77 results about "Rank correlation" patented technology

The invention belongs to the technical field of hydraulic engineering operation and maintenance, and relates to a prediction and early warning method for operation settlement of a steel dam gate on a complex foundation, in the operation process of the steel dam gate, a multi-parameter sensor network is used for collecting settlement monitoring data of different positions of the steel dam gate, and input characteristics and output characteristics are determined; sorting and screening different input features by using minimum redundancy and maximum correlation, and screening out the input features of which correlation coefficients are lower than a threshold value by using Spearman rank correlation coefficients; a normalization and Kalman filtering iteration prediction model is adopted to preprocess the screened input features, and multivariable time series data are obtained; inputting the multivariable time sequence data into the trained multivariable regression prediction model to obtain a real-time settlement prediction value of foundation settlement; and comparing the real-time predicted value with a set multi-level early warning threshold value, and carrying out corresponding-level early warning on the settlement exceeding the warning limit.

The invention discloses a photovoltaic power generation power prediction and power system scheduling method and a system for realizing the method. The method comprises four steps of data acquisition and preprocessing, similar day selection and training set construction, GA (Genetic Algorithm)-fuzzy RBF (Radial Basis Function) neural network modeling and power scheduling plan generation. The method comprises the following steps: selecting a meteorological variable with high correlation degree through a Spearman rank correlation coefficient, and constructing a time sequence feature; similar days are utilized to construct training samples, and the generalization ability of the model is improved; a fuzzy RBF network optimized by GA is adopted to enhance the nonlinear modeling precision; and inputting a prediction result into the power system simulation model, and generating an optimal scheduling instruction by adopting dynamic programming. According to the method, the photovoltaic power prediction accuracy and scheduling efficiency can be remarkably improved, the power grid stability is enhanced, and the new energy consumption capability is promoted.

The invention discloses an unmanned aerial vehicle subsystem health assessment method and system, and relates to the technical field of unmanned aerial vehicle health management, and the method comprises the steps: obtaining the data of a multi-source sensor of an unmanned aerial vehicle, screening the features strongly correlated with a fault state through a Spearman rank correlation coefficient, and obtaining key monitoring parameters; according to the self-adaptive mutation mechanism, the virtual evolution strategy and the double-objective optimization strategy, an improved multi-objective genetic algorithm optimization model is constructed, then key monitoring parameters are input, and a visual report of a health state assessment result and confidence rating is obtained. According to the method, the problems that traditional single-target optimization neglects the misjudgment rate, the parameter search efficiency is low and the evaluation credibility is insufficient are solved, accurate evaluation is achieved, the evaluation accuracy is improved to 84.86%, the key misjudgment rate is reduced to 0.009%, the convergence speed is improved by 40% compared with a traditional method, the safety and real-time performance of unmanned aerial vehicle health state recognition are remarkably optimized, and the method is suitable for popularization and application. The method is suitable for embedded system deployment.

The invention provides a deep compact feature representation network model based on a rank correlation induced dynamic mask mechanism, which is used for relieving confirmation deviation generated by distillation training due to excessive unlabeled data in semi-supervised learning and effectively compressing distribution of features in a potential space. According to the method, global and local connection between pixels is broken through a feature diffusion enhancement module, a cluster sensing neighborhood consistency module is introduced, and intra-class compactness and inter-class separability are enhanced by using adaptive metric learning to realize pixel-by-pixel feature representation; a rank correlation induction dynamic mask module is adopted, a category correlation mode is reserved in a rank matching mode, and high-probability errors are filtered out, so that the problem of rank relation loss caused by semantic consistency weakening probability alignment randomness is ensured. Evaluation on a plurality of data sets shows that knowledge is distilled to the student network through the teacher network, meanwhile, the correlation between the feature distribution geometric structure and the prediction space is optimized, and the method is obviously superior to an existing method, so that the generalization performance of the model is improved.

The invention relates to the technical field of power transmission and distribution of a power grid, in particular to a dynamic capacity increase risk prediction method for a high-voltage transmission line, a readable medium and equipment. The high-voltage power transmission line dynamic capacity increasing risk prediction method comprises the steps of collecting line temperature data of each point position of a corresponding power transmission line to form a data set; processing the data in the data set; calculating a rank correlation coefficient of the data; selecting the order of the fitting curve according to the absolute value of the rank correlation coefficient; obtaining a fitting approximate curve equation according to the order of the fitting curve; and performing prediction based on the fitting approximate curve equation, and outputting a prediction result. According to the scheme, the real-time temperature and the environment temperature of the temperature sensitive point position on the power transmission line and the current-carrying capacity data at the corresponding moment are measured, the corresponding relation algorithm of the difference value between the lead temperature and the environment temperature and the current-carrying capacity is constructed, and error feedback and adjustment are conducted at any time, so that the method has the advantages of being visual in calculation and rapid in adjustment; and the capacity increasing allowance and the safety in the subsequent short time can be predicted.

The invention relates to the field of emission control of coal-fired units, and provides an RF-LSTM-based desulfurization optimization control method and system for a coal-fired unit, and the method comprises the steps: carrying out the data screening of historical data of the coal-fired unit; selecting characteristic index parameters by utilizing a Spearman level correlation coefficient and a random forest algorithm based on the data after data screening; training an LSTM model by using the selected characteristic index parameters; inputting real-time operation data of the coal-fired unit into the trained LSTM model, and obtaining a predicted value of the SO2 emission concentration at the current moment; on the basis of the difference between the predicted value and the actual emission concentration, the slurry supply flow is controlled to increase or decrease. According to the method, on the premise that sulfur dioxide emission reaches the standard, the stability of emission concentration can be effectively improved, and desulfurization control optimization is achieved.

The invention discloses a microsatellite instability detection method and device based on a next-generation sequencing technology, and belongs to the technical field of biological detection. The method comprises the following steps: selecting MSI sites, and carrying out capture probe design and / or amplicon primer design based on any combination of the sites; selecting a plurality of samples, and calculating the missing fragment proportion and the normal fragment proportion of each sample at each MSI site; selecting a plurality of samples which are judged to be in an MSI-L or MSS state by a PCR method as baseline samples, and setting different depth levels for each baseline sample to construct a baseline; carrying out MSI state detection; selecting a sample of which the MSI state is MSI-H, and judging whether the sample is an abnormal sample or not through a Spearman rank correlation coefficient; and if yes, adjusting the base line and re-judging the MSI state. The method has high stability and high accuracy, does not need a control sample, and can be simultaneously applied to microsatellite instability detection of hybridization capture and amplicon sequencing data.

The invention relates to a method and a model for predicting the lifetime of a non-tumor peaceful patient based on clinical data. The method comprises the following steps of: collecting and sorting previous hospitalization clinical data of a non-tumor peaceful treatment death patient, and carrying out standardized pretreatment; carrying out attribute selection and reduction on the high-dimensional data from two perspectives of correlation and redundancy by using a maximum information coefficient (MIC); the method comprises the following steps: respectively carrying out attribute weighting by applying a Pearson correlation coefficient and a Spearman rank correlation coefficient; a weighted naive Bayes classification model is used for modeling; according to the method, clinical factors related to the lifetime of the non-tumor Anning treatment patient can be objectively analyzed and predicted, and guidance is provided for Anning clinical admission standard formulation and early and later treatment plans. The model can be realized in the form of a doctor workstation and the like, and the medical work efficiency and accuracy can be greatly improved.

The invention belongs to the field of methods for evaluating rare events in electronic engineering, and particularly relates to a failure area evaluation method applied to a high Sigma scene in the electronic engineering. A failure area evaluation method applied to a high Sigma scene in electronic engineering comprises the following steps: establishing an initial sample, establishing a prediction model, screening a high-risk sample, simulating the high-risk sample and calculating an offset delta d, and determining an approximate failure area sample, a secondary approximate failure area sample and a common area sample. Spearman rank coefficients are calculated for the samples of the three areas through a formula, different weights are given to the Spearman rank coefficients, and a final weighted rank correlation evaluation result is calculated. According to the evaluation method, through a sequencing function of a weighted rank correlation evaluation substitution model, the accuracy of sequencing representation of a high Sigma failure area is remarkably improved; and unnecessary simulation times are greatly reduced, and computing resources are saved.

The invention discloses a fractured-vuggy reservoir rock facies fine division method. The method comprises the steps that single well data of a to-be-measured area is acquired and preprocessed; carrying out depth homing on the required core physical property data; performing principal component analysis on the single well data to obtain dimension reduction and noise reduction fusion representation data; the data set is divided into a clustering data set and an inspection data set; carrying out optimal classification cluster selection on the clustering data set by adopting an elbow rule; classifying the data by adopting a fuzzy C-means improved clustering algorithm, and establishing an identification model of the fractured-vuggy reservoir rock facies; evaluating the identification model of the fractured-vuggy reservoir rock facies by adopting the contour coefficient; the identification model of the fractured-vuggy reservoir rock facies is inspected in combination with the inspection data set; outputting a test clustering result by using the test identification model of the fractured-vuggy reservoir rock facies, and performing sensitivity analysis by using a rank correlation matrix; and intersection is carried out by adopting the sensitive curves to obtain an identification chart of the fracture-vuggy reservoir rock facies.

The present invention relates to the field of data processing technology, and the present invention relates to a supply chain financial transaction security early warning method and system, comprising: obtaining financial data of the supply chain and its corresponding risk score, dividing the financial data into continuous data and discrete data, and calculating its importance coefficient. The importance coefficient is positively correlated with the Spearman rank correlation coefficient of the risk score, and is negatively correlated with the Pearson correlation coefficient in continuous data, the standardized mutual information in discrete data, and the point biserial correlation coefficient between continuous data and discrete data. An improved XGBoost model is used to predict the probability of risk type of financial data, and the risk type corresponding to the maximum value of the probability is the final risk type of the financial data. The present invention solves the problem of inaccurate risk assessment of supply chain financial transactions by improving the XGBoost model.

The invention provides an industrial load adjustable potential assessment method and device considering security constraints, and relates to the technical field of potential assessment. The method comprises the following steps: constructing a high-energy-consumption industrial user side load model and establishing a load curve; clustering the load curve by using a black hole entropy fuzzy clustering algorithm and determining a load interruption interval and a transferable interval; corresponding indexes are obtained according to the corresponding intervals, and multiple indexes such as safety consideration are introduced; an improved G1 method based on Spearman rank correlation is used for carrying out weighting on each index and determining the weight, and then the adjustable potential is evaluated. In this way, data support is provided for power supply enterprises to reasonably dispatch user side resources. The device comprises a first construction module, a second construction module, a weight value determination module and a calculation module, all the modules cooperate with one another to achieve the industrial load adjustable potential evaluation method considering the security constraints, and quantitative evaluation of the adjustable capability of industrial users is greatly achieved.

The present invention provides a signal detection method and system based on the anti-eigenvalue of the Spearman rank correlation matrix. The method of the present invention includes: obtaining observation signals received by multiple sensors; calculating the Spearman rank correlation coefficients of any two signals to form a Spearman rank correlation matrix; obtaining the eigenvalues of the Spearman rank correlation matrix and calculating the corresponding anti-eigenvalues; calculating a statistic based on the sum of the anti-eigenvalues; setting a detection threshold and comparing the magnitudes of the statistic and the detection threshold to complete the detection process. The present invention uses the Spearman rank correlation transformation of the signal to suppress the negative impact of impulse noise, designs a corresponding statistic by calculating the sum of the anti-eigenvalues of the Spearman rank correlation matrix, and thus realizes high-performance signal detection under impulse interference. The present invention has strong robustness against impulse noise and can perform signal detection in background noise containing impulse components.

The application discloses a cable branch box grounding fault detection method and system, the method comprises the following steps: calculating the effective value of each return line zero sequence current per power frequency cycle, determining the baseline mean and baseline standard deviation by using a sliding window, and identifying a sharp peak event and recording the occurrence time; obtaining the temperature change rate of the cable connection point of each return line, dividing the statistical period into multiple time windows, forming a sharp peak occurrence rate sequence and a temperature change rate absolute value mean sequence respectively, and calculating the Spearman rank correlation coefficient of the two sequences; according to the comparison result of the median of the rank correlation coefficient of the latest preset number of statistical periods and the determination threshold, it is judged whether there is an intermittent grounding fault in the corresponding return line. The application can utilize the correlation between the zero sequence current sharp peak distribution and the temperature change to identify the fault, and improve the stability and accuracy of the intermittent grounding fault detection.

The application discloses a power plant 5G private network elastic security data analysis method and device based on data mining, and relates to the technical field of power grid security data analysis. The method comprises the following steps: defining a power grid security data analysis target and an analysis range; building a data acquisition and transmission link, and collecting operation parameters, operation logs and fault record data of key equipment of the power grid in real time; transmitting the collected data to a collection server in a safe access area; pre-processing the received data by the collection server to obtain a time series data set of security data; performing preliminary correlation analysis on the time series data set by adopting a Spearman rank correlation method to obtain correlations between a plurality of security data state indexes; obtaining a time series data set corresponding to a strong correlation security data index based on the correlations between the indexes; and mining the association rules of the security data based on the time series data set by adopting a time series FP-Growth algorithm to obtain strong time series association rules. The application can improve the utilization efficiency of security data.

The invention relates to the technical field of environmental parameter analysis methods, and discloses a wind speed and temperature return period calculation method based on joint distribution, and the method comprises the following steps: obtaining wind speed and temperature data based on an on-site actual measurement system, and carrying out the fixed-period temperature extreme value and corresponding wind speed screening; determining a wind speed and temperature one-dimensional distribution model based on the goodness-of-fit index of regression analysis; and acquiring a wind speed and temperature joint distribution model by using an Archimedes Copula theory based on the rank correlation coefficient. According to the wind speed and temperature return period calculation method based on joint distribution, a method for accurately reflecting extreme temperature and corresponding wind speed distribution characteristics is provided to accurately estimate local environmental parameters; and the joint distribution characteristics of the most key factors in the environmental parameters are clarified by using the joint distribution of the two factors, and a plurality of return period calculation results capable of being used for different targets are provided, so that the widely developed engineering structure production and life activities are guided through accurate identification of the environmental parameters and description of the distribution characteristics.

The present invention discloses a method for fine division of fracture-vug reservoir rock facies, including: obtaining single-well data of a region to be measured and performing preprocessing; performing depth alignment on required core physical property data; performing principal component analysis on the single-well data to obtain dimension-reduced and noise-reduced fusion characterization data; dividing it into a clustering data set and a test data set; using the elbow method to select the optimal classification cluster for the clustering data set; using a clustering algorithm improved by fuzzy C-means to classify the data and establishing an identification model for fracture-vug reservoir rock facies; using the silhouette coefficient to evaluate the identification model for fracture-vug reservoir rock facies; combining the test data set to test the identification model for fracture-vug reservoir rock facies; using the test identification model for fracture-vug reservoir rock facies to output a test clustering result and performing sensitivity analysis using a rank correlation matrix; using a sensitivity curve to perform crossplotting to obtain an identification chart for fracture-vug reservoir rock facies.

The invention discloses a discrete element significant mesoscopic parameter screening method based on multi-model fusion. According to the method, based on a Plackett-Burman test, a Spearman correlation coefficient and a gradient boosting tree algorithm, the contribution degree, correlation and importance degree of a mesoscopic parameter relative to a response value are analyzed and calculated from the perspective of a linear regression model, a rank correlation coefficient and a decision tree, and the one-sidedness of a common method is made up; the Softmax function is adopted to quantify and fuse results of the three types of models, rationality and comprehensiveness of parameter screening are improved, complexity of subsequent parameter calibration tests can be effectively reduced, and powerful support is provided for rapid and accurate establishment of discrete element models.

The present invention discloses a combined prediction method for gas concentration based on dynamic optimization of indicators, including: obtaining historical monitoring data of multiple indicators affecting future gas concentration; calculating the Spearman rank correlation coefficient between each indicator and the gas concentration indicator respectively, dynamically optimizing the indicators according to the calculated Spearman rank correlation coefficient to obtain optimized indicators; calculating the feature matrix of each optimized indicator; inputting the feature matrix of each optimized indicator into a preset single-indicator prediction model one by one to respectively obtain the predicted values of the maximum gas concentration within a preset time period after the current moment corresponding to each optimized indicator; inputting the predicted values corresponding to each optimized indicator into a preset combined prediction model to obtain the final predicted value for predicting gas overlimit conditions. The present invention can achieve dynamic and accurate overlimit prediction of the gas concentration and its overlimit conditions in coal mine excavation and working faces.

The invention provides a low-rank associated multi-source target detection method based on adversarial data enhancement, which comprises the following steps of: acquiring image sets of different domains as two original data sets X and Y, and respectively inputting data in the two original data sets X and Y into an encoder for dimension reduction processing to obtain two new data sets X'and Y '; introducing a uniform clustering structure S by applying the incidence matrixes V and U, and learning the uniform clustering structure S from the shared features of the data sets X'and Y '; and through a pre-trained generative adversarial network, virtual data sets # imgabs0 # and # imgabs1 # are generated for the data sets X'and Y 'by using a clustering structure S, the generated virtual data and real data are used together to train a target detection model, and target detection is performed. According to the method, under the condition that the sample size is small, training is performed again under rich multi-source heterogeneous sample data through the generated virtual data, and the generalization ability of the model on data association tasks is improved.

This invention relates to a method and system for decoupling risk characteristics of natural electric fields on the surface based on the principle of order. The method includes: determining the spatial range of an electrode array based on the principle of spatial separability of risk sources and deploying the electrode array; acquiring multi-channel observation time series data of shallow surface natural potential through the electrode array; determining a time window based on the principle of spatial separability of risk sources; calculating the rank correlation coefficient between any two observation channels within the time window; mapping the calculated rank correlation coefficient to the corresponding field source state level according to a preset field source state level classification standard; calculating a decoupling index based on the normalized proportion of each field source state level and its preset risk weight; and directly quantifying the overall orderliness of the field sources through the decoupling index, thereby decoupling and identifying risk characteristics from the dynamic signal of the shallow surface natural electric field, effectively solving the core problem of identifying risk characteristics.

The invention relates to the technical field of transformers, in particular to a transformer state evaluation method based on a fuzzy mathematical theory. The method comprises the following steps: firstly, collecting original test data of a transformer, and performing normalization processing; and calculating the support degree and the lifting degree of the feature index based on the association rule, and judging the rationality of the feature index based on the lifting degree. And then, establishing a transformer state evaluation system according to the selected characteristic indexes, and dividing comment set grades of the transformer. Then, the basic probability of the feature index is determined using a dynamic fuzzy membership adjustment mechanism. And finally, using an improved D-S evidence theory based on a Spearman grade correlation coefficient to fuse the basic probabilities of different characteristic indexes to evaluate the comprehensive state of the transformer. The method can effectively improve the accuracy of transformer state evaluation, and can be used for guiding the actual maintenance arrangement work of the transformer.

The present application relates to the technical field of dynamic wireless power transfer (DWPT), and especially relates to a DWPT system electromagnetic exposure evaluation method and system considering human activity, aiming at the problem that the existing static evaluation method is difficult to take into account the time-varying characteristics of dynamic system and the difference of human activity distribution, the present application derives the time domain expression of the receiving coil current through establishing a circuit model, quantifies the influence of speed and load on the dynamic change of current; the electromagnetic region is classified into alert zone and waiting zone, and multiple height monitoring planes are set to correspond to the main organs of the human body; the magnetic field data set is obtained by laying measuring points, the regional average magnetic induction intensity, the maximum magnetic induction intensity and the magnetic field energy are calculated as the evaluation target; based on the rank correlation coefficient and the mutual information value, the redundant measuring points are removed, the core measuring point combination is screened by combining the NSGA-II multi-objective optimization, and the multi-objective high-precision synchronous prediction is realized by using the second-order ridge regression, so that the accurate characterization and low-cost measurement of dynamic electromagnetic exposure are realized.

The invention provides a time series data prediction method and device based on dynamic error modeling, and the method comprises the steps: predicting original time series data through an LSTM model, and generating a historical error sequence of a predicted value and a real value; carrying out CEEMDAN mode decomposition on the sequence, and extracting a plurality of error mode components with non-stationary characteristics; clustering the fusion components by using an affinity propagation algorithm to generate a representative error modal cluster after dimension reduction; analyzing the nonlinear correlation between the historical error sequence and each modal cluster based on a Spearman rank correlation coefficient, and obtaining a dynamic weight distribution result through normalization; and respectively predicting a future error value of each modal cluster through the parallel LSTM network, weighting and correcting a prediction result according to a dynamic weight, and generating a final prediction value. According to the time series data prediction method based on dynamic error modeling, parallel collaborative optimization of prediction and error correction is realized through explicit modeling error dynamic characteristics in combination with a CEEMDAN-AP joint optimization algorithm and Spearman correlation weight distribution, and the precision and real-time performance of time series prediction are remarkably improved.

This invention relates to a knowledge graph-based method, system, and medium for vehicular network (V2X) threat analysis. The method includes: U1. collecting data information of V2X entity objects and V2X threat intelligence; U2. based on the V2X threat intelligence data, extracting feature information of the V2X threat intelligence using a hierarchical clustering algorithm based on expected cross-entropy to obtain feature matrix data information of the V2X threat intelligence, and using a Spearman rank correlation coefficient-based cross-validation algorithm to characterize the correlation between entity objects and the feature matrix of the V2X threat intelligence to obtain data information on the correlation between V2X entity objects and threat intelligence. This invention not only solves the problem of not being able to comprehensively cover the identification of threats in all dimensions of V2X, but also provides a comprehensive and intuitive display of V2X threat intelligence from various angles, facilitating automotive security analysts to flexibly analyze the V2X security situation.

This application discloses a big data-based agricultural environmental monitoring method and system for wheat growth monitoring, relating to the field of agricultural environmental monitoring, including: dividing the monitoring area into time zones of equal size according to size and time; collecting leaf area index time series {LAI} ij} and wheat ear number time series {EPN ij The Mann-Kendall trend test was used to perform significance tests, and the test results were denoted as Trend. LAI and Trend EPN Perform rank transformation to obtain the rank value sequence {RLAI}. ij} and {REPN ij The correlation coefficient ρ is calculated using the Spearman rank correlation coefficient algorithm. s Nonlinear fitting was performed using the Gompertz function to obtain the growth rate parameter b and the maximum number of wheat ears K; Trend LAI Trend EPN , ρ s Using b and K as inputs, a weighted average algorithm is used to calculate wheat growth, yielding an evaluation index reflecting crop growth. Addressing the low accuracy of existing wheat growth monitoring technologies, this application combines the weights of various indicators to calculate a comprehensive evaluation index reflecting wheat growth, thus improving monitoring accuracy.

The invention discloses a multi-module coupled double-unit traction turnout fault intelligent diagnosis method and device. The method comprises the following steps: determining predicted fault types corresponding to a first extension set and a second extension set by adopting an integrated rank correlation metric algorithm; a predicted fault type combination is obtained, a preset conflict detection rule base is inquired based on the predicted fault type combination, and conflict detection rules meeting conditions are obtained; calculating to obtain a conflict detection value; if the conflict detection value is 1, a preset fault cause screening rule base is inquired based on predicted fault types corresponding to the first extension set and the second extension set, and a corresponding fault cause factor set is obtained; performing non-null logic judgment and cascade coding based on the fault-induced factor sets corresponding to the first extension set and the second extension set to obtain a coding result; querying a preset fault positioning coding table to obtain a corresponding fault cause range; and querying a preset maintenance support rule base, and outputting a fault processing suggestion. The accuracy and diagnosis depth of dual-computer fault analysis can be improved.

The invention discloses a combined heat and power generation unit load distribution prediction and optimization method, and the method comprises the following steps: S1, obtaining the historical data of a combined heat and power generation unit, carrying out the grouping of two working conditions: medium-pressure air exhaust and low-pressure air exhaust, carrying out the feature extraction through the combination of a Spearman rank correlation coefficient and mutual information, and carrying out the prediction of the load distribution of the combined heat and power generation unit; removing redundant features and low correlation features to obtain a target data set for model training; s2, constructing a TCN-ECA-BiLSTM prediction model, wherein the TCN-ECA-BiLSTM prediction model comprises a time sequence convolutional network layer, an efficient channel attention layer and a bidirectional long and short term memory network layer; according to the method, a TCN-ECA-BiLSTM prediction model and a TD3 reinforcement learning algorithm model are combined to form a prediction and optimization integrated framework, key feature reinforcement of the TCN-ECA-BiLSTM prediction model is matched with bidirectional sequence modeling advantages to reinforce prediction input after feature extraction, and the TD3 reinforcement learning algorithm model carries out distribution by considering a steam coal consumption rate and a power generation steam consumption rate. And real-time and reliable dynamic optimization is realized.

The application discloses a lung nodule classification method based on multi-task learning, comprising the following steps: 1, extracting lung nodule CT data and feature level labels; 2, calculating the Spearman rank correlation coefficient between lung nodule features, constructing and training a neural network so that the cosine similarity between the initial labels embedded in the output lung nodule features is approximately equal to the Spearman rank correlation coefficient; 3, designing an adaptive method to obtain a graph adjacency matrix to describe the correlation between lung nodule features; 4, constructing and training a multi-task model with GCN and 3D U-Net as the backbone network, containing an image fusion module and a cross-channel attention module, outputting the malignant degree classification result of the lung nodule, and giving the feature score and segmentation result to provide auxiliary information for the classification result. The application fully utilizes the correlation between lung nodule features to construct a lung nodule classification model, which can output multi-dimensional analysis results for lung nodules, so that the diagnosis result has high interpretability and credibility.

This invention discloses a multi-objective prediction method for power system sources and loads based on Laguerre polynomial theory, relating to the field of power system prediction and intelligent dispatching technology. The method includes the following steps: using Spearman Rank Correlation Coefficient (SRCC) to analyze the correlation of characteristic influencing factors of wind power, photovoltaic power, and power load; and using Robust Local Mean Decomposition (RLMD) to decompose the time series of wind power, photovoltaic power, and power load into high-frequency and low-frequency components to reduce their fluctuations; using Weighted Permutation Entropy (WPE) to analyze the complexity of the subsequences after RLMD decomposition, merging subsequences with similar complexity to reduce the model's prediction complexity; and constructing a hybrid Laguerre neural network prediction model using Laguerre polynomials. This invention is the first to simultaneously consider both accuracy and stability objectives in source and load prediction, selecting a compromise solution in the Pareto front using the MORUN algorithm, making the prediction results more applicable to power system dispatching scenarios with high robustness requirements.