284 results about "Concentration prediction" patented technology

The invention discloses a water source chlorophyll a concentration prediction model design method based on machine learning. The method comprises the following steps: acquiring chlorophyll a concentration data in a to-be-predicted region for a continuous period of time; carrying out data preprocessing on the chlorophyll a concentration data, and filtering high-frequency noise by adopting wavelet transform preprocessing; constructing a concentration prediction model, carrying out data preprocessing on chlorophyll a concentration data, and filtering high-frequency noise by adopting wavelet transform preprocessing; constructing different concentration prediction models, and inputting the processed chlorophyll a concentration data and physicochemical parameters into the prediction models to obtain a chlorophyll a concentration data prediction result; and comparing prediction results of different prediction models, and determining the prediction model. According to the prediction model design method, the WT-GRU model is adopted to preprocess the data through wavelet transform, the wavelet transform effectively extracts key time scale characteristics through signal decomposition, and the accuracy of chlorophyll a concentration prediction is remarkably improved.

The invention discloses a spectrum aliasing decoupling and concentration inversion method under the cross influence of multi-source environmental factors, belongs to the field of industrial process control and environment monitoring, and constructs an environment-spectrum collaborative fusion decoupling model for concentration prediction. The method specifically comprises the following steps: respectively collecting absorption spectrum signals of specified mixed gas at different temperatures, pressures and known concentrations, meanwhile, collecting environmental parameter data, constructing a multi-source data set, and carrying out denoising, dimension reduction and preprocessing on the multi-source data set; constructing a self-supervised feature extraction network for adaptive modulation of environmental parameters to realize deep fusion of spectrum and environmental information; the feature expression capability and generalization performance of the self-supervised feature extraction network are improved by using a self-supervised learning mechanism; and constructing a BPBO-GRNN self-adaptive concentration inversion optimization model for realizing inversion of mixed gas concentration and self-adaptive optimization of model parameters. According to the invention, high-precision concentration inversion and stable detection of the aliasing gas can be realized.

The invention belongs to the technical field of traffic pollution prediction, and particularly relates to a LightGBM-based multi-source data fusion pollutant concentration prediction method, which comprises the following steps of: 1, integrating and processing multi-source data, and constructing a plurality of feature sets; and 2, carrying out feature analysis on meteorological, pollutant and traffic index variables through correlation analysis and time sequence analysis, and providing support for modeling and model interpretation. And step 3, dividing a training test set according to a time sequence, and training and optimizing the LightGBM pollution prediction model through parameter tuning by taking minimization of RMSE as a target. And 4, verifying the performance of the model from prediction precision, spatial distribution and wind direction influence. And 5, analyzing and displaying the key driving factor and quantifying the contribution of the key driving factor. According to the invention, through feature engineering processing and model optimization of heterogeneous data such as weather and traffic, high-precision prediction and influence factor analysis of lane-level pollutant concentration are realized.

The invention provides an intelligent isotope gas concentration regulation and control method based on concentration prediction, which comprises the following steps: collecting multi-dimensional data in a closed box subjected to emptying treatment to construct a standardized control-response data sequence; predicting the isotope gas concentration through a neural network model to obtain the predicted isotope gas concentration and a long-term concentration drift trend estimation item; inputting the current standardized control-response data sequence into Transform to obtain current state representation, and generating a control instruction of valve opening and closing time and sucking pump power on the basis of the current state representation in combination with predicted concentration and target concentration; executing the control instruction, comparing the actual concentration with the predicted concentration, and coding the deviation into a deviation risk code through a multi-layer perceptron in combination with the control instruction; receiving the current state representation and the deviation risk code to establish a state-deviation mapping model, fitting deviation distribution parameters of the control deviation, and accumulating control experience to optimize a long-term strategy.

The invention discloses an SNCR outlet NOx concentration prediction method and system based on a random forest, and the method comprises the steps: S1, obtaining the operation data of an SNCR system, selecting a plurality of input variables affecting the outlet NOx concentration, and carrying out the normalization processing of the input variables; s2, carrying out maximum information coefficient analysis on the normalized data set, determining the delay time of each input variable relative to the NOx concentration of the outlet, and reconstructing the data set according to the delay time; s3, dividing the reconstructed data set into a training set and a test set, constructing a random forest model, and initializing model parameters; s4, optimizing parameters of the random forest model to obtain an optimal parameter combination; and S5, predicting by using the random forest model of the optimal parameter combination to obtain a predicted value of the NOx concentration at the outlet of the SNCR system. The method has the advantages of high prediction precision and the like.

The invention relates to the technical field of pollutant monitoring, in particular to an industrial solid waste blending combustion pollutant intelligent monitoring system which comprises a data fusion module, a diffusion judgment module, a concentration prediction module, a risk monitoring module and an early warning distribution module, and the trend of particulate matter and gas monitoring data is analyzed based on a blending combustion reaction furnace outlet and a smoke exhaust pipeline section. According to the invention, through combined processing of multi-point time sequence monitoring data, various pollutants and environmental parameters are collected in real time, precise spatial partitioning of a monitoring object, coupling analysis based on meteorological factors and spatial diffusion characteristics, dynamic identification of response differences of pollutants in each area of a factory, and through extraction of a concentration change interval and a risk trend signal, a real-time monitoring result is obtained. According to the method, emission risk types are subdivided, graded intervention information is generated, the response speed and the judgment capability of abnormal states of pollutants under complex environment disturbance are improved, dynamic adaptation and detailed management and control of an intelligent monitoring system are promoted, and high-reliability decision making of emission control in various scenes of a plant area is further supported.

The invention provides a water chlorophyll a concentration inversion method and system based on multi-modal data and a lightweight model, and relates to the technical field of water environment remote sensing evaluation. The method comprises the following steps: firstly, acquiring a Gaofeng No.5 satellite remote sensing image, a sentinel No.3 satellite image and ground actual measurement data, and completing image preprocessing and water body pixel extraction; constructing a hyperspectral index and an aquatic vegetation index, and fusing the hyperspectral index and the aquatic vegetation index with the water body temperature, the pH environmental factors and the spectral reflectivity to form a multi-dimensional feature sample set; a core feature subset is obtained through random forest and XGBoost coupling feature selection, and a lightweight student model is trained based on knowledge distillation; and constructing a to-be-predicted feature sample for the to-be-predicted time phase image and the environment factor, inputting the to-be-predicted feature sample into the lightweight student model to obtain a chlorophyll a concentration predicted value, and generating a spatial distribution map and a quality control map layer. According to the invention, high-precision, low-redundancy and efficient deployment chlorophyll a concentration inversion is realized.

The invention discloses a greenhouse gas concentration time sequence prediction method based on a sudden change perception attention mechanism. The method comprises the steps of data preprocessing, sudden change intensity sequence construction with boundary processing, time sequence feature coding, sudden change perception attention weight calculation, context vector generation and concentration prediction, model training and optimization and model prediction. The method aims to solve the problem that a standard deep learning model is slow in sensing and lagged in prediction for a sudden change event in a concentration sequence, and finally realizes high-precision prediction for future concentration change, especially a sudden concentration peak value by endowing the model with the capability of actively identifying and reinforcing the learning of a historical sudden change mode. The urgent demand for early warning of abnormal emission in practical application is met. The method is particularly suitable for processing foundation observation data with small resolution and even higher resolution, has the core value of improving the prediction capability of concentration dramatic change driven by sudden emission events, and can be widely applied to key scenes such as accurate carbon emission monitoring, environmental pollution early warning and climate model simulation.

The invention provides a water chlorophyll a concentration inversion method based on a hyperspectral remote sensing image, which relates to the technical field of remote sensing inversion and comprises the steps of image preprocessing, water pixel extraction, actually measured sample space registration, spectral feature construction, model score optimization, concentration prediction and the like. According to the method, an optimal scheme is screened in a feature-model combination through cross validation and a unified scoring function, the optimal scheme is applied to whole image calculation, and a chlorophyll a concentration spatial distribution map and a matched quality control map layer are output. The method has the characteristics of high precision, self-adaption and high engineering practicability, and is suitable for a water quality remote sensing inversion scene driven by multi-source hyperspectral data.

The invention discloses an LIBS rare earth multi-element quantitative detection method based on a deep learning model, and the method comprises the steps: constructing a multi-scale feature fusion deep learning architecture MSFF-Net, carrying out the weighted fusion of local features extracted by a 1D-CNN and long-range dependence features of BiLSTM modeling through a channel attention mechanism, and enhancing weak signal features in combination with a first-order derivative spectrum. Meanwhile, a synthetic data enhancement strategy based on a physical model is adopted to simulate a matrix effect and a spectral line overlapping scene, and spectral feature migration under different matrixes is achieved in cooperation with a matrix self-adaptive migration learning strategy. And after data acquisition and preprocessing, a trace element concentration prediction value is synchronously output through a multi-task learning framework. According to the method, the detection precision and the anti-interference capability of the LIBS technology on trace elements are effectively improved, and the method has remarkable application value in the industrial fields of mineral resource exploration, strategic metal recovery and the like.

The invention discloses a fermentation strain yield prediction method based on artificial intelligence, particularly relates to the field of biological fermentation production, and is used for solving the problems that in the prior art, due to the fact that the yield is difficult to directly map in concentration prediction, monitoring indexes are normal, the final yield deviates from expectation, and reliable strain comparison and production management decision cannot be supported. A historical batch process observation sequence and a material account sequence are aligned to generate a conversion efficiency characterization sequence to form a yield learning sample training yield prediction model, and the conversion efficiency characterization sequence is generated on batches and input into the model to obtain a yield prediction value. And calculating an accounting volume trajectory consistency parameter and a conversion contribution steady-state proportion parameter, inputting the parameters into a coefficient analysis model to obtain a decision credible coefficient, generating a yield risk judgment result for a yield prediction value based on the decision credible coefficient, and realizing a new-batch online prediction record.

The invention discloses a PM2.5 concentration prediction method and device, a medium and a product, and relates to the technical field of atmospheric environment monitoring and deep learning prediction.The method comprises the steps that laser radar vertical profile data containing extinction coefficients and depolarization ratio vertical profiles and a scalar time series data sequence are obtained; converting the laser radar vertical profile data sequence into a vertical sequence containing position information at any moment; adopting a multi-channel spatial feature extraction model, and obtaining a multi-channel spatial feature vector based on the vertical sequence; and after the scalar time sequence data sequence and the multi-channel spatial feature vector are spliced to obtain a fused feature vector, a PM2.5 concentration prediction value is obtained by adopting a time sequence prediction model. According to the method, the technical pain point caused by the fact that aerosol types cannot be distinguished and'high altitude-ground 'cross-layer physical association cannot be captured in the prior art is solved, so that accurate and robust prediction of the PM2.5 concentration on the ground is realized under complex weather conditions such as sand and dust transportation and boundary layer evolution.

The invention discloses a microfluidic nitration reaction platform based on in-situ Raman and a reaction optimization method. The platform integrates a microreactor, an in-situ Raman detection module, a spectrogram processing and quantifying module, an experimental optimization module and a central control module, and is suitable for monitoring and optimizing complex reactions under a non-mutual solution phase plunger flow system. According to the method, an indirect hard modeling (IHM) method is used for processing an overlapped Raman spectrogram, a standard curve is combined for establishing a multi-component concentration prediction model, and high-precision quantification of reactants and products is realized. According to the platform, a Gaussian process model is combined with dynamics prior, a multi-target Bayesian optimization process based on expected hypervolume improvement (EHVI) is driven, and Pareto optimal conditions in a temperature and residence time space are effectively explored. The system has the functions of automatic focusing sampling, spectrogram real-time decoupling, concentration inversion and experimental condition decision making, and is suitable for intelligent reaction development and optimization in the fields of fine chemical engineering, high-throughput reaction screening and the like.

The invention discloses a soil and groundwater pollution diffusion prediction method and system based on a time-space diagram neural network, and relates to the crossing field of environmental information modeling and artificial intelligence, and the prediction method comprises the following steps: S1, obtaining multi-source soil and groundwater monitoring data and external driving data; s2, generating a static hydraulic communication matrix and a dynamic convection dominant matrix, and constructing a multi-relation dynamic graph structure; s3, performing parallel processing on graph convolution operation of the static hydraulic connection matrix and the dynamic convection dominant adjacency matrix, capturing time dynamic characteristics, and obtaining a pollutant concentration prediction result; and S4, outputting structured prediction data and a visual product. By constructing a multi-relation dynamic graph structure fusing a hydrogeological physical mechanism and a dynamic flow field response, distortion of a pollution plume leading edge deduction trajectory caused by misjudgment of a spatial dependency relation can be avoided, so that the prediction precision of a non-uniform anisotropic pollution diffusion path is improved.

The invention provides a sediment concentration prediction method and system based on a freeze-thaw coupling physical constraint model, and relates to the technical field of sediment concentration prediction, and the method comprises the steps: obtaining hydrological data, meteorological data and various original data of soil and vegetation, and carrying out the preprocessing; constructing a benchmark sediment concentration physical model based on the multiple kinds of preprocessed original data, and constructing an input feature vector; inputting the input feature vector into a sediment concentration prediction ANN model, and outputting to obtain the sediment concentration; wherein in the training process of the sediment concentration prediction ANN model, reference sediment prediction and sediment mass conservation constraint of the physical model are introduced into an ANN model loss function, the loss function is enhanced by adopting a freeze-thaw weighting strategy, and the sediment concentration prediction ANN model in which the physical model and the ANN model are deeply fused is obtained. According to the invention, the stability and accuracy of prediction during extreme rainfall, runoff surge or frequent freezing and thawing are improved.

The invention relates to a gas concentration prediction method based on irregular sampling data multi-scale feature fusion, and belongs to the technical field of coal mine gas concentration prediction. The method comprises the following steps: S1, data preparation: carrying out data sequence division according to shift time set by a coal mine; s2, continuous time coding and feature fusion: performing potential space mapping on the divided sequence by using Shenchang differential; dividing multiple time windows, and performing multi-scale data feature extraction and fusion; s3, shift self-adaptive design: adopting a hybrid expert structure and a gating network to perform self-adaptive processing on coal mine shift changes, and outputting weighting and features by an expert layer; s4, failure fault-tolerant design: performing failure fault-tolerant processing, and compensating a data missing problem caused by common sensor abnormality to obtain reconstruction features; and S5, splicing the expert layer output features with the reconstruction features, inputting the spliced features into a prediction network, and predicting the gas concentration at the next moment. The accuracy of toxic and harmful gas can be remarkably improved.

The invention relates to the technical field of spectral analysis and biomedical detection, and discloses an SERS (Surface Enhanced Raman Scattering) spectrum quantitative detection method and system based on interpretable stacked ensemble learning. The method comprises the following steps: acquiring SERS spectral data of serum tumor marker standard substances with different concentration gradients; performing baseline correction and normalization preprocessing on the data, performing sparse feature selection by using an LASSO algorithm, and screening out key spectral features to construct a sample data set; constructing an interpretable stacking integration model, wherein the model adopts a base learner layer and a meta learner layer; training the model by using the training set, optimizing model hyper-parameters by using a cross validation strategy, and establishing a mapping relationship between spectral features and tumor marker concentrations; and collecting SERS spectral data of a to-be-detected serum sample, extracting key spectral features, inputting the key spectral features into the trained interpretable stacked integrated model, and outputting a concentration predicted value of the tumor marker in the to-be-detected serum sample. The method has the advantages of high precision, universality and molecular level interpretability.

The invention relates to a PM2.5 concentration prediction method based on RF-RFE and TCN-BiLSTM-Attention, and belongs to the technical field of atmospheric environment monitoring. According to the method, firstly, PM2.5 concentration monitoring data and data of related influence factors such as weather, other pollutants and time characteristics are preprocessed, recursive feature elimination (RFE) is combined with random forest (RF) for feature screening, and a time sequence input sequence is constructed; extracting local time sequence features of the sequence through a time convolutional network (TCN), and capturing bidirectional long-range time sequence dependence features of the sequence by using a bidirectional long-short-term memory network (BiLSTM); an attention mechanism is introduced to endow a BiLSTM output feature vector sequence with a differentiation weight, and key time sequence feature contribution is highlighted; and finally, outputting a PM2.5 concentration predicted value through a full connection layer, and completing model training and optimization based on the training set and the verification set. According to the method, the advantages of TCN, BiLSTM and Attention mechanisms are combined, the precision and stability of PM2.5 prediction are improved, and the method can be widely applied to atmospheric pollution early warning and environmental governance decision making.

The embodiment of the invention provides a dual-analyte concentration deep learning calculation method and system based on heterogeneous sensor data fusion, and relates to the technical field of analyte concentration detection. The method comprises the following steps: acquiring an image sequence acquired by an optical sensor and a time sequence signal acquired by an electrochemical sensor, and preprocessing the image sequence and the time sequence signal; extracting space-time correlation characteristics of the image sequence and dynamic time sequence characteristics of the time sequence signal; performing cross-modal feature fusion on the space-time correlation features and the dynamic time sequence features to obtain cross-modal fusion features; after the cross-modal fusion features are input into a shared encoder for encoding, concentration predicted values and uncertainty parameters of the two to-be-analyzed objects are output; performing Bayesian uncertainty quantization through Monte Carlo sampling, and outputting a concentration mean value and a confidence interval; and constructing a multi-objective function, and realizing end-to-end training optimization through the multi-objective function. The key technical problems of heterogeneous sensor data fusion, analyte cross interference and the like can be effectively solved.

The invention discloses a closed-loop control system and method based on deep-sea ore raising pipe ore pulp concentration monitoring, and the system comprises a water surface support platform control center which is used for outputting a control strategy; the hydraulic ore injector is used for receiving the control strategy and carrying out self-adaptive adjustment according to the control strategy; wherein the water surface support platform control center comprises an intelligent control unit and a controller, and the intelligent control unit is used for collecting operation parameter data of the hydraulic ore injector, processing the operation parameter data and predicting in real time to obtain a solid phase volume concentration predicted value; the controller is used for receiving the solid phase volume concentration predicted value and determining a control strategy according to the solid phase volume concentration predicted value; the intelligent control unit comprises a data acquisition module, a data processing module and a prediction module; the concentration of ore pulp in the pipeline can be more accurately controlled through the system, and the system can better adapt to complex working conditions such as submarine topography and ore supply quantity fluctuation.

The invention belongs to the technical field of artificial intelligence and environment monitoring, and particularly relates to an air pollutant concentration prediction method based on an ICLU-CWGAN model. The method comprises the following steps: acquiring and preprocessing data such as atmospheric pollutant concentration and weather, fusing an Inception multi-scale convolution thought with a ConvLSTM network, and constructing a multi-scale ConvLSTM network; the method comprises the following steps: under a Wasserstein generative adversarial network framework with a gradient penalty condition, taking a multi-scale ConvLSTM as a generator, taking an improved U-Net as a discriminator, and constructing an ICLU-CWGAN model; and completing model training by using the preprocessed training grid data, and predicting the atmospheric pollutant concentration. According to the method, the problems of weak long-time-sequence complex space-time correlation modeling capability, spatial distortion and unstable training in the prior art can be solved, so that the precision and the stability of atmospheric pollutant concentration space-time distribution prediction are improved.

The invention discloses a physical constraint-based multi-mode indoor carbon dioxide concentration prediction method, which comprises the following steps of: acquiring time sequence environment data of indoor carbon dioxide concentration, temperature and humidity, and obtaining indoor panoramic image data; performing time sequence feature extraction on the time sequence environment data, and extracting image high-level visual features through a pre-trained image model; in the multi-modal fusion process, a conflict decoupling gating fusion network is introduced to detect and suppress multi-modal time conflicts between static visual information and dynamic environment parameter changes, and contribution weights of an environment modal and a visual modal are dynamically adjusted through a self-adaptive gating mechanism; a time-delay perception cross-modal attention network is introduced to predict and model a physical delay characteristic between a visual event and an environmental parameter response, and cross-modal causal alignment is realized through lag distribution prediction and a weighting mechanism; and inputting the multi-modal features after conflict decoupling and time-lag alignment into a prediction network, and outputting a future indoor carbon dioxide concentration prediction result.

The invention discloses a sodium aluminate solution concentration prediction method and system based on dynamic weight, a processor and a storage medium, and belongs to the field of alumina production, S1, industrial process parameters are input into an improved Transform-based data-driven sodium aluminate solution concentration prediction sub-model, and a preliminary concentration prediction value is output; s2, outputting an error compensation value by using the sequence error prediction sub-model, and taking the error compensation value as a correction value of the initial concentration prediction value to obtain a corrected concentration prediction value; s3, utilizing the mechanism prediction sub-model to output a concentration prediction value under the physical constraint; and S4, carrying out dynamic weight fusion on the corrected concentration predicted value and the concentration predicted value under the physical constraint to obtain a final concentration predicted value. By the adoption of the sodium aluminate solution concentration prediction method and system based on the dynamic weight, the processor and the storage medium, high precision, high robustness and working condition adaptability of sodium aluminate solution concentration prediction are achieved.

The invention provides a mine dust concentration prediction and analysis method based on big data, and the method comprises the steps: building a real-time monitoring network of multi-working-face operation in a mine, obtaining the distribution data of dust clouds of each working face and tunnel concentration distribution information, carrying out the data collection according to a dust superposition effect, and obtaining the initial dust diffusion boundary of each region; according to the initial dust diffusion boundary data, the dust concentration superposition coefficient, the interaction influence radius and the diffusion rate of the dust cloud cluster during multi-working-face operation are analyzed, and the diffusion boundary change trend and the key influence area are determined; adjusting a resource allocation optimization scheme of each region according to a real-time change trend of roadway concentration distribution and a control instruction set, and determining operation parameters and distribution positions of dust control equipment; according to the updated concentration distribution state, whether an area with the dust superimposed effect exceeding the standard exists or not is recognized, if yes, control resources of ventilation equipment, dust removal equipment and a spraying device are redistributed, and a new control instruction set is determined.

A water pollutant detection method based on absorption-electrochemical characteristic fusion belongs to the technical field of environmental monitoring, and comprises the following steps: respectively acquiring absorption spectrum data and current-potential data through a spectrometer and an electrochemical workstation, highlighting key peak position characteristics by using an attention mechanism, and inhibiting redundancy and noise interference; extracting local and global features of the two types of signals by using a convolutional neural network, and splicing the local and global features to obtain joint representation; and cross-modal feature fusion is realized through a dynamic routing mechanism of the capsule network, and a spatial relationship and a hierarchical structure between features are captured. And finally, constructing a classification and regression double-branch output structure to realize pollutant type identification and concentration prediction. Compared with an existing single detection technology, the detection sensitivity of the low-concentration pollutants can be remarkably improved, the distinguishing problem of peak value overlapping is effectively solved, the selectivity and robustness of the system are enhanced, and therefore high-precision integrated detection of the water pollutants is achieved.

The invention discloses a sorghum canopy nitrogen concentration prediction method and device, a medium and a product, and relates to the technical field of nitrogen concentration prediction, and the sorghum canopy nitrogen concentration prediction method comprises the steps: obtaining a target multispectral image; the target multispectral image is a multispectral image comprising the canopy of the sorghum to be predicted in the current period; based on the target multispectral image, determining a background-removed multispectral orthographic image of the sorghum to be predicted; determining a plurality of optimal remote sensing variables and a plurality of optimal texture features of the to-be-predicted sorghum based on the background-removed multispectral orthographic image of the to-be-predicted sorghum; and inputting the plurality of optimal remote sensing variables and the plurality of optimal texture features of the to-be-predicted sorghum into the nitrogen concentration prediction model to obtain a predicted value of the canopy nitrogen concentration of the to-be-predicted sorghum in the current period. The method can be used for predicting the nitrogen concentration of the sorghum canopy in real time.

The invention discloses an odor concentration prediction method and system based on graph neural network and Monte Carlo coupling, and the method comprises the steps: obtaining the original data of the historical wind direction and wind speed of a meteorological station, carrying out the data preprocessing through a self-adaptive Monte Carlo algorithm based on Sobol global sensitivity, and obtaining a working condition directory data set; performing wind direction grouping on the working condition directory data set to obtain standardized meteorological data; based on the standardized meteorological data, constructing a graph data basic structure, obtaining two-dimensional node-level features, and determining species concentration; constructing an odor concentration prediction neural network model; and based on the odor concentration prediction neural network model, predicting the concentration of the species, and outputting an odor concentration prediction result. By considering the physical conservation of the vertical flux and the concentration, high-precision and real-time prediction of peculiar smell diffusion is realized. The peculiar smell concentration prediction method and system based on the graph neural network and Monte Carlo coupling can be widely applied to the technical field of peculiar smell concentration prediction.

The invention discloses a PM2.5 depth prediction method and system fused with a physical mechanism, and belongs to the technical field of PM2.5 concentration prediction.The method comprises the steps that spatial dynamic characteristics of expected changes of pollutants in space are extracted through a differential equation network based on a GCN; time dynamic features are extracted through a bidirectional time perception Mama model; according to the extracted spatial dynamic features and the time dynamic features, a feature fusion module is used for fusion to obtain a spatial-temporal feature fusion model; training a spatio-temporal feature fusion model by using the training set of the PM2.5 historical concentration; and performing prediction by adopting the trained spatial-temporal feature fusion model to obtain the PM2.5 concentration in the target area. According to the invention, the PM2.5 prediction precision can be improved.

The invention discloses a river hydrodynamic force and pollutant migration collaborative prediction method based on a coupling model. The method comprises the following steps: step 1, constructing an MIKE21 hydrodynamic field of a river channel; step 2, constructing a pollutant migration model; step 3, an LSTM-MIKE 21 coupling model is constructed; and 4, carrying out real-time prediction on a full-basin pollutant concentration field through the coupling model. According to the method, the physical mechanism model MIKE 21 and the data driving model LSTM21 are bidirectionally coupled, so that the precision and reliability of pollutant migration simulation are remarkably improved, and collaborative optimization solution of a hydrodynamic field and a pollutant concentration field is realized. Wherein an LSTM21 residual error correction mechanism effectively compensates a complex process which is not covered by a pure physical model, so that the prediction error of the peak concentration of pollutants is changed from gt to gt in a traditional method; 20% is reduced to lt; meanwhile, the conformity of mass conservation and non-negative concentration reaches 99.5% or above.