social media network by adopting the trained model. The method can effectively predict the occurrence time of the key node in the event propagation process of the social media network.