70 results about "Multivariate statistical" patented technology

The invention discloses a metabonomics data batch correction method based on multi-kernel learning, and belongs to the cross technical field of bioinformatics and analytical chemistry. According to the method, the multi-kernel learning technology is utilized, the advantages of different kernel functions are fused in a self-adaptive mode, a model conforming to data reality is constructed, complex drift characteristics of metabolite signals are accurately captured, and efficient and accurate normalization processing of metabonomics data is achieved. Compared with traditional data standardization methods such as SVR and LOESS, the method has the advantages that the performance is excellent in the aspect of reducing the metabolite peak intensity variability, and the data stability is remarkably improved. In the subsequent multivariate statistical analysis, the classification accuracy is greatly improved, the comparability among different batches of data is also remarkably enhanced, reliable data support can be provided for discovery of disease biomarkers, and the method plays a key role in large-scale metabonomics research.

The invention provides a comprehensive evaluation method for dissemination characteristics of minerals in coal of different particle sizes, and relates to the field of coal quality analysis. The method comprises the following steps: collecting coal analysis samples of different size fractions, and preparing coal polished sections of all the size fractions; the method comprises the following steps: collecting microscopic images, and constructing a representative microscopic image sample database of each size fraction; carrying out size fraction judgment, generating a component segmentation image on a single particle and a component segmentation image on a microscopic image, and extracting dissemination characteristic parameters of mineral substances in coal of each size fraction; key parameters are screened out, and comprehensive evaluation is carried out on the dissemination characteristics of the minerals in the coal of different size fractions. According to the method, an image processing technology and multivariate statistical analysis are combined, and qualitative and quantitative analysis of the dissemination characteristics of the minerals in the coal is achieved. Compared with an existing microscopic image analysis method, the method has the advantages that the human subjective influence and the labor intensity are remarkably reduced, and the analysis result has higher objectivity and reproducibility.

The invention relates to the technical field of electrochemical detection and food quality safety, and particularly discloses an edible oil rancidity degree rapid detection system based on an electrochemical sensor, comprising: a data acquisition module for receiving an original electrochemical response signal and performing format standardization, abnormal data rejection and integrity verification; the data preprocessing module adopts a statistical optimization algorithm to reduce noise, correct baseline drift, generate a unified dimension data set and complete stability verification; the feature analysis module is used for extracting rancidity related feature parameters through statistical correlation analysis, and carrying out fusion calculation after significance verification to obtain feature values representing rancidity; and the evaluation and judgment module inputs the characteristic value into a pre-training model, outputs a quantitative index through multivariate statistical calculation, dynamically sets a safety threshold in combination with the edible oil type, and outputs a rancidity judgment conclusion after comparison. The method depends on statistical data processing and analysis, the detection reliability and accuracy are improved, and the method is suitable for edible oil quality safety rapid screening and quality monitoring.

The invention discloses a method for analyzing influence of climate change and human activity on space-time evolution of water resources. The method comprises the following steps: acquiring meteorological information, hydrological information, land utilization information, soil information, social economic information and water resource information of a to-be-detected area; performing trend analysis on the water resource information to obtain a space-time evolution rule of the water resource information; performing attribution analysis on the time-space evolution rule to obtain influence factors of the time-space evolution of the water resource; obtaining the correlation degree of the influence factors and the space-time evolution rule, and completing the analysis of the climate change and human activity on the space-time evolution of the water resource. According to the method, the influence factors of water resource evolution of the to-be-detected area are researched through the multivariate statistical stepwise regression analysis model, the multicollinearity between the related influence factors of climate change and human activity is reduced through the characteristics of stepwise regression analysis, the calculation accuracy is improved, and the sustainable development of water resources is promoted.

The invention discloses a packaging machinery health state evaluation method and system based on multi-source data fusion, and relates to the technical field of multi-source data fusion, and the method comprises the steps: carrying out the fault root cause positioning according to a feature vector obtained through the decomposition of a main feature value, and obtaining a main evaluation dimension which causes the reduction of a comprehensive health degree index; according to the method, the evaluation dimension mapped by the component with the maximum absolute value in the feature vector corresponding to the main feature value is analyzed, accurate positioning of the fault root cause is achieved, the process effectively strips coupling correlation among dimension data on the basis of the multivariate statistical principle, the problems of misjudgment and missed judgment caused by judgment depending on experience of operation and maintenance personnel in the traditional technology are solved, and the accuracy of the fault root cause is improved. When the comprehensive health degree index of the equipment is reduced, the problem can be quickly locked from electrical, mechanical, process or environment dimensions, the troubleshooting time is greatly shortened, the maintenance response efficiency is improved, and a clear direction is provided for targeted equipment maintenance.

The invention relates to a characteristic marker combination for identifying honeysuckle tea varieties and an identification method thereof, and belongs to the technical field of variety tea identification. The characteristic marker combination comprises chlorogenic acid, swertioside, secoxyloganin, galuteolin, isochlorogenic acid A and isochlorogenic acid C. The invention further discloses a kit for detecting the secoxyloganin. The contents of the characteristic markers are simultaneously detected through a high performance liquid chromatography (HPLC) or liquid chromatography-mass spectrometry (LC-MS) technology, and different honeysuckle tea varieties can be rapidly and accurately identified by combining multivariate statistical analysis (such as principal component analysis or clustering analysis). The identification method provided by the invention has the characteristics of strong specificity, high stability and high sensitivity, is suitable for large-scale sample analysis, and can be widely applied to quality control, authenticity identification and market supervision of honeysuckle tea varieties.

The invention relates to the technical field of medical biology, and discloses a traditional Chinese medicine quality marker screening method and an angelica dahurica variety quality marker group, the traditional Chinese medicine quality marker screening method and the angelica dahurica variety quality marker group comprise the following steps: respectively crushing, extracting and detecting a newly bred angelica dahurica variety and a farm variety; carrying out pretreatment on mass spectrum data by adopting MZmine 2.5 software; and the coumarin components are screened by adopting a multi-stage mass loss window. According to the traditional Chinese medicine quality marker screening method and the angelica dahurica variety quality marker group, through integration of ultra-high performance liquid chromatography-tandem mass spectrometry and polygon quality defect filtering, extended virtual molecular formula library screening, characteristic molecular network, network annotation transmission and other calculation strategies, a plurality of coumarin components are successfully annotated; and multivariate statistical analysis is adopted, so that the quality marker of the newly cultivated angelica dahurica variety is obtained.

The invention relates to the technical field of industrial process soft measurement and quality control, and discloses a paper quality prediction method based on multivariate statistical latent variable fusion and space-time transformation, which comprises the following steps: acquiring space-time sequence data of a multi-source sensor in a papermaking process, constructing a space-time diagram structure reflecting a topological relation of equipment, and preprocessing. Then, multi-view latent variables are extracted through non-negative matrix factorization, independent component analysis and robust principal component analysis, attention fusion is conducted on the latent variables through an LV fusion module, and fusion latent variables are obtained; and inputting the fusion latent variable and original node data into a multi-scale convolution auto-encoder to obtain spatial feature embedding, and inputting the spatial feature embedding and the fusion latent variable into a space-time Transform module together to realize joint modeling of space correlation and time dependence. And finally, outputting a paper quality predicted value through a linear regression module. The method can achieve the accurate prediction of the paper quality under a high-dimensional and multi-noise working condition, and is suitable for online monitoring and modeling updating.

The application discloses a method for identifying original tea tree varieties by using characteristic metabolites, and particularly relates to the field of identifying tea tree varieties, which comprises the steps of sample preparation, metabolite extraction, liquid chromatography-mass spectrometry detection, data correction processing, characteristic metabolite construction and multivariate statistical clustering determination, etc.; in view of the internal metabolic characteristics formed in the long-term natural domestication process of the floating Liang chestnut leaf population, a high-dimensional characteristic metabolite vector is constructed by using the stable numerical distribution mode of the fresh leaves of the floating Liang chestnut on the sixteen characteristic metabolites; after the peak intensity data of each sample is corrected, filtered and structured, the obtained characteristic vector is input into the Euclidean distance hierarchical clustering model, so that the metabolic composition difference between the samples to be identified and the standard floating Liang chestnut leaf population is unsupervised statistically distinguished.

The invention discloses a strong wine authenticity identification method fusing multivariate statistics and metabolic feature extraction, relates to the technical field of strong wine authenticity identification, and overcomes the limitation that a traditional method only depends on peak area or mass-to-charge ratio for distinguishing by incorporating non-dominant structure difference in a metabolic map into an analysis model. A traditional method is difficult to distinguish metabolic differences caused by different processes or environmental changes, but according to the method, through multi-dimensional non-targeted metabolic characteristic analysis, the influence of the processing processes on the metabolic structures of the wine products can be accurately captured, and the distinguishing precision of the wine products of the same variety, different processing batches, the same raw material and different processes is remarkably improved. According to the method disclosed by the invention, the comprehensive isomerism index FCY is constructed by combining the structural isomerism variance and the map complexity factor, so that the difficulty in judgment caused by complex structures and hidden changes of wine samples is effectively solved.

The invention discloses a pyrazine-flavor-based multi-mode intelligent identification method for a Maotai-flavor liquor process. The method comprises the following steps: pretreatment: pretreating Maotai-flavor liquor samples of different brewing processes; collecting sensory data of electronic noses and electronic tongues of Maotai-flavor baijiu of different brewing processes; establishing a liquid phase detection method for the characteristic components of the alkylpyrazine in the Maotai-flavor liquor; verifying a component analysis method through precision, stability and repeatability experiments; component quantitative analysis based on liquid chromatography fluorescence detection; performing clustering and difference evaluation on Maotai-flavor liquor samples of different brewing processes by adopting multivariate statistical analysis; establishing a Maotai-flavor liquor process identification model: utilizing VIPgt in multivariate statistics; establishing different machine learning multi-classification models for different brewing processes of Maotai-flavor liquor according to the common characteristics of 1, and performing multi-model comparison; and based on the optimal performance classification model, establishing an SHAP model of the Maotai-flavor liquor brewing process identification model, and globally explaining the identification model by adopting an SHAP algorithm.

The invention discloses a credit card user marketing business distribution method and system, and the method comprises the steps: carrying out the analysis of the basic data of a credit card user through employing a multivariate statistical analysis method, and obtaining a first target user; classifying the first target users according to the regional information to form a plurality of second target users, and distributing the second target users to corresponding regional banks; determining a marketing mode according to the credit card use information of the second target user; determining a marketing scheme of the second target user according to the credit card consumption information of the second target user; and pushing the corresponding marketing scheme to the second target user according to the marketing mode of the second target user, evaluating the marketing effect, and optimizing the influence strategy according to the evaluation result, so that the method effectively improves the efficiency and accuracy of credit card customer management. Effective marketing can be implemented for stock customers, and the consumption permeability is improved. Banks are helped to realize business development, market competitiveness is enhanced, customer experience is improved, and customer satisfaction and loyalty are improved.

The application designs a satellite communication earth station monitoring protocol design method based on atlas clustering, which is suitable for the satellite communication earth station operation and maintenance control field, the classification processing and monitoring protocol planning and design of the controlled equipment and its parameters. The relationship and mode between the earth station monitoring data are revealed by mainly designing a multivariate statistical analysis and atlas clustering processing scheme of monitoring data, so as to better understand and identify different characteristics and behaviors, and realize the inductive design of the earth station monitoring protocol. The application has the characteristics of clear and easy to understand, good continuity, strong scalability and the like, and can especially solve the problems of non-standard satellite communication earth station monitoring protocol design process, disorder protocol content and poor continuity and the like.

The invention discloses a questionnaire system and method, and relates to the technical field of information processing, and the questionnaire system comprises a questionnaire design module, a questionnaire publishing module, a data collection module and a data analysis module. Wherein the data analysis module adopts a mode of combining descriptive statistical analysis with deep analysis such as correlation analysis, difference test, multivariate statistical analysis and text analysis to realize deep and comprehensive data analysis; the questionnaire method comprises the steps of questionnaire design, questionnaire release, data collection and data analysis, the questionnaire system and method can improve the efficiency and flexibility of questionnaire design, improve the accuracy and efficiency of data collection, guarantee the safety of system use and data privacy, remarkably improve the efficiency and quality of questionnaire work, and reduce the labor intensity of workers. And powerful support is provided for data collection and analysis in various fields.

The invention discloses a method and a system for constructing a multi-source data comprehensive safety evaluation model of an energy storage battery system, and the method comprises the steps: calculating the component score accumulated contribution degree and comprehensive component score of each index parameter by adopting a multivariate statistical analysis method; obtaining the risk score of each index parameter, calculating the cloud probability feature value of each risk level based on a cloud model, and forming a risk cloud probability data set; according to a Transform architecture based on a multi-head attention mechanism, an evaluation model is constructed; adjusting the number of heads in a multi-head attention mechanism according to the accumulated contribution degree of the component scores, and training the model by taking the comprehensive component scores and the risk cloud probability data set as an input data set and an output data set of the evaluation model; optimizing hyper-parameters of the trained evaluation model by adopting a subtraction optimizer algorithm improved based on golden sine by taking F1 score optimization as an optimization target; and the evaluation model after hyper-parameter optimization is used as a multi-source data comprehensive safety evaluation model of the energy storage battery system, so that the accuracy and the reliability of an evaluation result are improved.

The invention belongs to the field of pancreatic duct adenocarcinoma prediction, and provides a pancreatic duct adenocarcinoma prediction method and system based on oral flora. The pancreatic ductal adenocarcinoma prediction method based on oral flora comprises the following steps: acquiring relative abundance information of each biomarker in a potential biomarker set for diagnosing pancreatic ductal adenocarcinoma of a to-be-detected person; using the relative abundance information of each biomarker in the biomarker set and a pre-trained multivariate statistical model to obtain a probability value of suffering from pancreatic ductal adenocarcinoma; the method comprises the following steps: performing microbiome difference analysis on saliva, duodenal juice and pancreatic tissue samples of a pancreatic ductal adenocarcinoma patient group and a pancreatic benign disease patient group, and taking an intersection of differential bacteria in the samples to obtain a potential biomarker set for diagnosing the pancreatic ductal adenocarcinoma.

The embodiment of the invention discloses a multi-time-scale prediction method and system for cyanobacterial bloom, and the method comprises the steps: obtaining a cyanobacterial bloom spatial-temporal distribution prediction result through a pre-constructed hydrodynamic water quality bloom model, and obtaining a cyanobacterial bloom spatial-temporal distribution prediction result through a pre-constructed prediction model based on data driving based on a current chlorophyll a concentration sequence. Obtaining a chlorophyll a concentration prediction value sequence, and finally generating cyanobacterial bloom prediction information of the first time scale. And based on weather forecast data, obtaining a chlorophyll a concentration predicted value of a second time scale through a hydrodynamic water quality and water bloom model. Based on the historical environment monitoring data, through a pre-constructed multivariate statistical regression model, obtaining a cyanobacterial bloom intensity index; based on a historical chlorophyll a concentration monitoring sequence, a cyanobacterial bloom prediction result of a third time scale is finally obtained by analyzing a chlorophyll a concentration periodic change rule; the predicted periods of the first time scale, the second time scale and the third time scale are increased in sequence.

This invention discloses a power quality anomaly tracing method based on multivariate statistical decoupling, comprising: acquiring historical power quality data and relevant data at the time of analysis, and processing the data; identifying overall anomalies based on Hotelling T² control charts; if an overall anomaly is identified, performing anomaly statistical decoupling analysis based on Mason decomposition, and obtaining the order conditional statistics, unconditional statistics, and complete conditional statistics of each node, and determining whether each statistic reaches a significance level; when both the unconditional and complete conditional statistics of a node are significant, the node is the root cause; when the unconditional statistics of a node are not significant but the complete conditional statistics are significant, the node is a relational anomaly; when the unconditional statistics of a node are significant but the complete conditional statistics are not significant, the node is a cascading phenomenon; this invention improves the location efficiency, analysis accuracy, and decision-making pertinence of power quality operation and maintenance in distribution networks.

The application discloses a process quality detection method, system, equipment and medium based on wavelet packet decomposition and T2 control chart, which comprises the following steps: collecting original signals of multiple production characteristics of a chemical process in real time to obtain stable section signals; performing wavelet packet decomposition on the preprocessed stable section signals to obtain multiple sub-band signals; calculating multivariate statistical values of the sub-band signal data one by one and drawing control charts; calculating control limits of T2 statistics of the sub-band signals under normal working conditions; comparing the multivariate statistical values of the real-time sub-band signals with the control limits to determine whether a fault occurs and output a fault point; performing periodic adaptive updating, collecting the chemical process signals confirmed as normal working conditions in a period as a new training sample set at intervals of a preset time period, and repeatedly calculating the control limits of the new training sample set. The application improves the sensitivity and reliability of abnormal detection and the real-time monitoring capability of an industrial site.

The invention discloses a method for identifying native tea tree varieties by utilizing characteristic metabolites, and particularly relates to the field of tea tree variety identification, which comprises the steps of sample preparation, metabolite extraction, liquid phase mass spectrometry detection, data correction processing, characteristic metabolite construction, multivariate statistical clustering judgment and the like. The method comprises the following steps: constructing a high-dimensional characteristic metabolite vector by utilizing a stable numerical value distribution mode of fresh leaves on sixteen characteristic metabolites according to internal metabolic characteristics formed in a long-term natural domestication process of the castanopsis sclerophylla leaf group species; after peak intensity data of each sample is corrected, filtered and structured, obtained feature vectors are input into an Euclidean distance hierarchical clustering model, and unsupervised statistical judgment on the metabolic composition difference between the to-be-identified sample and the standard floating beam leaf group species is realized.

This invention proposes a method and system for attributing and identifying sediment flux evolution in high-altitude and cold mountainous areas, belonging to the field of sediment evolution attribution and identification technology. The method includes: using a joint mutation identification method to detect mutation points from multiple perspectives based on a driving factor dataset; using the Pettitt test to identify grade changes under distribution mutations to determine the year of distribution mutation; using the Mann-Kendall trend test to identify slope changes under trend mutations to determine the year of trend mutation; detecting mutation points based on the difference between the year of distribution mutation and the year of trend mutation, and dividing the evolution stages; and performing attribution analysis on all driving factors based on a three-stage progressive attribution analysis framework. This invention establishes a sediment attribution and identification framework for high-altitude and cold mountainous areas by integrating multivariate statistical learning and path modeling techniques, improving the stability and accuracy of attribution results, and providing decision support for soil and water conservation and regional ecological restoration.

The invention relates to the technical field of traditional Chinese medicine preparation production, in particular to a multivariable intelligent monitoring system for a pill fluidized bed drying process, which comprises a multi-parameter acquisition module, a multivariable data analysis processing module and an abnormity early warning and intelligent control module. Wherein the multi-parameter acquisition module acquires multi-dimensional heterogeneous monitoring data formed by process and material attributes in real time; the multivariable data analysis processing module identifies a parameter coupling relationship through multivariable statistical analysis based on the data to obtain an analysis result; and the abnormity early warning and intelligent control module establishes an abnormity discrimination model according to the coupling abnormity, early warning is carried out when coupling abnormity is detected, and a control strategy is automatically adjusted according to severity. According to the system, multi-parameter collaborative abnormal millisecond-level response can be realized, the steady-state maintenance rate is improved, and the problems of monitoring blind areas and control lag of a traditional method are solved.

The invention discloses a multi-agent cooperative behavior and result quality two-dimensional evaluation method and platform. The invention relates to the field of engineering education evaluation and civil engineering informatization, and provides a multi-subject collaborative behavior and engineering achievement quality two-dimensional evaluation method and platform. According to the method, for the multi-specialty joint graduation design of building-structure-construction and the like, firstly, cross-specialty collaboration process data of students are collected in a BIM collaboration platform, and a collaboration behavior time sequence model of student-role-operation-time-object is constructed; secondly, objective engineering quality indexes such as a structure safety reserve coefficient, a material use amount and a standardization degree are automatically extracted from a final building scheme, a structure scheme and a construction organization design; and then establishing a correlation model between the collaborative behavior characteristics and the result quality indexes by using a multivariate statistics or machine learning method, and calculating a collaborative contribution degree score for each student to realize quantitative evaluation of teams and individuals. The platform can also give an early warning of potential collaboration risks to teachers in real time, and provides an objective basis for joint completion teaching intervention and multi-dimensional score evaluation.

The invention discloses a neonatal early-onset septicemia risk prediction method and system based on maternal factors, and the method comprises the steps: firstly collecting maternal data, including maternal clinical and laboratory data; carrying out variable processing and screening on the maternal data to obtain significant variables related to the risk of neonatal early-onset septicemia; establishing a multivariable statistical prediction model based on the significant variables; and outputting a risk score of the neonatal early-onset septicemia by using the prediction model, wherein the risk score is used for individualized risk assessment. The method has the advantages that multivariate analysis can be integrated to assist clinical decision making, and the EOS risk can be predicted individually.

The application discloses a climate mode multivariate statistical downscaling method, a storage medium and equipment, relates to the technical field of climate mode statistical downscaling, and comprises the following steps: acquiring mode circulation information, measured precipitation and temperature information output by a global climate mode, and dividing the information into a training set and a test set; constructing a climate mode multivariate statistical downscaling initial model, including a circulation information compression module, a precipitation processing module, a temperature processing module, a temperature and precipitation output module; training the circulation information compression module, a new circulation information compression module retaining an encoder after training, and replacing the circulation information compression module of the initial model; using the training set and the test set to train and test the climate mode multivariate statistical downscaling model, inputting mode circulation information to be processed into the model after training and testing, and obtaining down-scaled precipitation and temperature information. The application avoids multiple loading of input data and waste of running time and resources.

The invention discloses a semiconductor valve analysis method and system based on multi-agent cooperation and an attention mechanism, and relates to the technical field of semiconductor equipment test and intelligent analysis, and the method comprises the steps: carrying out the denoising, missing value filling and test condition label matching operation of all-condition test data, and outputting a standardized test data matrix; fusing a multivariate statistical method and a space attention mechanism, and focusing a test data dimension which has obvious influence on the performance of the gas path system; performing dynamic weight distribution on historical data and real-time data of the multiple test items; and other intelligent agents are linked according to user operation feedback to dynamically adjust the attention focusing direction. According to the method, the problem of association fuzziness in traditional test data analysis is solved, and the precision of performance evaluation and prediction of the gas path system is improved.

The invention provides equipment, a method and a system for evaluating prognosis of cervical adenocarcinoma and squamous carcinoma and application of the equipment, the method and the system, and belongs to the technical field of medical evaluation. According to the invention, two biomarkers, namely adenylate kinase 4 and cavernous malformation 2 protein, for evaluating prognosis of cervical adenocarcinoma and squamous carcinoma are obtained through multivariate statistical analysis, difference screening, COX regression analysis and other biosignal analysis methods. Based on the biomarker, evaluation equipment, a computer execution method, a computer system and the like for evaluating prognosis of cervical adenocarcinoma and squamous carcinoma are constructed, and the evaluation equipment, the computer execution method, the computer system and the like are used for accurately judging prognosis conditions of cervical cancer patients.

The invention discloses a salvia miltiorrhiza quality evaluation method based on HPLC-DAD and multivariate analysis, and relates to the technical field of medicine quality evaluation, and the salvia miltiorrhiza quality evaluation method comprises the following steps: extracting effective components from salvia miltiorrhiza medicinal material samples collected from different regions by adopting an extraction solvent, the effective components at least comprise salvianic acid A sodium, rosmarinic acid, salvianolic acid B, cryptotanshinone, tanshinone I and tanshinone IIA, HPLC-DAD is adopted for chromatographic analysis, and the extracted effective components are subjected to multivariate statistical analysis. According to the system, the quality difference of salvia miltiorrhiza from different producing areas is evaluated, and scientific basis and data support are provided for standardized planting, producing area screening and clinical reasonable application of salvia miltiorrhiza medicinal materials.

The invention discloses a method for tracing the origin of nuts based on lipid components, which realizes high-sensitivity and high-coverage detection of nut lipids by combining a high-resolution mass spectrometry technology with multivariate statistical analysis data, screens out lipid markers with origin specificity, and constructs an accurate origin discrimination model. The method has high sensitivity, coverage and expandability, is suitable for various nuts, provides a reliable tool for nut producing area identification, and is especially suitable for geographical indication product protection, market order standardization and consumer rights and interests.

The invention relates to the technical field of water environment treatment and cyanobacterial bloom early warning, in particular to a cyanobacterial bloom induction factor identification and toxin risk grading evaluation system, which comprises a multi-source data acquisition module used for acquiring environment big data, hydro-meteorological data and historical water bloom monitoring data; and the data preprocessing and fusing module is used for cleaning the multi-source data and aligning the time sequence. According to the invention, environment big data, hydro meteorological data and historical water bloom monitoring data are acquired through the multi-source data acquisition module, and the data are integrated by combining a credibility weighted fusion algorithm of the data preprocessing and fusion module, so that the problem of inaccurate multi-source data integration is solved; and the key induction factor identification and dynamic weight determination module mines and screens potential factors through multivariate statistical analysis and association rules, and then adjusts weights according to real-time and historical data deviation degrees by a sliding window dynamic weight mechanism, so that the problem of disjunction between static weights and an actual water bloom driving mechanism in the prior art is solved.