12 results about "Nonlinear calibration" patented technology

This invention discloses a system and method for monitoring partial discharge and thermal characteristics of switchgear, comprising: a dual-spectrum imaging module that simultaneously acquires a sequence of ultraviolet pulse images and an infrared temperature field sequence; a spatiotemporal synchronization control module that generates a synchronization control signal and acquires operating condition information; and an edge fusion diagnosis module deployed at a local edge that extracts pixel-by-pixel features from the spatiotemporally aligned dual-spectrum images, constructs a joint feature vector matrix of partial discharge and temperature rise, and performs real-time nonlinear correction based on the operating condition feature vector by retrieving an operating condition feature correction model, then retrieves the spatiotemporal weight allocation rule corresponding to the current defect evolution stage, dynamically allocates spatial weights and temporal weights for weighted fusion, generates a fusion confidence heatmap, and thus achieves defect identification and level classification. This invention achieves accurate diagnosis of insulation defects in switchgear through a collaborative mechanism of adaptive operating condition correction and dynamic weight fusion.

This invention relates to the field of semantic processing technology and discloses a high-dimensional vector retrieval and semantic generalization system with cross-modal semantic alignment. The system includes: an index building unit for calculating the density weighting factor of vector clusters and storing it in index nodes; a query positioning unit for determining whether a query vector falls into a target cluster or a sparse region; an adaptive metric calibration unit for generating nonlinear correction coefficients based on the density weighting factor to dynamically adjust the distance metric weights; and a semantic expansion retrieval unit for calculating the shortest path in sparse regions based on index connectivity to aggregate retrieval results. This invention introduces a density weighting factor as metadata into the index structure to achieve adaptive nonlinear calibration of the retrieval metric standard according to local data density, thus solving the problem of retrieval accuracy imbalance caused by uneven distribution of heterogeneous modal data.

This invention discloses a harmonic mixing fractional phase-locked loop (PLL) assisted by a digital-to-time converter (DTC), comprising a reference path signal generation module, a mixing path signal generation module, and a harmonic mixing PLL loop module. The reference path signal generation module mainly consists of a multi-mode divider, a Delta-Sigma modulator (DSM), and a digital-to-time converter (DTC) to achieve initial cancellation of quantization noise. The mixing path signal generation module mainly consists of a prescaler to provide a suitable local oscillator signal. The harmonic mixing PLL loop module adopts a harmonic mixing feedback architecture, making the equivalent division ratio of the main loop 1. This invention eliminates the need for complex DTC nonlinear calibration, reducing power consumption and design complexity, while effectively suppressing quantization noise folding and achieving ultra-low phase noise output.

This invention relates to the field of photodetector calibration technology, and more particularly to a real-time nonlinear calibration method and apparatus for a single-photon detector. The method uses a relay circuit to enable the photodetector to simultaneously output an analog voltage signal and a photon pulse signal. Synchronous acquisition of the analog voltage signal and photon pulse signal is achieved through an analog acquisition unit and a photon counting unit. A data processing unit preprocesses the two signals and outputs a nonlinear correction factor curve. The apparatus for executing the above method is also disclosed, which can effectively achieve nonlinear correction of the photodetector under different temperature conditions, ensure the accuracy of photon counting within different dynamic response ranges, and significantly improve the overall performance of the photodetector and measuring instruments.

This invention discloses an automatic question difficulty assessment method that integrates visual features and textual semantics. Addressing the shortcomings of existing question difficulty assessment methods, which rely on manual annotation, are highly subjective, and are difficult to adapt to open-ended visual question answering, this invention constructs an unsupervised complexity modeling framework: After acquiring image-question pairs, a multi-visual language model is invoked to generate candidate answers; a semantic alignment network is used to calculate the semantic consistency score between the answer and the reference, simultaneously generating an Image Perception Quality Index (PQI); based on the performance of multiple models, the capability value is estimated and weighted to obtain the basic complexity; after PQI nonlinear correction and normalization, a continuous difficulty score is output, which can be mapped to a discrete difficulty level; simultaneously, a self-comparative evaluation mechanism is provided to achieve complexity ranking verification and calibration without manual labeling. This invention eliminates the need for manual difficulty annotation, takes into account the influence of textual semantics and image quality on difficulty, and provides a stable quantitative basis for adaptive learning, test paper generation, and model capability diagnosis.

The application discloses a kind of server memory parameter intelligent optimization method and device, the method is by collecting the characteristic data of service scene;According to the characteristic data, the corresponding target memory time sequence parameter is generated by the preset parameter mapping hybrid model, wherein the parameter mapping hybrid model includes the basic mapping layer for linear mapping and the feature interaction enhancement layer for nonlinear calibration;The target memory time sequence parameter is checked based on the service priority of service scene security, and after passing the security check, the target memory time sequence parameter is configured, the accurate matching of memory time sequence parameter and dynamic service scene is realized, and performance potential can be fully released under the premise of guaranteeing the stability of core business, can significantly improve the intelligent level and overall operation efficiency of business operation system.

The application discloses a two-dimensional galvanometer visual axis pointing coupling correction method, and mainly solves the problem that the visual axis pointing error calculated by the prior art is in the order of milliradians, and it is difficult to meet the high-precision correction requirement.The method comprises the following steps: step 1, establishing a visual axis pointing model for a two-dimensional galvanometer to be corrected; step 2, acquiring a linear model between the rotation angle of the two-dimensional galvanometer and the visual axis pointing angle of a two-stage stable pointing system based on the visual axis pointing model by using a linear approximation method; acquiring a pointing error based on the visual axis pointing model and the linear model; performing nonlinear correction on the pointing error by using a Volterra series to acquire a nonlinear correction relationship; step 3, acquiring a corrected pointing error by using the nonlinear correction relationship; and step 4, acquiring the control angle of the two-dimensional galvanometer based on the corrected pointing error.The application can reduce the visual axis pointing error from the order of milliradians to the order of microradians, the pointing calculation precision is improved by about 75 times, and the visual axis pointing calculation precision is significantly improved.

This invention discloses a wide-range current measurement dynamic compensation method and system for excitation system testing, belonging to the technical field of excitation system technology. This invention acquires the temperature of key chips in the signal conditioning module in real time, dynamically calculates zero-point offset and actual gain based on factory calibration parameters, performs digital filtering and noise reduction on the acquired voltage signal, and uses a single-iteration approximation method to complete initial current estimation and nonlinear correction, achieving full-range dynamic error compensation. The system includes a signal conditioning and acquisition module, a temperature acquisition module, a dynamic compensation processor, and an output module. This invention overcomes the shortcomings of traditional testing devices such as manual range switching, static calibration lag, and low accuracy at low currents. It can achieve continuous high-precision measurement over a wide range from milliamperes to tens of amperes, effectively suppressing temperature drift, noise, and nonlinear errors, significantly improving testing efficiency and stability, and is suitable for on-site maintenance and commissioning of hydropower station excitation systems.

The application discloses an ammonia quantitative monitoring method facing anisotropic response of an olfactory sensor and edge calculation, belongs to the technical field of gas monitoring, and specifically comprises the following steps: constructing an anisotropic response double-sensor array; collecting voltage signals and environmental temperature and humidity data in real time; performing anisotropic response logic gate determination on two voltage signals; storing voltage values and temperature and humidity data corresponding to effective signals into a sliding window buffer, extracting time domain statistical features and dynamic trend features, combining temperature and humidity to construct voltage-temperature ratio and voltage-humidity ratio as physical cross features, and obtaining feature vectors after standardization; inputting the feature vectors into a lightweight neural network model deployed at the microcontroller end, and outputting ammonia concentration values after nonlinear calibration and temperature and humidity compensation. The application realizes high-reliability real-time quantitative monitoring of ammonia by inhibiting environmental common-mode interference through anisotropic response logic gating, and realizing temperature and humidity adaptive compensation through physical cross features and an edge neural network.

This application provides a signal processing system for a lidar system, comprising: a detector, a sampling module, an amplification circuit, and a nonlinear correction loop. The detector converts the laser echo signal from an optical signal into an electrical signal. The amplification circuit amplifies the echo electrical signal according to a gain control signal sent by the nonlinear correction loop. The nonlinear correction loop sends the gain control signal to the amplification circuit, selects a sampling module from a set of sampling modules based on a first spectral characteristic, and samples the amplified echo electrical signal through the selected sampling module. This application solves the problem in related technologies where the gain of the amplification circuit is fixed or inflexible, affecting the accuracy of subsequent signal processing. It ensures the linearity of signal processing and the stability of signal strength, and to a certain extent avoids nonlinear distortion.

The application discloses a kind of electromagnetic pump flow control methods based on double duty ratio and current feedback compensation, it is related to flow control field, the electromagnetic pump flow control method based on double duty ratio and current feedback compensation includes: step S1, generates double duty ratio pulse width modulation signal, wherein first level PWM signal defines electromagnetic pump's macroscopic on-off timing to adjust the number of times of work per unit time, the beneficial effects of the present application are: the present application generates double duty ratio PWM signal, to first level coarse adjustment unit time work frequency and according to interval distribution on half wave, inhibit the low-frequency flow pulsation of traditional zero-crossing power regulation;By real-time acquisition coil current and closed-loop correction second level PWM duty cycle, effectively compensate the flow drift caused by coil temperature rise and voltage fluctuation;By querying nonlinear correction lookup table to obtain the optimized combination of first and second duty cycles, overcome the low-duty-cycle driving dead zone and high-duty-cycle magnetic circuit saturation nonlinearity.

This invention discloses a pXRF-based method for measuring the thickness of low-density materials, belonging to the field of chemical analysis technology. The method includes placing a standard sample containing a specific metal element in front of the pXRF instrument's measurement window; placing several low-density target samples with different known thicknesses one by one between the standard sample and the measurement window to obtain a set of thickness values ​​and a corresponding set of element content values; using a piecewise nonlinear regression method to fit calibration curve equations of different functional forms to establish a piecewise nonlinear calibration curve; placing and measuring a low-density target sample of the same type with unknown thickness to obtain new element content values, which are then substituted into the corresponding inverse function to calculate the thickness. This invention utilizes pXRF to achieve rapid, non-destructive, and low-cost measurement of the thickness of low-density materials. It is easy to operate, can be used long-term after a single calibration, and has high measurement accuracy. It is suitable for thickness determination of homogeneous low-density materials such as films, sample bags, glass, and wood panels.