548 results about "Hyperparameter optimization" patented technology

A system is provided for managing Internet of Things (IoT) networks. The system includes a learning module configured to employ machine learning models with hyperparameters optimized through a hyperparameter optimization process; wherein the process includes evaluating a set of hyperparameters against a performance metric to select optimal hyperparameters that enhance the adaptability and efficiency of dynamic membership functions within an adaptive fuzzy logic engine (AFLE).

Techniques for increasing the quality of output from large language models using reinforcement learning to select inference-time hyperparameters are disclosed. The large language model is configured with a set of values corresponding to a set of inference-time hyperparameters that are used to influence the output of the machine learning model after the model has been frozen. After obtaining a set of performance metrics that indicate the quality of the output, a reinforcement learning agent computes an adjustment for one or more of the hyperparameters, resulting in a modification of the hyperparameter values. Applying the new hyperparameter values, the large language model is then applied to a new set of input to generate a second output. The process iterates until the performance metrics associated with the output are satisfactory.

The invention relates to the technical field of typhoon prediction, and discloses a typhoon rapid enhancement prediction method based on time-space sequence and multi-modal feature fusion, and the method comprises the steps: constructing a multi-modal time-space sequence data set and an auxiliary data set based on typhoon optimal path data and multi-source satellite observation data; a unified manifold approximation and projection method is adopted to carry out dimension reduction preprocessing on the high-dimensional multi-modal space-time sequence data, and one-dimensional time sequence embedding representation of the typhoon observation sequence is generated; taking the one-dimensional time sequence embedded representation and the auxiliary data as independent input channels, and inputting a trained typhoon observation network model to predict a typhoon rapid enhancement probability; wherein the typhoon observation network model is a multi-mode time-space fusion deep learning architecture, the core of the typhoon observation network model is composed of a variational attention recurrent neural network, and hyper-parameter optimization is carried out through an improved Harris eagle optimization algorithm. According to the invention, accurate and robust identification of the typhoon rapid enhancement process is realized.

The invention discloses a multi-target intelligent optimization method and system for blasting parameters of a strip mine in a high-altitude cold region. The method comprises the following steps: carrying out data acquisition to obtain a parameter data set; performing data preprocessing on the parameter data set to obtain a feature sample set; constructing an initial blasting parameter model based on a machine learning algorithm, and performing hyper-parameter optimization on the model to obtain a blasting parameter model; a multi-objective optimization function is constructed: based on the multi-objective optimization function and the blasting parameter model, solving is carried out in combination with environmental condition constraints, and a pareto optimal solution set is obtained; according to the pareto optimal solution set, a representative solution is selected, a visual scheme is generated, and blasting parameter optimization of the strip mine in the high-altitude cold region is completed. According to the method, temperature, oxygen and frozen soil constraint conditions of the high-cold and high-altitude environment are introduced, blasting safety, lumpiness uniformity and the explosive utilization rate are considered at the same time through multi-target collaborative optimization, the method can adapt to the extreme environment, meanwhile, the one-sidedness of single-target optimization is avoided, and the intelligent level of blasting design and implementation is greatly improved.

The invention discloses a settlement amount time sequence prediction method and system for deep foundation pit excavation adjacent buildings. The method comprises the steps that original monitoring data of on-site building settlement are acquired; performing data preprocessing on the obtained original monitoring data; constructing a recurrent neural network; model input and output parameters are determined through principal component analysis; optimizing the recurrent neural network based on an optimizer; performing hyper-parameter optimization based on an optimization result; performing settlement time sequence prediction by using the optimized recurrent neural network; the system comprises a data acquisition module, a preprocessing module, a model construction module, an analysis module, an optimization module, a parameter optimization module and a prediction module. By constructing the settlement prediction model, dynamic modeling and accurate prediction of the settlement trend of the building in the excavation process of each stage of the foundation pit are realized; historical settlement monitoring data and multi-layer soil body excavation information are combined, and multi-source input parameters are introduced, so that the adaptability of the model to complex working conditions is enhanced.

The invention discloses a space-time carbon emission prediction method, system and device and a medium. The method comprises the steps of obtaining load historical sequence data; carrying out load flow calculation by adopting a first load flow analysis algorithm to obtain a load flow calculation result; based on the load flow calculation result, carbon flow analysis is carried out through power transmission between the nodes, and the carbon emission coefficient of each node is obtained; constructing a time sequence diagram neural network model, and performing preliminary training on the time sequence diagram neural network model; performing hyper-parameter optimization on the preliminarily trained time sequence diagram neural network model by adopting a first hyper-parameter optimization technology to obtain an optimized time sequence diagram neural network model; and inputting load sequence data and unit power output data at the current moment into the optimized sequence diagram neural network model, and carrying out online prediction on a carbon emission coefficient. According to the method, the real-time prediction of the carbon emission coefficient is realized based on the characteristic learning of the historical operation and monitoring data of the power system.

The invention provides a comprehensive energy consumption prediction method for a tilting type fuel gas aluminum melting furnace based on CPO-ITCN-GRU. The method comprises the steps that original energy consumption data generated in the running process of the tilting type fuel gas aluminum melting furnace and auxiliary characteristic data related to energy consumption are collected; comprehensive energy consumption conversion is carried out on the original energy consumption data, and denoising smoothing is carried out on the comprehensive energy consumption data obtained through conversion and auxiliary feature data related to energy consumption; dividing the denoised and smoothed data into a training set and a test set according to a preset proportion; constructing an ITCN-GRU comprehensive energy consumption prediction model, training the ITCN-GRU comprehensive energy consumption prediction model by using the training set, and performing hyper-parameter optimization by using a crown porcupine optimization algorithm in the training process; and performing performance evaluation on the trained ITCN-GRU comprehensive energy consumption prediction model by using the test set. According to the method, the capturing capability of the model on the local space-time characteristics and accurate hyper-parameter optimization are improved, so that the comprehensive energy consumption prediction accuracy of the tilting type fuel gas aluminum melting furnace is improved.

The invention discloses a hyper-parameter optimization method and system of an icing galloping prediction model based on meta-learning. The method comprises the following steps: collecting original data from meteorological monitoring equipment, a line state sensor and historical icing galloping records; uniformly aligning the collected original data according to time granularity to obtain multi-modal time sequence data; performing internal and external double-loop training on hyper-parameters of the constructed icing galloping time sequence prediction model by adopting the obtained multi-modal time sequence data, performing iterative training on internal loops of the internal and external double-loop training under the given hyper-parameters, and after model parameters converge, obtaining the icing galloping time sequence prediction model. And performing meta-learning-level hyper-parameter optimization on the basis of converged model parameters in the outer loop, and performing adaptive search on hyper-parameters on the premise of fixing an inner loop training normal form to obtain optimal hyper-parameters. According to the method, the key hyper-parameters of the model can be quickly and adaptively optimized under different meteorological scenes and line working conditions, the prediction precision and the convergence speed are improved, and the icing galloping risk of the power transmission line is warned in real time.

A system for classifying apple leaf diseases using deep learning and feature fusion, consisting of: A data input module, which includes a data storage module, is configured to store a data set created using various data sources, with the data set containing apple leaf images being derived from the data sets "Apple Leaf 9", "Kashmiri Apple Plant Disease" and "Plant Village Apple Leaf"; a data preprocessing module configured to perform preprocessing of the newly prepared dataset of apple leaves, wherein the data preprocessing module is configured to perform data decoding, expansion, resizing, segmentation, scaling and color conversion of the input image data; a feature extraction module that is operationally connected to the data processing module and is configured to receive preprocessed data and transfer the preprocessed data to one or more convolutional neural network models for feature extraction; a feature fusion module configured to combine the extracted features from the Convolutional Neural Networks to develop a fused feature vector; a hyperparameter optimization module configured to optimize the hyperparameters of the feature extraction models by implementing a particle swarm optimization algorithm; a classification module configured to classify the fused trait vector into 13 apple leaf disease classes using a random forest classifier; and a user interface connected to the classification module, configured to display the classification results.

The invention provides a lithium ion battery health state lightweight detection method based on a physical information neural network, and the method comprises the steps: collecting the time, voltage, current, temperature and state-of-charge data of a battery in a takeoff and landing stage discharge process, processing the data into takeoff and landing stage discharge time sequence data, and carrying out the detection of the lithium ion battery health state based on the takeoff and landing stage discharge time sequence data. The method comprises the following steps: designing characteristic factors related to battery aging, screening the characteristic factors by utilizing a Pearson's correlation coefficient and a grey relational degree algorithm to obtain optimal characteristic sequence data, inputting the optimal characteristic sequence data into a physical information neural network model constructed by two serially connected neural networks for training, and in the training process, obtaining the optimal characteristic sequence data. And performing hyper-parameter tuning on the two neural networks by adopting a Bayesian optimization algorithm, then performing fine tuning on the second neural network by adopting a hierarchical transfer learning strategy, and finally applying the trained physical information neural network model to battery health state detection. The method improves the quality of feature data, reduces the calculation complexity of features and models, and achieves the accuracy and reliability of the detection of the health state of the battery under the airborne condition.

The invention relates to the technical field of material nondestructive testing, in particular to a CFRP wing skin damage positioning system and method based on multi-modal signal processing and Bayesian optimization DSCN, and the method comprises the steps: carrying out permutation entropy analysis and Higuchi fractal dimension analysis on a signal data set received by a sensor; performing time-frequency feature extraction on the sensor receiving signal data set deviating from the reference; performing damage feature screening on the signal data in the effective time period; training the DSCN structure to obtain an initial CFRP wing skin damage positioning prediction model; performing hyper-parameter optimization on the initial CFRP wing skin damage positioning prediction model according to a Bayesian algorithm; and performing damage positioning prediction on a to-be-detected sample according to the final CFRP wing skin damage positioning prediction model to obtain a corresponding CFRP wing skin damage positioning result. According to the method, effective damage features can be accurately and efficiently extracted from complex signals, and the accuracy and efficiency of CFRP wing skin damage positioning are remarkably improved.

A train-track-bridge coupling response prediction method based on a sparrow optimization algorithm and a long short-term memory network comprises the steps that data such as train speed, axle load, track vibration acceleration, bridge strain and environment temperature are collected in real time through a multi-source sensor, and a multivariable time series data set is constructed after wavelet denoising and standardized preprocessing; and designing an LSTM network architecture on this basis, introducing an attention mechanism to dynamically allocate feature weights of each time step so as to enhance the ability to capture key signals in the track irregularity mutation and bridge resonance interval, and adopting a sparrow optimization algorithm to globally search an optimal combination of a hidden layer neuron number, a learning rate and a time step length in order to solve the problem of LSTM hyper-parameter optimization. Through the dynamic adaptive step length strategy balance algorithm, the early-stage global exploration and later-stage local development capabilities are balanced, the local convergence defect of a traditional grid search or genetic algorithm is avoided, the calculation efficiency can be remarkably improved, errors can be reduced, and the prediction precision can be improved.

The invention relates to the technical field of deep learning hyper-parameter optimization, in particular to a multi-target hyper-parameter joint optimization method and system for a deep learning model. The specific implementation process comprises the steps of obtaining a current hyper-parameter of a deep learning model, generating a learning track vector, and performing utility prediction to generate a learning utility projection; on the basis of the dominating relationship between the learning utility projection and the current Pareto optimal leading edge, pruning the disadvantage training task, updating the Pareto optimal leading edge by using non-dominating sorting, and constructing a Pareto strategy network; and carrying out topology congestion degree analysis on the updated Pareto optimal leading edge, generating a leading edge exploration bias vector, inputting the leading edge exploration bias vector into the Pareto strategy network to update hyper-parameter codes, and entering the next round of iteration. According to the method, through the multi-objective optimization algorithm and in combination with the dynamically updated Pareto strategy network, the problems that the hyper-parameter optimization process is long in time consumption, low in efficiency and prone to falling into local optimum are effectively solved, and the optimization efficiency and the solving quality are improved.

The invention discloses a corroded oil and gas pipeline residual strength prediction method considering a physical constraint loss function, and the method comprises the steps: collecting multi-source feature data of a corroded oil and gas pipeline, obtaining a residual strength measured value as a label, and constructing a training data set; an XGBoost regression model is combined with an SHAP interpretability analysis technology, and the influence degree and the influence direction of each feature on the residual intensity are quantified; constructing a neural network model, and determining an optimal architecture of a neural network by adopting a hyper-parameter optimization method; constructing a physical constraint term based on the influence degree and the influence direction of each feature, introducing the physical constraint term into a loss function of a neural network model, and forming a comprehensive loss function together with a data-driven loss term; and training the optimized neural network model by using a comprehensive loss function to obtain a final residual intensity prediction model. The method has the advantages that the prediction precision is improved, the model interpretability is enhanced, overfitting is prevented, and multi-source feature data are effectively integrated.

The invention discloses a light energy power station fault prediction system based on deep learning. The system comprises a data acquisition module used for reading equipment operation data from a sensor; the data preprocessing module is used for denoising, interpolating and standardizing the equipment data; the graph convolutional network construction module is used for constructing an equipment data graph structure and extracting features; the Lemap dimension reduction module is used for mapping the high-dimensional equipment features to a low-dimensional space; the time sequence modeling module is used for constructing a time sequence prediction model based on the low-dimensional features; the hyper-parameter optimization module is used for optimizing hyper-parameters of the time sequence model; the model verification module is used for evaluating the precision and response time of the fault prediction model; the model deployment module is used for deploying the prediction model to a monitoring system; the fault prediction and early warning module is used for monitoring in real time and generating fault early warning; and the continuous optimization module is used for regularly optimizing and retraining the fault prediction model. The method achieves the high efficiency of fault prediction of the light energy power station, remarkably improves the prediction precision and the reliability of equipment operation, and is widely suitable for equipment monitoring and early warning.

The embodiment of the invention discloses an air conditioner maintenance data classification method and system based on machine learning, and the method comprises the steps: integrating multi-source heterogeneous maintenance records generated in the maintenance process of air conditioner equipment, building a correlation index through a common identification field, and fusing dispersed data into a maintenance data set in a uniform format; performing hierarchical semantic analysis on unstructured texts in the set to generate structured semantic features, and performing time sequence feature extraction on structured data; then constructing a hybrid classification model training framework fusing semantic and time sequence features, and generating a maintenance data classification model through feature space alignment, dynamic weight distribution, hyper-parameter optimization and an early stop strategy; and finally, classifying newly-added maintenance records by applying the model, checking by combining an expert knowledge base, manually rechecking conflict results, and returning corrected data as an incremental sample back to the model to realize continuous optimization.

The invention relates to the technical field of water quality prediction, in particular to a river water quality prediction method and system based on a machine learning coupling hydrological model, and the method comprises the following steps: obtaining a data set, and dividing the data set into a training set and a test set; constructing a limit gradient lifting model, and performing hyper-parameter optimization on the model; performing correlation analysis on the water quality parameter set and the water quality index WQI value based on the trained limit gradient lifting model, and screening to obtain key water quality parameters influencing the water quality index WQI value; constructing an LSTM model and a soil and water evaluation tool model; simulating a future hydrological water quality process based on the soil and water evaluation tool model, and calculating and outputting a future water quality parameter result; and inputting a future water quality parameter result into the trained LSTM-WQI model to predict a future river WQI value so as to obtain a prediction result. According to the invention, the machine learning algorithm is coupled with the hydrological model to construct the water quality evaluation model which is convenient to use and adapts to local conditions, and efficient and accurate prediction of future water quality is realized.

The invention relates to the technical field of shield tunnel construction intelligent control, in particular to a shield tunneling machine tunneling speed intelligent prediction method based on multi-algorithm collaborative optimization. The method comprises the following steps: firstly, acquiring operation parameters and geological and environmental parameters of the shield tunneling machine in real time through a multi-source sensor and an industrial bus; carrying out localized cleaning, normalization and feature extraction on the data by utilizing an edge computing device; further, integrating particle swarm optimization, a genetic algorithm, a sparrow search algorithm and a starvation game search algorithm, and realizing global automatic optimization of the hyper-parameters of the bidirectional long-short-term memory network through a parallel independent optimization and result aggregation strategy; and finally, predicting the tunneling speed in real time by using the optimized model. According to the method, the problems of low prediction precision, weak model generalization capability, dependence on manpower on hyper-parameter optimization and the like caused by insufficient multi-source heterogeneous data processing capability are effectively solved, the prediction accuracy, the adaptability and the engineering practical value are improved, and reliable support is provided for safe and efficient propulsion of shield construction.

The invention discloses a weld defect identification method based on a dense connection convolutional network model, and the method specifically comprises the following steps: S1, constructing a dense connection convolutional network model, and embedding a coordinate attention module behind a transition layer of a convolutional network; s2, data acquisition and processing: acquiring an RGB image of the welding seam through an industrial camera, constructing a data set of the image, and performing image enhancement and standardization processing; s3, performing hyper-parameter optimization, and performing global optimization on the constructed model by adopting a Bayesian optimization algorithm; s4, performing model training and verification, and training a dense connection convolutional network model by using the optimized hyper-parameter combination; and S5, defect identification: inputting a to-be-detected welding seam image into the trained dense connection convolutional network model, and outputting a defect category and a positioning result. According to the method, the transition layer of the convolutional network is embedded into the coordinate attention module, so that the convolutional network model more accurately positions the welding seam position, and the detail features of the welding seam are extracted.

The invention discloses a seismic first arrival wave pickup method and device based on a multi-input attention mechanism network, multi-domain data and multiple network advantages are fused to realize first arrival automatic pickup, and compared with manual pickup, the efficiency is greatly improved, and the result is more objective; compared with a traditional automatic pickup method, the method is higher in precision and stability under different signal-to-noise ratios. Meanwhile, through hyper-parameter optimization, the generalization ability is high, reliable data can be provided for subsequent seismic processing, the advantages are outstanding under complex geological conditions, and the method is of great significance to geological analysis and resource exploration.

The invention provides an LSTM underwater robot modeling method based on TPE hyper-parameter optimization, and relates to the technical field of underwater robot modeling, and the method comprises the steps: carrying out the time sequence feature extraction through a memory unit comprising a forgetting gate, an input gate and an output gate, and predicting the state change amount delta Yt'at a t + 1 moment through an output layer; performing iterative optimization on the number of layers and the number of units of the LSTM network and training the model to obtain an optimized LSTM model; evaluating the optimized LSTM model through a multi-step cumulative prediction error, inputting an initial real state into the model to carry out T-step recursive prediction, updating a current state by utilizing a prediction state increment in each step, and finally calculating an average position error and an attitude error in a T-step window on a verification set; and selecting the hyper-parameter combination with the minimum comprehensive error of the verification set as a final model parameter, and completing the dynamic modeling of the underwater robot. According to the method, the problem that the model is inaccurate due to excessive parameterization in a nonlinear dynamic model can be solved.

The invention discloses a catalyst process design and optimization method and device, and belongs to the field of machine learning. The method comprises the steps of obtaining original process data in a catalyst production process, performing data cleaning and feature engineering processing, and generating a preprocessing data set capable of being used for modeling; screening a plurality of candidate machine learning models by using an AutoML technology, and performing hyper-parameter optimization on the baseline model; an incremental learning technology is adopted to update the optimized model online in real time so as to adapt to data fluctuation in the production process; and finally, the online updated model is applied to quality prediction and process parameter adjustment, so that closed-loop optimization of the production process is realized, and the product consistency and the production efficiency are improved. According to the method, the automation level and the intelligent quality control capability in catalyst production can be effectively improved.

The invention relates to an LSTM daily runoff prediction method based on MFF and NRBO, and the method comprises the steps: carrying out the variational mode decomposition of an original runoff sequence, and obtaining a plurality of intrinsic mode function components; hydrometeorological characteristics highly related to the runoff are screened through correlation analysis; fusing the intrinsic mode function component with the screened hydro meteorological features, and constructing a multi-dimensional information feature matrix; using a Newton-Raphson optimization algorithm to optimize hyper-parameters of the LSTM model; and taking the multi-dimensional information feature matrix as input, and performing daily runoff prediction by adopting the optimized LSTM model. The method has the beneficial effects that the Newton-Raphson optimization algorithm is applied to hyper-parameter optimization of the runoff prediction model for the first time, and the global search capability is enhanced, so that the prediction precision is improved. Meanwhile, multi-feature fusion is carried out on the intrinsic mode function component and the screened hydro meteorological features, and the multi-feature fusion is combined with hyper-parameter optimization, so that runoff prediction is more efficient and stable.

The invention discloses a water quality prediction method and system based on a Bayesian optimization coupling generative adversarial network, and belongs to the technical field of water quality prediction.The method comprises the steps that monitoring data in the water production process are collected, missing value elimination, data screening, abnormal value processing and standardization processing are conducted on the monitoring data in the water production process, and a water quality prediction result is obtained; obtaining a preprocessed water production process monitoring data set; the water quality prediction model is used for alternately training an effluent quality predictor and an added dosage estimator by adopting joint training; performing hyper-parameter optimization on the overall loss function by adopting a Bayesian optimization technology, and synchronously controlling a training process by adopting an early stop mechanism to obtain a water quality prediction model; and the water quality prediction model outputs water quality prediction result data after calculation and analysis based on the water production process monitoring data. According to the method, the water quality prediction model is formed through coupling of the Bayesian optimization technology and the generative adversarial network, and high-precision water quality prediction can be carried out in a production environment with complex water quality changes and uneven data distribution by adopting the model.

The invention discloses a role-driven business process automatic approval method based on reinforcement learning, and the method comprises the following steps: S1, collecting business process data, carrying out the preprocessing, and constructing a preprocessing data set; s2, constructing a graph data structure, extracting role interaction features, and generating a role dynamic weight matrix; s3, optimizing training by adopting an improved near-end strategy, and optimizing an approval path through shearing and regularization; s4, searching an optimal hyper-parameter by using a He-Make algorithm, and optimizing a learning rate, a discount, entropy regularization and a shearing range; s5, generating an optimal approval path, and dynamically adjusting role permission and manual intervention; s6, analyzing the approval data in real time, and updating the network and the weight matrix online; and S7, deploying a strategy network, and monitoring and dynamically adjusting the approval strategy in real time. According to the method, reinforcement learning and improved near-end strategy optimization are combined with the Hereum algorithm, and dynamic intelligent optimization of the approval process and role permission is realized, so that the approval efficiency and accuracy are greatly improved, and the manual intervention rate is reduced.

The invention designs a wind power prediction method based on improved time sequence fusion Transform (ITFT). According to the method, a Mama module is adopted to replace a traditional LSTM encoder-decoder structure, so that the long sequence modeling capability is remarkably improved; designing a wind speed prediction network (WFN) to generate future wind speed prediction as auxiliary input of the decoder; the improved EHO algorithm is applied to carry out hyper-parameter optimization, and chaos initialization, a fitness-distance balance strategy and a hybrid variation mechanism are fused. According to the method, the technical bottlenecks of an existing wind power prediction method in the aspects of precision, efficiency and interpretability are solved. Characteristic importance quantitative analysis is realized through a variable selection network, and a credible decision basis is provided for power grid dispatching. The method is suitable for wind power plant short-term power prediction, and the renewable energy consumption capability can be remarkably improved.

The invention discloses a water quality probability forecasting method based on Bayesian multi-time-sequence deep learning. The method comprises the following steps: S1, determining a forecasted water environment water ecological index, a driving index and a forecasting day number; s2, collecting time sequence data monitored by the forecasting indexes and the driving indexes, and after data preprocessing, constructing a data set required by model construction; s3, carrying out data division on the time sequence data, constructing a driving index forecasting model by adopting a multi-time sequence deep learning method, and carrying out parameter learning by selecting a Bayesian random discarding method; s4, performing effect evaluation on the accuracy and precision of the model, and adopting a hyper-parameter optimization method to improve the simulation forecast effect; s5, carrying out model training by adopting all data without segmenting the training set and the test set, carrying out water quality probability forecasting by utilizing the trained model, and outputting a forecasting mean value and a confidence interval; according to the method, the confidence interval is output while high-precision prediction is provided, and the scientificity and stability of prediction are improved.

The invention discloses an ocean buoy data filling method based on a space-time neural network. In order to solve the problems of insufficient spatial-temporal dynamic correlation modeling and low filling precision in the prior art, high-precision reconstruction of ocean buoy missing data is realized by coupling a graph convolutional network GCN and a gating cycle unit GRU in combination with spatial-temporal feature modeling. The method comprises the steps of data preprocessing; analyzing a data missing mode, and constructing a data sample by utilizing double mask matrix construction and a time window division strategy; based on multi-scale spatial feature extraction of GCN and long-term time-dependent modeling of GRU, model performance is improved through hyper-parameter optimization and Bayesian search; and carrying out model training, and evaluating the filling effect by adopting indexes such as mean square error. According to the method, the reconstruction precision of the space-time correlation missing value is remarkably improved, the generalization ability for a real missing mode is enhanced, and the method is suitable for real-time data processing of a large-scale ocean monitoring network and has high engineering application value.

The invention discloses a fault detection model training method, a fault detection method, a fault detection device and fault detection equipment, which are applied to the technical field of computers, and comprise the following steps: obtaining an improved butterfly optimization algorithm; the improved butterfly optimization algorithm is an algorithm for guiding the population to move towards the direction of the optimal solution in the optimization process to realize convergence; hyper-parameters needing to be optimized of the fault detection model are determined, the fault detection model is trained based on an improved butterfly optimization algorithm, and target optimal hyper-parameters are obtained; and constructing a fault detection model based on the target optimal hyper-parameter to obtain a target fault detection model. Compared with the low training efficiency of the current fault detection model, the improved butterfly optimization algorithm is used for guiding the population to move towards the direction of the optimal solution in the optimization process to realize convergence, and the target optimal hyper-parameter is determined, so that the randomness of the search process is kept, the controllability of the direction is enhanced, and the fault detection efficiency is improved. The population position is continuously close to the region where the optimal solution is located, so that the efficiency of hyper-parameter optimization is improved.

The power distribution network load flow calculation method based on the hyper-parameter optimization graph attention network comprises the steps of simulating and generating load flow operation data of a system according to historical wind and light output data and load data, and forming graph data by utilizing node characteristic data and topological information of a power distribution network. On the basis, a graph attention network (GAT)-based power flow regression model of the power distribution network is established, and spatial correlation characteristics among nodes are mined through a graph attention mechanism. In addition, when the topology of the power distribution network changes, the model can dynamically adjust the attention coefficient according to the new node connection relation, and rapid adaptation to the new topology is achieved. In view of the fact that the performance of the neural network highly depends on setting of hyper-parameters, an improved seagull optimization algorithm (ISOA) is introduced to carry out automatic optimization on the network hyper-parameters, so that an optimal GAT power flow regression model, namely, ISOA-GAT, is obtained. And finally, a Python simulation platform verifies that the method not only can quickly and accurately calculate the power flow of the power distribution network, but also has relatively strong topology generalization capability.