31 results about "Data approximation" patented technology

A computer implemented device is described that is adapted for improving memory efficiency for conducting machine learning on multi-dimensional vectors stored as specially configured data structures. The device includes network interfaces adapted for receiving data sets and communicating with data harvesting applications, as well as data storage and memory for storing the configured data structures. The multi-dimensional vectors and the system provide a data architecture that processes disparate data sets to programmatically extract features transformed from raw information, the extracted features stored in the form of data values suitable for conducting data approximations and neural network processing.

The invention discloses a self-adaptive regularized super-resolution image reconstruction method which can keep marginal definition, which mainly solves the problem that the prior method has edge fog in reconstruction of a degraded image. The method comprises the following steps: an imaging model is constructed; on the basis of an unconstrained objective function constructed by a Lagrangian multiplier method, gradient is increased to approach a bound term; the objective function is expanded; L1 norm is adopted to measure a data approximation term; a self-adaptive bilateral total variation model which can carry out local adaptive control on the smoothing effect is utilized to construct a self-adaptive regular term; a gradient approximation term is added to be as constraint of gradient consistency; edge information is kept; the self-adaptive regular term and a gradient consistency bound term are introduced as constraint conditions; an expanded Lagrangian objective function is constructed and optimized; and an optimized unconstrained objective function is utilized to reconstruct an image, thereby obtaining a high-resolution image of which the edge is kept. The method can keep image edge clear, can inhibit noise and is suitable for restoration treatment on the degraded image.

The invention relates to an exponential fit-adaptive Kalman-based ground-air electromagnetic data de-noising method. According to the technical scheme of the invention, for single-point electromagnetic data during the ground-air measurement, time windows are determined and then the data are segmented at equal logarithm interval based on the features of electromagnetic data in the ground-air time domain according to the approximate e-index attenuation law, and the time constant value of data in each time frame are extracted as fitting parameters. Meanwhile, the data in each time frame are processed based on the e-exponential fitting method, and fitting output results are adopted as predicted values to be input into a filter. After that, the adaptive scalar Kalman filtering method is applied to filtering the electromagnetic noise in data, and the filtered data are subjected to resistivity-depth imaging. Compared with the existing electromagnetic data filtering method in the ground-air time domain, the above filtering method of the present invention not only effectively suppresses the electromagnetic noise in electromagnetic data in the ground-air time domain, but also fully retains and enhances the effective information in measured data. Therefore, both the signal-to-noise ratio and the quality of electromagnetic data in the ground-air time domain are improved.

A spray injection analysis and nozzle configuration system is described having a user input unit that collects spray system input parameters and relays the collected parameters to a fluid performance matching unit and / or problem geometry unit for subsequent processing. The user inputs basic system parameters, including the desired spray fluid characteristics, to obtain suggested system configuration, including spray nozzle types and quantities. Accuracy of suggested spray nozzle type and configuration is increased via approximating the viscosity and / or surface tension parameters of the desired spray fluid with that of collected performance data. When a user already knows the desired spray nozzle type and associated system parameters, the user input unit routes this information to the problem geometry unit for creation of a problem geometry file, including calculation of the drop size distribution and spray velocity, and performance modeling via the fluid modeling unit.

The invention discloses an extensive makeup language (XML)-based method for synchronously updating increment of spatial data, which is used for establishing a spatial data bidirectional synchronous mechanism, realizing approximate consistency maintenance of multi-copy data, and solving the problems that: according to the common data synchronization method, offline environment is not supported, heterogeneous databases are not supported, and the synchronous object granularity and the synchronous process cannot be freely controlled and the like. Technologies such as NET, XML and the like are used as a basis of implementing the method for synchronously updating the spatial data. The system comprises a changed record collection module, a changed data organization module, a changed data conflict arbitration module and a changed data synchronization module, wherein the four functional modules have synergistic effects of monitoring, organizing and synchronizing the changed data to synchronously update the increment of the spatial data. According to the simple technical implementation method, the spatial data are effectively managed and automatically processed by computers, human-computer interaction is less, the time is saved, the work efficiency is improved, and the method has an extremely wide application prospect in the field of data consistency maintenance.

Optimizing data approximation analysis using low power circuitry including receiving a first set of data results and a second set of data results; charging a first capacitor on the circuit with a unit of charge for each of the first set of data results that indicates a positive data point; charging a second capacitor on the circuit with the unit of charge for each of the second set of data results that indicates a positive data point; applying a voltage from the first capacitor and a voltage from the second capacitor to a FET on the circuit, wherein a current flows through the FET toward an output of the circuit if the voltage on the first capacitor is greater than the voltage on the second capacitor and a difference in the voltage of the first capacitor and the second capacitor is greater than a threshold voltage of the FET.

Optimizing data approximation analysis using low power circuitry including receiving a first set of data results and a second set of data results; charging a first capacitor on the circuit with a unit of charge for each of the first set of data results that indicates a positive data point; charging a second capacitor on the circuit with the unit of charge for each of the second set of data results that indicates a positive data point; applying a voltage from the first capacitor and a voltage from the second capacitor to a FET on the circuit, wherein a current flows through the FET toward an output of the circuit if the voltage on the first capacitor is greater than the voltage on the second capacitor and a difference in the voltage of the first capacitor and the second capacitor is greater than a threshold voltage of the FET.

A mobile monitoring device and related systems and methods for monitoring a condition to which an object is exposed are disclosed. The monitoring device includes a sensor module for monitoring the condition and periodically producing sensor module data relating to the condition and a microcontroller. The microcontroller includes a sensor data component for receiving the sensor module data from the sensor module and a first transceiver component for receiving network manager data from a local network manager local to the monitoring device. The microcontroller further includes an associator component for comparing the network manager data with the sensor module data and, if the network manager data approximates the sensor module data, establishing an association with the local network manager.

A spray injection analysis and nozzle configuration system is described having a user input unit that collects spray system input parameters and relays the collected parameters to a fluid performance matching unit and / or problem geometry unit for subsequent processing. The user inputs basic system parameters, including the desired spray fluid characteristics, to obtain suggested system configuration, including spray nozzle types and quantities. Accuracy of suggested spray nozzle type and configuration is increased via approximating the viscosity and / or surface tension parameters of the desired spray fluid with that of collected performance data. When a user already knows the desired spray nozzle type and associated system parameters, the user input unit routes this information to the problem geometry unit for creation of a problem geometry file, including calculation of the drop size distribution and spray velocity, and performance modeling via the fluid modeling unit.

Optimizing data approximation analysis using low power circuitry including receiving a first set of data results and a second set of data results; charging a first capacitor on the circuit with a unit of charge for each of the first set of data results that indicates a positive data point; charging a second capacitor on the circuit with the unit of charge for each of the second set of data results that indicates a positive data point; applying a voltage from the first capacitor and a voltage from the second capacitor to a FET on the circuit, wherein a current flows through the FET toward an output of the circuit if the voltage on the first capacitor is greater than the voltage on the second capacitor and a difference in the voltage of the first capacitor and the second capacitor is greater than a threshold voltage of the FET.

The invention relates to a water quality prediction method based on Gaussian cloud transformation and a fuzzy time sequence and belongs to the technical field of artificial intelligence and water environment monitoring. The method mainly comprises the following steps: 1) discretizing historical observation values of water quality parameters into different granular layers through Gaussian cloud transformation; 2) calculating an approximate period length of water quality parameter historical data; 3) constructing a training set according to the approximate period length; and 4) calculating a prediction value by applying a fuzzy time sequence model. The water quality prediction method based on Gaussian cloud transformation and the fuzzy time sequence provided by the invention can be used for effectively processing a water quality data set with an uncertainty characteristic through Gaussian cloud transformation and prediction by the fuzzy time sequence. A prediction result is better in robustness; By blending the water quality data approximate period, noise data is removed by virtue of inherent characteristics of the water quality data, so that the prediction precision is prevented from being affected by the noise data, and the model is relatively high in self-adaptability to the data.

The invention relates to a big data similar searching method, in particular to a big data searching method based on sparse hash. The method is mainly used for conducting application development based on storage and searching of big data. The method includes first utilizing a sampling method to determine the size of a training set according to theory of a computer memory, learning the training set and learning a hash function for big data coding and binary coding of the training set, then conducting binary coding on the big data according to the learnt hash function. At the moment, online search application can be conducted, namely for one test case, first a binary code of the test case is obtained according to the obtained hash function, and then real-time search is conducted on the binary code of the big data. By means of the method, the big data searching time complexity is linear, the problem that manifold learning does not have an explicit function is solved, storage quantity of the big data is reduced to thousands of times, the method is easy to implement, and only some simple mathematical models are involved during code writing.

The high-dynamic weak 8PSK / 16PSK signal carrier capturing method disclosed by the invention is high in capturing result precision and good in real-time performance. The method is realized through the following technical scheme: a filtering sampling module firstly performs filtering sampling processing on a received signal, and writes data into a data ping-pong cache space in a positive sequence; the data processing unit reads ping-pong cache data from a data cache space of the ping-pong cache module in an inverted sequence; the change rate compensation module carries out carrier Doppler frequency change rate compensation on the sampled data, and sends an accumulation result to the FFT module for time-frequency transformation operation; the incoherent integration module is used for approximately obtaining absolute values of the complex data after time-frequency transformation and then carrying out incoherent integration; the spectrum windowing module performs cosine window windowing smoothing on the data after incoherent integration; the threshold median module judges whether capture succeeds or not by combining an integral result; and the resolving and correcting module corrects according to the median result and the Doppler change rate to obtain the carrier Doppler at the current moment, and completes carrier capture.

The invention discloses a maglev train levitation electromagnet attitude detection system and method, computer equipment and a storage medium. Vertical displacement, vertical acceleration, transversedisplacement, transverse acceleration, line movement speed and line real-time position are all acquired from a mounting position of a shell, so that detection data comes from the same position; detection data of different positions do not need to be approximately converted to the same position, errors generated by approximate conversion are avoided, and the detection precision is high; all parts in a signal acquisition assembly are started at the same time and transmitted through the same channel, it is guaranteed that all detection data are located on the same time axis, the problem that large phase errors occur during dynamic analysis due to the fact that different channels are adopted for obtaining data and time axes are not unified is solved, and the accuracy during dynamic analysis isimproved; and a signal conditioning and transmission module is adopted for data transmission, and a vehicle-mounted CAN network is not adopted, so that the sampling frequency and the sampling precision are improved.

A computer system for handling missing or infrequent data used for calculating an index is described. The computer system uses the data approximation scheme for calculating a credit volatility index. The computer system includes a memory configured to store at least one program and at least one processor communicatively coupled to the memory, in which the at least one program, when executed by the at least one processor, causes the at least one processor to receive data regarding options and, when data for a complete input data set is missing after a predetermined period, retrieve or generate estimated data points needed to calculate the index. The data points may be for credit default swap index derivatives using data regarding options on credit default swap index derivatives. The processor may generate a credit volatility index and transmit data regarding the credit volatility index.

The invention discloses a method for determining a rectification coefficient of a ship body to rudders of a double-propeller double-rudder ship, relates to the technical field of hydrodynamic force, and aims at the double-propeller double-rudder ship in a laboratory pool environment to obtain rotary motion constraint model test data by carrying out a maneuverability rotary motion constraint modeltest under different test working conditions. The rectification coefficient of the ship body of the double-propeller double-rudder ship in the hydrodynamic interference of the rudder can be determinedby performing data approximation and regression processing on the acquired data, the technical vacancy of the industry is made up, and the method is simple to operate, high in data correlation and good in engineering effect.

The invention is applicable to the technical field of data acquisition and monitoring control system configuration tool application, and provides a method and a device for quickly associating primitives in a graphic configuration tool. The method comprises the following steps: analyzing a graphic file to be associated, and obtaining data information of the primitives in the graphic file; judging whether primitives in the graphic file are associated or not; if the primitives in the graphic file are associated, associating the paths of the devices where the primitives are located; and if the primitives in the graphic file are not associated, associating the business objects of the primitives. According to the invention, whether the primitives to be associated are similar to the old associated data of the primitives is identified; the approximate primitives are quickly replaced by the old associated data, so that the association of the primitives can be quickly completed, a large amount of time is saved, the method is particularly suitable for the condition of existing old graphs, the association of the primitives can be quickly completed by only modifying the equipment paths of the associated primitives, and a large amount of time is saved.

The invention provides a DNA sequence query system based on shared data approximation. The system compresses DNA sequences from different archives into the same data approximation. For an interested DNA sequence, which archives the sequence exists in can be quickly inquired through the data infarct. The DNA sequence query system comprises three subsystems, namely a data preprocessing subsystem, adata approximate establishing and updating subsystem and a DNA sequence query subsystem. Useful information is extracted from a given DNA sequence; original data infringement is proposed to compress the DNA sequence data of the plurality of archives; and the DNA is inquired by utilizing the data infarct. The system can be used for DNA sequence query, and the archive library data of interest is further searched for further research by querying the archive library where the DNA of interest is located.

Optimizing data approximation analysis using low power circuitry including receiving a plurality of data bits each corresponding to a binary indication of a test result; placing each of the plurality of data bits on an approximation circuit, wherein each of the data bits is placed on the approximation circuit by applying, to a first capacitor during a set time period, a voltage corresponding to the data bit, and wherein placing each of the plurality of data bits on the approximation circuit results in a resulting voltage stored on the first capacitor; and determining a potential correlation of the test results by comparing the resulting voltage to a voltage threshold.

The present invention provides a method for extracting well seismic data based on Euclidean distance approximate matching, which includes: performing approximate seismic attribute matching, finding the coordinate point position closest to the target well, and extracting the seismic attribute value corresponding to the position approximately as the corresponding value of the target well position Seismic attribute values, and integrate all seismic attribute values ​​into an attribute set; carry out approximate matching of lithological data, through the start and end time range of the time window, and use time-depth conversion to convert the time range to the depth range corresponding to the target well, and then Calculate the lithology ratio within the depth range corresponding to the target well sample; match the extracted seismic attribute set with the lithology type to obtain a seismic attribute set with category labels. This method realizes the approximation of the target well position to the nearest grid position in the adjacent distance and the extraction of the seismic data of the target well. The numerical category labels and seismic attribute values ​​can provide sample point support for the training of the machine learning classification model.

The invention discloses a suspension electromagnet posture detection system, method, computer equipment and storage medium of a maglev train. The detection data can be obtained from the same installation position, so that each detection data comes from the same position, and there is no need to approximately convert the detection data from different positions to the same position, avoiding the error caused by approximate conversion, and the detection accuracy is high; all components in the signal acquisition component start at the same time, And it is transmitted through the same channel, which ensures that all the detection data are on the same time axis, avoiding the problem of large phase error in dynamic analysis caused by different channels to obtain data, and improving the dynamic analysis. Time accuracy; and the signal conditioning and transmission module is used for data transmission instead of the vehicle CAN network, which improves the sampling frequency and sampling accuracy.

The present invention provides a collective data environment detection system. The technical solution of the present invention is: the collective data environment detection system summarizes the data of various environmental monitoring departments, and then compares it with the data of the natural disaster period to make an early warning. Contact various environmental monitoring departments, build data transmission channels, collect data from various environmental monitoring departments in real time, analyze in real time according to the collected data, and generate data maps, and when a natural disaster occurs in a certain area, name the part of the data as a disaster Analyze the characteristics of the data during the natural disaster period and make special annotations. In the subsequent real-time data, carry out comparisons of special data annotations such as spontaneous combustion disasters. If similar data are found, then compare and verify data from other types of detection departments, and then Get accurate warnings. The beneficial effect of the present invention is that early warning of disasters can be made before spontaneous combustion disasters come, thus saving many lives and property. Benefit the country and the people.

The present invention is applicable to the technical field of configuration tool application in data acquisition and monitoring control systems, and provides a method and device for quickly associating graphic elements in a graphic configuration tool. The method includes: analyzing graphic files to be associated, and obtaining all The data information of the graphic element in the graphic file; judge whether the graphic element in the graphic file has been associated; if the graphic element in the graphic file has been associated, then associate the path of the device where the graphic element is located; If the graphics entity in the graphics file has not been associated, then associate the business object of the graphics entity. The present invention can quickly complete the association of graphic elements by identifying whether the graphic elements to be associated are similar to the old associated data of the graphic elements, and quickly replace these similar graphic elements with the old associated data, thereby saving a lot of time, and is especially suitable for If there are already old graphs, you can quickly complete the association of primitives by only modifying the device path of the associated primitives, thus saving a lot of time.

The invention provides a data-driven wind farm energy storage capacity planning method, which relates to the technical field of new energy power generation planning. First, collect relevant data such as wind power, energy storage devices, wind curtailment upper limit, transmission line capacity, etc., and then establish a wind farm operation optimization model with energy storage and a wind farm energy storage capacity optimization model based on distribution robust optimization. The energy storage capacity optimization model is converted into the form of mixed integer linear programming, and finally solved by a mathematical programming solver. The wind farm energy storage capacity planning method based on distribution robust optimization provided by the invention can make full use of the historical data of wind power, consider the error of the approximate probability distribution of the historical data, take into account the economy and conservatism, and is easy to program to find the optimal solution.

Disclosed are an auto-encoder and a method and medium for training the auto-encoder. The auto-encoder comprises: an encoder unit that maps input data into hidden variables in a hidden space; a constraint unit which is used for applying constraint to the hidden variables, so that the hidden variables are gathered around a central point in the hidden space; a decoder unit that decodes the hidden variables from the constraint unit into reconstructed data, in which the encoder unit, the constraint unit, and the decoder unit are trained such that the reconstructed data approximately matches the input data. According to the automatic encoder disclosed by the invention, the hidden variables of the similar data can be expressed in a local hidden space, so that the similar data is far away from thedissimilar data in the hidden space.

The invention provides a video content description model training method, an application method, equipment and a medium, and the training method comprises the steps: obtaining a target video and the content description of the target video, extracting a verb and a template from the content description, carrying out the coding processing and feature sampling of the verb and the template, and obtaining a first hidden variable feature, the method comprises the following steps: performing feature extraction on a target video to obtain a target visual feature, and performing data approximation processing on a priori hidden variable encoder and a language decoder according to a first hidden variable feature and the target visual feature to obtain a video content description model. Different descriptions are generated by using different expressions.

The invention discloses a high-dynamic weak 8PSK / 16PSK signal carrier acquisition method, which has the advantages of high precision of the acquisition result and good real-time performance. The present invention is realized by the following technical solutions: the filtering and sampling module first performs filtering and sampling processing on the received signal, and writes the data into the data ping-pong buffer space in positive order; the data processing unit reads the ping-pong buffer data in reverse order from the data buffer space of the ping-pong buffer module; The rate compensation module performs carrier Doppler frequency change rate compensation on the sampled data, and sends the accumulated results to the FFT module for time-frequency transformation operations; the incoherent integration module approximates the absolute value of the complex data after time-frequency transformation, and then performs non-coherent integration. Coherent integration; the spectral windowing module performs cosine windowing and smoothing on the data after incoherent integration; the threshold median module combines the integration results to determine whether the capture is successful or not; the solution correction module corrects according to the median result and the Doppler rate of change Obtain the carrier Doppler at the current moment to complete the carrier acquisition.