6results about How to "Reduce forecast bias" patented technology

The application provides obtaining a plurality of training data sets, the training data sets comprising one-to-one corresponding microfluidic chip parameter combinations and chip chamber fluid average temperature sequences, the parameter combinations comprising chip thickness, solution volume, ambient temperature, denaturation temperature, annealing temperature and extension temperature, and the chip chamber fluid average temperature sequence being a sequence of spatial average temperature of a chip chamber fluid domain changing with time; obtaining a temperature prediction optimization model, an input of the temperature prediction optimization model being the training data set, and an output being a heater set temperature sequence; and inputting the training data set into the temperature prediction optimization model to obtain the heater set temperature sequence. The method realizes reverse prediction of the heater set temperature curve, and improves prediction accuracy of the model under complex thermal cycle conditions.

The application relates to the technical field of machine learning, and particularly discloses a spatial local XGBoost machine learning model based on multi-scale geographical weighting, which comprises the following modules: a feature decoupling module for constructing a double-channel input structure of geographical features and non-geographical features, realizing feature decoupling and fusion through a multi-scale spatial weight matrix; a bandwidth allocation module for dynamically determining a bandwidth scale in an XGBoost tree splitting process and establishing dynamic weights of tree levels; a constraint gain module for generating a splitting point marked with a scale; a contribution decoupling module for extracting splitting features and correlating geographical features and non-geographical features, realizing salted prediction and contribution index extraction; and a verification optimization module for optimizing model parameters through multi-scale heat maps and spatial autocorrelation analysis. The model can capture geographical spatial effects of different scales, improve spatial local prediction accuracy, and be applied to a scene with spatial heterogeneity in soil salinization analysis.

The application discloses a fusion-based adaptive pupil interaction method and belongs to the technical field of pupil positioning, and the steps of the method comprise the following steps: HOG feature extraction cooperates with an SVM model to predict key points, lock the iris position and calculate the centroid; based on the screen ratio of a display, a spatial calibration model with the same ratio as the display screen is constructed, interpolation algorithms are used to optimize pupil data, the corrected coordinate data is projected to the display, the actual position of the line-of-sight point in the screen is determined, the parameter vector of the model is iteratively updated through an L-BFGS-B optimization algorithm, the coordinate data is iteratively optimized through an HBG algorithm after mapping, and the fixation center point is determined. The application has the beneficial effects that: in view of the uncertainty of the pupil position, a calibration model is constructed by using device parameters, the data is finely calibrated by using interpolation and optimization algorithms, the error is greatly reduced, the prediction deviation is further reduced through the iterative optimization of the HBG algorithm, and the system has the advantages of simple structure, high lightweight degree, high positioning accuracy and fast response speed.

The application discloses a hydrological model differentiation calibration method based on runoff coefficient grading and coupling rainfall-melting-snow-runoff process, and comprises the following steps: collecting continuous hydrological data of a target basin and preprocessing; dividing out flood events of each flood according to the shape of the flow process line to form a flood sample set; calculating the runoff coefficient of each flood; determining a grading threshold according to the statistical distribution of all sample runoff coefficients, dividing the flood sample set into low and high runoff coefficient subsets; introducing a snow melting module into a traditional hydrological model; independently calibrating the model coupled with the snow melting module by using the two subsets respectively; obtaining the antecedent soil moisture of each flood, and determining the antecedent soil moisture threshold corresponding to the runoff coefficient grading threshold; in the flood forecasting stage, the antecedent soil moisture of the basin is obtained in real time, compared with the threshold, and the corresponding parameter set is dynamically selected and called for forecasting; the application improves the precision and stability of flood simulation and forecasting.

This invention discloses a multi-objective optimization method based on a GCN proxy model, which mainly addresses the problems of high computational complexity, low optimization efficiency, and low prediction accuracy in cascading failure simulation during critical node detection in complex networks. First, initialization and objective function construction are performed, setting relevant algorithm and model parameters, and constructing a dual objective function for attack cost and attack failure effect. A complex network dataset is generated, downloaded, and preprocessed. Second, addressing the low computational and optimization efficiency caused by multiple traversals of the entire network in each evaluation of cascading simulations during critical node detection in complex networks, this invention constructs and trains a multi-branch attention GCN proxy model. Through multi-dimensional feature extraction, multi-task learning, and group calibration, the prediction accuracy of the number of cascading failure nodes is ensured. Then, a multi-objective optimization algorithm fusing GA and PSO is used iteratively, combined with the GCN proxy model to predict the objective function, improving optimization efficiency. The convergence and diversity of solutions are balanced through GA global search and PSO local optimization. Next, the Pareto optimal solution is calibrated and verified to ensure the relative error is within a reasonable range and to verify accuracy. Finally, the experimental results are output and archived. This invention utilizes the GCN proxy model to assist in the fusion of multi-objective optimization algorithms, significantly improving optimization efficiency and prediction accuracy, and enabling precise detection of key nodes in complex networks.

The present application relates to the technical field of lithium ion battery, and provides a GRU-EC-GAPSO lithium ion battery SOH prediction method, which comprises the following steps: extracting a health factor from voltage and time data in a local SOC interval during charging and discharging of a lithium ion battery, inputting the GRU model to perform SOH prediction, calculating an error sequence of a predicted value and an actual value, training an EC model by using the error sequence, and obtaining a GRU-EC model; using a GAPSO algorithm to optimize parameters of the GRU-EC model, and obtaining a GRU-EC-GAPSO model; and applying the trained GRU-EC-GAPSO model to different types of lithium ion battery data to perform SOH prediction. The present application has high prediction accuracy and good robustness in lithium ion battery SOH prediction.