218 results about "Neural network modeling" patented technology

The invention discloses a PCBA circuit board welding spot detection method based on multi-modal data fusion, and relates to the technical field of electronic manufacturing quality detection.The PCBA circuit board welding spot detection method comprises the steps that a distributed sensing network is constructed, multi-modal data are collected, welding spot information is obtained in an omnibearing mode, and time-space alignment of the multi-modal data is carried out; performing feature extraction on the multi-modal data, dynamically weighting each modal feature through an attention mechanism, and highlighting key defect characterization; a welding spot spatial topological graph is constructed by using a graph neural network, and a spatial relationship between welding spots is modeled. By integrating optical, X-Ray, thermal, mechanics, electricity and other multi-dimensional data, the information limitation of single-mode detection is broken through, the complementation of different mode data is utilized, the attention mechanism is combined to dynamically weight each mode feature, the complex defect is accurately identified, the graph neural network is utilized to model the welding spot space topological relation, the associated defect is further captured, and the defect detection accuracy is improved. And the defect classification accuracy is improved.

The invention relates to the field of digital twinborn modeling, and discloses a digital twinborn visual modeling method and system based on a neural network, and the method comprises the steps: collecting the multi-dimensional perception data of a target physical system, carrying out the format unification and normalization processing of the original data of a sensor through a heterogeneous data fusion module, and obtaining a data fusion module; constructing a high-dimensional input feature set for neural network modeling by combining a structured embedding algorithm; carrying out stability pre-evaluation on the constructed input feature set, and screening core features by adopting a disturbance sensitivity analysis mechanism; a dynamic residual feedback mechanism is introduced to carry out enhanced training on the preliminary twin model, and modeling is carried out on the space-time dependency relationship of different components through a graph neural network; tracking a stability index of visual representation in real time in a model training process; and according to a multi-dimensional visualization result output by the final twin model, performing interpretation in combination with an industrial scene semantic rule base. The method has the advantage of improving the practicability of the twin model in the industrial scene.

The invention discloses an industrial equipment fault prediction method based on multi-modal data, and belongs to the technical field of specific calculation models, and the method comprises the steps: carrying out the preprocessing according to the collected multi-modal data of the operation of industrial equipment, so as to unify the format of the multi-modal data, and obtaining the structural data; extracting features of the structured data one by one according to data categories, and obtaining a multi-modal fusion feature through a dynamic fusion mechanism; according to the multi-modal fusion features, a fault prediction classification score is obtained through a deep neural network model to perform fault prediction; and when the drift parameter of the multi-modal data is greater than a preset threshold value, performing incremental training on the deep neural network model through the multi-modal data collected in real time to update parameters of the deep neural network model. Through multi-modal data unified processing, dynamic feature fusion, deep neural network modeling and an online learning mechanism, the problems of insufficient multi-modal data fusion, prediction uncertainty quantization deficiency, poor model adaptability and the like are solved.

The invention discloses a multi-modal visual arrangement recommendation method, belongs to the technical field of artificial intelligence and data visualization crossing, and realizes visual arrangement recommendation based on multi-modal input analysis, a dynamic mixed recommendation model and an intelligent optimization algorithm. Comprising the following steps: multi-modal intention analysis: realizing intelligent analysis of multi-modal input through combined use of a base model and a fine tuning model, realizing high-precision intention classification in combination with a pre-training language model and a domain adaptation fine tuning technology, and triggering dynamic prompt word recommendation; performing intelligent layout generation: performing global optimization of component space allocation by adopting a genetic algorithm, performing business rule adaptation by combining a constraint solver, and modeling an interaction relationship between components by utilizing a graph neural network; and dynamic mixed recommendation: constructing a three-level recommendation architecture including collaborative filtering, content matching and reinforcement learning. According to the method, a closed-loop recommendation process of user intention-intelligent recommendation-feedback optimization is realized, and the intelligent level of visual arrangement and the user experience are remarkably improved.

The invention provides an intelligent data production method and system based on a graph neural network and adaptive learning, and the method comprises the steps: collecting a text, a time sequence, an image and sensor data in real time, and converting unstructured data into a structured JSON format through a format analysis tool; extracting text entity features based on a large language model, modeling a cross-modal entity association relationship in combination with a graph neural network, and fusing multi-source heterogeneous data through a dynamic knowledge graph; the detection threshold is dynamically optimized by adopting a reinforcement learning algorithm, and the false alarm rate is reduced in combination with a double-track verification mechanism; and predicting a data trend by using the space-time diagram convolutional network, and outputting an interpretability analysis report through the generative large model. According to the invention, a multi-protocol adaptation engine and a distributed stream processing architecture are adopted, and a protocol analysis layer automatically identifies a plurality of industrial protocols, so that the manual adaptation time is reduced; and real-time synchronization of production line-level data is realized through edge node parallel acquisition and compression transmission.

The invention discloses an attention state recognition neural network modeling and reasoning method based on an electroencephalogram sequence. The core is to construct and train a deep neural network model suitable for electroencephalogram signals so as to realize intelligent recognition and classification. Firstly, wavelet transformation and time-frequency feature extraction are carried out on electroencephalogram time sequence signals, and a multi-dimensional input tensor is generated in combination with channel position information; and inputting the feature into a deep network fusing spatial convolution, gating circulation and a residual connection structure, and extracting spatio-temporal joint features. A cross-time-step attention mechanism and a dynamic loss adjustment strategy are introduced in a training stage, so that the discrimination capability of the model on an alertness state transition region is improved. The final model can conduct reasoning on electroencephalogram data of any length, and a classification label and a confidence score are output and used for measuring classification reliability. The method focuses on construction and optimization of a specific calculation model, reflects application characteristics of an intelligent algorithm in cognitive state recognition, and belongs to an intelligent calculation method with a neural network as a core.

The invention discloses a distributed power supply access-oriented power distribution network fault distance measurement method and system, and relates to the technical field of power systems, and the method comprises the steps: collecting the electrical parameters and operation states of distributed power supply access nodes in a power distribution network in real time, building a dynamic manifold model based on an ecological niche theory, and carrying out the calculation of the dynamic manifold model; adaptively adjusting manifold learning neighborhood parameters according to power fluctuation data included in the electrical parameters, and updating node ecological niches to reconstruct a dynamic manifold model; based on the reconstructed dynamic manifold model, fault features are extracted from three scales of a current harmonic component, a feed line inter-harmonic propagation path and whole network voltage influence, and a three-dimensional feature vector is generated through fusion of a graph correlation algorithm; based on the expanded fault sample library and the power fluctuation data, constructing a fault transfer relation model to predict a ground fault and a short circuit risk area; a fault source is modeled by adopting a topological neural network, and a fault point distance measurement value is output through state prediction and strategy deduction.

The invention provides an online monitoring method and system based on a power transmission line, and relates to the technical field of power transmission line monitoring. According to the method, the heterogeneous sensing terminal, edge calculation, the graph neural network and Bayesian reasoning are combined, multi-source data acquisition, state identification and risk prediction are realized, the fault diagnosis accuracy and the risk early warning capability are improved, and the intelligent level and the safety guarantee capability of power transmission line operation are enhanced; the monitoring data is analyzed in real time through an edge calculation unit to generate a state label, a potential fault mode is recognized by combining graph neural network modeling space-time relevance, a risk factor library is further constructed, and a real-time fault probability graph is generated based on a Bayesian network. And dynamic identification and early warning of risk types such as wire strand breakage, icing overrun and mechanical fatigue can be realized.

The invention relates to the technical field of voice processing, can be applied to service system platforms of medical health, financial science and technology, communication and the like, and discloses an audio compression and reconstruction method, device, equipment and medium. Residual vector quantization and inverse quantization are carried out to generate basic spectrum features; and compensating a high-frequency component by adopting a spectrum expansion network, predicting a time domain error in combination with a long-short-term memory network, generating a time domain residual compensation signal, and superposing the time domain residual compensation signal with the reconstructed time domain audio signal to obtain a target audio signal. Through residual vector quantization and neural network modeling, the audio reconstruction quality under low-bit-rate compression is improved, spectrum detail reservation and time domain error compensation are optimized, meanwhile, the calculation complexity is reduced, and the method is suitable for high-sampling-rate and resource-limited scenes.

The invention provides a feeder terminal fault detection method, system and device, a medium and a program product, and the method comprises the steps: obtaining multi-source heterogeneous data comprising feeder terminal operation data, a topological graph structure of a power distribution network and external sensing data, the multi-source heterogeneous data comprises at least one kind of structured or unstructured time sequence data, and the external sensing data comprises at least one kind of structured or unstructured time sequence data; the external sensing data comprises meteorological data, geographic space information and historical fault records; and preprocessing the multi-source heterogeneous data, inputting the preprocessed time-aligned multi-source heterogeneous data into the trained time-space diagram neural network model to extract features and perform joint modeling, and outputting a fault type classification result and a fault probability distribution result corresponding to each feeder terminal node in the power distribution network. According to the method, a multi-source heterogeneous data fusion and time-space diagram neural network modeling mechanism is introduced, so that the model has a dynamic response capability to complex environment changes, and the identification precision of a potential fault mode is effectively enhanced.

The invention relates to a collaborative obstacle avoidance method and system based on multiple unmanned aerial vehicles, and the method comprises the steps: enabling each unmanned aerial vehicle to scan an environment, capturing dynamic obstacle information, carrying out the modeling of a LSTM neural network, generating a future motion trajectory probability graph, and converting the future motion trajectory probability graph into a dynamic thermodynamic diagram; then, traditional path key points are upgraded to space-time key points, and an obstacle avoidance path is planned according to the time-space key points; broadcasting a path key point sequence by each unmanned aerial vehicle, detecting conflicts and triggering priority arbitration, and adding a waiting ring to the unmanned aerial vehicle subjected to arbitration delay; meanwhile, continuously monitoring the environment, and immediately triggering emergency obstacle avoidance if the dynamic obstacle trajectory is not consistent with the prediction; according to the method, comprehensive perception is realized through multi-source dynamic data acquisition and fusion, a thermodynamic diagram is generated by space-time trajectory modeling, collaborative path planning and conflict detection arbitration are carried out, paths can be dynamically adjusted and optimized according to environmental changes, obstacle avoidance strategies are adjusted, and the obstacle avoidance efficiency is improved by means of emergency obstacle avoidance and safety redundancy design. And efficient and safe collaborative operation of the multiple unmanned aerial vehicles in a complex dynamic environment is ensured.

The invention discloses an intelligent charging and discharging management method and system for a lithium battery, and the method comprises the steps: collecting multi-dimensional parameter data of the lithium battery, and generating a data set with a timestamp; and analyzing the data features based on the battery type classification model, and determining a battery type identifier. And utilizing the recurrent neural network to model the relevance between the capacity attenuation and the health state, and predicting the residual capacity and the health score. If the capacity or health score is lower than the threshold value, extracting the environment temperature and the load demand to generate a temperature-load feature vector; and matching the candidate strategy set from the pre-established index database through the hash table. And distributing weights according to health scores and load demands, sorting strategy efficiency and life influences by adopting a linear regression model, and screening an optimal strategy. And if the strategy calculation complexity exceeds the equipment capability, iteratively optimizing parameters by utilizing a genetic algorithm, simplifying a strategy instruction, generating a charging current, a voltage curve and a discharging rate control instruction, and executing the charging current, the voltage curve and the discharging rate control instruction in real time by a battery management system. The method prolongs the service life of the battery and improves the charge-discharge efficiency.

The invention belongs to the technical field of PM10 concentration prediction, and discloses a PM10 concentration prediction method based on a space-time diagram neural network and an expert hybrid model, and the method comprises the following specific steps: S1, time feature extraction (RTAF): the PM10 concentration is influenced by a plurality of time factors, including short-term fluctuation, medium-term trend and long-term trend; a dynamic multi-modal weighted graph is constructed, meteorological factors, geographic positions and historical pollution similarities are coded into features of edges and nodes, a PM10 spatial propagation mechanism is modeled based on an adaptive graph neural network, a residual attention fusion module is introduced into the model in the time dimension, multi-scale time dependence features are effectively extracted, and the time-dependent features are extracted. According to the method, a long-term trend and a short-time fluctuation process are captured, finally, dynamic modeling and expert selection are performed on a complex PM10 propagation mode by using an expert hybrid network, the prediction robustness and generalization ability are improved, the model fully fuses a PM transmission mechanism and a depth space-time modeling ability, and high-precision prediction of the PM10 concentration in the next 24 hours is realized.

The invention discloses a power equipment defect detection method based on image recognition, and belongs to the technical field of power equipment defect detection, and the method comprises the steps: carrying out the defect simulation based on physical mechanism driving according to an equipment three-dimensional model and physical field simulation parameters, and obtaining a defect simulation data set; according to the defect simulation data set and the real inspection data, training a cross-modal deep learning network based on physical law constraint to obtain a defect identification model; performing time-space diagram neural network modeling according to the historical time sequence inspection data and the defect identification model to obtain a state evolution model; and inputting inspection data acquired in real time into the equipment health state evolution model, and performing online reasoning to obtain a defect detection result. The problems that an existing electrical equipment defect detection method excessively depends on scarce real defect samples, the generalization ability for complex working conditions is weak, and the defect evolution trend prediction ability is lacked are solved.

The invention relates to the technical field of capacitor manufacturing and state monitoring, in particular to a capacitor aging curve optimization method based on deep learning, which comprises the following steps: step 1, multi-modal data acquisition and standardization processing; 2, modeling and training a physical constraint neural network; 3, executing a dynamic control and maintenance strategy: dynamically adjusting the switching frequency of the inverter based on a comparison result of the residual life prediction value output in the step 2 and a preset threshold value, and enabling the ripple frequency to deviate from a capacitor impedance sensitive frequency band; and according to the aging mode classification result output in the step 2, triggering a corresponding temperature sampling frequency adjustment or bus voltage reduction operation. Through fusion of multi-modal data, a physically constrained neural network model and a dynamic control strategy based on a prediction result, the aging process of the capacitor is successfully optimized, the use efficiency of the capacitor is improved, and the life cycle of the capacitor is effectively prolonged.

The invention provides a construction method and system of a rock-soil shear strength parameter prediction model, and relates to the technical field of model construction.The construction method comprises the steps that CT scanning data of a real rock-soil sample is obtained and preprocessed, and an irregular three-dimensional particle set for geometric modeling is obtained; performing particle surface roughness modeling analysis on the three-dimensional particle body set to obtain a surface roughness coefficient set of the three-dimensional particle body set; performing particle size cumulative distribution function segmentation processing and fractal dimension control on the surface roughness coefficient set to obtain a structure grading sample set; constructing a THMC multi-field coupling model based on the structure grading sample set to obtain a working condition particulate matter physical response parameter set; and performing a simulation experiment on the working condition particle physical response parameter set, and performing deep neural network modeling processing based on an experiment result to obtain a strength response model capable of predicting the cohesive force and the internal friction angle of the rock-soil body under any working condition input. According to the invention, the prediction efficiency and accuracy are significantly improved.

The invention relates to the technical field of communication, in particular to a high-precision positioning method based on 5G. Comprising the following steps: collecting signal data through a 5G base station, building a hierarchical positioning model by using a graph neural network modeling environment and signal features, and optimizing the weight in real time; a mixed near-far field algorithm is used for cooperatively processing signals, and the positioning precision is improved. An error correction model is constructed based on transfer learning, and adaptive optimization in a complex environment is realized. A user position is dynamically tracked through sparse representation learning, prediction is performed in combination with Kalman filtering, a positioning result is updated in real time by using a 5G network, feedback optimization is performed, and high-precision positioning is ensured. According to the method, the positioning precision is improved, the environmental adaptability is enhanced, an error correction mechanism is perfected, a complete high-precision positioning system is formed, and the positioning method has higher reliability and efficiency in a complex environment.

The invention discloses a digital twinborn monitoring method for a self-evolution concrete-filled steel tube arch bridge, and the method comprises the steps: collecting multi-source monitoring data through deploying a multi-mode sensor at a key component of a bridge, and constructing an initial structure topological graph; and dynamic evolution and anomaly detection of the structure topology are realized by using a graph neural network and a Transform structure. Intelligent recognition and fault diagnosis of a structure state are realized by constructing a multi-physical field five-dimensional tensor and a health knowledge graph. And finally, generating and optimizing a structure alternative scheme by adopting a graph generative adversarial network and a genetic algorithm, verifying the performance of the scheme through finite element simulation, and realizing intelligence of bridge health monitoring and maintenance decision making. According to the method, real-time sensing, dynamic modeling and intelligent diagnosis of the structural state of the concrete-filled steel tube arch bridge are achieved by fusing multi-modal sensor data, multi-physical field coupling analysis, graph neural network modeling and an intelligent reasoning mechanism, the real-time performance and accuracy of monitoring are improved, and a scientific basis is provided for structural optimization design of the bridge.

The invention relates to the technical field of ship design optimization, and discloses a hull shape optimization method based on neural network modeling. In the data acquisition stage of the method, initial appearance parameters and hydrodynamic performance data of a ship body are obtained, the appearance parameters comprise geometric dimensions and shape features, and the performance data comprise resistance coefficients and wave-making resistance values. In the neural network construction stage, a neural network model with a multi-layer perceptron structure is trained by using collected data, weights are updated through a back propagation algorithm, and a nonlinear mapping relation between appearance parameters and hydrodynamic performance indexes is established. In the shape optimization stage, the trained neural network model is used for carrying out iterative adjustment on the shape of the ship body, fluid dynamic performance indexes are recalculated through the model after each adjustment until preset convergence conditions are met, and finally optimized ship body shape data are output. According to the method, partial complex calculation is replaced by the neural network, and intelligent optimization of the hull appearance is realized.

The invention relates to an inertial sensor sensitive structure design method based on a generative adversarial network algorithm, and the method comprises the steps: defining geometric parameters and material parameters of an inertial sensor sensitive structure, and generating a three-dimensional geometric configuration of the inertial sensor sensitive structure; performing finite element simulation on the three-dimensional geometric configuration to obtain a simulation result containing performance indexes; adjusting the geometric parameters and the material parameters for a plurality of times, generating a large number of simulation results, forming a simulation database, and forming a sample data set formed by structure parameter combination and target performance pairing according to the simulation database; a generative adversarial network is adopted, the sample data set is input into the generative adversarial network for training, a reverse structure generation model is obtained, and the reverse structure generation model is adopted for prediction; through an innovative framework of simulation data driving, neural network modeling and closed-loop optimization, a systematic solution is provided for sensor development which is high in performance, low in cost and rapid in iteration.

The invention discloses an equipment working state recognition method based on a voiceprint recognition model, and relates to the technical field of industrial equipment operation state recognition. The equipment working state recognition method based on the voiceprint recognition model comprises the following steps: collecting operation audio waveform data of target equipment, extracting acoustic representation data containing parameters such as short-time energy, a frequency spectrum centroid, a spectrum flux, MFCC and a zero crossing rate, inputting the acoustic representation data into a pre-trained voiceprint recognition model to extract voiceprint feature representation vectors, and carrying out voiceprint feature representation on the target equipment; according to the method, the audio signal is divided into the frames, the acoustic features such as short-time energy, spectrum centroid, spectrum flux, MFCC and zero crossing rate are extracted, the inter-frame evolution relation is modeled in combination with the bidirectional neural network, and the attention mechanism is introduced to highlight the key frame segment, so that the real-time performance of the audio signal is improved, and the real-time performance of the audio signal is improved. The recognition capability of working conditions such as fuzzy state boundary or unobvious transition is effectively enhanced, and the time sequence analysis and state judgment precision is improved.

The invention belongs to the cross technical field of artificial intelligence and image recognition, and discloses a cursive script image sequence recognition method for modeling by fusing a stroke structure, and the method comprises the steps: preprocessing and normalizing a cursive script image, eliminating interference information, normalizing the form of the cursive script image, and guaranteeing the stability and effectiveness of a subsequent processing flow; based on stroke decomposition and feature coding of structural analysis, stroke-level structural features are extracted through character skeleton modeling, and cursive writing characters are converted into time series data; modeling a cursive script recognition neural network fused with structural semantics, and recognizing a stroke time sequence by adopting a BiLSTM neural network; an identification output and result optimization module is designed to ensure the final identification quality; through application of scene integration and platform deployment interface design, practical deployment of the model and butt joint of a front-end platform are realized, and a complete cursive script recognition ecology is constructed. According to the method, not only is the structured representation capability of the cursive script image improved, but also the recognition precision and generalization performance of the model on complex handwriting are remarkably enhanced.

The invention specifically relates to a photovoltaic building design and control system based on a multi-modal neural network, and relates to the technical field of building energy saving, renewable energy utilization and artificial intelligence application. Comprising a multi-modal data acquisition and fusion module, a multi-modal neural network modeling and optimization design module, a real-time control strategy generation and execution module and a digital twin platform and continuous learning module. According to the method, global optimization is designed, powerful nonlinear fitting and feature fusion capabilities of the multi-modal neural network are utilized, multi-dimensional complex factors such as climate, buildings, users and a power grid are comprehensively considered, the optimal BIPV integration scheme which is high in power generation efficiency, small in influence on building performance and good in economical efficiency is rapidly generated, and the design efficiency and the scheme quality are remarkably improved.

The invention discloses a multi-task adaptive learning method based on dynamic strategy switching. The method comprises the following steps: S1, constructing a multi-task learning model, and initializing a task encoder, a feature extraction network and a decoding module; s2, encoding an input sample into a task encoding vector; s3, extracting task feature representation; s4, inputting an improved Cross-pitch structure, and carrying out the feature cross fusion of the improved Cross-pitch structure; s5, inputting a strategy scheduling controller, and generating fusion weight and path configuration based on loss change, gradient difference and feature distance; s6, executing nonlinear cross fusion to generate shared feature representation; s7, generating a task prediction result; s8, calculating loss and updating parameters; and S9, circularly training until convergence. According to the method, adaptive regulation and control of a fusion strategy and dynamic optimization of a feature sharing structure are realized, and the method is suitable for a multi-task neural network modeling scene.

The invention discloses an error compensation method of a strain type six-dimensional force sensor based on multi-source sensing information fusion, and belongs to the technical field of intelligent sensors. The error compensation method comprises the following steps: data acquisition; preprocessing and fusing data; constructing an error prediction model by adopting a time sequence neural network; training data construction and model training; performing online error compensation and feedback; and verifying, adjusting and optimizing the system. Based on fusion of multiple sensors (IMU, a temperature module and a timer) and time sequence neural network modeling, combined online compensation of errors caused by gravity, temperature drift and time drift of the six-dimensional force sensor is achieved, the method has the advantages of being high in compensation precision, high in response speed and high in adaptive capacity, and the application requirements of various high-precision measurement and control systems are met.

The invention relates to an inertial sensor sensitive structure design method based on a neural network interpolation algorithm, and the method comprises the steps: defining geometric parameters and material parameters of an inertial sensor sensitive structure, and generating a three-dimensional geometric configuration related to the inertial sensor sensitive structure; performing finite element simulation on the three-dimensional geometric configuration to obtain a simulation result containing performance indexes; adjusting the geometric parameters and the material parameters for a plurality of times, generating a large number of simulation results, forming a simulation database, and forming a sample data set formed by input parameter combination and target performance pairing according to the simulation database; a feedforward neural network is adopted, the sample data set is input into the feedforward neural network for training, an interpolation prediction model is obtained, and the interpolation prediction model is adopted for prediction; through an innovative framework of simulation data driving, neural network modeling and closed-loop optimization, a systematic solution is provided for sensor development which is high in performance, low in cost and rapid in iteration.

The invention provides an intelligent bill recognition method based on a deep neural network model, and relates to the technical field of intelligent bill recognition based on the deep neural network model.The intelligent bill recognition method comprises the steps that various bill data are collected, image features and text features are extracted respectively and preprocessed, and complete multi-modal data support is provided for subsequent modeling; secondly, modeling image features by using a convolutional neural network and modeling text features by using a deep neural network, splicing the image features and the text features into a joint feature vector, and deeply fusing information of an image and a text modal through a cross-modal attention mechanism to enhance multi-modal relevance; and finally, designing a dynamic weight adjustment mechanism, adaptively adjusting the optimization weights of the cross entropy loss of the classification task and the mean square error loss of the field analysis task based on the cross entropy loss of the classification task and the mean square error loss of the field analysis task, and balancing optimization conflicts in a multi-task scene.

The invention relates to the technical field of industrial artificial intelligence and intelligent manufacturing, and provides a defective product root cause analysis method and system based on a multi-modal space-time diagram neural network. The method comprises the following steps: S1, constructing a dynamic factory knowledge graph; s2, multi-modal space-time diagram neural network modeling and risk prediction are carried out; s3, root cause interpretation based on an attention mechanism; and S4, reinforcement learning parameter optimization of man-machine cooperation. According to the method, the whole medicine production system is abstracted into a dynamic'factory knowledge graph ', and unified modeling is performed on complex space-time association in the graph by using a multi-modal space-time graph neural network; on the basis, autonomous and safe optimization of key process parameters is realized through a reinforcement learning module fused with expert knowledge feedback, so that a complete closed loop from sensing, diagnosis, decision making and optimization is formed.

The invention discloses an injection molding defect reason traceability system based on a knowledge graph, and the system comprises a multi-source data collection and processing module which is used for collecting and processing data in an injection molding production process; the knowledge graph construction module is used for extracting entities and relationships to generate a time-varying heterogeneous knowledge graph; the two-channel heterogeneous graph neural network modeling module is used for constructing a two-channel heterogeneous graph neural network on the time-varying heterogeneous knowledge graph; the defect traceability path searching and scoring module is used for executing traceability path searching and scoring and sorting candidate paths; the evidence fusion and cause sorting module is used for calculating a defect cause confidence coefficient and outputting a defect cause sorting result; the process adjustment scheme module is used for outputting a process adjustment scheme based on the defect cause sorting result; and the knowledge graph dynamic updating module is used for collecting an execution result of the process adjustment scheme and writing the execution result back to the time-varying heterogeneous knowledge graph. According to the invention, knowledge graph reasoning is adopted to realize injection molding defect cause intelligent traceability.

The invention discloses an intelligent fault diagnosis system for a cold and heat source center, and belongs to the technical field of fault judgment of heating, ventilation and air conditioning equipment. The system collects an operation feature vector X (t) in real time, and a state signal database stores data; a signal processing module of the upper computer subsystem adopts empirical mode decomposition X (t) to generate an IMF sequence and a residual sequence rn (t), a characteristic analysis module judges chaotic characteristics by calculating the maximum Lyapunov index of each IMF, and a neural network modeling module selects a modeling path according to chaotic / non-chaotic marks to generate a predicted value. The result integration module fuses the multi-path prediction results, compares the multi-path prediction results with a preset threshold interval, and outputs normal, early warning or fault state marks; and the result display module dynamically visualizes the state curve and the maintenance suggestion. According to the system, real-time judgment and early warning of abnormal operation of the cold and heat source equipment are realized, and decision support is provided for preventive maintenance.