240 results about "Drift detection" patented technology

The invention discloses a circuit board AOI detection result analysis method based on an intelligent algorithm, and the core of the method is to carry out the high-precision synchronous collection, denoising normalization processing and time sequence alignment of a detection image and multi-dimensional process parameters, and extract joint features through an attention mechanism. A process parameter-detection result correlation distribution model under each process scene is established, model output abnormal drift detection and tracing reasoning are realized, the detection model and the process parameters are dynamically optimized in combination with interpretability AI, the method realizes efficient tracking, attribution and adaptive optimization of detection misjudgment caused by process parameter variation, and the detection accuracy is improved. And the stability of the production process of the SMT detection system is improved.

The invention relates to the technical field of smart factory decision-making, and discloses a digital twin-driven smart factory AI intelligent decision-making system, which comprises a data acquisition and preprocessing module used for acquiring and preprocessing real-time and historical data of a production line; the twinborn modeling identification module is used for constructing a twinborn body and carrying out parameter identification and uncertainty quantification; the alignment evaluation credible module is used for comparing twin prediction data with real data and generating availability marks and trust scores; the strategy simulation optimization module is used for generating candidate strategies and risk evidences based on the constraints and the key performance indicators; the grayscale online publishing module is used for screening and grayscale publishing a strategy according to the trust score and the performance result; the operation evaluation auditing module is used for collecting operation data and generating an auditing packet; and the drifting detection updating module is used for detecting distribution drifting and recalibrating and updating the twinborn body and the strategy library. According to the invention, the reliability, robustness and continuous optimization of the production decision of the smart factory are realized.

The invention discloses a self-adaptive calibration system and method for a combustible gas sensor. The self-adaptive calibration system and method are suitable for a catalytic combustion type combustible gas sensor. The system comprises a basic calibration unit, a real-time monitoring unit, a drift detection and compensation unit and an output adjustment unit. The basic calibration unit is used for carrying out basic calibration on the sensor and determining an initial sensitivity coefficient, zero offset and response time of the sensor; the real-time monitoring unit is used for monitoring real-time parameters of the sensor under different environmental conditions; the drift detection and compensation unit is used for establishing a sensor performance parameter drift model and calculating a calibration compensation coefficient; the output adjusting unit is used for adjusting the voltage output value of the sensor in real time according to the calibration compensation coefficient; the method can adapt to environment change for real-time calibration, effectively compensates sensor drift, improves measurement precision and reliability, prolongs the service life of the sensor, reduces maintenance cost, and improves system safety.

The invention discloses an environment-adaptive transformer oil sample Raman spectrum detection method and related equipment, and relates to the field of optical sensing systems. The method comprises the following steps: collecting oil sample Raman spectrums and environmental parameters in multiple operation scenes, and constructing a multi-scene spectrum characteristic model and a standard fingerprint database; pre-processing and denoising parameters are adaptively set based on the environmental perception vector, and baseline correction and joint denoising are carried out on the original spectrum; scene discrimination is carried out by fusing the characteristics of peak position, peak height, peak width, integral area and the like, a scene-related component standard spectrum dictionary is generated, and the concentration and confidence of each target component are obtained by adopting constrained spectral line unmixing and quantitative calibration; and driving the fingerprint database and the model to update in combination with quality control indexes such as spectral shape relevancy and residual errors and a drift detection result. The system is composed of a Raman spectrum acquisition module, an environment monitoring module and a data processing module, and can improve the robustness and quantitative precision of Raman detection of transformer oil in a complex environment.

The invention relates to the technical field of transformer substation intelligent inspection, in particular to a meta-learning-based small-sample transformer substation equipment state adaptive inspection system, which comprises a state acquisition module for acquiring the current feature vector and environmental parameter data of a target node; the drift detection module is used for comparing environment parameters to judge data drift and dynamically adjusting a confidence coefficient threshold value; the risk assessment module inputs the feature data into a meta-learning model to output an initial risk probability, and generates an effective risk probability based on threshold filtering; the blind area measurement module is used for acquiring unobserved nodes and calculating system state blind area entropy; the scheduling decision-making module is used for comparing the blind area entropy with a threshold value and generating an entropy reduction bottom instruction or a self-adaptive routing inspection distribution instruction; the strategy updating module is used for extracting an actual inspection result and feeding back to the model for parameter updating; according to the invention, the scheduling difficulty when the resources are limited is solved, and the self-adaptive capability of the system under different environment interferences is improved.

The invention discloses a key sensor short-time abnormal distribution drift detection method in a unit start-stop process, and belongs to the technical field of gas turbine power plant financial supervision and artificial intelligence, and the method comprises the steps: synchronously triggering multi-channel signal collection through a main clock, and achieving noise suppression and data pre-screening through the combination of first-order difference and threshold filtering; constructing a nonlinear weighted feature matrix, and fusing a time attenuation coefficient and a shafting acceleration factor to enhance the transient feature expression capability; generating a sensor association graph based on double-threshold determination of weighted Pearson's correlation coefficients and mutual information, and dividing stable subgroups by using an incremental label propagation algorithm; designing a double-layer Cluster-GCN model, aggregating subgroup internal characteristics in the first layer, introducing a fuel valve position-acceleration comparison gating mechanism in the second layer to correct a global edge weight, and generating a node embedding vector sensitive to working condition change; gaussian kernel density estimation and an instantaneous deviation index of embedding similarity are fused, and a historical sliding mean value and subgroup connectivity analysis are combined, so that sensor faults and working condition abrupt changes are distinguished.

The invention provides an AI-driven equipment health state assessment method and system, and relates to the technical field of intelligent operation and maintenance. The method comprises the following steps: acquiring equipment operation data, and performing time and dimension unification and quality control to form a multi-source operation data set and an environment context; generating an initial state feature based on the mechanism feature library, and obtaining a general representation through self-supervised pre-training; executing calibration learning by using a preset health label, and establishing a fusion evaluation model containing time sequence consistency and physical boundary constraint; carrying out distribution alignment and uncertainty estimation on the basis of scene differences to obtain alignment characterization and credibility scores so as to optimize a model gating strategy; performing joint mapping on the new data, outputting health index, fault probability and residual life estimation, and generating a root cause clue; lightweight online updating is executed under drifting detection, health indexes and root cause clues are written back to a mechanism feature library, early warning levels and maintenance suggestions are generated, and therefore complete-cycle intelligent sensing and self-adaptive optimization of the equipment state are achieved.

The invention discloses a version knowledge graph reasoning method and system based on large language model enhancement, and relates to the technical field of dynamic knowledge graphs, and the method comprises the steps: employing a semantic drift detection and compensation mechanism, comparing context coding differences of same entities of different versions in an initial knowledge graph, recognizing drift, and dynamically adjusting entity embedding vectors, outputting a compensation update atlas; an LLM enhanced multi-hop reasoning algorithm is adopted, multi-hop reasoning is carried out on the compensation updating map, symbol reasoning, vector reasoning and context reasoning are carried out in parallel in each hop, and an entity relation reasoning result is obtained through dynamic weight fusion; and superposing an entity relationship reasoning result to the compensation updating graph through a cloud collaborative node, adding an entity edge and automatically maintaining a version history log, and obtaining a reasoning fusion version knowledge graph. Through a multi-hop reasoning algorithm enhanced by a large language model, the deep semantic mining capability and reasoning accuracy of a cross-version entity relationship are effectively improved.

The invention discloses a deep learning correction system for an MEMS sensor, and relates to the technical field of sensor correction, and the system comprises a multi-source data collection module which collects various data in real time, verifies and caches the data, and transmits the data; the feature extraction and analysis module processes the data and extracts features, and transmits the features to the drift detection modeling and motion impact discrimination module; modeling, calculating, monitoring sudden change and synchronizing information; the event type output result is judged; the correction decision execution module executes correction accordingly, and is internally provided with self-checking and fine tuning functions to guarantee the stability of the system; according to the invention, technologies of multi-source data acquisition, multi-dimensional feature analysis, coupling dynamic regression, deep convolutional neural network and the like are fused, so that comprehensive sensing and accurate drift detection of the sensor are realized; and through collaborative operation of an attention fusion joint discrimination algorithm and the like, event types are accurately distinguished and targeted correction is performed, the system stability is maintained, and the measurement precision and the practical value are improved.

The invention provides an AI knowledge base construction method and system for multi-source data integration, and belongs to the technical field of knowledge base construction. The method comprises the following steps: adaptively pulling unstructured data at an API-free site through a reversible crawler of a heterogeneous data access framework; the mode drift detector senses field change in real time and updates a protocol in a hot plug mode within 500ms, so that zero-stop synchronization is realized; carrying out weighted resolution on conflicts according to source credibility index attenuation, and generating version stamps with digital watermarks and consanguinity URI (Uniform Resource Identifier) for records; uniformly mapping to a shared semantic space through cross-modal comparative learning, and constructing an incremental hypergraph knowledge graph which can be evolved by probability weight; and extracting an optimal sub-graph by using generative adversarial reinforcement learning, writing the optimal sub-graph into a time sequence knowledge warehouse by using a zero knowledge evidence chain and carrying out block chain solidification, and outputting a verifiable certificate. By the adoption of the AI knowledge base construction method and system for multi-source data integration, end-to-end automatic, high-credibility and traceable large-scale knowledge base construction is achieved.

The invention belongs to the technical field of data processing, and discloses a multi-source heterogeneous data fused remote sensing map dynamic database construction method and system, and the method comprises the steps: firstly obtaining multi-source heterogeneous remote sensing data of a target area, extracting remote sensing time sequence data of a land parcel after preprocessing, and obtaining a remote sensing time sequence; constructing a time sequence semantic track containing a key semantic feature time sequence for each land parcel; identifying a semantic evolution trend by using a semantic drift detection model, and generating an abandoned suspicious score; the static factors, the dynamic factors and the prior knowledge of the land parcels are fused, a structure causal graph is constructed by means of a causal discovery algorithm, and abandoned causal credibility is reasoned based on a graph neural network model; and automatically generating abandoned land labels for the land parcels of which the double scores exceed the threshold value, and finally writing the land parcel information into a dynamic database. And the problems of construction lag and information incompleteness of the dynamic database are solved.

The invention discloses a machine learning-based flight wheel block removal time prediction method, and relates to the technical field of flight prediction, and the method comprises the steps: collecting flight preorder state data, airport resource distribution data and meteorological data in real time, and generating an original data set through multi-source heterogeneous data fusion; constructing spatio-temporal features including a preorder flight delay propagation chain, a stand-vehicle conflict energy matrix and a meteorological attenuation factor, and screening and optimizing a feature set through distribution drift detection; training and verifying the Bayesian depth quantile regression model, and outputting a prediction result with a confidence interval; and combining the airport Internet of Things positioning feedback optimization feature set and parameters to generate a prediction deviation diagnosis report. According to the method, a preorder flight delay propagation chain and a stand-vehicle conflict energy matrix are constructed, flight dynamics, resource allocation and weather attenuation factors are embedded into a unified spatial-temporal feature space, and the problem of feature information loss caused by data isolation is solved.

The invention discloses a data-driven perfusion process automatic adjustment method based on machine learning, and the method comprises the steps: collecting and preprocessing multi-source time sequence data, and generating an alignment feature vector sequence; performing perfusion stage division and stage coding vector generation based on the aligned feature vector sequence; constructing a three-layer liquid state machine model, and determining a stage liquid pool activation sequence; stage gating coding is executed, the liquid pool is driven to generate a dynamic state, and cross-stage migration is completed; integrating the dynamic state sequence, generating a joint state, and inputting a multi-task readout layer to output adjustment parameters; and executing distribution drift detection, topology updating and parameter calibration, and outputting final perfusion adjustment parameters. Through a data driving method based on stage topology modeling, liquid state machine dynamic evolution and a multi-task readout mechanism, accurate prediction, risk identification and self-adaptive adjustment of the perfusion process are achieved, and the perfusion quality and long-term operation stability are improved.

The invention relates to the technical field of data processing, and discloses a model dimension-based new energy vehicle value retention rate evaluation method and system, and the method comprises the steps: constructing three-table data, and carrying out the preprocessing; and fusing the static features and the time sequence features according to the type identifier to generate a feature sequence. And inputting the static features into a static feature sub-network to obtain a static representation, inputting the time sequence features into a time sequence feature sub-network to obtain a hidden representation of each time step, and obtaining a time sequence representation based on attention weighting. And performing drift detection on the input distribution and the model residual distribution. And splicing the static representation and the time sequence representation, inputting the spliced representation into a quantile prediction sub-network, outputting two value retention rate prediction results of different quantiles, and obtaining a value retention rate quantile set. And obtaining an insurance amount evaluation suggestion value based on the value retention rate quantile set and the new vehicle purchase price. According to the method, fine-grained description of different type feature differences is ensured, and the accuracy and reliability of a prediction result are improved.

The invention relates to a business data processing method and system, equipment and a medium. The method comprises the following steps: preprocessing multi-source heterogeneous cross-domain economic data to generate a standardized data stream; semantic drift in the standardized data flow is detected in real time, and a detection result is generated; semantic alignment judgment is carried out based on the detection result, and a dynamic alignment signal is generated; performing incremental training of a mapping model by using the dynamic alignment signal to generate an updated cross-domain mapping model; and finally, performing mapping conversion on the data stream based on the updating model, and outputting service data with unified semantics. By adopting the method, the semantic change in the economic data can be responded in real time, the problems of semantic drift detection lag, long model updating period and high maintenance cost in the traditional technology are effectively solved, and the accuracy and timeliness of cross-domain economic data processing are remarkably improved.

The invention discloses a configuration drift detection and consistency coordination method during operation of a Changsafety chain, and relates to the technical field of distributed system configuration management, the method comprises the following steps: deploying a main controller and a query-side double container in an Operator controller Pod, setting a client tool and a hierarchical cache in the query-side, and obtaining on-chain configuration by the main controller through a localhost interface; in the Recencile cycle, configuration drift is judged through standardization processing and hash comparison, and a change source is recognized in combination with self-defined resource metadata and chain side change metadata; coordination is executed according to four synchronization modes of CR forcing, chain observation, bandwidth period CR forcing and manual arbitration; according to the scheme, the problems of state fuzziness, error coverage and the like caused by dual fact sources are solved, low-delay detection and flexible coordination are achieved, and configuration consistency during running of the long safety chain is guaranteed.

The invention discloses an AI-based online data analysis method for medical molecular sieve oxygen production equipment. The method comprises the following steps: step 1, collecting multi-source operation monitoring data of the medical molecular sieve oxygen production equipment; 2, constructing a task prompt vector to obtain a normalized feature vector; 3, inputting the normalized feature vector into an improved TabPFN model to obtain a hidden space representation, and generating an enhanced representation in combination with a retrieval result of the prototype memory bank; 4, performing drift detection on the enhanced representation; 5, inputting the robust representation into a multi-task decoder; and 6, executing event label judgment, and outputting online analysis result data. According to the invention, real-time analysis of multi-source data and advanced identification of abnormal trends are realized, and the method is suitable for equipment state monitoring and risk early warning in hospital wards, rehabilitation centers and long-term oxygen supply scenes.

The invention discloses an intelligent and automatic test case generation method based on a large language model, and relates to the technical field of large language model application, and the method comprises the following steps: collecting a software demand description file and an interface standardization file of a project, carrying out semantic analysis and mapping, generating a logic constraint set, and establishing a path mapping table; inputting the path mapping table into a large language model, generating a test scene and assertion, and executing semantic drift detection; when semantic drift is detected, a drift report is generated, a reverse correction process is executed, a patch prompt is generated, and the generation chain is executed again; and after the semantic distance of the prompt chain is detected to reach a convergence state, outputting a stable test case set. According to the method, the generation structure is controlled through the deterministic prompt sequence, so that the stability and controllability of test scene and assertion generation are realized; and through semantic drift detection and a reverse correction mechanism, real-time correction of semantic offset of the prompt chain is realized, and the accuracy and stability of the test case are improved.

The invention relates to the field of press machine detection, and particularly discloses a press machine anomaly detection method, which comprises the following steps of: segmenting continuous force and displacement signal flow into independent stroke data segments, and extracting multi-dimensional feature vectors containing statistics, form and frequency domain information from the stroke data segments so as to comprehensively describe the state of each stamping stroke. Then, inputting the current stroke feature vector into an acute anomaly detector based on a heterogeneous auto-encoder, and sensitively capturing an emergent and isolated abnormal event in a high reconstruction error form by learning a reconstruction mode of normal data; besides, a chronic reference drift detector based on a Wasserstein distance is utilized, and systematic slow drift caused by equipment wear or working condition change is identified by analyzing distribution change of feature vector flow within a period of time. And finally, through an intelligent decision-making and arbitration mechanism, the acute abnormal score and the drift signal are fused, and a real equipment fault is identified.

The invention discloses a cloud platform automatic inspection method and system based on multi-dimensional intelligent analysis. By collecting resource metadata, performance indexes, logs and change events, a unified event object containing resource identifiers, event timestamps, association keys and credibility scores is generated, and out-of-order rearrangement and layered missing compensation are completed based on event time water level lines. Second-level and minute-level multi-granularity dynamic baselines are constructed for the key indexes, drift detection is carried out, and the exception score is calculated by fusing the super-boundary amplitude, the change rate, the duration and the cross-index consistency. And constructing a resource dependence topology with credibility and aging attenuation, introducing a time-delay consistency constraint to carry out contribution degree attribution and cut candidate root causes, fusing a Bayesian network and a fault knowledge graph and combining a historical fault library to update a prior probability positioning root cause on line, and outputting an interpretable link. Implempotent self-healing execution, acceptance and rollback closed loop are realized according to risk access control, the positioning accuracy is improved, and the fault recovery time is shortened.

Various systems and methods are presented regarding detecting data drift. The data of interest can be batches of utterances received at an interface (e.g., a chatbot). The batches of utterances can be compared with topics present in training data utilized to train a data classifier (e.g., an autoencoder), wherein topics identified in the batches of utterances that are not present in the training data can be considered to be novel topics. The greater the presence of novel topics in a batch of utterances, the greater the divergence of the batch of utterances from the content of the training data. The novel topics can be identified and subsequently applied to the training data such that the data classifier can be re-trained with the novel topics, thereby causing the data classifier to be contemporaneous with the novel topics. In an embodiment, the utterances can be short streams of text, symbols, and suchlike.

The invention provides a self-adaptive weighted hybrid modeling method and system for data drift sensing. The method comprises the following steps: acquiring multi-dimensional time series data and performing feature interaction analysis and screening to generate a fusion factor set; inputting the fusion factor set into a mixed structure comprising at least one ensemble learning model and at least one sequence model for training; based on the verification set and the dynamic evaluation indexes, determining fusion weights of all models in the mixed structure, and constructing a dynamic weighted mixed model; and setting a data distribution drift detection mechanism, comparing the distribution difference between a current data window and a historical reference data set, and automatically starting a retraining process of the hybrid structure to update the model when the difference reaches a trigger condition. According to the method, through cooperation of data drift perception and an adaptive weighting mechanism, the problems that an existing hybrid model is rigid in fusion strategy and lags behind a retraining mechanism are solved, and the robustness and long-term effectiveness of the model in a non-stationary data stream are remarkably improved.

The invention discloses a method and a system for acquiring the output power of a turbine, relates to the technical field of turbine monitoring, and provides a five-stage closed loop aiming at a ship high-temperature wet vibration environment, and millisecond-stage alignment torque, rotating speed, fuel oil, exhaust and vibration signals are synchronously acquired in a multi-source manner; secondly, a credibility matrix is constructed through null drift detection, wavelet denoising and consistency verification, and initial power is obtained through Kalman fusion; performing real-time comparison by using a digital twinborn model, and adaptively adjusting the weight of the sensor to output a correction power curve; then decomposing the virtual-real residual into slow drift, transient and harmonic factors, and dynamically modifying the sensitivity and the sampling rate to realize source end self-compensation; and finally, performing windowed compression, partitioning and salting hash on the correction curve, quickly confirming the right of the three-node lightweight block chain, and writing the fingerprint back to a twin model to realize minute-level tamper-proof traceability. The method has the advantages of high-resolution measurement, rapid deviation correction and real-time credible evidence storage.

The invention belongs to the technical field of vehicle networking safety, and discloses a vehicle-mounted CAN network ECU identification and intrusion detection method based on incremental online learning, and the method comprises the steps: S1, collecting a differential voltage signal on a vehicle-mounted CAN bus; s2, preprocessing and obtaining time domain features and frequency domain features of the differential voltage signals to obtain a CAN voltage data set, and constructing a data set with labels; s3, constructing an incremental learning classification model based on the initial data; s4, inputting streaming data to be detected to the incremental learning classification model, and monitoring ECU classification performance indexes of the streaming data through a drift detection algorithm; and S5, if the data distribution drift of the streaming data is detected, updating the incremental learning classification model and returning to the step S4. According to the method provided by the invention, the problem that the ECU recognition accuracy is reduced due to data drift in a dynamic environment is effectively solved, and the vehicle-mounted network security is improved.

The invention belongs to the technical field of malicious traffic detection and network security, and provides a malicious traffic detection model robustness enhancement method and system based on reinforcement learning and incremental learning. The method mainly solves the problems that an existing malicious traffic detection model is insufficient in robustness, poor in confrontation sample compliance, prone to catastrophic forgetting in a dynamic threat environment, weak in adaptability and the like. According to the main scheme, the method comprises the following steps: constructing state action space joint modeling, a three-order composite reward function and a compliance action mask mechanism through reinforcement learning in combination with a Transform strategy network, and generating an adversarial sample with high concealment and high compliance; an MMD concept drift detection mechanism is used for sensing attack mode evolution, a hierarchical EWC parameter protection strategy is combined, drift sensing-parameter protection collaborative incremental learning is achieved, and self-adaptive updating of a detection model is completed; an intelligent malicious traffic defense framework with high adversarial robustness and continuous environmental adaptability is constructed through the process, and meanwhile, the model performance can be verified through a related mechanism, so that the defense effect is ensured.

The invention relates to the technical field of data processing, in particular to a topological consistency risk analysis and monitoring early warning method for a tower inclination state, and the method carries out the time alignment and quality control of multi-modal data, such as an inclination angle, a wind speed and direction, temperature and humidity, positioning and an equipment state. Constructing a time-varying adjacency matrix based on line topology, geographical adjacency and wind field elements, and normalizing the time-varying adjacency matrix to form a graph model; the inclination angle or the trend of the inclination angle serves as a graph signal, space consistency measurement is calculated, and linkage with sliding window trend detection of a single rod and EWMA and ADWIN drift detection is carried out; identifying segment-level or region aggregation anomalies by synthesizing image smoothing energy, adjacent rod difference and spatial self-correlation; and outputting early warning grades under wind direction, wind speed, duration time window and data quality gating, and adaptively adjusting a threshold value, a time window and a model weight during drifting. The method can reduce false alarm and missing alarm, improves early warning accuracy and interpretability, and is suitable for centralization or edge cloud collaborative deployment.

Microprocessor for a scale system for a mobile storage carrier operates in three states: motion, stable, and fault where stability is determined based on load cell signal variations or external sources and a fault state follows a stable state in response to signal drift in one or more load cells.

The invention discloses a power distribution load prediction method and system, and relates to the technical field of graph calculation, and the method comprises the following steps: obtaining load data and auxiliary features of a user, and constructing a load portrait; according to the load portrait, selecting a model subset from a pre-training model library, migrating the selected model, and outputting an initial load prediction result; constructing a load association graph according to the spatial adjacency and behavior similarity between the users, and spreading load characteristics in the load association graph; regulating and controlling the model of the target user according to the propagated load characteristics in combination with an initial load prediction result; obtaining a user load data distribution change after regulation and control, and identifying a behavior mutation; through multi-model integration, individual fine tuning, load association graph construction, incremental learning and drift detection, the problems of low precision, poor dynamic adaptability and slow response to sudden change in traditional power distribution load prediction are effectively solved.

The embodiment of the invention provides an atmospheric pollutant early warning method based on multi-source isomerism, which belongs to the technical field of data processing, and specifically comprises the following steps: step 1, multi-source isomerism data encryption access and transmission; step 2, data quality control and missing measurement processing; step 3, performing space-time alignment and scale transformation; 4, feature engineering and variable construction; 5, carrying out fusion modeling and pollutant prediction; step 6, dynamic threshold and risk scoring; 7, carrying out online learning and drift detection; step 8, uncertainty evaluation and interpretability output are carried out; and step 9, early warning release and closed loop iteration. Through the scheme of the invention, the response capability and the early warning accuracy are improved.

The invention discloses a multi-dimensional fusion sensor data anomaly detection method based on a hydrogen-based shaft furnace, and belongs to the technical field of metallurgical equipment monitoring. The method comprises the steps of data preprocessing, single-variable anomaly detection, multivariable consistency detection, anomaly type judgment and alarm output. The single variable detection adopts a rolling median absolute deviation method, drift detection, continuous tiny variable threshold and border crossing detection; the multivariate detection calculates reconstruction residuals of each sensor through principal component analysis (PCA) and identifies overall consistency anomalies. And the system judges the fault type according to the comprehensive score of the multiple detection results and outputs a detailed alarm record and a statistical report. According to the method, multi-algorithm fusion analysis is carried out on time sequence data collected by a plurality of sensors in the operation process of the shaft furnace, so that the abnormal states of the sensors are accurately recognized. The method can be widely applied to real-time monitoring of the hydrogen-based shaft furnace smelting process, the accuracy and robustness of anomaly detection are improved, and false alarms and missing alarms are reduced.