50 results about "Diagnostic model construction" patented technology

The invention discloses a cross-working-condition bearing fault diagnosis method based on a multi-module combination model, and the method comprises the following steps: 1, collecting bearing vibration time domain signals when different bearing faults are generated under different working conditions, dividing a source domain training set with fault type labels, and obtaining a source domain training set; the target domain training set and the target domain test set are not provided with fault type labels; step 2, constructing a diagnosis model, wherein the diagnosis model comprises a feature extraction module, a fault classifier module, a global domain confrontation module, a condition domain confrontation module and a feature weighted alignment module; 3, inputting the source domain training set, the target domain training set and the target domain test set into a diagnosis model for model training; and 4, carrying out bearing fault identification by adopting the trained diagnosis model. The generalization ability of the bearing fault diagnosis model is improved.

The invention discloses a children cough variant asthma diagnosis system based on a decision tree, and relates to the technical field of asthma diagnosis, the children cough variant asthma diagnosis system comprises a diagnosis management platform, the diagnosis management platform is in communication connection with a data acquisition module, a diagnosis model construction module, a diagnosis reasoning module and an intervention suggestion module, and the modules are in electric signal connection; the data acquisition module is used for collecting and preprocessing child patient clinical data including symptoms, signs, medical history and multi-dimensional examination data. By integrating the multi-dimensional examination data and the medical history, symptom change trend and family history information of the child patient, the condition of the child patient can be comprehensively and deeply analyzed, reasoning and judgment are performed by applying a branch structure of a decision tree and accurately combining various characteristics, so that subjective and empirical deviations in a traditional diagnosis method are effectively avoided, and the diagnosis accuracy is improved. The accuracy of CVA diagnosis is remarkably improved, a child patient is helped to obtain correct treatment in time, and misdiagnosis and missed diagnosis can be effectively reduced.

The invention relates to the technical field of photovoltaic energy storage system fault diagnosis, and discloses a photovoltaic energy storage cabinet fault diagnosis method and system based on deep learning. The invention discloses a photovoltaic energy storage cabinet fault diagnosis method based on deep learning, and the method comprises the steps: carrying out the pre-training of unmarked photovoltaic energy storage cabinet monitoring data through a physical constraint comparative learning algorithm, and constructing a feature representation space with physical significance; training a feature encoder by using the marked sample and constructing a prototype diagnosis model; constructing a comparative interpretation generator, and providing visual fault interpretation by generating an anti-fact sample; constructing a feature significance mapping technology; a feature, symptom and reason three-layer interpretation framework is constructed, and a diagnosis result is converted into a form which is easy to understand by maintenance personnel; constructing an interactive diagnosis interface; according to the invention, through comparing learning pre-training and a few-sample fine tuning strategy, a novel fault can be learned and identified only by a small number of marked samples, and the data acquisition and marking cost is reduced.

The invention discloses a digital orthodontic treatment planning system, and relates to the technical field of orthodontics. The system comprises a data acquisition module, a three-dimensional model construction module, a case database module, a diagnosis model construction module and a treatment planning module. The data acquisition module is used for acquiring oral scanning and CBCT scanning data and generating an oral three-dimensional model; the three-dimensional model construction module performs fusion processing on the two types of models to construct a three-dimensional gum model; the case database module is used for constructing a case database on the basis of soft tissues, CBCT projection measurement hard tissues and three-dimensional gum model characteristics; the diagnosis model construction module constructs an intelligent diagnosis model through a self-adaptive deep neural network and generates a treatment scheme set; and the treatment planning module combines patient data to judge malformation, formulate a scheme, and output a feasible treatment scheme after collision detection, model adjustment and scheme evaluation.

The invention discloses a gastric cancer diagnosis model construction method, system and device and a medium. The method comprises the following steps: acquiring a volunteer clinical data set; extracting a single index and a composite index of each volunteer from the volunteer clinical data set; calculating the importance of each index on gastric cancer diagnosis, and determining final feature combination data based on the importance corresponding to each index; and constructing a plurality of gastric cancer diagnosis models based on a machine learning model, training each gastric cancer diagnosis model based on the final feature combination data, performing performance evaluation on each trained gastric cancer diagnosis model by using verification set data, and determining an optimal gastric cancer diagnosis model based on a Marius correlation coefficient and an F1 score evaluation result. The model is evaluated again by using test set data; and performing gastric cancer diagnosis of the to-be-diagnosed person based on the optimal gastric cancer diagnosis model. According to the technical scheme, clinical data and routine examination indexes of a clinical laboratory are integrated, and the efficiency of gastric cancer diagnosis is improved.

The invention discloses a method for constructing a diagnostic model for identifying leaf diseases. The method comprises the following steps: constructing a leaf disease original image set; preprocessing the leaf disease original image set to generate a leaf disease image sample set; the content of the leaf disease image sample set comprises leaf disease classification, leaf disease stages and a corresponding image set; the method comprises the following steps: defining a network model of which the basic structure is ResNet50, and adding an IB module to form a ResNet50-FIB structure; the IB module is used for performing multi-scale fusion and enhancement processing and outputting deep semantic features and spatial detail information, the deep semantic features are used for matching leaf disease classification, and the spatial detail information is used for matching leaf disease stages; the output layer outputs disease classification and disease stages according to image recognition; and training the network model, and constructing and generating a diagnosis model for identifying leaf diseases. According to the technical scheme, the bottlenecks of a traditional model in precision and speed balance, early disease misjudgment and cross-crop adaptability can be broken through, and the method has industrial popularization potential.

The invention discloses an ADHD auxiliary diagnosis model construction method, control equipment and a program product, relates to the field of medical artificial intelligence, and obtains a high-performance ADHD auxiliary diagnosis model through a small sample. The method comprises the following steps: preprocessing acquired sample data, screening out a predetermined number of indexes with the importance higher than that of a diagnosis result in the sample data to obtain a sample set, and dividing the sample set to obtain a training set and a test set; and constructing a logistic regression classifier model, training the logistic regression classifier model by using the training set, and testing the trained logistic regression classifier model by using the test set to obtain the ADHD auxiliary diagnosis model. According to the method, the problem that an ADHD prediction model depends on the sample size is solved, and the feasibility of clinical application is improved.

The invention discloses an oral cancer diagnosis model construction method, which is based on DeepLabv3 +, increases an auxiliary branch to provide position information for a diagnosis model and increases SCConv attention so as to realize segmentation processing of oral cancer multi-tissue category pathological images. The method comprises the following steps: acquiring sample set labeling data sets ORA-TCGA and ORA-Nine for training a diagnosis model, classifying data set tissues into tumors, epithelium, interstitial substances, lymphatic vessels, skeletal muscles and vessels, and performing diagnosis model training and prediction by using OCMS-Net. The diagnosis model is based on OCMS-Net and comprises a main branch A and an auxiliary branch B, the input of the main branch A is an oral cancer pathological image, the output of the main branch A is a segmentation mask, and the segmentation mask represents different tissue types with different colors; the input of the auxiliary branch B is an oral cancer pathological image, and the output of the auxiliary branch B is a binary segmentation map and a nuclear pixel distance map of a cell nucleus.

The invention discloses a kiwi fruit drought stress diagnosis model construction method, and relates to the field of kiwi fruit drought stress diagnosis, and the method comprises the steps: carrying out the terrain partitioning of a planting region, and obtaining a plurality of different differentiation planting sub-regions; determining the deployment density and the burying depth of a soil moisture sensor, and collecting time sequence data representing the soil moisture condition of each sub-region; collecting water physiological state time sequence data of the kiwi fruits, and constructing a physiological feature data set; based on an LSTM neural network model, outputting a kiwi fruit drought stress grade probability; and verifying a diagnosis result of the model by adopting a confusion matrix, calculating three evaluation indexes including accuracy, precision rate and recall rate, and dynamically updating and optimizing the model. The method has the advantages that the technical span of kiwi fruit drought stress from'uniform monitoring 'to'partition accurate diagnosis' is realized, real drought and disease stress are effectively distinguished, and reliable technical support is provided for intelligent agriculture.

The invention relates to the technical field of fan blade fault diagnosis, and discloses fan blade fault diagnosis model construction and fault diagnosis methods and devices, and the fan blade fault diagnosis model construction method comprises the steps: constructing a fan blade finite element model according to geometric data, material data and load boundary data of a target fan blade; simulating a blade fault by using the fan blade finite element model to obtain a fault simulation vibration signal and fault strain data; the fault and actual simulation vibration signals and the fault and actual strain data are input into a fan blade fault diagnosis model for multi-scale fault feature extraction, and multi-scale fault features are obtained; based on the local maximum mean value difference loss and the classification loss, the multi-scale fault features are input to a classification module of the fan blade fault diagnosis model for training, the fault diagnosis model is trained through simulation data, strain data and multi-scale feature extraction, and the accuracy of the fault diagnosis model is improved.

The application discloses a DNA methylation site marker for detecting hepatocellular carcinoma, a multiplex dPCR kit and a diagnostic model construction method. The DNA methylation site marker comprises cg02829688, cg13080379, cg03760839, cg10703826, cg12664119, cg16990168 and cg23371746 sites. The application designs a primer probe group for the above sites, constructs a multiplex dPCR kit of a double reaction system, realizes quantification of target DNA methylation sites, takes the methylation levels of the 7 sites as characteristic variables, adopts an XGBoost algorithm and / or an LR algorithm to construct a diagnostic model. Experimental verification shows that the model can effectively distinguish hepatocellular carcinoma patients from liver cirrhosis, chronic hepatitis B, metabolic dysfunction-related fatty liver disease patients and healthy individuals, especially shows good efficiency in auxiliary diagnosis of early hepatocellular carcinoma, and provides a new technical scheme for clinical diagnosis and screening of hepatocellular carcinoma.

The invention discloses a fault diagnosis method and system applied to an impressed current cathodic protection system. The method comprises the following steps: acquiring multi-source time sequence data, namely sensor data sequence data, under different fault working condition categories from an ICCP monitoring system in real time or periodically through preset different types of sensors; performing data preprocessing on the multi-source time series data to obtain preprocessed time series data; constructing a fusion algorithm model based on a deep learning classification model long short-term memory network LSTM + Attention, and obtaining an artificial intelligence diagnosis model by preprocessing time sequence data; constructing a hybrid fault diagnosis model based on an artificial intelligence diagnosis model and an expert rule engine; and according to the hybrid fault diagnosis model, realizing fault diagnosis correspondingly used for the impressed current cathodic protection system. According to the invention, the technical problems of insufficient fault diagnosis precision and low efficiency in the fault diagnosis process of the ICCP system in the prior art are solved.

The invention provides a thyroid nodule auxiliary diagnosis model construction method and system based on multi-modal data and a storage medium. The method comprises the following steps: acquiring multi-modal thyroid nodule medical data; performing classification processing on the thyroid nodule medical data of each mode by using a single-mode classification model to obtain each thyroid nodule intermediate classification result corresponding to the thyroid nodule medical data of each mode; and carrying out weighted fusion on all the thyroid nodule intermediate classification results to obtain thyroid nodule benign and malignant classification results based on the multi-modal thyroid nodule medical data. According to the embodiment of the invention, the single-mode classification model is adopted to perform classification processing on the medical data of each mode, so that the step of converting the multi-mode data into the structured data is omitted, and the data processing efficiency is higher; in addition, the thyroid nodules can be diagnosed, monitored and treated, clinicians are assisted in improving the diagnosis efficiency of the thyroid nodules, and clinical burdens are reduced.

The invention discloses a diagnostic model construction method based on a novel network toxicology research normal form, and belongs to the field of health risk assess.The diagnostic model construction method specifically comprises the following steps that firstly, a potential poison and disease target library is constructed, and common potential targets between ATBC and fracture healing damage are screened and recognized; iI, constructing a PPI network to identify a core target, and analyzing functions and pathway enrichment of the core target; iII, identifying and screening the biomarkers, and constructing a diagnostic model to evaluate the diagnostic value of the biomarkers; according to the method, the evaluation precision of ATBC on the potential risk of fracture healing is remarkably improved, the conformation limitation of traditional static docking is broken through, stability evaluation in the time dimension is provided, the limitation of a rigid docking method is effectively overcome, meanwhile, visual representation of the dynamic behavior of a compound is provided, and the accuracy of the dynamic behavior of the compound is improved. The method provides a quantitative basis for mechanism analysis of molecular interaction, provides an effective technical means for subsequent high-risk substance screening, and is simple and convenient to operate and wide in application range.

The invention discloses a kidney essence deficiency diagnosis model construction and optimization construction method and system, and the method comprises the steps: carrying out the grouping processing of basic clinical information data, and obtaining the baseline data of kidney essence deficiency syndromes; performing standard equalization processing on the kidney essence deficiency syndrome baseline data to obtain kidney essence deficiency syndrome standard data, performing grouping analysis on the kidney essence deficiency syndrome standard data to generate an optimal model training set, and calculating and determining a kidney essence deficiency syndrome score of each group of kidney essence deficiency syndrome standard data after grouping; according to a diagnosis model structure, the kidney essence deficiency syndrome score and the quantification interval of the kidney essence deficiency syndrome score, constructing an initial kidney essence deficiency diagnosis model; and training the initial kidney essence deficiency diagnosis model based on each group of kidney essence deficiency syndrome standard data and the optimal model training set, and obtaining an optimal kidney essence deficiency diagnosis model according to a training result. By applying the method and the system provided by the invention, the diagnosis and prediction level and accuracy of the kidney essence deficiency syndrome are improved.

The invention relates to the technical field of municipal drainage pipeline data intelligent detection, and particularly discloses a drainage pipeline intelligent diagnosis method, which comprises the steps of collecting multi-modal data of a target pipeline, and generating a standardized multi-modal data set; obtaining source domain data, constructing a depth migration architecture, forming an initial diagnosis model, and outputting a first diagnosis result; the initial diagnosis model is actively learned and adjusted, a multi-mode collaborative diagnosis model is generated, and a second diagnosis result is output; bidirectional calibration is set, and a multi-mode collaborative diagnosis model is optimized; constructing a multiphase flow simulation model, and outputting a virtual sample; setting a reinforcement learning environment, reinforcing the multiphase flow simulation model, and outputting a final diagnosis result; in combination with geographic information, expert diagnosis is introduced, a multiphase flow simulation model learns again, and a closed loop is formed; and inputting the standardized multi-modal data set into the multi-modal collaborative diagnosis model to realize intelligent diagnosis of the drainage pipeline. According to the method, multiple modes are deeply migrated and learned, a full-link feedback mechanism is formed, and the engineering practicability is high.

The invention discloses a sliding bearing fault diagnosis model construction method and device, equipment and a medium. The method comprises the steps of generating a training set based on historical operation data of a sliding bearing and corresponding multi-dimensional target features; constructing a convex quadratic programming problem of a support vector machine by using the training set according to an interval maximization strategy, and converting the convex quadratic programming problem into a second-order cone programming problem; and solving the second-order cone programming problem by using a solver to obtain an optimal sparse weight vector and offset, and obtaining a sliding bearing fault decision function based on the optimal sparse weight vector and offset to obtain a sliding bearing fault diagnosis model so as to carry out sliding bearing fault diagnosis. By fusing the multi-source operation data and the multi-dimensional target characteristics of the sliding bearing, the defect of single-dimensional data is overcome, and the accuracy of fault state diagnosis is improved; and moreover, the convex quadratic programming problem of the support vector machine is converted into the second-order cone programming problem for efficient solving, so that the solving efficiency is improved.

The invention belongs to the field of medicine models, and provides a method for constructing a diagnosis model of glomerular crescent bodies of lupus nephritis patients. Comprising the steps of target collection data acquisition, statistical data analysis, inter-group comparison, biological index importance calculation and screening, candidate variable secondary screening, crescent body diagnosis model construction and crescent body diagnosis column graph construction. According to the method, the XGBoost algorithm is utilized to perform initial screening on the pre-prepared variables, and then the unconditional logistic gradual forward regression analysis is utilized to perform secondary screening, so that reliable screening of the crescent marker variables is realized, and in addition, the screening efficiency of the variables is improved; by converting the crescent body diagnosis model into the crescent body diagnosis column diagram, the crescent body diagnosis process is simplified, and the intuition of the model is improved.

The invention relates to an intelligent quality diagnosis model construction method for small sample industrial data. The method comprises the steps of performing data set construction preprocessing according to original image data of an industrial production site to obtain a preprocessed qualified product data set and a defective product data set; according to the preprocessed qualified product data set, a feature extraction backbone network is obtained through generative adversarial reconstruction network unsupervised training; according to the preprocessed qualified product data set and the feature extraction backbone network, through defect sample generation processing, a synthetic defect image and a corresponding pseudo label are obtained; according to the defective product data set, the feature extraction backbone network, the synthetic defect image and the corresponding pseudo tag, a quality diagnosis model is obtained through supervised model training; and inputting a to-be-detected industrial image into the quality diagnosis model for reasoning, and outputting a quality diagnosis result. The method can solve the problem of rare industrial field defect samples, and improves the quality diagnosis accuracy.

The invention discloses a bearing fault diagnosis and interpretable learning method, system, device and medium, and belongs to the technical field of bearing fault diagnosis and learning, and the method comprises the steps: obtaining and preprocessing source domain bearing data and target domain bearing data, and obtaining a feature data set; constructing diagnosis models according to the feature data set, and screening out an optimal diagnosis model from the diagnosis models; constructing a hierarchical diagnosis and verification framework, and diagnosing the target domain bearing data according to the optimal diagnosis model to obtain a diagnosis result; and observation learning is carried out on a diagnosis result through level analysis. The problems that misdiagnosis occurs in cross-domain diagnosis in the prior art, fault data and normal data are difficult to distinguish, and black box characteristics of deep learning are difficult to distinguish are solved.

This invention provides a method and system for fault diagnosis of marine engine lubricating oil systems based on transfer learning from the simulation domain to the real domain. The method includes: simulating faults under different operating conditions based on a constructed simulation model of the marine engine lubricating oil system to obtain simulated fault signal responses; acquiring real-domain fault data of the engine lubricating oil system under different operating conditions based on an engine experimental platform; obtaining simulation domain fault data using an encoder-decoder based feature fusion structure; obtaining a fault diagnosis model of the engine lubricating oil system based on dual transfer learning of data transfer and algorithm transfer; and performing real-time fault diagnosis of the marine engine lubricating oil system based on the engine lubricating oil fault diagnosis model to obtain diagnostic results. This invention achieves transfer learning from simulation domain fault features to real-domain fault features, thereby solving the problems of high construction cost and insufficient generalization performance of diagnostic models.

The invention relates to the technical field of cable detection, and discloses a cable insulation state diagnosis model construction method, a cable insulation state diagnosis method and a cable insulation state diagnosis device. Combining a plurality of base learners including a gradient boosting decision tree algorithm model, a lightweight gradient elevator algorithm model, an extreme gradient boosting tree algorithm model and a support vector machine algorithm model to construct a target base learner combination model, and using an improved frost ice optimization algorithm to optimize a mixed kernel extreme learning machine as a meta learner; therefore, the finally constructed cable insulation state diagnosis model can give consideration to different feature selection logics and model advantages, the diagnosis precision and generalization ability are effectively improved, information redundancy is reduced, and the problem of performance limitation of a single model is solved.

The application discloses a sliding bearing fault diagnosis model construction method and device, equipment and medium. The method comprises the following steps: generating a training set based on historical operation data of a sliding bearing and corresponding multi-dimensional target features thereof; constructing a convex quadratic programming problem of a support vector machine by using the training set according to an interval maximization strategy, and converting the convex quadratic programming problem into a second-order cone programming problem; obtaining an optimal sparse weight vector and a bias by using a solver to solve the second-order cone programming problem; obtaining a sliding bearing fault decision function based on the optimal sparse weight vector and the bias to obtain a sliding bearing fault diagnosis model and to perform sliding bearing fault diagnosis. By fusing multi-source operation data of the sliding bearing and multi-dimensional target features thereof, the shortcomings of single-dimensional data are made up, and the accuracy of fault state diagnosis is improved. Furthermore, the convex quadratic programming problem of the support vector machine is converted into the second-order cone programming problem for efficient solving, and the solving efficiency is improved.

The present invention discloses a method for constructing a category tree and a multi-granularity fault diagnosis model for thermal power equipment faults, as well as a hierarchical zero-sample diagnosis method. The present invention designs a category tree construction mechanism for data attribute fusion, utilizes data similarity to fuse attribute similarity, constructs a fault category tree from the bottom up, and constructs a multi-granularity fault diagnosis model, designs a layer-by-layer knowledge transfer strategy, regards the coarse-grained task model parameters as knowledge, provides guidance for fine-grained tasks, and works together with the designed attention module to improve the multi-granularity fault diagnosis model's ability to distinguish similar and unseen faults, thereby obtaining a trained multi-granularity fault diagnosis model. Utilizing the trained model, a probability fusion strategy is designed to integrate the coarse-grained task diagnosis results into the fine-grained task, thereby achieving zero-sample fault diagnosis. The present invention can mine the multi-granularity characteristics of fault types and improve the accuracy of similar and unseen fault classification in zero-sample fault diagnosis of industrial processes.

The invention relates to a primary sicca syndrome noninvasive diagnosis system and device and a medium, and the system comprises an obtaining module which is used for obtaining salivary gland CT image data of a user; the diagnosis module is used for inputting the salivary gland CT image data of the user into a diagnosis model to obtain the prediction probability that the user suffers from the primary sicca syndrome; when the diagnosis model is constructed, on the basis of salivary gland CT image data of a plurality of patients, the parotid gland and the submandibular gland in the salivary gland CT image data of the patients are automatically segmented by adopting an automatic segmentation model to obtain gland regions; extracting radiological features from the gland region, and performing data enhancement on the extracted radiological features through a conditional variation auto-encoder to generate a balanced positive and negative sample data set; and based on the balanced positive and negative sample data sets, training a classification model by adopting an integrated learning strategy to obtain a diagnosis model, and analyzing the diagnosis model by utilizing an SHAP value analysis mode. According to the invention, the accuracy and precision of diagnosis can be improved.

The application is suitable for the field of artificial intelligence technology, and provides an incremental training method based on knowledge anchor point reinforcement, a medical diagnosis model construction method and equipment. The incremental training method comprises the following steps: based on a target model trained by completing a historical task, the semantic correlation and knowledge contribution of each model parameter in the target model and the core knowledge of the historical task are calculated respectively to obtain knowledge anchor point parameters; in the incremental training process of the target model, the change amplitude threshold of the knowledge anchor point parameters is dynamically adjusted in combination with the real-time update state of the model global parameters and the correlation between new and old tasks; the change amplitude of each knowledge anchor point parameter in the incremental training process is monitored in real time, and when the change amplitude of any knowledge anchor point parameter exceeds the change amplitude threshold of the knowledge anchor point parameters, the incremental training is paused and the knowledge anchor point parameter recovery correction is performed, thereby ensuring efficient learning of new task knowledge of the model, and significantly improving the retention accuracy of the core knowledge of the historical task and the model training stability.

The application discloses a kind of planetary roller screw fault diagnosis model construction methods based on federal learning and light weight model, it is related to the field of fault diagnosis.Vibration data of planetary roller screw normal state and fault state are collected, and data set is constructed;Data preprocessing;Lightweight model SResNet18 is built;Based on federal learning framework and light weight model, planetary roller screw joint fault diagnosis modeling is carried out;Finally, the size and complexity of the model, the accuracy of the model and the training time of the model under the federal learning framework are evaluated.The application effectively solves the problem that there is no planetary roller screw fault diagnosis model construction method at present;Under the premise of ensuring data privacy, the data of each client is fully utilized to jointly establish the planetary roller screw fault diagnosis model;The lightweight model SResNet18 proposed in the application reduces the training time of the model under the federal learning framework, and solves the problem of high transmission cost of federal learning.

The application provides a marker composition, a kit, a diagnostic model construction method and application for identifying different types of breast fibroepithelial tumors, and specifically belongs to the technical field of pathological molecular diagnostic products. The marker composition comprises any one or two or more of the following markers: MAL2, KRT8, DSP, cg00515756, cg24852561, cg00686880, cg22449901, cg09025210, cg00819233 and cg04404381. The diagnostic model formed by the marker composition provided by the application can well distinguish PT and FA and different grades of PT, and is superior to the previously reported diagnostic model.

The embodiment of the invention provides an instrument fault diagnosis model construction method and device, a processor and a storage medium, and relates to the technical field of faults, and the method comprises the steps: generating a first sample data set of an experiment instrument and a second sample data set of a comparison instrument corresponding to the experiment instrument; determining a fault label according to the first sample data set; generating a fault diagnosis sample data set according to the first sample data set, the second sample data set and the fault label; and training the initial instrument fault diagnosis model through the fault diagnosis sample data set to obtain a target instrument fault diagnosis model. And performing fault diagnosis on the to-be-diagnosed instrument by using the target instrument fault diagnosis model so as to quickly obtain a fault diagnosis result of the to-be-diagnosed instrument. According to the invention, fault diagnosis and classification can be realized without adding extra hardware equipment, and the accuracy is high.

The invention relates to the technical field of coal mine fully-mechanized coal mining equipment, and discloses a digital twinning-based cutting drum cutting pick monitoring method, which comprises the following steps: data acquisition: acquiring three-way stress, temperature signals and vibration signals of each welding position; building a multi-channel diagnosis model, specifically comprising a vibration signal analysis model and a temperature signal analysis model, and training the models; building a virtual model, including a geometric model, a cutting pick-coal rock interaction physical model and a drum dynamics model, performing inversion and fusion on the models, and then performing model updating and application; and based on the vibration signal analysis result, the temperature analysis result and the overall stress condition of the cutting pick and the cutting roller, whether the cutting pick is abnormal or not is judged, and monitoring is achieved. According to the method, through organic combination of multi-sensor fusion, artificial intelligence diagnosis and digital twinborn technologies, the core pain points of poor real-time performance, weak scene adaptation, lack of predictive ability and the like of a traditional method are solved, and the method has extremely high engineering application value and popularization prospects.