17 results about "Nonlinear normalization" patented technology

The invention relates to the technical field of power system safety and stability control, and discloses a new energy field station broadband oscillation risk early warning method based on voltage dynamic indications, which comprises the following steps: acquiring instantaneous value data in a new energy electric field station to obtain electric quantity time sequence data; processing the electrical quantity time sequence data according to the target sliding window and outputting risk characteristic quantity; performing nonlinear normalization processing on the risk characteristic quantity by using a weighted fusion technology, and generating a single broadband oscillation risk index reflecting the broadband oscillation level of the new energy station according to a normalization processing result; and carrying out difference comparison on the single broadband oscillation risk index and a dynamic risk early warning threshold, and triggering an oscillation risk early warning signal and implementing oscillation risk early warning when a difference result meets a target warning requirement. According to the method, the defense gateway is moved forward by capturing the dynamic symptoms before oscillation occurs, so that the fundamental conversion from post-event to beforehand is realized, and a precious decision window period is won for operation adjustment.

The invention discloses a man-machine co-driving detection method and device based on steering wheel torque and a medium. The method comprises the steps that a steering wheel torque accumulated value of a target vehicle is obtained; performing nonlinear normalization processing on the steering wheel torque accumulated value to obtain a torque confidence coefficient, and if the torque confidence coefficient is greater than or equal to a driver take-over intervention threshold value, calculating to obtain a comprehensive confidence coefficient by combining the torque confidence coefficient and the correction confidence coefficient; and if the comprehensive confidence is greater than or equal to the driver take-over final judgment threshold, judging that the driver of the target vehicle intervenes in driving. According to the man-machine co-driving detection method and device based on the torque of the steering wheel and the medium, the extensive judgment mode of a single torque threshold value is abandoned through a progressive technical means in accordance with the real operation characteristics of a driver, layer-by-layer accurate screening and verification logic is formed from signal collection to comprehensive judgment, and the detection accuracy is improved. The problem that whether a driver actually intervenes in vehicle driving operation is difficult to accurately recognize and judge when auxiliary driving is started can be solved.

The invention discloses a regional consumption demand calculation method based on e-commerce platform data, and belongs to the technical field of regional consumption demand calculation. The problem that a regional consumption demand calculation method is inaccurate in calculation is solved. The method comprises the steps of collecting basic data, obtaining preprocessed regional transaction data through an abnormal value self-adaptive suppression and neighborhood spatial interpolation method, carrying out spatial association, and calculating a regional commodity consumption characteristic entropy value; decomposing a time trend and a space trend in the consumption data through a periodic function and a space overflow model, and calculating a decomposed consumption trend and a decomposed daily average consumption trend; considering the influence of price elasticity and logistics timeliness, and calculating a consumption demand coupling factor; performing spatial heterogeneity correction through a geographical weighted kernel function to obtain a corrected consumption demand coupling factor; the regional commodity demand index is calculated through nonlinear normalization and supply constraint. Compared with a traditional theoretical demand calculation method, the method has the advantage that the landing conversion rate is improved.

The invention relates to the technical field of gene detection, and discloses a mycobacterium tuberculosis drug resistance gene detection data analysis system which comprises a data import module, a signal correction module, a feature extraction module, a multi-stage interpretation module, a result fusion module and a report generation module. The system performs baseline drift correction, abnormal point detection and nonlinear normalization on a fluorescence intensity value output by a detector, extracts multi-dimensional features such as peak intensity, a mutation ratio, a consistency coefficient and an in-batch difference index, and constructs a dynamic threshold function to realize wild type and mutation type layered judgment. A drug resistance risk conclusion is generated through drug resistance characteristic matrix mapping, a standardized detection report is output, and intelligent analysis of a detection result and automatic report generation are achieved. Through multi-stage signal correction, multi-dimensional feature extraction and dynamic threshold interpretation, intelligent analysis and standardized report output of a mycobacterium tuberculosis drug resistance gene detection result are realized, and data judgment accuracy and clinical application efficiency are improved.

The invention discloses a spectrum signal detecting and positioning method and system based on unsupervised reinforcement learning, and belongs to the technical field of wireless spectrum sensing and electronic countermeasure. In order to solve the problems of low-power and low-SNR signal automatic detection and positioning in a complex electromagnetic environment, the method comprises the steps of nonlinear normalized spectrum mapping, anchor frame heuristic initialization, multi-source state coding and attention fusion, PPO-based continuous action reinforcement learning, a multi-dimensional learnable reward function and a multi-round dynamic detection strategy. And adaptive detection and track positioning of the frequency spectrum signal are realized. According to the method, high-precision detection and time-frequency track positioning of various types of signals in a low SNR environment can be realized without manual labeling, and the robustness and universality of spectrum sensing are improved.

The invention discloses an industrial robot dynamic feedback normalization reinforcement learning optimization method and system. The method comprises the following steps: S1, constructing an experience playback buffer area D, Q network parameter theta and a target network parameter theta-in an industrial robot memory; s2, the industrial robot executes an action based on a control instruction output by the current Q network, and an original feedback signal rraw and a next state st + 1 are observed; s3, performing nonlinear normalization processing on the original feedback signal rraw; s4, if the industrial robot triggers a system failure protection mechanism, determining a negative system failure penalty term p; s5, the complete state transition tuple of the final feedback signal rfinal is stored in an experience playback buffer area D; s6, sampling training data from the experience playback buffer area D, calculating a target Q value, and updating a Q network parameter theta; and S7, when a preset network updating condition is satisfied, synchronizing the Q network parameter theta to the target network theta-, and updating the parameters alpha t and beta t in the nonlinear normalization based on the gradient statistics.

The application discloses a breast reconstruction type recommendation system and method based on multi-modal data, relates to the technical field of breast reconstruction type recommendation, and aims to solve the technical problems of traditional operation type recommendation depending on linear superposition evaluation of a single index, being unable to quantize the nonlinear complex mechanism of metabolic and vascular injury, and leading to inaccurate risk evaluation and lack of targeted recommendation, the system comprising a feature calculation and analysis module, wherein the feature calculation and analysis module comprises a clinical information structured unit, a metabolic and vascular injury index calculation module, an image feature unit and a test feature unit. Through the metabolic and vascular injury index calculation module, the nonlinear normalization function, the standardization function, the interactive effect analysis algorithm and the Sigmoid function are integrated, the true risk difference between patients with short disease duration but poor blood glucose control and patients with long disease duration but good blood glucose control is effectively distinguished, the risk evaluation is more in line with the pathophysiological law, and a scientific quantitative basis is provided for operation type selection.

The application provides a lightweight anti-UAV detection method and device based on frequency-space fusion, which comprises the following steps: acquiring visible light and infrared images and performing registration preprocessing, performing spatial domain mapping and fast Fourier transform on the visible light and infrared images to obtain amplitude spectrum and phase spectrum, generating fused amplitude spectrum, performing complex reconstruction on the fused amplitude spectrum and the original phase spectrum, mapping back to the spatial domain through inverse Fourier transform to obtain a frequency domain fusion feature, splicing the frequency domain fusion feature and a local feature to form a final fusion tensor, introducing a ghost operator to perform ghost convolution to extract a multi-scale feature, equivalent to a boundary box as a high-dimensional Gaussian distribution, calculating a distance metric and performing nonlinear normalization on the distance metric to obtain a similarity index, obtaining a total loss function based on the similarity index and multiple losses, and training an UAV detection model based on the total loss function, the multi-scale feature and a data set to detect UAVs through the model.

The application discloses a kind of abrasive water jet tunneling surrounding rock damage real-time imaging method and system, comprising: obtaining the real-time space coordinates of nozzle end in roadway three-dimensional coordinate system, and collecting water pressure data in high-pressure fluid pipeline system;Vibration response signal is collected by acceleration sensor and acoustic emission sensor fixed on nozzle;The vibration response signal is band-pass filtered and denoised and water pressure nonlinear normalization correction;Wavelet packet decomposition is carried out to the corrected signal, and the characteristic vector including signal energy gravity frequency, vibration amplitude root-mean-square value, wavelet packet energy entropy and each frequency band energy proportion is extracted;Mahalanobis distance of characteristic vector is calculated based on reference lithology characteristic library, and surrounding rock damage is judged in combination with high-frequency band energy proportion drop trend, and damage index is calculated;Damage index is associated with space coordinates, and the three-dimensional damage thermodynamic diagram of roadway surrounding rock is generated.The application realizes the real-time perception and imaging of surrounding rock damage under non-contact condition.

The invention discloses an inter-well relevancy calculation method based on multi-attribute layered coupling, and relates to the technical field of oil reservoir development. The calculation method comprises the following steps: S1, data input and target well layer depth standard division; s2, constructing a multi-attribute joint empirical cumulative distribution function and performing nonlinear depth mapping; s3, carrying out hierarchical nonlinear normalization and interpolation; and S4, layered weighted coupling relevancy calculation is carried out. According to the method, multi-attribute information is systematically fused, layered nonlinear depth mapping and a weighting strategy are combined, multi-level and multi-attribute comprehensive correlation analysis between the target well and the adjacent well is achieved, the accuracy and applicability of correlation calculation are improved, and the requirement for inter-well data comparison is met.

This invention belongs to the field of detection technology, specifically disclosing a method and system for detecting chloride ions in high-concentration samples used in the salt-making industry. The detection method includes the following steps: (1) cyclically taking samples from the process pipeline to ensure real-time sample availability; (2) diluting the sample to ensure that the concentration of the sample falls within the optimal detection range; (3) inputting the diluted sample into the titration system, performing titration using potentiometric titration, outputting the first derivative DE / DV in real time, and using a nonlinear normalization function to normalize the original DE / DV sequence to obtain S. norm Value, according to S norm Adjust the speed of the titration pump, S norm The point corresponding to the maximum value is the titration endpoint. The titration volume V2 of the silver nitrate standard solution at this point is recorded, and the chloride ion concentration in the pipeline sample is calculated. This invention stably maps the wide-range fluctuating DE / DV signal to a fixed interval, solving the core problem of inaccurate endpoint determination and ensuring the accuracy and reliability of the detection results.

The present application relates to the technical field of inverter output current normalization processing, and provides an inverter current normalization method based on balanced distortion, which comprises the following steps: sequentially performing filtering processing and compression processing on inverter output current to obtain current amplitude I ; determining the positive and negative directions of original phase current and generating a sign factor S ; performing weighting processing on the current amplitude I by using a preset sensitivity coefficient factor p ; and performing normalization processing on the inverter output current amplitude after weighting processing by using a nonlinear normalization function. The present application can better balance the distortion degree of different amplitude currents, and can effectively improve the sensitivity of fault characteristics, especially under light load conditions. By uniformly mapping the current value with amplitude variation to a specified interval, reliable diagnosis of inverter single tube open circuit fault under variable load conditions is realized, and misjudgment caused by load variation is avoided.

The application discloses a product carbon mark management system and method based on a cloud platform, and relates to the field of environmental information management and green supply chain.The method comprises the following steps: collecting carbon emission data of each stage from the whole life cycle of a product, generating standardized emission data through nonlinear normalization; generating stage carbon synergy factors based on node collaborative analysis and sensitivity evaluation, and constructing a cross-enterprise carbon synergy matrix; calculating stage carbon impact factors and damping coefficients to generate comprehensive stage emission factors; calculating cumulative sensitivity accordingly to form a whole life cycle carbon footprint; identifying abnormal stages and dynamically correcting in combination with historical data to generate carbon mark stability; and finally generating a carbon mark and storing it through a blockchain to realize a cloud platform management closed loop.Through efficient collection and standardization of whole life cycle carbon emission data, multi-layer collaborative analysis and dynamic correction, and credible storage and management closed loop of the carbon mark, the accuracy, reliability and traceability of the product carbon mark are realized.

The present application relates to the technical field of intelligent evaluation of road engineering, and particularly relates to a method for predicting the aging degree of pavement material under the action of environmental temperature and humidity cycles, which specifically comprises the following steps: obtaining time-series temperature and humidity data and aging performance indicators through laboratory accelerated aging test, converting the time-series temperature and humidity data and aging performance indicators into effective cumulative exposure through time decay weight and temperature and humidity synergistic strengthening function; constructing a label vector by using nonlinear normalization and adaptive correction, and training a neural network model in combination with a multi-resonance feature vector; and outputting the predicted value of the aging indicator in the original physical dimension by means of neural network model prediction and reverse normalization processing. The present application realizes the quantitative characterization of the nonlinear cumulative effect of temperature and humidity cycles, improves the accuracy of aging prediction, and is suitable for service life evaluation and maintenance decision of pavement material.

The application discloses a kind of two-way MPPT interconnection's energy storage product mutual charging efficiency optimization method, it is related to portable energy storage equipment interconnection power supply and efficiency optimization technical field, including the following steps: after starting energy storage equipment, equipment EMS system self-checking and acquisition battery SOC, cell temperature, two-way MPPT module operating state, by communication module transmission to associated control APP;Control APP runs equipment interconnection adaptability evaluation model, carries out nonlinear normalization to the data acquisition, judges equipment interconnection adaptability;In the application, using two-way MPPT direct interconnection link, equipment interconnection adaptability evaluation model, coupled charging and discharging parameter optimization algorithm, load standby power collaborative algorithm and abnormal adaptive processing mechanism, combined with machine learning and MPPT switching optimization technology, solve the problems of traditional AC inverter mutual charging low efficiency, equipment collaborative ability is insufficient, load power supply is disturbed, operation is complicated and potential safety hazard.

The invention relates to the technical field of data processing, and discloses a big data-based professional talent background information investigation method, which comprises the following steps of: firstly, carrying out structured atomic segmentation on a resume full text, and extracting an experience set containing time metadata and text content; secondly, logarithmic nonlinear normalization is carried out on the experienced time sequence span, and the endogenous semantic information quality is calculated; further generating a post time sequence-semantic entropy flux reflecting the information flow rate in unit cognitive time, and constructing a global reference manifold based on the historical trajectory of the job seeker; and finally, measuring and calculating the deviation degree of the local experience relative to the personal benchmark, generating an abnormal index, comparing the abnormal index with a preset threshold value, and outputting a deterministic binary judgment result. According to the invention, the precise detection of the beautification of the resume is realized by capturing the illegal feeling of the forged resume in time and content density.

The invention discloses a four-dimensional dynamic adaptive gating fusion method for mine communication, and belongs to the technical field of radio transmission. Comprising the following steps: step 1, carrying out four-dimensional state sensing to obtain a four-dimensional channel state vector s; 2, performing nonlinear normalized mapping on the four-dimensional channel state vector s; 3, inputting the normalized features in the step 2 into an ultra-lightweight neural network for weight decision making; and 4, applying stability constraint to the weight decided in the step 3, and realizing self-adaptive fusion of the output of the physical model and the data model. According to the method, the four-dimensional state of the channel can be sensed in real time, the signal compensation parameters are dynamically adjusted according to the four-dimensional state sensed in real time, the method plays a leading role in signal compensation, and the bit error rate in a mine communication scene is remarkably reduced. The problems that a traditional hybrid drive architecture fusion strategy cannot adapt to second-level violent changes of the underground channel state and is difficult to meet the real-time requirement of underground intrinsic safety equipment are solved.