116results about How to "Realize the structure" patented technology

The invention relates to a PDF extraction method and system based on deep learning and layout analysis, and the method comprises the steps: obtaining a PDF file, carrying out the element recognition of a scanning version file and a text version file through an OCR technology and a deep learning model, and enabling the elements to comprise a text, a picture, a table and a formula; dividing a page of the PDF file into a plurality of logic blocks according to the identified elements, and determining a hierarchical relationship and a sequence; deleting elements without substantial help from the identified elements, and extracting the content of the document; performing page layout analysis according to the document content, and reordering and organizing each logic block to obtain a layout; document contents are classified and analyzed, and then classified storage is carried out; and according to user requirements, splicing the classified and stored document contents in a corresponding layout to generate a document in a target format. Compared with the prior art, efficient and accurate PDF extraction is achieved, and meanwhile the important structure and content of an original document are reserved.

The invention provides a middle and primary school multi-person interactive teaching method and system based on a VR technology, and relates to the technical field of virtual reality, and the method comprises the steps: generating a model through deep learning based on pre-obtained teaching aid model data, determining subcomponents, combining a teaching aid generation engine, and constructing a virtual teaching aid; a color image and a depth image are obtained through a structured light depth camera, two-dimensional pixel coordinates are obtained, depth information is combined, a three-dimensional skeleton curve of limb movement of a student is generated, an optical motion capture result is obtained, an inertial motion capture result is obtained through inertial drift calibration based on the three-dimensional skeleton curve, and a limb movement three-dimensional track of the student is reconstructed; generating a force feedback signal by solving a kinetic equation through a collision detection algorithm based on the three-dimensional trajectory of the limb movement of the trainee; and rendering the virtualization model into a virtual teaching scene, synchronizing the limb movement three-dimensional track and the force feedback signal in real time, and generating a multi-person collaborative interaction process of the virtual teaching aid.

The invention discloses a multivariable decision processing method and device, a storage medium and electronic device.The method comprises the steps that an initial multivariable decision rule system configured by a user side based on standard DSL grammar rule information is obtained, and the standard DSL grammar rule information is the DSL grammar rule information preset for a multivariable transaction decision development scene; determining all multivariable decision rule combinations corresponding to the initial multivariable decision rule system, determining a reference transaction job flow corresponding to the multivariable decision rule combinations, and constructing a multivariable combination instance library for the initial multivariable decision rule system, and performing engineering language conflict detection processing on the initial multivariable decision rule system based on the multivariable combination instance library to obtain a target multivariable decision rule system after conflict detection, and performing transaction operation process decision processing on the transaction data by adopting the target multivariable decision rule system.

The invention discloses a coal mine production management knowledge question-answering method based on natural language processing, and the method comprises the following steps: S1, collecting environment data and user behavior data, and constructing a data set; s2, performing real-time processing on the data set; s3, constructing a user portrait by using a density-based spatial clustering algorithm; s4, constructing a knowledge graph, constructing a cognitive model by adopting a graph convolutional network and an improved Transform, and performing reasoning in combination with logistic regression and an XGBoost algorithm; s5, constructing an environment simulation model through historical data analysis; s6, generating a personalized natural language answer according to the user portrait, the knowledge graph and the environment state prediction; s7, collecting user feedback data, and adjusting model parameters and an optimization strategy by using an adaptive learning algorithm; and S8, storing the plan template by using a relational database, and continuously optimizing the plan template through historical data analysis. Natural language processing and deep learning are utilized, coal mine production management intelligent question answering is achieved, and the method has the advantages of being efficient and high in individuation.

The invention discloses an adjustable rapid constant-force packaging and bonding mechanism, and relates to the technical field of semiconductor packaging. A bonding head body is movably mounted in the mechanism main body along a first linear direction; the two sides, in the second linear direction perpendicular to the first linear direction, of the bonding head body are connected with the mechanism body through a negative stiffness structure and a linear positive stiffness structure respectively. The linear positive stiffness structure is connected with a guide head of the flexible guide mechanism; a sliding table mechanism is mounted on the mechanism main body; a sliding table body of the sliding table mechanism is connected with guide heads on the two sides through a connecting rod mechanism. The connecting rod mechanism is used for converting the displacement of the sliding table body in the first linear direction into the displacement of the guide head in the second linear direction; a thread adjusting part is movably mounted on the mechanism main body; and the thread adjusting piece is used for adjusting the displacement of the bonding head body in the first linear direction by rotating the thread adjusting piece. The mechanism designed by the invention has the function of quickly entering a constant force range, and meanwhile, different constant forces can be adjusted to be output.

The invention relates to the field of semiconductor packaging, and provides a semiconductor packaging structure and a forming method thereof, and the method comprises the steps: providing a glass substrate which is provided with a plurality of discrete glass through hole interconnection structures; an inorganic dielectric layer and a first interconnection structure located in the inorganic dielectric layer are formed on the upper surface of a glass substrate through the BEOL technology in the front-end manufacturing process of a semiconductor, and the density of the formed first interconnection structure is larger than that of a second interconnection structure formed on the lower surface of the glass substrate subsequently. The feature size of the first interconnection structure is smaller than that of a subsequently formed second interconnection structure; forming an organic passivation layer and a second interconnection structure located in the organic passivation layer on the lower surface of the glass substrate by adopting an RDL process in a semiconductor back-end process; and mounting at least one first semiconductor chip on the upper surface of the inorganic dielectric layer. The method improves the wiring density of the first interconnection structure on the upper surface of the glass substrate.

The invention discloses an academic information conflict resolution method based on evidence intensity index grading. According to the method, all conflict viewpoints and traceability chains thereof are reserved by adopting a'full-amount retention, non-merging and non-deleting 'strategy; calculating a multi-dimensional weighted evidence intensity index of a comprehensive journal influence factor, an introduced frequency and an author academic influence for each literature; judging the conflict severity according to the difference value of the evidence intensity indexes of the two conflict parties, wherein the smaller the difference value is, the higher the severity is; modeling a conflict cluster into an auxiliary node layer in a graph database, and superposing the auxiliary node layer on a core graph without interfering with an ontology framework; conflict severity ranking results drive calibration reduction of downstream weight calculations. The method can be used for information conflict management in academic knowledge graph construction, and interpretable conflict severity grading reference is provided.

The invention provides an MMC-CLCC novel DC power transmission system structure and a control method, and relates to the technical field of high-voltage DC power transmission. The method comprises the following steps: establishing an MMC-CLCC hybrid DC power transmission system topology architecture; the method comprises the following steps: determining a full-bridge and half-bridge sub-module proportion of an MMC in an MMC-CLCC hybrid direct-current power transmission system topology architecture by considering a direct-current fault ride-through demand and a possible power flow reversal demand; determining a control strategy of the MMC-CLCC hybrid direct-current power transmission system based on the topological characteristics of the topological architecture of the MMC-CLCC hybrid direct-current power transmission system; and determining an on-line power flow reversal control strategy of the MMC-CLCC hybrid direct-current power transmission system according to the actual engineering capacity power flow reversal demand. The method provided by the invention solves the problem that the applicability of large-scale new energy delivery and a scene with multiple direct current drop points in a load center area is insufficient.

The invention discloses an electric power standard structuring method based on an image recognition technology, and the method comprises the steps: obtaining all characters of an electric power standard file through the OCR character extraction of an electric power standard PDF file; then, through an OCR (Optical Character Recognition) self-healing mechanism, the accuracy of identifying all characters of the power standard file by the model is improved, all accurate characters of the power standard file are obtained, then power standard meta-information is extracted, and then power standard text content entries are carried out; by deeply researching the storage form and characteristics of the electric power standard file and combining the technologies of image recognition, natural language processing and the like, accurate recognition, content extraction and structured storage of the electric power standard file are realized, and powerful support is provided for informatization construction and efficient operation of the electric power industry.

The wine cabinet or the refrigeration cabinet with the hidden door hinges comprises a cabinet body, a door body, the door hinges and cushion blocks, hinge cavities with side openings and rear openings are formed in the upper corner and the lower corner of one side of the door body, and each hinge cavity comprises a rotating hole with a lower opening and a fixing cavity; the door hinge comprises a fixed seat plate, a shaft seat plate and a hinge rotating shaft arranged on the shaft seat plate; the cushion block is in a right-angle shape and comprises a cushion part and a fixing part. The fixed seat plate is fixed with the front panel of the cabinet body, the shaft seat plate is rotationally matched with the hinge cavity, and the hinge shaft is rotationally matched with the rotating hole; the fixing part and the fixing cavity are embedded and fixed, and the cushion part is attached to the hinged cavity bottom plate and is in clearance fit with the shaft seat plate. The shaft seat plate is hidden in the hinge cavity, and the fixed seat plate is arranged in a gap between the door body and the cabinet body. Due to the adoption of the structure, the door hinge and the rotating hole are hidden and not exposed, and the top surface of the door body is flush with the top surface of the cabinet body, so that the overall appearance of the wine cabinet or the refrigeration cabinet is flat and attractive; and meanwhile, the door body is supported by the hinge rotating shafts on the two door hinges together, so that the door body is stressed in a balanced manner.

The application provides a multi-agent cooperative task state management and abnormal backtracking method and system, relates to the technical field of artificial intelligence, and comprises the following steps: obtaining interaction data, state data and execution logs, constructing an initial task knowledge graph based on an ontology model, establishing a bidirectional causal chain by using a Bayesian network to form a dynamic knowledge graph, analyzing state migration rules when an abnormality is detected, identifying a propagation node, constructing an abnormal propagation subgraph, generating an influence chain, determining a recovery priority and triggering state rollback. The method can quickly trace the abnormal source, effectively control abnormal diffusion, and improve system robustness and task recovery efficiency.

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 embodiment of the invention discloses a home service support method and device based on a large language model, and a medium, is applied to the technical field of smart questions and answers, and is used for solving the problems that the existing answer data query efficiency is low and the matching degree is low. The method comprises the following steps: processing an original service document based on a preset vectorization processing engine to obtain a vectorization index table; processing the problem of the current user through a preset vectorization processing engine, and obtaining document content corresponding to the processed problem of the current user based on the vectorization index table; determining a current home service scene through the corresponding document content so as to call a prompt word template of the current home service scene; and analyzing the corresponding document content and the current user question based on the cue word template, generating a cue word question, inputting the cue word question into a preset large language model, obtaining answer data, and outputting the answer data through a corresponding output template.

The application discloses a hospital department layout generative intelligent planning method and system, relates to the technical field of hospital department layout data processing, and comprises the following steps: acquiring hospital layout digital data, and generating hospital layout condition coding. A hospital department hard constraint graph is constructed based on the hospital layout condition coding, and a hospital department constraint residual error vector is calculated. The hospital department constraint residual error vector is input into a generative layout model, and a hospital department digital layout scheme is output. Through the closed-loop processing of the hospital layout condition coding, the hospital department hard constraint graph, the hospital department constraint residual error vector and the generative layout model, the hospital department layout generation is converted into a data processing flow that is codeable, traversable, feedbackable and iterative, and the beneficial effects of improving the layout generation constraint consistency, the data traceability and the digital scheme output property are achieved.

The invention relates to the technical field of high polymer materials, in particular to a high-flexibility polyester fluoropropylene dispersion and a preparation method thereof.The preparation method comprises the steps that linear saturated polyester diol is taken, an activating agent is added to conduct end group activating treatment on a polyester main chain, then a catalyst is added to conduct grafting treatment, and bridged polyester is obtained; introducing fluoropropylene gas into the reaction kettle of the bridged polyester, and carrying out copolymerization treatment to obtain a copolymer; cooling the copolymer to a preset value, adding a surfactant to perform coating treatment on the copolymer, and slowly adding deionized water to perform shearing treatment to obtain an aqueous dispersion; and adding tetraethoxysilane into the aqueous dispersion, and carrying out hydrolysis treatment under the condition that the pH value is 4.3-4.6, so as to obtain the high-flexibility polyester fluoropropylene dispersion. According to the invention, the elongation at break, bending tolerance and cracking resistance of the dispersion after film formation are improved, and the hidden danger of material failure under high-flexibility scenes such as flexible electronic base materials and bendable coatings is solved.

The present invention relates to the technical field of computer software for building knowledge models, and in particular to a computer application system and a generation method for quickly building and generating knowledge models, the method comprising: for objects corresponding to objects in the real world, using a metamodel as a member to build a first model, then using the first model as a member to build a second model, and then using the second model as a member to build a third model, and so on, until an object model is obtained; for object models corresponding to different selected objects, setting the model relationships of the members therein through a graphic editor, and marking the model relationships with knowledge tags to form a knowledge graph, the knowledge tags contain logic scripts that can be called to complete calculations and controls; defining process components in the form of blocks, using the process components as components of the object model, and generating a knowledge model that reflects the real world. The system comprises a model composition definition module, a model relationship definition module, and a model behavior definition module.

The present application relates to the field of optical materials, and particularly relates to a manganese-based crystal material with temperature-triggered reversible structure transition and adjustable light-emitting color and a preparation method and application thereof. The red light-emitting phase has a chemical formula of (C6H8N)3Mn2Cl7.3H2O, belongs to a monoclinic system, and a space group is P2(1) / n P2 1 / n ; the green light-emitting phase has a chemical formula of (C6H8N)2MnCl4, belongs to a monoclinic system, and a space group is P2(1) / n C2 / c . The present application provides a manganese-based crystal material with temperature-triggered reversible structure transition and adjustable light-emitting color, which has a chemical formula of (C6H8N)3Mn2Cl7.3H2O at room temperature and is converted into (C6H8N)2MnCl4 after being heated to 60 DEG C. The crystal material can be obtained by using a simple solvent evaporation method, and the preparation method is simple. The reversible removal and adsorption of crystal water by temperature or humidity can realize the reversible structure transition and light-emitting switching between the water-containing red light-emitting crystal and the anhydrous green light-emitting crystal.

This application discloses a method and system for determining pollution causes based on the correlation between atmospheric diffusion conditions and emission contributions, relating to the field of atmospheric pollution cause identification. The method includes: aligning meteorological, emission, and receptor observations to a unified spatiotemporal reference to obtain multi-source baseline data; calculating a diffusion stability index characterizing vertical mixing and horizontal transport using meteorological monitoring, generating time-varying diffusion condition weights; generating a time-response diffusion kernel from the emission source to the receptor point using a diffusion model under unit emission conditions, and constructing diffusion-contribution correlation information with observed concentrations to characterize the temporal coupling strength; establishing a diffusion-contribution dynamic optimization model, using diffusion condition weights to constrain the coupling relationship, and iteratively solving for time-varying contribution coefficients based on the convergence target of simulated and observed concentrations, outputting the dynamic contribution of each source and the pollution cause. Thus, time-varying diffusion conditions are explicitly introduced into the source contribution inversion framework, achieving stable separation and reliable attribution of source contributions.

The invention relates to the technical field of industrial equipment monitoring, in particular to a power equipment fault online monitoring system for polyester staple fiber production, which comprises a feature extraction module, a dynamic analysis module, an evolution prediction module, an anomaly recognition module, a feature construction module and a diagnosis matching module. The system collects operation data of power equipment in real time, extracts time sequence features and calculates a dynamic offset vector, and further predicts a state evolution track and identifies preliminary abnormality. The feature construction module traces back to extract feature fragments associated with the anomaly, fault component signals of different physical meanings are obtained through multi-level signal deconstruction and input into a fault feature spectrum generator, and a three-dimensional fault feature spectrum reflecting the fault evolution law is constructed. And the diagnosis matching module performs mode matching on the three-dimensional map and a preset fault knowledge base to determine a specific fault type and an evolution stage thereof. According to the method, early discovery of fault germination, accurate identification of fault types and dynamic judgment of fault evolution stages are realized.

This application provides a method for regenerating ternary cathode materials based on acid etching and molten salt lithium replenishment, as well as the regenerated electrode material and battery. The method includes the following steps: pretreating the failed ternary battery cathode material with an organic acid solution; mixing the pretreated failed ternary battery cathode material, dopant, and lithium-containing molten salt, and performing a first-stage low-temperature molten salt heat treatment at a first specified temperature; further subjecting the material after the first-stage low-temperature molten salt heat treatment to a second-stage high-temperature heat treatment at a second specified temperature for 2-4 hours; and washing and drying the product after the second-stage high-temperature heat treatment to obtain the regenerated ternary cathode material. The wet acid etching process selectively removes rock salt phase impurities and contaminants from the electrode material surface, achieving interface cleaning and activation; the molten salt method for lithium replenishment using multi-element synergistic doping introduces multiple beneficial elements for bulk phase doping while replenishing lithium, thereby simultaneously strengthening the material's crystal structure and interface stability.

The present application relates to the technical fields of air filtration material and nanocomposite material, and discloses a PVA / chitosan nanofiber air filter element prepared by electrospinning and a preparation method thereof.The filter element comprises an electrospun nanofiber felt, which is composed of a PVA matrix and chitosan nanofibril (CSF) dispersed in the PVA matrix, and the mass ratio of CSF to PVA is 0.5wt% to 10wt%.The present application introduces positively charged rigid CSF, which not only gives the filter element a particle removal efficiency as high as 97% and an extremely low pressure drop, but more importantly, CSF effectively improves the structural stability of the hydrophilic PVA matrix in a high humidity environment as a rigid skeleton, effectively solves the pain point of the traditional filter element that is easy to collapse after absorbing water, and can be quickly biodegraded in the natural environment after being discarded, and has environmental protection and commercial application value.

The utility model discloses an aluminum honeycomb panel with reinforcing ribs, which relates to the technical field of aluminum honeycomb panels and particularly comprises an aluminum honeycomb top plate, hanging lug plates are fixedly mounted on two sides of the aluminum honeycomb top plate, and connecting holes are formed in two sides of one side surface of each hanging lug plate. An aluminum honeycomb panel is fixedly installed on the surface of one side of the hanging lug plate through connecting holes, hanging lug pieces are arranged on the two sides of the aluminum honeycomb panel, the upper surface of the aluminum honeycomb panel is of a hollow structure, a binder layer is arranged on the upper surface of the aluminum honeycomb panel, and an aluminum honeycomb core is fixedly installed on the upper surface of the aluminum honeycomb panel through the binder layer. The V-shaped parts at the bottoms of the angular pressure dividing frames are arranged between the transverse reinforcing ribs in a staggered mode in a downward pressing mode, so that the V-shaped parts are stressed to be opened and abut against the surfaces of every two transverse reinforcing ribs during stress downward pressing, stretching distortion and stretching bonding of the same-direction S-shaped reinforcing rib honeycomb cavity can be effectively avoided, stress complementation and symmetry are achieved, and the strength of the honeycomb cavity is improved. And stress balance and structure symmetry are realized.

The application relates to the field of artificial intelligence and computer vision, and provides a three-dimensional point cloud scene target positioning method based on a natural language instruction, which comprises the following steps: screening a relation triple containing an object category in a three-dimensional visual scene; determining a hierarchical attribute of the object category and constructing a hierarchical knowledge graph; analyzing a given natural language description to obtain a plurality of two-order semantic triples; adding the hierarchical attribute to the object in the two-order semantic triple to obtain a hierarchical two-order semantic triple; determining an initial visual feature of an object in a three-dimensional point cloud scene to be recognized; outputting a prediction score of the object; screening the objects corresponding to the prediction scores from high to low to obtain candidate objects; outputting a matching score of the candidate objects and a first high-order semantic triple; and determining a target object in the three-dimensional point cloud scene to be recognized. The application improves the prediction accuracy and has stronger robustness.

The invention relates to the technical field of knowledge maps, in particular to a ship design reference data intelligent recommendation method and system based on a knowledge map, and the method comprises the steps: constructing a multi-dimensional ship knowledge map; analyzing the received design task to obtain a design basic parameter and a target performance index, constructing a task related sub-graph, extracting a task constraint condition, and generating a target index semantic vector; carrying out graph structure representation learning on the task related sub-graphs by utilizing a graph neural network coding model, and carrying out feature fusion on the task related sub-graphs and target index semantic vectors to generate comprehensive design reference vectors; and based on the comprehensive design reference vector and the task constraint condition, matching historical cases in the knowledge graph to obtain a design reference recommendation result. According to the method, by constructing the multi-dimensional ship knowledge graph, design basic parameters, performance indexes and constraint conditions are systematically managed; historical design cases can be intelligently matched, high-correlation design reference results are provided, and design efficiency and accuracy are improved.

This invention relates to a fully localized hybrid retrieval and knowledge graph construction method and system for standard gas R&D and production, belonging to the interdisciplinary field of artificial intelligence and industrial control. The method includes: acquiring multi-source heterogeneous documents and performing text parsing, mixed Chinese and English word segmentation, and terminology standardization; employing a hybrid retrieval strategy that weights and fuses BM25 keyword retrieval and vector semantic retrieval to obtain fused retrieval results; automatically extracting entities and relations from the fused retrieval results to generate triplet data; constructing a knowledge graph based on the triplet data, with nodes bearing source document identifiers and version tags; and deploying the embedded model, text units, semantic vectors, and knowledge graph on a local server. This invention achieves the fusion of precise matching and semantic recall, ensures core data security, transforms unstructured knowledge into a computable and traceable knowledge network, and significantly improves the efficiency and intelligence level of knowledge retrieval in standard gas R&D and production.

The invention relates to a power transmission and transformation reduction verification method and system based on a knowledge graph, and belongs to the technical field of power transmission and transformation project review. The method comprises the following steps: acquiring review file data of a power transmission and transformation station, extracting text content according to a natural language processing technology to construct a knowledge graph sub-graph, and generating a design review standard file list; performing text word segmentation based on the design review standard file list to obtain a structured review template; inputting the review file data of the power transmission and transformation station into a model, and verifying the review key points of the structured power transmission and transformation project in combination with the file integrity and the structured review template; and taking the review key points as cue words, writing the cue words into an intelligent review verification model, and generating a power transmission and transformation review report.

This invention provides a method, system, device, and storage medium for smart contract vulnerability detection, belonging to the fields of blockchain security and artificial intelligence technology. It includes extracting an abstract syntax tree, control flow graph, and data dependency graph from the original smart contract code; obtaining a composite graph object based on the abstract syntax tree, control flow graph, and data dependency graph; processing the composite graph object based on node semantics and function call relationships to obtain a hierarchical heterogeneous graph; constructing a heterogeneous graph Transformer network; using a heterogeneous graph Transformer encoder to process the hierarchical heterogeneous graph for encoding node and edge information, obtaining node representation vectors that fuse semantic and structural information; extracting multi-scale features from the node representation vectors; fusing the extracted multi-scale features using learnable weights; and processing the fused multi-scale features using a base classifier cluster to obtain vulnerability type prediction results and confidence levels. This invention achieves full-process optimization from structural modeling and semantic fusion to decision reasoning, significantly improving the accuracy and interpretability of smart contract vulnerability detection.

The invention relates to the technical field of new energy and intelligent scheduling, in particular to a shared electric vehicle solar power conversion network optimization system, which comprises a data acquisition module, a prediction module, an available energy and inventory constraint generation module, an inventory state management module, a linkage scheduling module and an execution and update module. According to the invention, the system collects and predicts solar power generation data, battery replacement demand data and battery state data in a rolling time window, constructs an available energy constraint including an energy lower bound, an energy expectation and an energy upper bound, and generates an inventory constraint based on a battery replacement demand; a battery state is divided and updated to form an inventory state, a charging scheduling instruction and an operation and maintenance supply scheduling instruction are generated under a constraint condition, operation and maintenance supply arrival time is associated with inventory constraint establishment time, and then rolling updating is performed according to an execution result, so that collaborative scheduling of a battery switching network under a solar energy supply condition is realized.

This invention belongs to the field of engine turbocharger testing technology and discloses a method and system for evaluating the transient response of engine turbochargers. The method includes: test preparation, operating condition setting, transient excitation and data acquisition, calculation of speed increment time, calculation of intake pressure to establish efficiency indicators, and two-dimensional diagnostic evaluation. This invention effectively solves the technical problem that existing evaluation parameters cannot distinguish the root cause of response lag, and has advantages such as accurate positioning and strong guidance. It is particularly suitable for the research and optimization of turbocharger systems with stringent transient response requirements, such as hydrogen fuel cell internal combustion engines.

The invention discloses an intelligent talent recruitment screening method based on multi-source data fusion, and relates to the technical field of big data management, and the method comprises the steps: collecting candidate multi-source heterogeneous data, carrying out the feature extraction, and generating a multi-modal initial feature vector; fairness related attribute data and historical recruitment result data of candidates are collected and combined with the skill confidence feature set, and a recruitment decision causal graph model is constructed; and based on a recruitment decision causal graph model, performing causal effect estimation and anti-fact reasoning by taking the multi-source verification confidence as a weight, performing fair correction calculation on candidate ability evaluation represented by the skill confidence feature set, and outputting a fair correction talent sorting list. According to the method, the multi-source evidence cluster is constructed in the shared semantic space, and the multi-source verification confidence is calculated, so that the unified management and credibility quantification of the candidate multi-source heterogeneous data in the skill dimension are realized, and the controllability and traceability of a skill assessment result in a big data management scene are improved.