2649 results about "Feature selection" patented technology

A method for structuring the latent space of an autoencoder is provided. The method includes analyzing natural language descriptions related to input data; creating language-guided libraries that categorize and abstract data features based on the analyzed descriptions; mapping input data into the categorized and abstracted features within the latent space of the autoencoder; and training the autoencoder to minimize reconstruction loss while adhering to the structure imposed by the language-guided libraries.

The invention discloses a high-resolution radar echo extrapolation prediction method fused with satellite data, and the method specifically comprises the following steps: firstly, inputting historical radar echo sequence preprocessing at a previous T moment, including denoising, normalization processing and data set segmentation, and obtaining cleaned data; then, through a deterministic modeling method (SimVP), a fuzzy prediction sequence of a future T duration is obtained, then a variational auto-encoder (VAE) maps an original radar echo image and the fuzzy prediction sequence to a low-dimensional potential space, and two-stage diffusion modeling is carried out on the basis; in the first stage, a space-time converter (ST-Translator) is used to extract space-time evolution characteristics of radar echoes; in the second stage, satellite data at the corresponding time of the previous T moment is input, preprocessing including normalization processing, feature selection and data set segmentation is carried out, cleaned data is obtained, and the influence of the satellite data is dynamically adjusted in the diffusion process by adopting a multi-source fusion denoising network Fsrform so as to make full use of satellite information; and finally, inversely transforming output results of the two stages into a pixel space to obtain a high-resolution radar echo extrapolation prediction result of the future T duration. According to the invention, computing resource consumption can be effectively reduced, and the precision and detail fidelity of short temporary rainfall prediction are improved.

The invention discloses a multi-field part size and appearance defect intelligent detection system, and relates to the field of image analysis. The system comprises an image acquisition module, an image preprocessing module, a feature extraction module, a defect identification and size measurement module, a data processing and analysis module, an automatic control module and a man-machine interaction module. According to the method, CNN and LBP, Hough transform and SIFT algorithms are fused, high-level semantics and bottom-level texture / geometric features are considered, 2304-dimensional fusion feature vectors are formed through feature splicing, the feature extraction integrity of parts in multiple fields is improved, texture detail features can be accurately extracted, geometric shape features can be accurately obtained, and the method is suitable for large-scale popularization and application. Meanwhile, the adaptive feature selection mechanism dynamically optimizes the feature combination according to the detection result, the problem of calculation redundancy is reduced, and the detection precision is ensured while the detection efficiency is improved.

The invention discloses a network security analysis early warning system based on artificial intelligence, and the system comprises a data collection layer which captures full flow based on DPI, aggregates firewall logs, terminal behaviors and threat intelligence, and constructs a structured data pool; through TLS fingerprint identification of AI driving, the encrypted traffic is penetrated, and a sampling strategy is dynamically adjusted in combination with reinforcement learning. The intelligent analysis layer is used for carrying out cross validation on known threats and abnormal behaviors; the time sequence CNN extracts encrypted traffic features, and a novel threat detector is rapidly generated by using historical attack fragments in combination with a meta-learning framework; sHAP value driving dynamic feature selection and optimization feature vector input; the decision-making early warning layer is used for fusing multi-source features through a Bayesian network and generating 0-100 score risk scores; a self-adaptive threshold module is combined to adjust a score threshold in real time, and a high-risk event is pushed; the collaborative response layer is used for triggering a preset decision tree, deploying a GAN dynamic honeypot to trap an attacker and reversely tracing; the Neo4j visually restores the attack path, and blocking is executed after the threat is confirmed by a progressive response mechanism.

The invention discloses a typhoon wave forecasting method based on integrated machine learning. The method comprises the following steps: firstly, integrating historical typhoon wave data, meteorological data and marine environment data; preprocessing the data, including integration, cleaning, vacancy filling and standardization, and performing multi-source data completion by adopting a K-nearest neighbor algorithm and a spline interpolation method; secondly, screening key characteristic parameters through a Pearson's correlation coefficient, and reinforcing nonlinear correlation representation in combination with a mutual information method; then, constructing an integrated prediction model containing an LSTM (Long Short Term Memory), an XGBoost (X Goose Boost) and a Transform; and finally, dividing a training set and a verification set by adopting a dynamic time sequence division strategy, optimizing model hyper-parameters, and completing training and testing of the typhoon wave height prediction model. According to the method, the data sparsity problem is solved through multi-source data fusion and feature selection optimization, the generalization ability is improved through an integrated model architecture, and compared with a traditional single model, the training period is remarkably shortened, and the forecasting precision and timeliness are improved.

The invention discloses an antagonism prompt detection method and device based on a large language model and a medium, and belongs to the technical field of artificial intelligence safety. The technical problem to be solved by the invention is how to overcome the defects of static rule lagging, high manual maintenance cost and insufficient context understanding in the security protection process of a large language model in the prior art, and dynamic, real-time and high-precision antagonism prompt detection is realized. According to the technical scheme, the method comprises the following steps: feature extraction: comprehensively analyzing semantic information, structural information and context information in a user text, extracting semantic features, structural features and context features, performing Min-Max normalization on the semantic features, the structural features and the context features, and splicing the semantic features, the structural features and the context features into a 128-dimensional joint vector, performing feature selection on the joint vector through an L1 regularization logistic regression model, compressing to 10-dimensional core features, and removing redundant information; carrying out resistance scoring; and dynamically defending.

The invention discloses a metering error correction method and system based on electric power data acquisition, and the method comprises the steps: detecting an operation state through a self-inspection and data acquisition module, and obtaining environment data, electrical parameters and error information; key feature data are selected and standardized through a data preprocessing module; an LSTM model and an SVR model are constructed through a model establishment and fusion module, and prediction results are fused to generate error compensation output; real-time error compensation is carried out on a combined prediction result by utilizing an ARIMA model and error time sequence processing through a dynamic error compensation module; the data smoothing and evaluation module is used for smoothing the voltage data and evaluating the error compensation effect; an error compensation effect is monitored through a monitoring and optimizing module, and feature selection and model parameters are optimized regularly; through the fault diagnosis and emergency module, an abnormal condition is detected, and a corresponding standby scheme is triggered. According to the invention, the accuracy and anti-interference capability of electric power metering are effectively improved, and the method is suitable for complex and changeable metering environments.

The invention relates to the technical field of industrial equipment health monitoring and fault diagnosis, and discloses a centrifugal pump intelligent detection method and system based on multi-source data. The centrifugal pump intelligent detection method based on the multi-source data comprises the steps that centrifugal pump multi-source heterogeneous data are collected and preprocessed; extracting multi-scale features and constructing an optimal feature subset through hierarchical feature selection; constructing a multi-level dynamic Bayesian network to establish a probabilistic reasoning framework; calculating a feature credibility weight based on the signal quality, the feature stability and the diagnosis correlation; executing multi-source evidence fusion and Bayesian state reasoning; confidence self-calibration and parameter self-adaptive adjustment are realized through historical diagnosis result feedback; the method can effectively cope with the complex and changeable environment of an industrial site, can still keep stable diagnosis performance under the condition that the sensor part loses efficacy or the data quality is uneven, improves the accuracy and reliability of fault diagnosis, and provides powerful technical support for predictive maintenance of industrial equipment.

The invention relates to a satellite-to-ground network adaptive image semantic communication method based on an NOMA technology, and the method comprises the following steps: S1, building a satellite-to-ground semantic communication frame integrated with the NOMA technology: carrying out the power superposition and serial interference elimination of different user transmission contents through the NOMA technology, a coding and decoding process is optimized through deep joint source channel coding based on hierarchical vision using a shift window; s2, establishing an adaptive semantic communication model: dynamically adjusting coding parameters based on a target bandwidth ratio and a signal-to-noise ratio of a satellite-ground channel; s3, proposing a dynamic weight adjustment algorithm: optimizing model convergence speed and transmission performance tradeoff under different channel conditions; and S4, designing a feature selection module: performing priority ranking on the image features, and selectively transmitting important features according to bandwidth resources. According to the method, the adaptive semantic communication model (ASeC-NOMA) which can be dynamically adjusted according to the target bandwidth ratio and the signal-to-noise ratio is constructed, and a large number of model storage parameters are saved while the transmission performance is ensured.

The invention provides a sewage treatment plant terminal effluent carbon source quality detection and evaluation method, which realizes full-flow dynamic evaluation and regulation of carbon source quality through on-line monitoring and intelligent algorithm fusion. According to the method, an online water quality full-spectrum detector is used for collecting original spectrum data flow, and after preprocessing such as variational mode decomposition denoising and mutual information feature selection, organic matter content quantification, variation trend analysis and anomaly detection are completed in combination with algorithms such as a support vector machine and an autoregressive moving average model. An entropy weight method is introduced to dynamically adjust the weight of the evaluation model, model parameters are optimized based on a gradient descent algorithm, a process adjustment instruction is generated through reinforcement learning and fuzzy logic, and an automatic system is linked to execute regulation and control. According to the method, the problems of hysteresis and singleness of traditional offline analysis are solved, multi-dimensional real-time evaluation, abnormal quick response and process dynamic optimization of the quality of the carbon source are realized, the sewage treatment efficiency and the effluent quality stability are improved, and a technical support is provided for continuous standard reaching of the quality of the carbon source.

The invention discloses a visual language multi-modal fusion method based on parameter-free cross attention, and belongs to the field of computer vision. The implementation method comprises the following steps: using a fixed pre-training language model as a trunk, using a visual encoder to extract image features, and calculating a cross attention weight between language query and visual features through a parameter-free activation function, replacing a plurality of groups of learnable projection matrixes introduced by a traditional cross attention module, and significantly reducing the model parameter scale. A multi-scale visual feature generation mechanism based on pooling operation is introduced, and rich visual semantic prompt information is provided for a language model. A dynamic feature selection module is designed in combination with cross attention, visual areas corresponding to all text tokens are screened, low-correlation areas are discarded, only visual content more contributing to the current language context is reserved, accurate information matching and efficient fusion between modals are achieved, and the accuracy and efficiency of information fusion are improved. And the performance of the visual language model in tasks such as image-text question answering, image generation and multi-modal instruction understanding is improved.

The invention discloses an abnormal transaction behavior detection method and system based on a block chain, and aims to improve the transaction security of the block chain. According to the method, a DBSCAN algorithm and an isolation forest algorithm are combined, and abnormal transaction behaviors are identified through clustering analysis and random feature selection. Wherein the DBSCAN algorithm is used for preliminarily screening abnormal transaction behaviors, and the isolation forest algorithm is used for further refining detection and evaluating abnormal scores of transactions. According to the invention, the false alarm rate is effectively reduced, the detection coverage and accuracy are improved, and technical support is provided for block chain transaction security.

The invention relates to the technical field of fault diagnosis, and particularly discloses a bearing fault diagnosis method based on adaptive feature migration fusion. According to the method, an adaptive feature migration fusion algorithm is utilized, and multi-mode signals of acceleration, load, temperature and rotating speed are comprehensively processed, so that accurate fault diagnosis of the embedded intelligent bearing under a complex working condition is realized. Aiming at the problems that a traditional diagnosis method is difficult in signal feature extraction and large in data distribution difference under different working conditions, the invention provides a self-adaptive feature migration fusion algorithm, and the robustness and expression ability of fault features in different working conditions and data sets are improved by constructing a migration learning model and combining a feature selection and fusion mechanism. Compared with the dependence of a traditional deep learning algorithm on large-scale labeled samples, the method can efficiently utilize a small amount of label data, the requirements for hardware computing power and data resources are remarkably reduced, and an efficient and flexible embedded bearing fault diagnosis method is provided for online fault diagnosis of an industrial site.

The invention relates to the technical field of energy consumption prediction, in particular to a building envelope energy-saving effect prediction method, which specifically comprises the following steps: deploying a sensor in a building maintenance structure to collect related data, and constructing a data set; marking the collected data to form a training set; a physical constraint decoupling normalization method is adopted for the data in the training set to generate features after decoupling normalization; a weighted feature vector is generated by adopting an attention mechanism guided by physical prior; constructing an energy-saving effect prediction network, and inputting the weighted feature vectors into the network for prediction; optimizing the network to obtain a trained network; newly-collected data is processed and then input into the trained network, the final prediction probability of each energy-saving grade is output, and the grade with the maximum probability is taken as a prediction result. According to the method, the collected data is preprocessed and then input into the network, so that the defects of inaccurate data processing, unreasonable feature selection and the like can be overcome, and the accuracy of a prediction result is improved.

The present disclosure describes computer-implemented methods and systems for automating application processing with dynamic data collection and augmentation related to applicants' behavior. The method includes receiving a plurality of applications with corresponding information, then aggregating, storing, and preprocessing data related to the applicants' behavior. The method also includes a machine learning model including a training dataset, a feature selection module, a hyperparameter tuning module, and a prediction model. The method includes predicting the application outcome based on the application information and behavior data and generating an underwriting decision based on the prediction. The method further includes providing underwritten applications for display and receiving a selection of a subset of the underwritten applications.

The invention relates to a crop drought degree prediction method based on unmanned aerial vehicle remote sensing monitoring. The method comprises the following steps: S1, data collection: collecting a multispectral image and a thermal infrared image of a farmland in real time through a remote sensing sensor; s2, image preprocessing: carrying out preprocessing operation on the collected multispectral image and thermal infrared image; s3, class specific feature selection: dividing the farmland into different classes according to the types, growth stages and expected drought degree grades of the crops, decomposing a multi-class classification problem into a plurality of dichotomy problems, and constructing a deep learning model for feature learning and importance evaluation for each dichotomy problem to obtain a class specific feature selection result; a class specific feature set for each class is formed, and class specific features for different classes are fused to form a comprehensive feature set; s4, model construction and training: constructing a drought degree prediction model according to the comprehensive feature set; and S5, drought degree prediction: realizing real-time monitoring and prediction of drought according to the real-time data and the prediction model.

The invention relates to an epileptic seizure detection method and system based on self-attention mechanism and GRU-LSTM fusion, and the method comprises the following steps: (1) carrying out the preprocessing of original electroencephalogram signal data, and extracting time domain and nonlinear features; (2) randomly dividing the data set into a training set, a test set and a verification set; (3) constructing a deep learning model fusing a self-attention mechanism, a gating circulation unit and a long short-term memory network; (4) extracting fusion features from the trained deep learning model, and inputting the fusion features into a support vector machine classifier to perform epileptic seizure and non-seizure classification; and (5) outputting a classification result through multi-modal feature fusion, long and short term dependence modeling and adaptive feature selection. The method has the advantages that a self-attention mechanism, a gating circulation unit, a long-short-term memory network and a support vector machine classifier are combined, deep features of electroencephalogram signals are extracted through a multi-stage processing flow, and finally epileptic seizure and non-seizure classification is carried out.

The invention relates to the technical field of machine learning, and discloses a decision tree data model establishment method, which comprises the following steps of: acquiring heterogeneous data sources such as a structured data table, a time sequence data stream and graph structure data through distributed nodes, sampling the time sequence data stream by using a dynamic sliding window, and vectorizing the graph structure data through a graph embedding algorithm; a multi-stage feature selection model is constructed to screen features, and a dynamic decision tree generation framework adopting an adaptive splitting criterion is established based on the features. A tree structure is adjusted by applying a multi-objective optimization algorithm, and the performance is improved by introducing an incremental pruning mechanism. And the online model updating module monitors data distribution change, reconstructs a local sub-tree in good time, and injects noise to protect data privacy in combination with a differential privacy protection mechanism. According to the method, heterogeneous data is effectively processed, the model classification precision is improved, the complexity is reduced, the generalization ability is enhanced, the model can be updated online, and the data privacy is protected. The electronic equipment calls related instructions to execute the method, and efficient data processing and analysis can be achieved.

The invention relates to a multi-modal understanding optimization method based on fine-grained feature extraction and global information integration. The method comprises the following steps: segmenting an input image into a plurality of local image blocks, extracting local fine-grained visual features and global visual features, and interacting the local fine-grained visual features and the global visual features based on an attention mechanism to obtain local context features; secondly, performing feature fusion on the local fine-grained visual features to generate fused visual features; mapping the fused visual features to a semantic space which is the same as the text features of the large language model to obtain projected visual features, and dynamically screening through attention weight based on the text features to obtain key visual features; and fusing the key visual features with the text features to generate joint feature representation, and inputting the joint feature representation into a large language model to generate a semantic analysis result. According to the method, global information and dynamic feature selection are introduced, so that the ability of the model to understand multi-modal content in a complex scene can be improved, and the model calculation overhead is reduced.

The invention relates to the technical field of wind power equipment monitoring, and discloses a wind power gear box intelligent fault early warning method and system based on machine learning. The method comprises the steps that multi-source monitoring data such as vibration signals, temperature data and oil analysis data of the wind power gear box are acquired, and multi-scale operation characteristics are extracted through time-frequency conjoint analysis; key fault sensitive features are determined through an adaptive feature selection algorithm, and a dynamic fault feature weight matrix is constructed in combination with a historical fault case library; multi-modal data fusion is adopted to generate an enhanced fault feature set, and modal decomposition is carried out on the enhanced fault feature set to obtain a trend component and a fluctuation component; a fault evolution feature space is constructed by using a deep neural network based on two components, then a fault development mode is identified by using a time sequence mode matching algorithm, and finally a graded early warning signal is generated according to a matching degree with a preset mode, so that fault features can be comprehensively captured, and safe operation of a wind power gear box is ensured.

The invention relates to the technical field of power data fusion, in particular to a data fusion mining method and system based on multi-modal power cross-domain. The method comprises the following steps: acquiring a multi-source heterogeneous power data set; performing tensor decomposition on the multi-source heterogeneous power data set to obtain a power core feature tensor set; performing feature selection and reconstruction on the power core feature tensor set to obtain a reconstructed power feature data set; performing domain adaptive feature mapping on the reconstructed power feature data set to obtain a cross-domain power feature mapping data set; performing domain difference elimination on the cross-domain power feature mapping data set to obtain a domain alignment power feature data set; performing attention weight calculation on the domain alignment power feature data set to obtain a cross-domain fusion power feature data set; according to the method, the utilization efficiency of multi-source data in the power system can be remarkably improved.

The invention discloses a CNN and Mama-based remote sensing image semantic segmentation method and system, and belongs to the technical field of remote sensing image semantic segmentation, and the method comprises the steps: carrying out the preprocessing of an obtained remote sensing image, and obtaining a plurality of image blocks; and carrying out semantic segmentation on the image blocks by adopting a remote sensing image semantic segmentation model to obtain a pixel-level classification result. Wherein the remote sensing image semantic segmentation model comprises an encoder, a multi-scale residual space pyramid pooling module and a decoder; the encoder adopts a ResNet network to extract multi-level features; the multi-scale residual spatial pyramid pooling module integrates different scale features from the CNN encoder; the decoder adopts a CSMamba block to capture long-range dependency, and performs feature selection in combination with a channel and a space attention mechanism. And a multi-output supervision module is combined on the output after the encoder characteristics are fused in each stage of the decoder, and the decoder is supervised to gradually generate a semantic segmentation map of the remote sensing image.

The invention provides a depression state assessment method and device based on heart rate variability characteristics and a parallel neural network. Real-time electrocardiosignals are collected in real time through wearable equipment and transmitted to a mobile terminal, the mobile terminal transmits the electrocardiosignals to a cloud server, and work of signal preprocessing, heart rate variability (HRV) feature extraction, feature selection and data enhancement and depth model construction and optimization is carried out on the cloud server. And evaluating the depression state of the subject. The cloud outputs an evaluation result to a display screen of the mobile terminal to be displayed, the depression state of the subject is finally displayed, and the four evaluated depression states are healthy, mild depression, moderate depression and severe depression; in addition, the terminal also supports depression state historical record query and key HRV feature tracing.

The invention relates to the technical field of brain-computer interfaces, in particular to a multi-mode AI glasses vision-electroencephalogram cooperative control method, device and equipment. The method comprises the following steps: constructing a multi-modal data acquisition framework for acquiring a visual scene image and an electroencephalogram signal, generating an alignment data stream to identify a target feature, and generating an attention map; time sequence features are extracted, and intention features are recognized; constructing a feature matrix, performing modal alignment on the feature matrix, performing feature fusion based on an alignment result of modal alignment, and generating a unified representation; performing time sequence segmentation, extracting associated features, constructing a state sequence based on the associated features, marking conversion nodes by using the state sequence, and generating a dynamic mode according to the conversion nodes; performing information analysis by using a dynamic mode, determining a modal weight, performing feature selection based on the modal weight, performing classification mapping on a feature selection result, and generating a control sequence; and generating an interaction instruction according to the control sequence, and completing vision-electroencephalogram cooperative control of the multi-modal AI glasses.

The invention discloses a battery health diagnosis and analysis method based on big data. The method comprises the following steps of performing real-time monitoring and data acquisition through various sensors and data sources; performing multi-source data fusion based on feature selection of the multi-source data to obtain a key feature set; creating an adaptive deep learning model, and inputting features in the key feature set into the adaptive deep learning model to obtain a battery health state prediction result; generating a battery health score based on the battery health state prediction result and the real-time data of the battery; based on the real-time battery health score, an early warning threshold value is adjusted, a dynamic alarm is generated, and corresponding maintenance suggestions and early warning measures are provided according to different health scores; and continuously optimizing the adaptive deep learning model according to historical battery health data and a model prediction result. According to the method, the data weight is dynamically adjusted based on the data quality, the sensor precision and the influence degree of the sensor precision on health prediction, and more accurate battery health state prediction can be realized.

The invention discloses a method and a system for monitoring the reliability of a photovoltaic converter in a special plateau environment. The method comprises the following steps: firstly, acquiring electrical quantity, temperature quantity, environment quantity and operation quantity, filtering abnormal data, and realizing accurate alignment of multi-frequency signals in combination with a dynamic time warping algorithm; secondly, constructing a plateau sensitive feature set, and optimizing feature quality through physical consistency check and three-stage feature selection; then, an IGBT thermal fatigue equation and a capacitance aging equation are fused to establish a health index evolution model, a Bayesian physical information neural network is used for prediction, and a high-reliability confidence interval is output through Monte Carlo sampling. And finally, dynamically correcting the residual life based on the comprehensive environment factor, and triggering a hierarchical maintenance decision according to the health index state, the residual life and the confidence interval width. The service life prediction precision of the photovoltaic converter in the plateau environment is remarkably improved, the operation and maintenance cost is effectively reduced, and the equipment operation reliability is enhanced.

The invention discloses an intelligent construction site safety evaluation method and system based on data elements, and relates to the technical field of computer platform load balancing, and the method comprises the steps: collecting construction site multi-source heterogeneous data, and carrying out the preprocessing of the construction site multi-source heterogeneous data; constructing a safety evaluation model of time sequence modeling and graph structure fusion through combination of a long short-term memory network and a graph neural network, performing feature selection and dimension compression operation on the processed multi-source heterogeneous data of the construction site to obtain an input feature tensor, and inputting the input feature tensor into the safety evaluation model; and calculating a comprehensive score according to a model output result, and generating a classification result and data feedback through an edge node. According to the intelligent construction site safety evaluation method based on the data elements provided by the invention, combined modeling of time sequence dependence and space association is realized through a deep modeling structure fusing LSTM and GNN, and the risk identification precision under multi-source heterogeneous data is effectively improved.

The invention discloses a transformer fault detection device based on a fuzzy logic algorithm, and the device comprises a data collection module which obtains the operation original data of a transformer in real time through combining the dissolved gas in oil with the temperature, vibration, current and voltage; the data preprocessing module is used for carrying out missing value processing, noise removal and abnormal value detection and processing on the original data; the feature extraction module is used for realizing dynamic feature selection based on data analysis provided by the data preprocessing module; the fault identification module is used for carrying out abnormal waveform judgment on current, voltage, temperature and vibration parameters through a threshold calculation unit and carrying out threshold adjustment based on an optimization algorithm; and the fault detection module triggers the alarm unit or maintains a normal working state according to an identification result of the fault identification module. According to the invention, through monitoring analysis and timely alarm notification, accurate and efficient monitoring of the transformer fault is realized, and stable operation and long-term reliability of the transformer are ensured.

The embodiment of the invention provides a barrel service life prediction method and device based on gradient enhancement and a quantile recursive network, and the method comprises the steps: collecting the multi-source sensor data of the whole life cycle of a barrel, and carrying out the preprocessing, and forming a standardized data set; constructing an XGBoost model, performing feature importance analysis on the standardized data set through the XGBoost model, screening key features based on an analysis result to obtain a data set after feature selection, and performing optimization training on the XGBoost model through gradient lifting and regularization methods; dividing the data set after feature selection into a training set and a test set; in combination with a quantile regression method, constructing a life prediction model based on a quantile recurrent neural network, training the life prediction model by using the training set, and training and optimizing model parameters through a bifurcated sequence to obtain an optimized barrel life prediction model; and inputting a test set into the model, outputting a residual service life prediction result containing a prediction value and a confidence interval, and evaluating the prediction performance.

The invention relates to the crossing field of hydraulic engineering hydrological monitoring technology and machine learning prediction technology, discloses a sediment concentration prediction method based on deep learning, and aims to solve the problems that in existing sediment concentration prediction, hyper-parameter manual tuning is low in efficiency, key feature attention is insufficient, local and time sequence information is difficult to consider by a single model and the like. Accurate prediction is realized through five core modules: a data preprocessing module performs missing value filling, abnormal value processing and derivative feature generation on hydrological data; the feature selection module screens key features based on mutual information; the time sequence construction module generates time sequence data through a sliding window; the hyper-parameter automatic optimization module adopts Bayesian optimization iteration to obtain an optimal hyper-parameter; the CNN-LSTM-attention prediction module fuses CNN local feature extraction, bidirectional LSTM time sequence dependence capture and multi-head self-attention mechanism key feature focusing capability, is suitable for scenes such as river channels and channels, and provides efficient decision support for hydrological regulation and control.