13results about How to "Reduce model error" patented technology

The invention relates to the field of environment monitoring, and discloses an industrial source VOCs distribution analysis system based on a laser radar, which comprises a data acquisition module, a spectrum extraction module, a preliminary modeling module, an inversion optimization module, an intelligent analysis module, a visualization module and a dynamic monitoring and early warning module. The data acquisition module acquires VOCs data of an industrial source area and processes the VOCs data; the spectrum extraction module is used for extracting spectrum characteristics of VOCs from the processed laser signals; the preliminary modeling module estimates the VOCs concentration; the inversion optimization module globally optimizes the preliminary model in combination with an atmospheric transmission model, a nonlinear inversion algorithm and a genetic algorithm; the intelligent analysis module further improves the concentration prediction precision; the visual module is used for displaying the spatial distribution of the VOCs; and the dynamic monitoring and early warning module gives an early warning for abnormal concentration change. According to the invention, high-sensitivity, real-time monitoring and accurate distribution analysis of industrial source VOCs are realized, and a scientific basis is provided for pollution control and risk management of an industrial environment.

The invention provides a composite insulator electric field distortion detection method based on an electroluminescent effect, and the method comprises the steps: firstly, employing silicone rubber as a matrix material, optimizing a formula and a preparation technology, and preparing an electroluminescent composite coating material which is highly sensitive to electric field intensity and has excellent adhesiveness and environmental stability; secondly, the unmanned aerial vehicle is used for carrying a high-resolution camera, and comprehensive and clear insulator surface images are obtained in combination with the optimized route design; then, multivariable regression prediction is performed on image data through an algorithm, local and global characteristics of electric field distribution are accurately captured, and modeling errors are reduced. And finally, comparing an actually measured electric field distribution curve with a theoretical curve, analyzing electric field distortion, and diagnosing potential defects by using a feature matching technology. According to the invention, high detection accuracy can be maintained in different environments, the efficiency and precision of electric field distribution detection are greatly improved, and reliable technical support is provided for maintenance and safe operation of power equipment.

The invention discloses a regional relative troposphere delay modeling method based on a sliding window, and belongs to the field of satellite navigation and positioning. The method comprises the following steps of: calculating ZTD difference values between any point and a reference point on the same latitude in a sliding window, and taking the minimum value in the ZTD difference values between the any point and the reference point within 80km-100km as a threshold value; in the same sliding window, establishing a function model between the ZTD difference value and the longitude according to the ERA5 data, predicting coefficients of the function model at different moments at the same reference point by using a BP neural network, and further calculating to obtain the ZTD difference value per hour; or in the window moving process, the BP neural network is used for predicting the coefficient of the predicted point through the coefficient of one datum point, and then the ZTD difference value between the predicted point and the surrounding points is obtained through calculation; and comparing the ZTD difference value with a threshold value, and if the ZTD difference value is smaller than the threshold value, directly carrying out double-difference method correction. According to the method, the troposphere delay difference value can be subjected to high-precision modeling and prediction.

The application relates to an optimal design method of a modular axial flux permanent magnet synchronous wind generator, and belongs to the field of wind power generation. The generator comprises a stator and a rotor; torque pulsation and torque density are used as optimization targets, and stator slot depth, rotor permanent magnet thickness and rotor permanent magnet pole arc coefficient are used as to-be-optimized design variables; constraint conditions of the design variables are determined, a Latin hypercube sampling experiment is carried out, a DNN proxy model of the optimization target and the to-be-optimized design variables is established; a Bayesian optimization algorithm is used to optimize hyperparameters of the proxy model, and optimal hyperparameters are obtained; the proxy model adopting the optimal hyperparameters is trained to obtain an optimal proxy model, and based on the optimal proxy model, an NSGA-II optimization algorithm is used to optimize the to-be-optimized design variables, and finally, an optimal combination of the to-be-optimized design variables is determined. The application can reduce modeling errors, improve calculation efficiency, robustly process multi-target trade-off under complex parameter coupling conditions, and effectively improve the performance of the modular generator.

The invention discloses a cosine similarity weighted multi-source statistical optimal near-field acoustic holography method, system, medium and equipment, and the method comprises the steps: arranging M microphones in a wavelength region away from a sound source, and obtaining the near-field holographic surface sound pressure pH of a housing structure; selecting planar, cylindrical and spherical unit wave functions based on the sound source appearance features to construct a multi-source wave function model; calculating the cosine similarity between the multi-source wave function model of each order and the holographic surface sound pressure pH, and updating based on the cosine similarity to obtain a weighted multi-source wave function model; and solving a transfer coefficient of the weighted multi-source wave function model of the holographic surface and the reconstruction surface through a Tikhonov regularization method, thereby realizing high-precision reconstruction of the multi-source sound field.

The application provides a wind turbine generator control method and device and a wind turbine generator. The method comprises the following steps: obtaining an initial control rate of the wind turbine generator based on a constructed dynamic model of the wind turbine generator and a selected reference model of the wind turbine generator; obtaining time delay information in a system of the wind turbine generator, calculating a time delay observation value of the wind turbine generator based on the time delay information, and using the time delay observation value to describe a dynamic response characteristic of the wind turbine generator, to represent an interference received by the wind turbine generator in a running state and a model error of the wind turbine generator; and substituting the time delay observation value into the initial control rate to obtain a target control rate, and using the target control rate to control the wind turbine generator. The method can reduce the influence of the working condition interference of the wind turbine generator on the control effect, compensate for the system modeling error, and improve the effectiveness and stability of the wind turbine generator control.

This invention discloses a furnace temperature control method, relating to the field of intelligent control of electronic packaging welding processes. The method includes: collecting real-time temperature data and welding process parameters of the substrate to be welded, and acquiring post-weld deformation data; aligning and fusing the real-time temperature data, process parameters, and deformation data to extract key features of the furnace temperature curve, and constructing a data record set by associating them with the deformation of corresponding measuring points; based on the data record set, establishing a quantitative prediction model using regression analysis or machine learning algorithms; under preset process constraints, optimizing the furnace temperature curve setpoint with the predicted warpage as the optimization objective, calculating the corresponding predicted warpage using the quantitative prediction model, generating the optimal furnace temperature curve setpoint, and sending it to the welding equipment for execution. By constructing a quantitative prediction model using temperature data from sensitive points and post-weld deformation, the optimal furnace temperature curve is generated through reverse optimization to suppress welding warpage.

The invention discloses a cross-medium aircraft water strike bounce trajectory generation method based on proxy model prediction, and belongs to the field of aircraft trajectory planning. According to the method, a cross-medium aircraft trajectory generation problem is modeled into a nonlinear programming problem with constraint conditions, solving is carried out based on a pseudo-spectral method, a feasible air section flight trajectory meeting constraints is obtained, and the reliability of cross-medium aircraft trajectory generation is improved. Latin hypersquare design is used as a sampling method, input distribution can be accurately reconstructed through sampling comparison with limited iteration times, samples with higher space coverage rate are provided for construction of an agent model, modeling errors caused by the fact that a simplified model is used by a cross-medium aircraft are reduced, and the modeling accuracy is improved. And the prediction result of the state quantity of the aircraft after the water striking section has relatively high accuracy. According to the method, nonlinear mapping is carried out on the state quantities before and after the water striking section through the proxy model constructed through finite element software simulation, an accurate water striking section kinetic equation does not need to be deduced, and the calculation complexity caused by modeling is remarkably reduced.

The invention discloses a boiler performance prediction method based on operating condition division and neural network, which relates to an online prediction method for coal-fired boiler performance, and solves the technical problem that the prediction of boiler performance under atypical operating conditions is not accurate enough and is prone to distortion. The key points of the technical solution are to collect boiler operation data and boiler performance indicators to obtain original data sets, the boiler performance indicators include boiler efficiency and NOx emission concentration, the boiler efficiency is indirectly obtained through a preset efficiency calculation module, and the NOx emission concentration is obtained by a continuous emission detector; abnormal data and non-steady-state data are eliminated through data preprocessing to obtain a static data set; secondly, the boiler operating condition division standard is obtained through a K-means clustering method, and the error square sum is used as an evaluation indicator to evaluate the effectiveness of the clustering method; finally, a data pool is established for each boiler operating condition, and a BP neural network model is established separately for each operating condition to achieve the purpose of accurately predicting boiler performance.

This invention relates to the field of unmanned aerial vehicle (UAV) swarm control technology, and more particularly to a method, system, medium, and software product for simulating and modeling phase trajectory of a UAV network system. The method includes system and task analysis, construction of interaction rule matrices, state analysis and phase vector space construction, approximate calibration of the stability domain, determination of the ultimate impulse, simulation of disturbance impulse injection, and data segmentation and transition zone marking, realizing a closed-loop process from modeling to disturbance assessment. Trajectory prediction is performed using a NARX neural network, combined with Gaussian white noise augmentation and multi-round simulation to achieve approximate calibration of the stability domain and determination of the ultimate impulse, and automatically outputs a structured trajectory dataset. This invention improves modeling accuracy, simulation efficiency, and system security.

This invention discloses a method for identifying damage to aircraft smart skin based on complex modal expansion and Kalman filtering. The steps include: acquiring strain values ​​at the location of fiber optic strain sensors; constructing a finite element model of the aircraft smart skin and extracting modal features; constructing a complex modal expansion equation; solving for the full-field strain value of the aircraft smart skin based on the complex modal expansion equation; and obtaining the damage identification result of the aircraft smart skin. This invention uses a finite element model obtained by discretizing a proportional 3D model of the aircraft smart skin to solve for the full-field strain value. It employs a finite fiber optic strain sensor to expand the full-field strain in real time and performs damage identification based on the full-field strain. The finite element model has a simple structure and does not require iterative optimization, effectively improving identification efficiency and robustness while ensuring identification accuracy.

The invention discloses a robot friction compensation and external force estimation method based on an ANFIS, and the method comprises the steps: enabling a joint to walk through a static friction-Stribeck-viscous friction full speed range in a controlled environment, collecting the angular speed and motor torque, carrying out the preprocessing, training an ANFIS model, and obtaining a friction torque estimation value which can be output in real time; introducing the estimated value as feed-forward compensation into a generalized momentum observer, and calculating an external torque initial value; denoising is carried out through median filtering and Butterworth low-pass filtering, and a clean external torque signal is obtained; and finally, establishing an AR (1) noise model under the working condition of no external force, setting a dynamic collision threshold value by using the variance and the safety margin, and judging the collision if the signal exceeds the threshold value. Only a joint encoder and motor current are used in the whole process, an additional force sensor is not needed, and accurate and real-time external force sensing and collision protection can be provided for dragging teaching and man-machine cooperation.

The present application relates to the cross field of coal mine geological modeling and artificial intelligence vision application, and provides a coal mine gas occurrence three-dimensional geological modeling method and system based on AI vision, which comprises: acquiring multi-source data of a mining area in a coal mine, including geological basic data, engineering dynamic data, gas monitoring data and underground visual image data; performing low-light enhancement processing on the visual image data to obtain an enhanced image; based on the enhanced image, using a deep learning model to identify coal rock features, outputting semantic segmentation results containing coal rock interfaces, geological structures and fissures, and corresponding geological geometric parameters; taking the semantic segmentation results as spatial geometric constraints, combining the geological data, engineering data and gas monitoring data, and performing data fusion through a spatial interpolation algorithm to construct a three-dimensional geological model reflecting the gas occurrence state. The present application can realize high-precision modeling, dynamic updating and safety management and control of the gas occurrence state in the mining process in a coal mine.