850results about How to "Improve interpretability" patented technology

The invention discloses an automatic causal structure generation method based on semantic representation and logical reasoning of a large language model. The method comprises the following steps: acquiring an input text; performing semantic coding and clustering on the obtained input text by utilizing a large language model, and establishing a candidate causal variable set; causal relationship detection is carried out on the established candidate causal variable set based on anti-fact intervention and do-calculation; performing causal direction judgment, and generating a directed acyclic causal graph meeting logic consistency; and on the basis of the generated directed acyclic causal graph, natural language interpretation is generated by using a large language model, and logic consistency closed-loop verification is carried out. According to the method, automatic generation from the natural language to the causal structure is realized, the causal variable set is automatically extracted and constructed from the unstructured natural language text, the defects that variables need to be manually defined and modeling depends on field experts in the existing causal modeling process are avoided, and the labor cost and professional threshold of causal structure construction are remarkably reduced.

The invention discloses an unmanned aerial vehicle cluster task allocation method based on a large language model optimization genetic algorithm, and belongs to the field of computers. The method comprises the following steps: setting a specific chromosome coding mode; generating a multi-constraint initial population; calculating fitness to quantify the advantages and disadvantages of individual genes of the population; when the optimal individual meets the requirement or the maximum iteration round is reached, ending; retaining the optimal individual as a filial generation; generating a batch of new filial generation individuals by the large language model, and fusing the new filial generation individuals with the current filial generation population; calling an optimized large language model to analyze individual chromosome semantics, and outputting an evolutionary potential score; obtaining an individual comprehensive selection probability by integrating the fitness and the evolution potential score, and executing a selection operation; and selecting individuals based on the individual comprehensive selection probability to carry out crossover and mutation operation to generate offspring. The large language model is embedded into the core link of the genetic algorithm, and the algorithm efficiency is improved by improving the population diversity of the genetic algorithm in the unmanned aerial vehicle cluster task allocation scene.

The invention discloses a hydrological sequence missing data complementation and trend prediction system and method, and belongs to the technical field of water conservancy information. The system comprises a data acquisition and preprocessing module, a spatial-temporal feature fusion module, a physical constraint completion module, an uncertainty prediction module, a credible evidence storage and verification module and a visual tracing module which are connected in sequence. The method comprises the following steps: collecting and preprocessing multi-source hydrological data; combining the spatial-temporal features through a dual-channel network to generate joint representation; generating and checking a completion sequence based on joint representation and coupling physical constraints such as a unit line method and a Manning formula; performing uncertainty trend prediction by using an integrated predictor; the key data of the whole process are linked and stored, and are automatically verified through an intelligent contract; and full-life-cycle visual tracing is provided. According to the method, the complementation robustness and accuracy in a high-missing-rate scene are effectively improved, the physical consistency of a complementation result is ensured, and credible evidence storage and transparent decision of the whole process are realized.

The invention belongs to the technical field of software engineering, particularly relates to an automatic driving system performance enhancement method based on an expert hybrid architecture, and aims to solve the core problems that an end-to-end automatic driving system is confronted with semantic fuzziness to cause unreliable decision, multi-task interference hinders optimization planning, too long reasoning delay increases driving risks and the like. According to the method, an ExpertAD framework is provided, task key features are amplified through a perception adapter (PA), and the relevance of scene context understanding is guaranteed; related driving tasks are dynamically activated through a sparse expert mixture (MoSE), and task interference is minimized; and in combination with a customized training loss function, collaborative optimization of planning effectiveness and reasoning efficiency is realized. Experiments show that compared with an existing method, the method has the advantages that the average collision rate is reduced by 20%, the reasoning delay is reduced by 25%, higher multi-skill planning capacity is achieved in rare scenes (such as accident handling and first-aid vehicle avoiding), and good generalization is achieved for unseen urban environments.

The invention relates to the technical field of slope disaster monitoring, in particular to a high and steep slope geological disaster monitoring method and system based on multi-source data fusion. The method comprises the following steps: tracking surface deformation through PS-InSAR time sequence analysis and atmospheric phase correction to obtain surface deformation characteristics; identifying the apparent diseases of the slope, and extracting spatial and temporal distribution characteristics of the apparent diseases of the slope; analyzing spatio-temporal evolution characteristics under geologic structure constraints by utilizing slope rock mass structure characteristics and the earth surface deformation characteristics; according to underground water level dynamic and potential sliding surface weakening features and the slope apparent disease spatial and temporal distribution features, extracting disaster-causing key factor features after cooperative correction; and combining the disaster-causing key factor characteristics, the true three-dimensional geological environment background characteristics and the spatio-temporal evolution characteristics to analyze the comprehensive early warning grade and the potential instability mode of the landslide. Scientific support is provided for high and steep slope geological disaster prevention and control, and engineering construction and operation safety can be guaranteed.

The invention discloses a building pipeline intelligent flaw detection and leakage positioning system, which relates to the technical field of building pipeline detection and comprises a data acquisition module, a multi-modal fusion module, a defect identification module, a leakage positioning module, a digital twinborn module, a health prediction module and a decision control module. The data acquisition module is used for deploying multiple types of sensors along a building pipeline, acquiring acoustic emission signals, infrared thermal imaging images, pressure fluctuation data and fluxgate detection signals in a running state of the pipeline, and outputting original sensing data; the multi-modal fusion module is connected with the data acquisition module and is used for performing feature extraction on the original sensing data, constructing a multi-modal feature vector, dynamically adjusting each modal weight based on environmental parameters and outputting a weighted fusion feature vector; and the defect identification module is connected with the multi-modal fusion module and is used for inputting the weighted fusion feature vector into a deep neural network model.

The invention provides a clinical decision knowledge graph construction method and system, and the method comprises the steps: extracting standardized entities corresponding to diseases, symptoms and diagnosis and treatment elements from medical knowledge data; constructing a clinical concept knowledge graph for representing a medical concept logic relationship and a causal relationship according to the semantic association relationship and the causal dependency relationship among the standardized entities; a diagnosis event, an examination event and a treatment event related to the patient are extracted, link evidences among the events are determined based on the event chain relation among the events, and a clinical event knowledge graph used for representing the disease course evolution process of the patient is constructed according to all the link evidences; and performing knowledge element fusion based on an entity association relationship between the clinical concept knowledge graph and the clinical event knowledge graph, and generating a target knowledge graph for clinical decision analysis. By adopting the scheme of the invention, the cross-map fusion of the static medical concept knowledge and the dynamic disease course event chain relationship can be realized, and the clinical decision knowledge structure with the reasoning ability can be constructed.

The invention discloses an automatic driving test and evaluation method and device based on a large language model, and solves the problems that the pertinence of a test scene is insufficient, the result understanding depends on manpower, the evaluation interpretation is weak and the like in a traditional method. By means of semantic analysis, logical reasoning and domain knowledge support of a large language model and in combination with an automatic driving test domain knowledge graph, generation of a structured test portrait of a tested system, construction of a test demand list, generation of a semantic test scene description file, interpretable test result attribution, multi-dimensional comprehensive evaluation and report output are completed in sequence. And a test portrait is updated through report feedback to form a closed loop. The method is suitable for function verification and performance evaluation of an automatic driving simulation test, can also be used for post-processing analysis of real vehicle test data, realizes automation of a whole test process and interpretability of an evaluation result, and improves test precision and efficiency.

The invention discloses a two-stage space-time diagram network and multi-source information fusion time sequence probability prediction method and device, and relates to the technical field of artificial intelligence and big data analysis. The method comprises the following steps: acquiring historical wind speed sequence data and corresponding target variable sequence data; preprocessing the data, and constructing a training sample according to a preset time sliding window; constructing a time sequence probability prediction framework of the two-stage space-time diagram network and multi-source information fusion; based on the fluctuation correlation of the historical sequence data, constructing an adjacent matrix between nodes; inputting the training sample and the adjacent matrix between the nodes into a wind speed prediction model, training the model through a designed threshold perception loss function of a wind speed-power nonlinear relationship, and outputting a multi-node wind speed prediction probability of a first stage; and inputting the multi-node wind speed prediction probability and the multi-source environmental factor data into the gradient boosting tree model for second-stage prediction, and outputting a multi-node wind power prediction result. According to the invention, prediction errors can be reduced.

The invention provides a multi-parameter water quality data fusion analysis method and system. The method comprises the steps that water quality parameters are collected to form a three-dimensional data cube; constructing a space-time tensor model by using Tucker decomposition and a graph convolution network, and generating a core tensor matrix; constructing a dynamic constraint library and embedding the generative adversarial network; training a generative adversarial network by using Transform and physical constraint loss, generating synthetic data and verifying the synthetic data; performing space-time fusion by using meta learning weight distribution and Bayesian deep learning to generate a weight matrix and a confidence interval; missing data are restored through physical constraint interpolation and Gaussian process regression, and SHAP and LIME interpretation and path diagrams are generated; and performing real-time analysis by using an edge-cloud collaborative architecture to generate an intelligent report. Through physical constraint modeling, dynamic weight distribution and edge-cloud collaborative architecture, the problems that synthetic data violates physical laws, weight staticization, response lag and insufficient interpretability are solved.

The invention relates to the technical field of bearing fault diagnosis, in particular to a bearing fault diagnosis method based on a neural network and multi-criterion preference consensus, and the method comprises the steps: data collection and feature extraction, data preprocessing, fault diagnosis model matching, preliminary diagnosis and probability generation, final diagnosis and probability generation, and bearing state decision making. According to the method, a multivariate decision-making auxiliary model fusing a physical marginal value function and a neural network attention mechanism is provided, and a reinforcement learning driven consensus achievement process is combined, so that adaptive optimization of a multi-expert fault diagnosis result is realized; the problems that an existing bearing fault diagnosis method cannot deal with complex working conditions, is insufficient in physical interpretability, depends on a single decision mode and mostly adopts static weight distribution are solved, the accuracy, robustness and interpretability of bearing fault diagnosis are improved, and high-precision and low-round group fault diagnosis is achieved.

The invention provides a multi-source security knowledge fusion driven security risk management large model construction method and application. The method comprises the following steps: constructing a campus security knowledge graph and a semantic vector database based on multi-modal campus security data; performing approximate nearest neighbor search in the semantic vector database based on user query to obtain context evidence related to the user query, splicing the context evidence with the user query to obtain a retrieval enhancement prompt, the user query being campus security risk information of any mode; and training based on the campus security knowledge graph to obtain a security risk management large model, and inputting the retrieval enhancement prompt into the security risk management large model to obtain a risk solution. According to the scheme, the attention confidence, the feed-forward network confidence and the knowledge alignment confidence of all levels are aggregated in the reasoning process to generate the comprehensive confidence score, so that the illusion problem caused by no data reference in the model reasoning process is effectively avoided.

The invention discloses a load-oriented self-adaption double-source waste heat mode switching control method, particularly relates to the technical field of industrial process control and waste heat comprehensive utilization, and aims to solve the problems that heat supply mode switching in an existing double-source waste heat system depends on artificial experience, waste heat supply cannot be matched in a self-adaption mode along with load change, and the working efficiency is low. The waste heat utilization rate is low, the heated load temperature fluctuation is large, and the operation stability is poor. A heat exchange topology model is established in a double-source waste heat system, a load heat demand track and a waste heat capacity fingerprint are synchronously constructed in a rolling time window, and a heat supply mode sequence is optimally selected by taking a load heat demand satisfaction degree, a waste heat utilization degree and a mode switching cost as indexes under an equipment operation constraint and a mode switching constraint. Therefore, the effects of automatically matching the heat load and the waste heat supply under the condition of dynamic change of the working condition and the load, giving consideration to temperature stability and efficient utilization of the waste heat, reducing manual adjustment and improving the system energy efficiency and the operation stability are achieved.

The invention discloses a communication base station lithium battery fault diagnosis method and device and electronic equipment, relates to the technical field of lithium battery monitoring and fault diagnosis, and is used for solving the problems that the communication base station lithium battery fault diagnosis is inaccurate, and the fault source is difficult to locate. According to the invention, multi-dimensional data acquisition, including voltage, current, temperature and electrochemical impedance spectroscopy, is carried out through the battery management unit; extracting static and dynamic inconsistency characteristics based on the collected data, and calculating a comprehensive inconsistency index; similarity matching is carried out by combining the impedance deviation spectrum with a fault feature library, and the fault type and grade are identified; comprehensively judging a fault source through fault monomer space positioning, electrical topological correlation analysis, impedance spectrum deep analysis and thermal behavior anomaly analysis; and the accuracy of fault diagnosis and root positioning is obviously improved.

The invention provides a power load prediction method and device, equipment and a medium. Relates to the technical field of power load prediction. The method comprises the following steps: respectively mapping historical time sequence data from different data sources of a to-be-predicted power system into single-source time sequence characteristics representing overall time sequence change characteristics of the corresponding data sources, and modeling a dependency relationship of the single-source time sequence characteristics corresponding to the data sources, obtaining time sequence embedding representing overall dynamic evolution of the to-be-predicted power system; historical time sequence data from different data sources are converted into natural language texts expressing data source attributes, time sequence characteristics and power load associated information, the natural language texts corresponding to the data sources are coded, and prompt embedding including semantic information of the data sources is obtained; fusing the time sequence embedding and the prompt embedding to obtain cross-modal enhanced embedding; and decoding the cross-modal enhanced embedding to obtain a power load prediction result of the to-be-predicted power system.

The invention relates to an engineering potential safety hazard detection method based on a multi-modal large model. The method comprises the following steps: acquiring an on-site monitoring image of a target scene; identifying a key entity in the field monitoring image, and extracting positioning information of the key entity; generating a depth map based on the on-site monitoring image; based on the positioning information of the key entity, combining with the depth map, associating the positioning information and the depth information for the key entity, and generating a scene description text; based on the on-site monitoring image, the depth map corresponding to the on-site monitoring image and the scene description text, constructing a multi-modal input packet; inputting the multi-modal input packet into a learned engineering safety domain knowledge base and a visual language large model subjected to instruction fine tuning, and outputting a hidden danger judgment result; the instruction fine tuning of the visual language large model is based on a hidden danger inference chain data set, the data set has hidden danger images corresponding to hidden danger items in an engineering safety field knowledge base and annotation data of each hidden danger image, and the annotation data comprises step result data corresponding to each hidden danger detection step.

The invention discloses an image fusion method based on a double-branch feature decoupling auto-encoder, and belongs to the field of computer vision and image processing. According to the method, for the problems of modal pollution and structural distortion in infrared and visible light image fusion, structural semantic information and high-frequency texture details of a source image are extracted respectively by constructing a content feature coding module and a detail feature coding module, and feature decoupling is achieved. And performing deep fusion on the decoupled features by using an adaptive feature weighting mechanism, and reconstructing a fused image with infrared target saliency and visible light detail definition through a shared decoder. According to the method, an end-to-end two-stage training strategy is adopted for optimization, complex prior or post-processing is not needed, the effects of improving the fused image contrast, edge preservation and target recognition performance are remarkable, and the method is suitable for the fields of weak light monitoring, intelligent perception, unmanned system navigation and the like.

The invention discloses a material attribute classification method based on multispectral and polarization information, and the method comprises the steps: firstly achieving the real-time collection and processing of image data through hardware cooperation, constructing a training data set containing multispectral, different polarization angles and RGB images, and providing a label for each sample; secondly, samples in the training data set are preprocessed, stokes parameters are calculated, and input data are obtained through the stokes parameters; and then constructing a feature encoder of a three-branch network structure, processing input data and calculating physical reconstruction loss, carrying out adaptive weighted fusion on output of the feature encoder, and respectively predicting a material category and a surface state through a dual-task output head. According to the method, high-accuracy material and surface attribute identification can be realized under the conditions of complex materials, mirror reflection and fine surface difference.

The invention discloses an interactive load prediction method based on fusion of prior knowledge and data driving, and the method comprises the steps: reconstructing a load sequence into a trend load sequence and a disturbance load sequence through modal decomposition, and constructing a prior knowledge base of tag-feature weight according to SHAP interpretability analysis; and through a prior knowledge base and feature contribution analysis, channel-level weight learning is carried out by using an automatic correlation determination module, and key input features are obtained. And inputting the reconstructed load sequence and the key input features into an improved Transform model, mapping a priori knowledge base into a priori knowledge matrix, embedding the priori knowledge matrix into the model, and generating a load prediction result. And carrying out interpretability analysis on a prediction result, and feeding back and updating an analysis result to a priori knowledge base to realize dynamic collaborative updating of prediction and the knowledge base. According to the invention, organic combination of expert knowledge and a data-driven model is realized, and the precision and stability of load prediction are improved.

The invention discloses a data auditing method and device based on artificial intelligence, electronic equipment and a storage medium, relates to the technical field of artificial intelligence, can be specifically applied to the financial and medical fields, and meets high requirements of financial and medical industries on compliance and auditing tracking while improving auditing efficiency and reducing labor investment. The method comprises the steps that in response to a data auditing request, target data to be audited are preprocessed, and a hierarchical tree structure used for describing text content and layout position information of the target data is obtained; based on the large language model, generating a field extraction cue word matched with the data type of the target data, guiding the large language model to perform field extraction on the hierarchical tree structure by utilizing the field extraction cue word, and obtaining a field extraction result output by the large language model; and performing abnormal field auditing and abnormal field evidence labeling on the field extraction result by utilizing a pre-constructed knowledge graph to obtain an auditing result of the target data, and outputting the auditing result.

The invention discloses a large-model-driven urban railway newly-opened station flow prediction method, and relates to the technical field of urban railway traffic operation management. The technical problems of low prediction precision, poor interpretability of a traditional model and insufficient reasoning flexibility in a complex urban scene caused by lack of historical passenger flow data of a newly opened station of an urban rail are solved. The method comprises the following steps: firstly, collecting pre-processed passenger flow related multi-dimensional data, and then constructing a multi-dimensional reference system to retrieve a target station similar reference or extract a macroscopic index reference; judging a scene category through a rule engine, calling a corresponding template, and filling information to generate a structured cue word; inputting the cue word into the large language model, and outputting a content prediction value and a result of a natural language reasoning chain according to the reasoning thinking chain template; and finally, checking the result, and completing optimization of the method based on feedback. The method has a good application prospect in the technical field of urban rail transit operation management.

The invention discloses a data-mechanism-knowledge combined driving dam safety state self-diagnosis method and system, and the method comprises the steps: obtaining dam operation state data, extracting a state quantity based on a pre-constructed computable evidence graph, and generating a monitoring evidence; performing inversion correction on mechanism model parameters under the priori constraint of the atlas, and generating mechanism evidence; the monitoring evidence and the mechanism evidence are accessed to a probability graph model for joint inference, and arbitration correction is carried out when a conflict triggering condition is met; and carrying out credibility calibration on the diagnosis result by adopting a conformal prediction method based on the historical calibration sample set, and outputting a security state diagnosis conclusion containing a confidence interval. According to the method, the problems of multi-source evidence conflict, mechanism parameter distortion and low diagnosis result credibility are effectively solved, and credible and automatic diagnosis of dam safety is realized.

The invention belongs to a near-infrared spectrum phenotype analysis technology in the field of agricultural breeding, and discloses a corn near-infrared multi-environment yield prediction method and system based on spectrum structure and character adaptive gating. The method aims at solving the problems that in the prior art, the relation between spectrum segments is not modeled, the sensitivity difference of characters and environments to wavelengths is difficult to distinguish, physical priori and multi-environment constraints are lacked, and consequently the prediction precision and interpretability of complex characters are insufficient, and an efficient and low-cost tool is provided for multi-environment phenotype selection and corn breeding decision making. The method comprises the core steps of data preprocessing, multi-scale spectral feature embedding, spectrogram construction and graph attention coding, feature fusion, character adaptive wavelength gating, multi-task regression, physical constraint model training prediction and the like. The system is composed of a corresponding function module and a training and model management module, key spectrum information can be fully mined, and a high-precision result can be stably output. The method can be used for multi-environment yield and quality prediction, online near-infrared monitoring and phenotype selection decision making of crops such as corn, can also be popularized to multi-mode breeding big data analysis and spectrograph waveband optimization, and is good in interpretation and wide in application prospect.

The invention discloses an electromechanical equipment fault cross-domain diagnosis method and system based on minimum category confusion, and the method comprises the steps: selecting an external public data set or labeled historical data as a source domain data set, and then collecting the real-time operation data of electromechanical equipment to be diagnosed, taking the real-time operation data of the electromechanical equipment to be diagnosed as a target domain data set; constructing a fault diagnosis model; performing joint adversarial training on the fault diagnosis model by using the preprocessed source domain data set and target domain data set, and performing lightweight processing and parameter solidification after training is completed to obtain a final fault diagnosis model; and deploying the final fault diagnosis model to an edge computing equipment end, and diagnosing the preprocessed real-time fault data of the electromechanical equipment by using the final fault diagnosis model to obtain a diagnosis result. According to the method, the multi-scale feature extractor is adopted to perform multi-scale feature extraction and fusion, so that the feature extraction capability is improved, and missing of key features is avoided.

The invention discloses a feature analysis and prediction method for drawing a depression patient based on a convolutional neural network, and belongs to the field of artificial intelligence and mental health. The method comprises the following steps: acquiring drawing image data of a depression patient and a healthy control group, and constructing a standardized data set; preprocessing such as denoising, size normalization and edge enhancement is carried out on the image; constructing a convolutional neural network model comprising four convolutional layers, two pooling layers and a feature fusion layer, and automatically extracting key features such as line definition, spatial layout and detail richness by adopting a variable receptive field mechanism and an attention module; correlation analysis is carried out in combination with scores of a clinical evaluation scale, and the model is optimized through transfer learning and a focus loss function; and generating a risk assessment report. According to the method, objectivity and accuracy of depression screening are remarkably improved, subjective evaluation deviation is reduced, visual decision support is provided for clinicians, and early recognition and intervention of depression are achieved.

This invention provides a method for hypertension risk assessment, relating to the field of epigenetic detection technology. It includes: performing event alignment and residual modeling on a qualified signal set to establish a base probability at the locus level and forming baseline features through neighborhood consistency screening; constructing a tunneling sensitization model and contextual attention, fusing discriminative and generative evidence to obtain enhanced features; quantifying IGF2BP3 and aligning and fusing it with the enhanced features; constructing an IGF2BP3-mediated metabolic network based on this, performing time-series modeling and stability assessment, and extracting final-state features; fusing the final-state risk vector with multi-gene scores to complete adaptive grading and compliant report generation. This method is robust, interpretable, and easy to deploy.

The invention discloses a charging station dynamic recommendation method and system based on multi-dimensional real-time data, and relates to the technical field of big data analysis, and the charging station dynamic recommendation method based on the multi-dimensional real-time data comprises the steps: obtaining charging data of all users according to charging requests of the users, the charging data comprises historical order data, real-time charging gun data, charging pile data and charging station data; based on the charging data of all users, constructing a multi-source heterogeneous relational graph of an associated order, a charging gun, a charging pile and a charging station, and constructing a corresponding initial feature vector for each node in the multi-source heterogeneous relational graph; and updating all charging station node features based on adjacent order nodes, order node information of each user, charging gun node features and charging pile node features so as to obtain final all charging station node features, calculating recommendation indexes of all charging stations, and recommending the charging stations to the users. And the recommendation result can better meet the real intention of the customer and the station operation collaboration.

The application discloses a candidate drug ranking method and system based on topological coding supervision reconstruction, and belongs to the field of bioinformatics. The method comprises the following steps: step S1, obtaining drug design data; step S2, constructing a drug-biological entity association matrix; step S3, calculating a drug side topological matrix for describing the connection relationship between drugs and a biological entity side topological matrix for describing the connection relationship between biological entities according to the association matrix; step S4, constructing a finite-order topological filter and a spectral stable topological filter to form a bidirectional topological evidence field; step S5, obtaining a reconstruction supervision matrix through a supervised reconstruction objective function; and step S6, obtaining candidate drug ranking based on the reconstruction supervision matrix. The application can improve the high-order mechanism recognition ability, robustness and interpretability in drug-target prediction, drug repositioning, candidate molecule screening and molecular generation result ranking.

The application belongs to the technical field of artificial intelligence, and specifically discloses a power customer service model optimization method and system based on artificial intelligence, which obtains a support set and a query set for large model small sample learning, uses the support set and the query set to perform meta-training on a large model, retrieves an auxiliary set to perform inference test on the pre-trained large model for power business scene problem instances, and then performs reinforcement learning on the pre-trained large model based on a comprehensive strategy reward value according to the inference steps and the inference result obtained through the test, so as to obtain a strategy optimized large model to infer and optimize the solution to the power business scene problem in actual application. The application uses small sample learning technology to realize fine-tuning of the inference ability of the large model, and uses comprehensive strategy feedback to perform reinforcement learning, thereby improving the explainability of the inference process of the large model and the accuracy of the inference result, and providing more efficient decision support for power customer service question answering of the large model.

The present application relates to the technical field of heat conduction coating life prediction, and particularly relates to an inorganic high-temperature-resistant heat conduction coating life prediction method and system, which comprises obtaining a heat pipe outlet temperature sequence, a reactor core power sequence, a heat pipe spatial position, a heat pipe adjacency relationship and a cold end boundary category; dividing heat pipe groups according to the heat pipe spatial position and the cold end boundary category; performing power coupling correction on the heat pipe outlet temperature sequence according to the reactor core power sequence to obtain a corrected outlet temperature sequence; obtaining a reference heat pipe group through spatial decoupling screening and consistency screening in a heat pipe group to which a target heat pipe belongs; calculating static difference characteristics and dynamic difference characteristics according to the corrected outlet temperature sequence, and obtaining an interface thermal resistance degradation index through weighted summation; and determining the remaining life according to the change trend of the interface thermal resistance degradation index and life determination conditions. The present application can reduce temperature response deviations caused by power disturbance and spatial coupling, and realize accurate identification and life prediction of heat conduction coating heat conduction capacity degradation.