42 results about "Linear algorithm" patented technology

The invention discloses an augmented reality head-up display system and a display method thereof, relates to the technical field of auxiliary driving, and aims at complex road conditions and diversified driving requirements. The comprehensive fusion analysis of the driving environment and the driver state is realized by sequentially executing the four steps of environment perception and multi-source data acquisition, situation analysis and information complexity judgment, dynamic content screening and display strategy generation and real-time monitoring and self-adaptive feedback closed loop. The method comprises the following steps of: marking and normalizing multi-dimensional sensing data to form an environment factor and a driving load factor, and judging the current information complexity by using a nonlinear algorithm; thirdly, screening the most critical content from the information pool according to the priority, and performing refined arrangement according to the display position and style; and if the HUD configuration deviation is monitored and detected, dynamic strategy updating is executed, so that the AR-HUD always ensures the accuracy and the conciseness of information display in different scenes, and the driving safety and the user experience are improved.

The invention belongs to the technical field of energy storage operation monitoring, and particularly relates to a user side energy storage operation monitoring analysis method based on source-network-load cooperation, which comprises the following steps: acquiring data of an energy storage system, collecting historical data, and fusing multi-dimensional data; predicting a power grid demand and a user load within the time t, and optimizing and adjusting a charging and discharging strategy of the energy storage system in real time by adopting a linear algorithm; comprehensively evaluating the operation benefit of the energy storage system; the method comprises the following steps: performing feature extraction on historical data of an energy storage system, combining real-time operation data of the energy storage system, comprehensively predicting potential risks in operation of the energy storage system, and obtaining a risk prediction result; based on the real-time change of the power grid demand and the user load and the battery state and risk prediction result of the energy storage system, the charging and discharging operation strategy is automatically adjusted, and the stable operation of the system is ensured. The power grid demand and the user load can be predicted more accurately, and potential risks can be found and predicted in time.

The invention provides a drive circuit for a frequency converter load, and relates to the technical field of frequency converters. Comprising a control module, a driving module, a power supply module, a signal acquisition and processing module and a load matching module, and the load matching module has impedance matching, tuning matching and shaping filtering functions and has universal adaptive driving capability for most types of motors and ultrasonic transducers; the power supply module adopts a high-speed MOS tube, the driving module adopts a high-speed driving chip to adapt to the high-speed MOS tube, the frequency application range is large, the DDS waveform generation module is adopted to realize PWM wave output of different frequencies, the change step length of the frequency is extremely low, and the change is smooth; a variable step size algorithm and an LADRC algorithm are combined, and the working resonant frequency of the variable frequency load can be automatically and accurately locked within a certain frequency range. A linear LADRC algorithm is adopted, behaviors of a driving circuit are adjusted through real-time measurement and estimation, and the method can adapt to changes and uncertainty of a motor and an ultrasonic transducer in a complex working environment.

The invention discloses a drainage pipe network operation intelligent monitoring system based on multi-source data fusion, and relates to the technical field of drainage pipe network intelligent monitoring. According to the method, through collaborative design of the dynamic weight matrix and the ChebNet-LSTM fusion model, limitation of a traditional algorithm on nonlinear feature capture and time-space correlation analysis of the drainage pipe network is broken through; the dynamic weight matrix is combined with node topological data and abnormal case data, so that the association strength between the nodes can be dynamically adjusted, and the complex working condition of the pipe network can be accurately adapted; according to the ChebNet-LSTM model, space correlation features are extracted through a ChebNet layer, long-term and short-term time sequence dependence is captured through an LSTM layer, the problem that a traditional linear algorithm cannot fit a non-linear scene is solved, the limitation of single-dimension feature analysis is broken through, pipe network data space-time collaborative modeling is achieved, multi-task output can be achieved while the calculated amount is reduced to meet the real-time monitoring requirement, and the real-time monitoring efficiency is improved. And the abnormity pre-judgment accuracy and the tracing precision are greatly improved.

The invention relates to a Stewart six-dimensional force sensor, six branch chain rods distributed in an inclined manner are arranged between an upper platform and a lower platform, each branch chain rod comprises an upper rigid rod, an upper cross flexible hinge, a measuring unit, a lower cross flexible hinge and a lower rigid rod which are sequentially connected from top to bottom, the upper rigid rod is fixedly connected with the upper platform, and the lower rigid rod is fixedly connected with the lower platform. And the lower rigid rod is fixedly connected with the lower platform. A calibration device is used for calibrating a Stewart six-dimensional force sensor. The invention relates to a Stuart six-dimensional force sensor calibration method, which comprises the steps of off-line calibration and training and on-line measurement and calculation. The Stewart six-dimensional force sensor has the advantages of being large in rigidity, high in bearing capacity, stable in structure, fast in dynamic response, high in precision, good in isotropy and the like, and solves the problem that a traditional linear algorithm cannot process complex nonlinear errors. The measurement precision is reduced due to serious inter-dimensional coupling; and the complexity of a high-precision model is too high and the high-precision model is difficult to run in real time in an embedded system.

The present disclosure relates to a method and apparatus for determining a signal detection network. The method includes: determining sample communication parameters for communication between a sample transmitter and a sample receiver via a sample intelligent reflector surface (IRS) (S101); and training an initial neural network based on a training sample set composed of the sample communication parameters to obtain a signal detection network, wherein the input of the initial neural network is the sample communication parameters, and the output of the initial neural network is an estimate of the transmitted signal of the sample transmitter (S102). According to the present disclosure, the initial neural network can be trained based on the training sample set to obtain a signal detection network. Since the neural network is not a linear algorithm and is not limited to operating on linear relationships, it can be effectively used for signal detection even in complex communication environments with many nonlinear factors, and a high-precision estimate of the transmitted signal can be obtained.

The invention relates to the technical field of automatic control and intelligent driving, and discloses a precise macro-micro composite driving system control method based on a self-learning nonlinear PID algorithm. The method comprises the steps of collecting real-time operation data to judge nonlinear dynamic characteristics, obtaining a deviation value, extracting nonlinear characteristics and adjusting control parameters, switching to a self-learning mode to update a model, compensating interference and the like. By monitoring and compensating the nonlinear characteristics and external interference of the system in real time, the response speed and control precision of the system are improved, the dependence on external experience parameters is reduced, and the system adapts to complex and changeable working environment requirements.

The invention discloses a spectrum simulation nonlinear algorithm and system based on DMD modulation. By establishing a working state, wavelength distribution, an energy weight and a target spectrum matrix, constructing an operation rule and considering DMD surface illumination nonlinearity, automatic fitting of different skewness spectrum shapes is realized, skewness errors are reduced, and spectrum simulation accuracy is improved; a matrix operation system reduces fitting complexity and improves simulation efficiency. And a BP neural network optimized by a genetic algorithm is adopted to fit a spectrum minimum unit, so that the self-adaption and optimization capability of the algorithm is enhanced, and the spectrum is accurately fitted. A 450nm-1000nm spectrum shape fitting experiment is completed by using a specific experimental device, the effectiveness and the reliability are verified, and the method has wide application prospects in multiple fields and can meet spectrum simulation requirements of different scenes.

The application discloses a quantum linear solving method and device based on quantum adiabatic linear algorithm, and a medium, and relates to the technical field of quantum computing. The method comprises the following steps: obtaining a linear system to be processed, and constructing a Krylov subspace matched with the linear system; determining a subspace equation set of the linear system based on the Krylov subspace according to the linear system and the Krylov subspace; constructing a quantum circuit corresponding to the quantum adiabatic linear algorithm according to the subspace equation set, and solving an approximate solution of the linear system in the Krylov subspace according to the quantum circuit. The application can combine the subspace method with the quantum adiabatic linear algorithm, perform dimension reduction processing on the original high-dimensional linear equation set in the quantum adiabatic linear algorithm through the subspace method, so that the dimension of the linear equation set actually solved in the iteration process is smaller than the dimension of the original linear equation set. Therefore, the demand of the quantum discrete adiabatic linear algorithm for computing resources is reduced, and the quantum discrete adiabatic linear algorithm can be implemented in an NISQ chip and a classical computer.

The application discloses a feed formula determination method and device, electronic equipment and storage medium. According to a plurality of feed information and animal growth required nutrient structure information, a target constraint condition is determined, and a preset first linear algorithm and a second linear algorithm are used to generate first feed formula information and second feed formula information that satisfy the target constraint condition according to the feed information, so as to preliminarily obtain the feed formula through the linear algorithm. And the preset genetic algorithm is used to combine and iterate the first feed formula information and the second feed formula information until the preset iteration termination condition, and third feed formula information is obtained, so that the genetic algorithm is used to effectively avoid the situation that the linear algorithm falls into a local optimal solution and cannot obtain a global optimal solution, the nonlinear space solution of linear programming processing is made up, and the reliability of the feed formula optimization result is improved.

An apparatus, system and method for providing a may include at least a phacoemulsification surgical console having a customizable non-linear custom aspiration mode. The console may include at least an aspirator; a foot pedal; and a non-transitory computing code resident on a computing memory associated with a computing processor which, when executed by the processor, causes to be executed the steps of: receiving a percentage actuation of the foot pedal; calculating, including from a non-linear algorithm, a percentage actuation for the aspirator corresponded to the received percentage foot pedal actuation; and dictating the calculated percentage aspirator actuation to the aspirator.

The application relates to the field of power electronics, and particularly provides a LLC resonant converter soft-start control method and device and an LLC resonant converter, wherein the method comprises sequentially performing three-stage soft-start control, wherein in the first stage, a fixed dead zone is used to gently establish resonance for a first time length; in the second stage, a non-linear algorithm based on multiple hiccup periods and a decreasing quantity increasing with periods is used to smoothly adjust the dead zone; and in the third stage, the dead zone is decreased to a target value at a two-stage speed of slow first and then fast; fault detection including at least start timeout and input voltage abnormality is performed in parallel in the whole process; start completion is judged based on multiple conditions of total time, dead zone value and voltage ratio; and when the driving dead zone is updated, protection of not being lower than a safety minimum value is forced to be performed. Through the above common control, the converter is smoothly and quickly started, and the start impact and voltage stress are effectively inhibited.

The invention discloses an observable method and system for a unified log index and tracking data, and the method comprises the steps: collecting various types of original observation data, carrying out the unified formatting, constructing a column type storage structure, and carrying out the hierarchical storage; then optimizing data layout by utilizing column-level metadata and a query mode feature library, and implementing a multi-level index strategy; meanwhile, through data cleaning, structured enhancement and contextual information association, a causal atlas between heterogeneous observation data is established, and observation data with rich contexts is generated; and finally, a high-performance query processing system is constructed based on a unified query engine and a sub-linear algorithm, and progressive result return and visual analysis are realized. According to the method, the data retrieval efficiency and the analysis depth are remarkably improved, operation and maintenance personnel can more quickly position the root cause of the system problem, and the average time of fault diagnosis is effectively shortened.

The invention provides a building air conditioner flexible load regulation and control method and electronic equipment, and the method comprises the steps: carrying out the processing through a linearization algorithm based on the quality of an indoor building wall, the heat capacity of the wall, a second derivative model and a fourth derivative model, and obtaining a building wall temperature model at a target moment; based on the building wall temperature model at the target moment, the indoor building air quality, the air heat capacity, the first derivative model, the second derivative model and the third derivative model, processing is conducted through a linearization algorithm, an indoor air temperature model in the building at the target moment is obtained, and then a temperature comfort index at the target moment is introduced; a user temperature comfort constraint is constructed according to the indoor air temperature model in the building at the target moment, so that the overall regulation resource consumption and the user experience can be fully considered for the pre-constructed regulation resource consumption function associated with the rated power of the air conditioner under the constraint, and the situation that the regulation resource consumption and the user experience are influenced due to relatively single factors considered by regulation can be avoided. And the use experience of the user cannot be ensured.

The invention discloses a multi-microgrid integrated energy system scheduling method, device and equipment. The method comprises the following steps: acquiring micro-grid basic data corresponding to the multi-micro-grid integrated energy system; according to the micro-grid basic data, a minimum cost model corresponding to the multi-micro-grid integrated energy system is constructed, and the minimum cost model at least comprises an energy purchase cost algorithm, a carbon transaction cost algorithm, a carbon sequestration income algorithm, a wind curtailment and light curtailment cost algorithm, an electric energy interaction cost algorithm and an operation and maintenance cost algorithm. Determining micro-grid scheduling data corresponding to the multi-micro-grid integrated energy system based on a segmented linear algorithm according to the minimum cost model, and performing scheduling processing on the multi-micro-grid integrated energy system based on the micro-grid scheduling data to determine dynamic scheduling data of the multi-micro-grid integrated energy system, and scheduling the multi-microgrid integrated energy system according to the dynamic scheduling data so as to ensure efficient, safe, economical and environment-friendly operation of the system.

The invention relates to the technical field of data analysis, in particular to an Internet of Things time sequence analysis method based on a self-adaptive hypergraph neural network, which comprises the following steps: acquiring a plurality of human activity identification data, determining a selected data volume according to data fluctuation, determining activity label similarity according to information association degree, and obtaining a time sequence analysis result; according to the activity label similarity, human activity identification data is selected as training data according to the overall similarity and the selection similarity or according to the amplitude change rate; according to the motion complexity of the training data, determining whether to increase and adjust the scale feature level to determine a target scale feature level; and the features of the points are connected through Tensor, and classification is carried out through Linear to obtain a prediction category result corresponding to the training data. The method can improve the recognition precision of the human activity recognition task.

The embodiment of the invention discloses a fitting precision detection method and device and a storage medium. The fitting precision detection method comprises the steps that at least two images are acquired and preprocessed, and the at least two images are images of different angle areas of the plane object; for each image, capturing according to a preset capturing frame, performing linear fitting on the captured area according to a Hough linear algorithm to obtain a fitted linear segment, screening the fitted linear segment according to a preset screening condition, and taking the screened linear segment as the protective film edge of the corner area corresponding to the current image; and calculating the laminating precision of the protective film according to the position information of all the screened straight line segments.

The invention relates to the technical field of image processing, in particular to a compensation dispersion transfer entropy-based causal analysis method, and aims to accurately capture an instantaneous causal relationship and improve the accuracy of a constructed brain network. The method comprises the following steps: acquiring an original time sequence, and normalizing the original time sequence by adopting a normal cumulative distribution function to obtain a normalized time sequence; and performing dispersion processing on the normalized time sequence, and endowing each normalized data point with an integer in a preset integer range through a linear algorithm to obtain a dispersion time sequence. And performing information transmission quantitative analysis on the dispersive time sequence based on the information entropy, and evaluating the uncertainty degree of information flow transmission from the source system to the target system and the uncertainty degree of information flow transmission from the source system to the target system under the constraint of a preset condition. And introducing a compensation mechanism to optimize the uncertainty evaluation result.

The invention discloses a vacuum infusion cyclic extrusion pressure control device and method based on a double-layer linear algorithm, and belongs to the technical field of composite material manufacturing, the device comprises double sets of independent hardware systems, a double-pipeline control system and a pressure control box; the method comprises a pressure control method and a compaction rate control method. The compaction rate control method comprises the steps that in the cyclic extrusion process, the compaction rate is accurately controlled by adjusting the opening proportion of a vacuum electromagnetic valve of a second pipeline of the control device, and a double-layer linear algorithm is adopted; the pressure control method adopts segmented pressure control, comprises high-pressure section control and low-pressure section control, and is controlled by a first pipeline or two pipelines of the control device together. The pressure control precision reaches + / -0.1 kPa, and the compaction rate control precision is within + / -6%; the control precision can be kept within the atmospheric pressure change range of 97.5-102.5 kPa, and the system adapts to different geographical and meteorological conditions; the full-range control from 0.5 kPa to atmospheric pressure and the compaction rate control from 1 kPa / s to 5.5 kPa / s can be realized.

The invention discloses a method for early warning of aerodynamic instability of an inlet of a hypersonic aircraft engine. After normalizing the pre-processed wall pressure data characteristics, a binary classification of start-up and non-start-up state data is performed based on a linear SVM algorithm, an optimal classification surface is determined as a non-start-up boundary of the current state of the inlet, and an η% stability margin is defined to determine its non-start-up warning boundary. The non-start-up boundary is used to realize instability judgment of the inlet. On this basis, the back pressure multiple attribute of the wall pressure data along the way is trained and evaluated based on a BP neural network, and the wall pressure along the way at the non-start-up boundary and its warning boundary is used as a regression input. The non-start-up state of the inlet and the back pressure multiple boundary value of its warning state are predicted through a BP neural network model, which can be used to realize aerodynamic instability early warning of a hypersonic aircraft engine within a specific Mach number and flight altitude range. A new feasible scheme is provided for early warning of aerodynamic instability of the inlet of a hypersonic aircraft engine.

The invention provides a simulation engine multi-parameter cooperative control method, device and equipment, a medium and a program product, and the method comprises the steps: obtaining the load change information of a simulation engine; under the condition that the load change information is larger than a preset threshold value, the operation state of the simulation engine is determined to be a transient state, a closing instruction is sent to a control module of an exhaust gas recirculation valve, and a first control instruction is sent to a control module of a variable nozzle turbine; the opening degree of the variable nozzle turbine is reduced to a first preset opening degree and is increased to a second preset opening degree after being kept for a preset time length, and the opening degree of the variable nozzle turbine is used for determining transient air inlet data transmitted to the simulation engine; and a second control instruction is sent to the fuel injection module, so that transient fuel injection data of the fuel injection module is controlled to reach first preset fuel injection data within the total control duration by using a stage-linear algorithm, and the fuel-oxygen equivalence ratio, obtained through simulation based on the transient air inlet data and the transient fuel injection data in the transient state, of the simulated engine is reduced.

The invention discloses a multi-region sheet resistance optimization calculation method and system based on voltage scanning, and belongs to the field of electronic device design and control. By inputting area parameters and power density parameters of a multi-region device and combining voltage scanning and sheet resistance constraint verification, global optimization matching of sheet resistance of each region is realized. Compared with a traditional local optimization method, global solution space search is achieved based on the O (N) linear algorithm, local optimal traps are avoided, the calculation efficiency is improved by more than 10 times, the square resistance uniformity standard deviation can be reduced to below 0.3 and is verified in distributed thin film heater design, and the effect of remarkably improving temperature uniformity and device reliability is achieved.

The invention discloses a method for predicting the minimum miscible pressure of CO2 and crude oil by using machine learning, and the method comprises the steps: carrying out the high-precision modeling based on a linear SVM algorithm, building a correlation prediction formula, and constructing a correlation prediction model; and training and optimizing the correlation prediction model by using the collected data, and establishing a stacked integrated model in combination with a random forest regression model RF and the correlation prediction model to realize accurate prediction of the CO2-crude oil minimum miscible pressure MMP. According to the method, the time and the cost for determining the CO2-crude oil MMP value can be reduced to the maximum extent. Therefore, a reliable CO2-crude oil MMP value can be provided through prediction, and the miscible phase type of CO2 oil displacement is determined, so that the design of a CO2 oil displacement scheme is improved, the optimal design of an injection-production scheme is guided, and the crude oil recovery rate is increased.

Computer processing and outcome prediction systems and methods used to generate algorithm time prediction polynomials, inverse algorithm time prediction polynomials, determine race conditions, determine when a non-linear algorithm can be treated as if it were linear, as well as automatically generate parallel and quantum solutions from classical software or from the relationship between monotonic attribute values.

This invention provides a drive circuit for inverter loads, relating to the field of inverter technology. It includes a control module, a drive module, a power supply module, a signal acquisition and processing module, and a load matching module. The load matching module has impedance matching, tuning matching, and shaping filtering functions, providing universally adaptable drive capabilities for most types of motors and ultrasonic transducers. The power supply module uses high-speed MOSFETs, and the drive module uses a high-speed drive chip to adapt to the high-speed MOSFETs, offering a wide frequency range. A DDS waveform generation module is used to achieve PWM wave output at different frequencies, with extremely low frequency change steps and smooth transitions. Combining a variable step-size algorithm and a LADRC algorithm, it can automatically and accurately lock the operating resonant frequency of the inverter load within a certain frequency range. The employed linear LADRC algorithm adjusts the behavior of the drive circuit through real-time measurement and estimation, adapting to the variations and uncertainties in the complex working environments of motors and ultrasonic transducers.

The present invention discloses an improved arc length method for efficiently solving nonlinear thermal boundary problems, and belongs to the field of high-performance computing simulation technology. The method first needs to establish a geometric model to be analyzed, and then the geometric model to be analyzed is meshed with any unit type, and then the boundary conditions and load conditions of the model to be analyzed are set, and finally the above conditions are input into the algorithm solver. In the solution, the size of the second norm of the tangent increment obtained in the prediction stage of adjacent incremental steps is determined. The arc length calculation method of the algorithm prediction stage is selected according to the determination result. Thereby, the algorithm solution process is accelerated under the premise of realizing reliable solution of heat transfer nonlinear problems. The present invention combines heat transfer theory, finite element method, and nonlinear algorithm control strategy, improves on the theoretical basis of the existing arc length method, and uses the excellent numerical calculation ability of computers as a carrier to solve thermal nonlinear problems, which not only improves the numerical accuracy of the calculation results, but also improves the speed of solving thermal nonlinear problems.

The invention discloses a rapid approximate block coding quantum circuit construction method and a linear equation set solving method and device. The solving method comprises the following steps: obtaining a linear equation set Ax = b; constructing a time-dependent Hamiltonian H (f (s)) of adiabatic evolution, wherein the eigenstate of the initial Hamiltonian H (0) is expressed as bgt; the eigenstate of the target Hamiltonian H (1) is expressed as A-1bgt; ; a discrete adiabatic evolution quantum circuit is constructed, the discrete adiabatic evolution quantum circuit has an evolution operator UH (f (s)) corresponding to the time-dependent Hamiltonian H (f (s)), the evolution operator UH (f (s)) continuously acts on the discrete adiabatic evolution quantum circuit, and the initial quantum state corresponds to the eigenstate bgt of the adiabatic evolution initial Hamiltonian H (0); the final quantum state corresponds to the eigenstate A-1bgt of the adiabatic evolution target Hamiltonian H (1); . The invention provides a specific quantum circuit implementation method of sparse Hamiltonian fast approximate block coding, and realizes an adiabatic quantum calculation linear algorithm based on fast approximate block coding to complete the solution of a linear equation set.

The invention provides a neck-worn sensor that is a single, body-worn system that measures the following parameters from an ambulatory patient: heart rate, pulse rate, pulse oximetry, respiratory rate, temperature, thoracic fluid levels, stroke volume, cardiac output, and a parameter sensitive to blood pressure called pulse transit time. From stroke volume, a first algorithm employing a linear model can estimate the patient's pulse pressure. And from pulse pressure and pulse transit time, a second algorithm, also employing a linear algorithm, can estimate systolic blood pressure and diastolic blood pressure. Thus, the sensor can measure all five vital signs along with hemodynamic parameters. It also includes a motion-detecting accelerometer, from which it can determine motion-related parameters such as posture, degree of motion, activity level, respiratory-induced heaving of the chest, and falls.

The application provides a blood flow signal denoising method based on PSO-EMD optimized wavelet threshold, which comprises the following steps: collecting photon signals scattered by tissues by using a photoelectric detector, processing the signals by a digital correlator, and obtaining blood flow signals by N-order linear algorithm calculation; decomposing the blood flow signals by EMD to obtain IMF signals; selecting a wavelet threshold denoising algorithm, and dynamically screening the threshold by using PSO, taking the signal-to-noise ratio as a target fitness function, and selecting optimal parameters by considering the target adaptation principle and the effective convergence principle of the PSO algorithm; removing noise from high-frequency IMF components in the IMF signals according to the wavelet threshold denoising algorithm of the optimal parameters, and superimposing and reconstructing the high-frequency IMF components, low-frequency IMF components and residual components to obtain denoised blood flow signals. The application can effectively remove noise and maximize the details of the effective information of the blood flow signals.

This invention relates to the field of wireless communication transmission technology and discloses a data transmission method for weak network environments. The method includes: acquiring the channel impulse response of the first and second symbol periods of a wireless channel; extracting multipath delay parameters, signal amplitude parameters, and discrete time axis coordinates; mapping the parameters to a spatiotemporal parametric coordinate system; generating first and second channel feature surfaces; calculating the curvature variance of the first channel feature surface; when the curvature variance is lower than a preset correlation distortion threshold; extending the second channel feature surface along the evolution gradient trend to generate a predicted channel feature surface; thereby determining the equalizer tap coefficients and compensating for data symbol distortion. This invention enhances transmission reliability in highly dynamic weak network environments by locking the gradient divergence path generated by the linear algorithm in a Doppler shift environment through geometric surface physical continuity constraints.