423 results about "Model fitting" patented technology

The invention belongs to the technical field of three-coordinate measurement, and particularly discloses a three-coordinate measurement method based on incomplete curved surface fitting, which comprises the following steps: hierarchically extracting geometric features from incomplete point clouds, and dividing a workpiece to be measured into a plurality of ordered curved surfaces; adding a curved surface label to each point cloud data, and generating a plurality of ordered point cloud sets based on the curved surface labels; generating virtual points based on the geometric features to fill the missing region; reconstructing a three-dimensional sub-model of each ordered curved surface based on the ordered point cloud set, the virtual points and the edge point cloud; and fitting the three-dimensional sub-model to generate a continuous curved surface model, and obtaining workpiece measurement parameters based on the continuous curved surface model. By extracting geometric features, classifying point clouds and generating virtual points, the limitation that a traditional method depends on a prior model is broken through, unknown or non-standard geometry can be processed, global continuous modeling of incomplete point clouds is achieved through ordinal curved surface division, three-dimensional sub-model reconstruction and continuous curved surface fitting, and the accuracy of measurement parameters is ensured.

The invention provides an adaptive welding seam detection and three-dimensional reconstruction method based on deep learning and binocular vision. The adaptive welding seam detection and three-dimensional reconstruction method comprises the steps of S1, collecting samples and making a training data set; s2, the picture of the sample to be welded is processed, a feature region is recognized, the image quality of the region to be welded is analyzed and evaluated through wavelet transform and local variance, and the noise level and the contrast ratio are calculated; s3, dynamically generating edge detection parameters and model fitting parameters according to the image quality; s4, using an edge detection algorithm to extract edge point cloud of the welding seam area; s5, performing RANSAC linear fitting, weighted least square fitting and polynomial curve fitting on the edge point cloud in parallel; s6, selecting an optimal fitting result based on an image quality adaptive dynamic scoring model; and S7, carrying out three-dimensional coordinate conversion in combination with the three-dimensional matching model IGEV-Stereo, and outputting a final welding seam three-dimensional coordinate. According to the invention, automatic detection of the position and size of the welding seam can be efficiently and accurately realized.

The invention discloses a shield segment dislocation detection method and system fusing visual and geometric features, and the method comprises the steps: employing a mobile track three-dimensional laser scanning system, and rapidly obtaining the three-dimensional point cloud data of a tunnel segment; projecting the three-dimensional point cloud data of the tunnel segment to a two-dimensional image according to the scanning parameters and the image preset resolution, and mapping the laser point reflection intensity into a pixel gray value; shield segment inter-ring joints and shield segment in-ring joints are extracted respectively, and segment joint positioning is completed; and according to the seam positioning information, extracting local point clouds at the two sides of the seam, respectively carrying out circular model fitting, calculating the height difference of circular models at the two sides at the seam, and obtaining an in-ring slab staggering value. According to the method, the visual information of the point cloud reflection intensity and the geometric information of the spatial position are fused, rapid and accurate shield segment in-ring slab staggering detection is realized, and the efficiency and the accuracy are high.

The invention discloses a power grid facility three-dimensional model fitting method and system based on panoramic point cloud, and relates to the technical field of three-dimensional model fitting. The method comprises the following steps: collecting panoramic point cloud data and high-definition image data of a power grid facility, and carrying out data preprocessing; importing a BIM model of a power grid facility, and establishing a homologous coordinate system with the panoramic point cloud data; performing data alignment on the panoramic point cloud data and the BIM model, and extracting key features of the power grid facility from the aligned point cloud data; adjusting vertexes and edges of the BIM model based on the extracted key features to enable the BIM model to be matched with the point cloud data, and ensuring the precision and reliability of the BIM model through consistency verification; and generating a three-dimensional model consistent with an actual power grid facility based on the corrected BIM model for real-time monitoring and defect detection. The operation and maintenance risk caused by data deviation can be reduced, human intervention is reduced, and the automation level of the system is improved.

The invention provides a whole blood donor low hemoglobin delayed blood donation risk prediction method, which comprises a data normalization and abnormal value processing module, a classification feature coding module, a category balance and model fitting module, a hyper-parameter adjustment module and a model interpretation and deployment module. According to the prediction method provided by the invention, data normalization and targeted feature coding are carried out on blood donation records and hemoglobin detection data, an SMOTE oversampling algorithm is used for class balance and then fitting of a Light GBM model is carried out, and a Hyperband hyper-parameter adjustment method is used for adjusting and optimizing SMOTE and Light GBM parameters. And storing the tuning model for feature importance interpretation, model packaging and network deployment, thereby realizing the effects of optimizing the whole blood donor invitation strategy and enhancing the continuity and stability of blood supply.

The invention provides an estuary coast water quality forecasting method based on a hydrodynamic water quality coupling model. The method comprises the following steps: S1, collecting basic data required for constructing an estuary coast hydrodynamic water quality model; s2, establishing and verifying an estuary coast hydrodynamic water quality coupling model based on an FVCOM-ERSEM model; s3, utilizing the verified model to simulate wave and water flow coupling motion and pollutant space-time distribution in the estuary coast area within a set time; s4, key factors influencing estuary coast hydrodynamic force and pollutant transportation are concluded through monitoring, literature and investigation; s5, sensitivity analysis is carried out, and the spatial-temporal change rule of pollutants in the estuary coast area under different factors is studied; and S6, analyzing the influence of the change of various factors on the pollutant concentration by taking the water quality parameters as indexes. According to the method, influences of engineering construction, pollutant discharge and other factors on water quality parameters are identified through quantitative analysis, and the model fitting degree is improved by adopting a high-precision grid technology.

A system and method include tuning hyperparameters for an ARIMA model using a derivative free approach by determining a set of initial hyperparameter values, fitting an ARIMA model to the set of initial hyperparameter values, selecting a tuning method for the set of hyperparameters, responsive to selecting a single-objective method, computing a first objective function value from time-series data applied to the ARIMA model based on the set of initial hyperparameter values, or responsive to selecting a multi-objective method, computing at least a second objective function value and a third objective function value from the time-series data applied to the ARIMA model based on the set of initial hyperparameter values, determining whether a stopping criterion for tuning the set of hyperparameters has reached, responsive to determining that the stopping criteria has reached, outputting a set of tuned hyperparameter values.

The present disclosure provides a model training method and device, and a data processing method and device. The method of the present disclosure uses a reference model to screen an existing instruction dataset for an instruction whose model fitting difficulty meets a preset condition, and the method can screen the instruction dataset for a challenging instruction having high model fitting difficulty as a seed instruction. A plurality of similar instruction samples are generated by means of expansion on the basis of the seed instruction, thereby obtaining more challenging instruction samples, a training set comprising the instruction samples and reference responses for the instruction samples is constructed, and knowledge distillation can be achieved by using the reference model on the basis of the existing instruction dataset, thereby obtaining a training set containing higher-quality instruction data. Furthermore, using the training set to train a deep learning model enhances the capability of the deep learning model to handle more complex and challenging tasks, and improves the performance of a trained target model.

The invention provides a high-precision optical axis direction resolving method for a two-degree-of-freedom large-stroke reflecting mirror. The method comprises the following steps: firstly, determining an incident angle of the reflecting mirror at an initial zero position according to the mounting position of the two-degree-of-freedom large-stroke reflecting mirror; secondly, an analytical expression of the relation between the two rotation angles of the reflector and the change angle of the direction of the optical axis of the reflected light is established; dispersing the change angle of the visual axis direction of the reflected light within a limited range to obtain a plurality of groups of angle combinations of the reflected light; for each combination, calculating according to an analytical expression to obtain a reflector rotation angle, and constructing a relational database between the reflector rotation angle and a reflected light visual axis pointing change angle; and finally, carrying out model fitting on the database, and carrying out accurate control on the visual axis pointing of the two-degree-of-freedom large-stroke fast reflector by utilizing the model. According to the invention, polynomial fitting decoupling of the resolving relation between the rotation angle of the two-degree-of-freedom large-stroke fast reflector and the change angle of the direction of the optical axis of the reflected light is realized.

The invention provides a Chinese ancient book entity extraction method based on a deep active learning strategy, and relates to the field of natural language processing. According to the method, the active learning strategy and the entity extraction model based on the deep neural network are subjected to model integration, so that the model fitting capability of a deep network learner can be brought into full play, and the problems of insufficient annotation data and relatively high expert annotation cost are further relieved by fully utilizing the improved active learning strategy. Meanwhile, data pool selection is carried out by adopting a query strategy fusing uncertainty, minimum boundary property and multi-spatial feature characterization, so that the learning efficiency is effectively improved. Furthermore, for an entity extraction network model, a double-encoder structure is designed, a global encoder based on pre-training learning and a local encoder based on character-level embedding are integrated, and fused semantic representation of sentences is obtained, so that more comprehensive context semantic knowledge of the sentences is comprehensively captured; therefore, the performance of the entity extraction model based on the deep neural network is improved.

The invention discloses a degraded cultivated land soil organic matter three-dimensional mapping method fusing remote sensing, near-earth sensing and historical data, and relates to the technical field of digital soil mapping, the method is based on an INLA-SPDE three-dimensional model framework, remote sensing, near-earth sensing and historical data are fused, and high-precision three-dimensional mapping is achieved. The method comprises the following steps: collecting organic matter data and near-sensing visible near-infrared reflection spectrum data of different depths of a soil profile at each sampling point in a research area, then screening visible near-infrared reflection spectrum characteristic wavebands significantly related to organic matters, collecting and integrating historical data, and completing standardization processing on remote sensing and near-earth sensing data; dividing a modeling set and a verification set, calculating a range value based on a semi-variable function, determining triangular mesh parameters, and constructing an SPDE model; the near sensing data and the remote sensing data serve as covariables and are fused into the INLA-SPDE three-dimensional model, and a soil organic matter three-dimensional space distribution diagram is generated after model fitting and uncertainty is quantified. According to the method, the problem that the vertical change of the soil attribute cannot be accurately simulated by a traditional method is solved, and high-precision and high-efficiency three-dimensional mapping of the regional scale soil organic matter in the horizontal and vertical dimensions is realized.

The invention relates to a machine learning attribution analysis method introducing causal information, and belongs to the field of statistical learning and machine learning, and the machine learning attribution analysis method introducing causal information comprises the following steps: (1) data preprocessing; the method comprises the steps of preprocessing original data, including missing value processing, abnormal value processing and the like, and then performing standardization processing to ensure that the scales of all features are consistent so as to unify different magnitudes and units; (2) training model fitting data; a model suitable for problem types is adopted to carry out data fitting on the processed features, feature sorting is carried out in each iteration of model training to screen irrelevant features, attribution analysis is carried out to calculate the stability of each attribution analysis, and the stability is taken as an index to guide when the model stops training; (3) performing causal inference on the features of the model obtained in the step (2) on the data set preprocessed in the step (1); (4) introducing the causal information obtained in the step (3) into model training, designing a new loss function, and training again on the basis of the parameters of the previous model; and (5) the causal information obtained in the step (3) is used for guiding attribution analysis of the model obtained in the step (4), and reliable model interpretation is further obtained.

The invention provides a soil quick-acting potassium measuring method and system and a medium, and relates to the technical field of soil data measurement, and the method comprises the following steps: extracting near infrared spectrum data of soil to be measured; solving Euclidean distances between one row and other rows in the spectrum matrix, and taking the number of distances smaller than a preset threshold value as the number of neighbors of the row; traversing the spectrum matrix, constructing a neighbor number array, normalizing a density vector, and transposing and assigning the density vector as an initial weight of a competitive adaptive reweighted sampling (CARS) method; determining an optimal regularization parameter through a Bayesian optimization method, fitting a Lasso model, obtaining a coefficient matrix of the Lasso model, and determining a regression coefficient of each spectral feature; and screening spectral features with non-zero regression coefficients in the near infrared spectral data of the soil to be detected to form an effective feature matrix. According to the method, the calculation complexity of the CARS method is reduced, parameter setting is simplified, overfitting is prevented, the model fitting precision is improved, and the prediction performance of soil available potassium measurement is improved.

The invention relates to the technical field of data processing, and particularly discloses a data analysis method and device fusing a large number rule, equipment and a medium, and the method comprises the steps: determining a minimum convergence amount through hierarchical aggregation; calculating a cumulative sample mean trajectory and carrying out convergence diagnosis; performing noise attenuation weighting based on LLN convergence characteristics; performing distribution and component decomposition under steady moment constraint; performing consistency check and re-extraction robustness of LLN guidance; lLN-constrained model fitting and uncertainty calibration output are carried out; according to the method, based on noise attenuation weighting of LLN convergence characteristics, samples which are close to a steady state obtain greater influence in estimation, and unstable or sparse samples are naturally weakened, so that self-adaptive suppression of heterogeneity and small sample noise is realized, and the robustness of overall estimation is improved; steady moment constraint is introduced into distribution / component decomposition, it can be guaranteed that components obtained through decomposition are consistent with observation convergence characteristics in the aspect of high-order statistics, and component mismatching caused by extreme values or local fluctuation is reduced.

The invention relates to the technical field of urban environmental pollution, and particularly discloses a method for establishing a refined attribution model of urban atmospheric environmental pollution. The method comprises the following steps: S1, collecting and preprocessing multi-source data, analyzing a spatial and temporal change rule of air quality, and defining change characteristics of human activity intensity; s2, researching an influence factor parameterization method of the urban air quality, performing optimal variable screening, and selecting an optimal variable according to a contribution rate and incorporating the optimal variable into a model; s3, performing repeated sampling simulation on the original observation data to solve an average regression coefficient and a root-mean-square error of a simulation result so as to evaluate the uncertainty of model fitting, and comparing the difference between a predicted value and the concentration of the urban air quality observation station to perform precision verification; and S4, quantitatively analyzing the cause of urban air quality pollution according to the obtained quantitative attribution model. According to the technical scheme, the difference of urban environmental pollution levels can be quickly clarified, and a supporting effect on early warning and intervention of heavy pollution events is achieved.

The invention relates to a quadruped robot laser odometer method based on ground point optimization, hardware and application, and the method comprises the steps: obtaining an original point cloud collected by a laser radar, carrying out the preprocessing, dividing a polar coordinate region, jointly extracting ground plane feature points and ground geometric edge feature points based on normal vector and local curvature features, and carrying out the calculation of the ground plane feature points and the ground geometric edge feature points; ground geometric edge feature points are utilized to optimize ground model fitting; optimizing a point set of the ground based on a time sequence reference plane of the ground; the optimized ground points are removed, and non-ground point features are extracted from the remaining point clouds; establishing feature association between the feature points in the current frame and the target point cloud, and screening effective matching pairs; a cascade weight optimization objective function associated with the ground point plane constraint, the non-ground point line feature constraint and the surface feature constraint is constructed, and continuous pose transformation of the quadruped robot is solved; hardware is realized based on the method, and the method is applied to real-time pose estimation and environment mapping of the quadruped robot in outdoor unstructured terrains and complex dynamic environments.

The invention discloses a character string data read optimization method based on a B + tree and a learning index. A learning index generated based on a machine learning (ML) model focuses on integer data of a fixed size, but is lack of consideration for character strings of variable lengths and is poor in performance in character string data. In order to solve the problem, the invention provides a character string slicing method which can convert a character string with a variable length into an integer with a fixed size for prediction and extract a common prefix from a character string key to reduce a model fitting error. According to the method, the characteristics of character string data are fully considered, complex multi-dimensional linear regression is simplified into one-dimensional linear regression, and the model calculation overhead is reduced. Besides, the invention provides a self-adaptive strategy to select a proper index to manage the calculation cost, the method for dynamically selecting the index nodes can utilize the advantages of the two index types to the maximum extent, and the calculation overhead is effectively reduced while the index efficiency is ensured.

The invention relates to the technical field of computer data processing, and discloses a financial reimbursement data processing method, device and system.The financial reimbursement data processing method comprises the following steps that multi-modal data of a reimbursement voucher is obtained, and the multi-modal data is stored in a block chain; the multi-modal data comprises pictures, voices and videos of reimbursement vouchers; recognizing characters on the picture based on a convolutional neural network, and extracting and analyzing character information of key frames of the voice and the video by using a natural language processing model; and performing feature extraction on the character information to obtain character feature information, judging whether the character feature information falls into a reimbursement directory, and performing local model fitting according to the character feature information. According to the method, the dynamic verification rule is generated based on the historical reimbursement data, the bill anomaly probability is calculated in real time through the anomaly coefficient formula, and the anomaly detection accuracy is improved by combining the error coefficient and keyword matching degree and the historical record relevance index.

The invention discloses a numerical control machine tool autonomous tool setting method based on 3D visual guidance. The method comprises the following steps: firstly, acquiring three-dimensional point cloud data of a cutter and a workpiece by using a 3D visual sensor and preprocessing; then, segmenting the point clouds through a region growing algorithm, realizing the separation of the tool point clouds and the workpiece point clouds, and extracting the point clouds of each cylindrical section of the workpiece; thirdly, based on a cylinder fitting algorithm, cylinder model fitting is conducted on the segmented tool and workpiece point cloud, and geometric parameters such as the axis direction, the radius and the center coordinate are obtained; calculating a relative position relation between the tool and the workpiece according to the geometric parameters, generating a tool setting path and converting the tool setting path into an NC code; and finally, the NC code is transmitted to a grinding machine control system through a grinding machine communication module, and tool setting operation is completed. The automatic tool setting device has the advantages that the relative position relation between the workpiece and the tool can be automatically measured in the machining process, automatic and rapid tool setting is completed, and the machining quality and efficiency are guaranteed.

The invention relates to the technical field of data processing, and discloses a hydrocephalus correlation analysis method and system based on multi-dimensional data. The method comprises the steps that a feature matrix is generated by obtaining and standardizing hydrocephalus multi-dimensional data, after the matrix is discretized, an association rule mining algorithm is adopted to extract an association mode, a rule set is obtained in combination with knowledge graph verification, data clusters are divided based on mahalanobis distance clustering, mixed effect time sequence model fitting parameters are established for all the clusters, and a mixed effect time sequence model is obtained. And calculating a target sample attribution cluster and predicting a symptom improvement trajectory and a confidence interval. According to the method, automatic integration, credible association rule extraction, precise subtype division and individualized symptom trajectory prediction of the multi-dimensional data of the hydrocephalus patient are realized.

The invention discloses a complex system fault twinborn deduction method based on virtual-real dynamic fusion, and the method comprises the steps: firstly designing a preprocessing scheme of multi-modal data fusion, and carrying out the spatial-temporal feature unification and standardized state vector construction of heterogeneous sensing data; secondly, establishing a real-to-virtual digital mapping model, deducing a multilevel system state based on a dynamic Bayesian network, and introducing a weighted adaptive gradient correction mechanism to improve the fitting precision of the model; and finally, proposing a fault evolution evaluation method of reflecting real from virtual, and realizing hierarchical identification and early warning triggering of the system operation state. By introducing a digital twinning thought, establishing a virtual model synchronously evolved with an entity system, and constructing a whole-process twinning deduction framework of perception, modeling and evaluation, the method can significantly improve the accuracy, response efficiency and robustness of fault prediction of a complex system.

The invention discloses a power distribution network area low voltage prediction method, and relates to the technical field of voltage prediction, and the method comprises the steps: S1, data preprocessing; s2, establishing a GA-BP neural network load prediction model; s3, GA-BP load prediction model simulation is carried out; s4, establishing an LSTM neural network voltage prediction model; s5, performing model simulation and result analysis; and S6, voltage prediction and treatment suggestion. According to the power distribution area low voltage prediction method, firstly, a GA-BP neural network is utilized to predict area loads, influence factors such as temperature, humidity, date types and different moments in one day are fully considered, and predicted load data are utilized as input to construct an LSTM model to predict area voltages; the transformer area low voltage is effectively predicted through the GA-BP-LSTM combined model, the model is high in fitting degree and low in prediction error, the low voltage degree of a user is reflected in real time, and more information support is provided for follow-up transformer area low voltage treatment.

The invention discloses a hybrid swarm intelligence algorithm-based photovoltaic parameter modeling method, which relates to the technical field of photovoltaic parameter modeling, and is characterized in that a Latin hypercube design is utilized to determine a sampling range of feature weights, n feature weight vectors are generated, a parameter space is effectively covered, the quality and diversity of population initialization are improved, and the modeling efficiency is improved. And the global exploration capability of the algorithm is enhanced. Through dynamic judgment of a real-time photovoltaic power prediction value and fitness, whether an error range of a current model is acceptable or not is quickly evaluated, so that unnecessary iterative calculation time is reduced. And an adaptive optimization mechanism: in a local optimization stage, dynamically updating particle positions by using a firefly algorithm, and cooperatively improving optimization precision and search efficiency in combination with a balance index fitted by a balance prediction model, so that the stability and convergence speed of the algorithm in the optimization process are ensured. And through comparative analysis of the balance index and the evaluation threshold, a relatively optimal particle position is selected.

The invention belongs to the field of shale oil exploration evaluation, and discloses an LgR method logging source rock evaluation model fitting parameter prediction method and a shale type shale oil resource quantity calculation method and system. The prediction method comprises the following steps: screening mudstone data in a logging curve; determining a plurality of combinations of possible values of fitting parameters K, a and b, and calculating corresponding mudstone baseline parameters; aiming at different combinations of possible values of fitting parameters K, a and b and corresponding mudstone baseline parameters, according to the screened mudstone data, utilizing an LgR method well logging source rock evaluation model to respectively calculate and obtain organic matter abundance fitting values; and based on the organic matter abundance fitting value and the organic matter abundance measured value, evaluating a fitting effect, and screening out a parameter value combination with the optimal fitting effect as a predicted fitting parameter of the well with the organic matter abundance measured value in the work area. According to the method, all possible parameter value combinations can be comprehensively screened, the influence of subjective factors is eliminated, and undetermined parameters are objectively and accurately predicted.

The invention discloses a weld size measurement and defect detection method based on base point cloud features. The method comprises the following steps of weld point cloud data collection, point cloud processing, equidistant slice projection, weld section model fitting, feature point extraction, size measurement and defect calculation. The method provided by the invention can measure the size and defect of the welding seam in real time, and is high in accuracy and strong in applicability. According to the method provided by the invention, welding seam size measurement and defect detection are simultaneously realized in a manner of extracting the welding seam point cloud features, the welding seam forming condition and the quality state are directly reflected, and the problems of low efficiency, large data fluctuation, non-objective judgment and the like of traditional manual welding seam forming quality detection are solved.

The invention discloses a tunnel central axis rapid extraction method based on point cloud data, and belongs to the technical field of tunnel engineering. According to the method, technologies such as global principal component analysis, local segmentation processing, RANSAC robust fitting and spline interpolation are innovatively combined, comprehensive analysis of the overall structure and local details of the tunnel is achieved, an efficient local splicing strategy is innovatively provided, the data processing amount is effectively reduced, the algorithm efficiency is improved, and the method is suitable for large-scale popularization and application. Segmentation processing and elliptic cylindrical surface model fitting are innovatively adopted, and adaptability to complex tunnel shapes is improved. According to the method, many limitations in the prior art are overcome, local features and global features are concerned at the same time, and a more efficient, more accurate and more reliable solution is provided for extraction of the central axis in tunnel engineering.

The invention discloses an unmanned aerial vehicle autonomous information perception path optimization method fusing sparse Gaussian estimation and RLSAC, and particularly relates to the technical field of unmanned system multi-target tracking, and the method comprises the steps: employing Kalman filtering to carry out the optimization processing of target initial data obtained by a ground laser radar, achieving the filtering of the noise of a target, obtaining the dynamic trajectory of the target, and obtaining the optimal path of the target. Therefore, the output insufficiency of the RLSAC to the dynamic trajectory is made up. The data after Kalman filtering processing is used as original input data of a multi-unmanned aerial vehicle target tracking I PP algorithm based on information path planning, the own unmanned aerial vehicle adopts the I PP algorithm to track a target in real time and transmit target related data to provide input data for RLSAC, the RLSAC is used to process the input data, model fitting is carried out, and the RLSAC is used to carry out target tracking. And the RLSAC outputs rewards of different geometric morphology models adopted by the own unmanned aerial vehicle during mode recognition, and the model which is more consistent with the output of the RLSAC is determined by calculating the confidence coefficient.

The invention discloses a photovoltaic power prediction method for Gaussian mixture error probability distribution compensation based on initial value optimization, and the method comprises the steps: carrying out the fitting of a distribution rule of a photovoltaic power preliminary prediction error through employing a Gaussian mixture model (GMM) based on initial value optimization according to a photovoltaic prediction error, and carrying out the calculation of the distribution rule of the photovoltaic power preliminary prediction error; and compensating a predicted value of the error to a point prediction result to realize interval prediction of photovoltaic output. As environmental factors such as illumination intensity, temperature, visibility and the like have great influence on a prediction result of photovoltaic power generation output, historical photovoltaic output data are divided into a plurality of different weather scenes by adopting a K-means clustering algorithm (K-means) based on an irradiance index, an optimal initial value data set selected by inputting a whale optimization algorithm (WOA), and the prediction result of the photovoltaic power generation output is predicted by using a prediction result of the photovoltaic power generation output according to the optimal initial value data set selected by the whale optimization algorithm (WOA). And then a Gaussian mixture distribution model is adopted and an expectation maximization (EM) method is combined to solve a model optimal parameter so as to carry out modeling analysis on probability distribution of photovoltaic prediction errors, and a model fitting result is used to correct a photovoltaic power generation power point prediction value under a specified confidence level interval so as to obtain a final prediction interval. According to the scheme provided by the invention, simulation analysis is carried out by taking photovoltaic power data of a certain photovoltaic power station in Jiangsu province as an example, and a result shows that the method provided by the invention has better fitness and reliability.

The invention provides a composite material bearing structure storage life evaluation method considering an acceleration stress lower limit, and the method comprises the steps: 1, carrying out the initialization modeling: enabling the service life of a product to obey the logarithmic normal distribution, and building a three-parameter inverse power law acceleration model of the service life feature theta and the stress level S; 2, estimating parameters: estimating unknown parameters by applying a maximum likelihood estimation method for the three-parameter inverse power law acceleration model; 3, model selection: comparing the calculated likelihood function values, wherein the larger the likelihood function value is, the more stable and more accurate the model is; meanwhile, the variation coefficients of the acceleration factors are calculated in combination with test data for comparison, and the model with the small variation coefficient is more accurate; and 4, accelerated life evaluation: giving reliability, and evaluating the storage life of the three-parameter inverse power law model. The method is suitable for accelerated test evaluation of a product with a threshold effect; the three-parameter inverse power law model provided by the invention can show a better model fitting effect when the acceleration effect does not meet the linear characteristic.

Methods, systems, and computer storage media for providing iterative data processing optimization using an iterative data processing optimization engine in a data intelligence system are described. Iterative data processing refers to handling data where the processing steps are repeated multiple times, across multiple views or modalities, to train machine learning models, filter and score data or generate output. The iterative data processing optimization engine employs expectation step machine learning models that are simple but with fast language models to efficiently and effectively probe and analyze data, while iteratively refining maximization step machine learning models that are optimized and fast to approximate the probing mechanism of the expectation step machine learning models more efficiently, for example, using metadata, external information, and compressed representation. The iterative data processing optimization engine can operate based on an agentic framework using lightweight artificial intelligence (AI) agents to perform model fitting, featurization, and report generation autonomously.