209 results about "Sigmoid function" patented technology

The invention discloses a fault root cause positioning method and system for a server cluster, and relates to the technical field of network fault diagnosis. According to the method, nanosecond-level synchronous acquisition of micro-service call chains, container indexes, physical nodes and network data is realized through a precise time protocol, and a consistent data set is constructed through entity association and standardized processing; a service-resource topological graph is dynamically constructed, and an inter-service calling edge weight model is innovatively designed: a real-time load factor and a historical fault index attenuation sum processed by a Sigmoid function are fused, and the weight is periodically updated to accurately quantify the inter-node influence intensity; converting the topological graph into a Bayesian network; when a fault occurs, a three-level assembly line compression alarm is adopted, frequent item sets are mined through bitmap indexes and parallel FP-Growth, and strong causal association item sets are screened in combination with topological edge weights and KL divergence; strong causal alarm is taken as evidence, probabilistic root cause sorting is output through reverse random walk sampling, and high-precision positioning of complex distributed system faults is achieved.

The invention discloses an aquaculture disease prediction method based on multi-modal data fusion, and relates to the technical field of aquaculture, and the method mainly comprises the steps: collecting aquaculture data containing structured data and unstructured text data; inputting the structured data into a TabTransform model, carrying out column embedding and modeling of a context relationship between features through a multi-layer Transform encoder, and outputting a structured feature vector; inputting the unstructured text data into a pre-trained BERT encoder, and extracting an output vector marked by the CLS as a text semantic feature; and splicing the structured feature vector and the text semantic feature into a fusion feature, inputting the fusion feature into a full connection layer, and predicting the incidence probability of each target disease through a Sigmoid function. The prediction accuracy is improved, and the problems of single prediction dimension and weak generalization ability in the prior art are effectively solved.

The invention provides a coal mine comprehensive guarantee capability dynamic coupling early warning method and system, and relates to the technical field of coal mine safety production assessment, and the method comprises the steps: constructing a multi-dimensional fusion assessment system comprising a first-level index and a second-level index according to a coal mine comprehensive guarantee demand and an assessment target; collecting corresponding multi-source heterogeneous index data under each dimension in real time; a two-stage dynamic weight distribution mechanism is adopted, an improved entropy weight method and a working condition sensing factor are combined, and the final weight of each index at the current moment is calculated; constructing a four-dimensional feature vector and forming a multi-dimensional tensor, calculating a comprehensive score of each dimension, and generating a coal mine comprehensive support capability index through weighted fusion and Sigmoid function mapping; and performing graded early warning according to the comprehensive support capability index of the coal mine and outputting an early warning signal. According to the method, dynamic coupling evaluation of multi-dimensional risks such as coal mine technology, safety, personnel and system can be realized, comprehensiveness and real-time performance of coal mine safety guarantee are improved, and closed-loop management of risk early warning and resource configuration is effectively supported.

The invention relates to the technical field of distributed micro-grids, and discloses a distributed micro-grid grid-connected protection and power balance intelligent regulation and control method, which comprises the following steps of: firstly, acquiring multi-source operation data of a micro-grid, extracting characteristics by using an intelligent analysis model to obtain an operation characteristic sequence, and calculating a grid correlation coefficient; dimensionality reduction is carried out on correlation matrixes of correlation coefficients of different regional power grids, key feature vectors are extracted, and regulation and control weights of the regional power grids are obtained through Sigmoid function conversion. And dynamically adjusting the reference regulation and control points of the operation feature sequence according to the weights, generating regulation and control parameters and updating regulation and control records. In addition, a multi-dimensional power grid regulation and control space is constructed to monitor abnormity, and a joint optimization model is established to solve an optimal regulation and control scheme. According to the method, the operation state can be accurately analyzed, and the grid-connected protection capability and the power balance regulation and control level of the distributed micro-grid are effectively improved.

The invention discloses a forest fire risk prediction method based on multi-source data, and belongs to the technical field of forest fire prevention. Aiming at the problem of low prediction precision caused by one-sided information of a single data source and insufficient multi-source data fusion in the prior art, the method is realized by the following steps: acquiring micrometeorological data including temperature and humidity, wind speed and air pressure, and image data including an infrared image and a visible light image; the data of the mountain fire-prone area comprises historical fire frequency, vegetation type and topographic information; uTC + 8 time synchronization and WGS84 coordinate system space calibration are carried out on the data, and missing values and abnormal values are processed; carrying out feature layer fusion by adopting an attention mechanism, and extracting core features such as a temperature and humidity coupling index and vegetation dryness; spatial correlation features are captured through CNN, a time sequence trend is captured through LSTM, a mountain fire occurrence probability is output by using a Sigmoid function after decision-making layer fusion, and a result is calibrated in combination with sub-region features. Through multi-source data deep fusion and spatial-temporal feature collaborative analysis, the accuracy and timeliness of forest fire risk prediction are improved, a new data source can be expanded and accessed, and the method is suitable for a complex forest fire prevention scene.

The invention discloses a photovoltaic power station defect detection method based on thermal imaging data, and relates to the technical field of artificial intelligence, and the method comprises the steps: S1, collecting the thermal imaging data of a photovoltaic module under different working conditions, and carrying out the preprocessing of the thermal imaging data, and obtaining a thermal imaging data set; s2, based on the thermal imaging data set, dynamically dividing a temperature interval and calculating an adaptive bandwidth, and performing weight distribution and color band center weighted fusion by using a Sigmoid function to generate an adaptive pseudo-color image capable of enhancing local contrast; s3, constructing a double-branch defect detection model based on the original temperature data and the pseudo-color image; and S4, performing defect detection on the real-time thermal imaging data based on the double-branch defect detection model to obtain a defect detection result of the current photovoltaic power station. The method can be effectively applied to large-scale inspection operation and maintenance of the photovoltaic power station.

The invention discloses an AI generated image detection method and device and a storage medium, and the method comprises the steps: obtaining a to-be-detected image, and inputting the to-be-detected image to an image encoder of a counterfeit detection model; extracting a depth visual feature vector of the to-be-detected image through the image encoder; performing linear operation on the depth visual feature vector and a preset weight parameter matrix, and superposing a preset offset parameter in the linear operation process to obtain an intermediate scalar value; the middle scalar value serves as input of a Sigmoid function, and a scalar confidence coefficient score is calculated; and judging whether the to-be-detected image is an AI generation image or not according to a comparison result of the scalar confidence score and a dynamic judgment threshold. According to the method, depth features are extracted through cross-modal alignment training, linear mapping efficient reasoning is carried out, the generalization ability of an unknown generation model is remarkably improved, dynamic threshold value self-adaptive adjustment is combined, lightweight deployment is achieved while high precision is guaranteed, and the actual application requirement is met.

The invention relates to a heart disease occupational health risk assessment system based on an RF-MLP model, and belongs to the technical field of health risk prediction. In order to solve the problem that prediction performance is limited due to the fact that an existing method is insufficient in the aspects of feature selection and model optimization, the invention provides a collaborative prediction scheme fusing recursive feature elimination (RFE), a random forest (RF) and a multi-layer perceptron (MLP). According to the technical scheme, the method comprises the steps that a data collecting and preprocessing module cleans and normalizes physiological indexes; the feature selection module calculates feature importance scores through an RFE-RF algorithm to screen key indexes; the random forest optimization module constructs a decision tree cluster; the MLP module is used for realizing nonlinear mapping through ReLU and Sigmoid functions; and the model fusion module integrates the output of the double models to generate final prediction. The system remarkably improves the prediction precision and generalization ability, provides high-robustness support for early warning of heart diseases of occupational people, and promotes disease prevention to be converted from passive response to active intervention.

The invention discloses a wavelet attention shadow removal method based on soft prior guidance, and belongs to the field of image processing. The method comprises the following steps: constructing a multi-scale wavelet attention codec architecture, decomposing an input image into a low-frequency component representing illumination characteristics and a high-frequency component containing texture details by using discrete wavelet transform, and establishing multi-scale frequency domain representation; proposing probabilistic soft prior modeling, adaptively generating a continuous differentiable probability distribution diagram by using a convolutional sub-network and a double-slope Sigmoid function, and guiding differentiated repair of high and low frequency features in a dynamic weight mode; and cross-domain illumination correction based on gray world hypothesis is introduced, spatial domain enhancement is performed on a frequency domain restoration result, illumination consistency of cross-domain processing is ensured, and finally a shadow-free image is output. According to the method, wavelet domain feature optimization is guided through the probabilistic soft mask, high-quality shadow removal is realized, and the robustness and visual quality of shadow removal are remarkably improved while the light weight of the model is kept.

The invention discloses a multi-dimensional information-based adaptive gateway traffic scheduling method, which comprises the following steps of: acquiring server performance indexes, network delay, bandwidth and geographic position data in real time, and establishing a historical database; performance load weights are generated through weighted calculation of CPU, memory and disk indexes, and routing priorities are dynamically adjusted by adopting exponential function mapping; performing normalization processing on network bandwidth, delay and geographic distance, and calculating a service access weight to optimize a transmission path in combination with a step function; analyzing historical traffic data based on an LSTM model to predict a future trend, and generating a traffic prediction weight through a sigmoid function to realize pre-equalization; and combining the three types of weights, weighting to generate dynamic routing configuration, eliminating abnormal nodes in real time in combination with a health check mechanism, triggering high-load early warning and linking a predictive balancing strategy. According to the method, multi-dimensional index collaborative decision-making is realized, routing distribution is dynamically optimized, network delay is effectively reduced, and system throughput and stability in a high-concurrency scene are improved.

The invention discloses a robot path planning method for warehouse logistics cluster operation, and belongs to the technical field of path planning. The method comprises the following steps: constructing a two-dimensional model diagram by using a grid method; the ant colony algorithm is improved; an improved ant colony algorithm is adopted to plan an operation path for each logistics robot; a local obstacle avoidance strategy is carried out through a dynamic window algorithm, and whether obstacles exist around is detected; and the collision type is judged, and a corresponding collision avoidance strategy is adopted until all the robots safely and correctly arrive at the preset target point. According to the method, the pheromone concentration is initialized by introducing the Sigmoid function, blind search in the early stage is avoided, the self-adaptive slope parameter is introduced, and the steep degree of the Sigmoid function is dynamically adjusted according to environmental characteristics; meanwhile, the ant colony algorithm and the dynamic window algorithm are fused, parameters are dynamically adjusted according to the number of iterations, the obstacle rejection weight is introduced, the algorithm is prevented from falling into local optimum, and global path planning and local dynamic obstacle avoidance are achieved.

The invention discloses a reentry attack detection system and method based on a time sequence transaction aggregation graph network, and the method comprises the steps: constructing a time sequence multilateral directed transaction graph through a transaction graph generation module, constructing and enhancing the node and edge representation of a block chain transaction graph through a multilayer mechanism in a time sequence transaction aggregation graph network construction module, and the attack detection module carries out weighted fusion on the node features, the time sequence features and the structural features and then inputs the fused features into a logistic regression classifier, a probability value of a reentry attack behavior is calculated through a Sigmoid function, classification judgment is carried out according to a set threshold value, and a reentry attack detection result is obtained. The reentry attack detection method provided by the invention has a real-time monitoring capability, can accurately identify an ongoing reentry attack behavior, and effectively prevents further loss of contract funds through an instant interception mechanism. The method has high adaptability to a continuously evolving block chain transaction environment, and a novel unknown reentry attack mode can be continuously and effectively detected.

The invention provides a quantitative risk assessment method and system for an ice rock collapse-river blockage-dam break flood disaster chain. The method comprises the following steps: acquiring an ice rock collapse risk grade index by adopting a BP neural network improved by an Adam optimizer; if the ice rock collapse occurs, correcting and calculating an ice rock collapse movement path based on a D8 algorithm in combination with a rock-soil friction coefficient, correcting the volume in real time through an erosion-accumulation dynamic balance model, and constructing a river blocking risk grade index in combination with the river entering volume and speed; the stability of the dam body is evaluated through a dynamic catchment area model and a Sigmoid function, and the dam break risk is calculated in combination with the ratio of the volume to the area; and if the dam break disaster does not occur in the barrier dam, integrating the ice rock collapse risk grade, the river blocking risk grade and the dam break risk grade index, and evaluating the comprehensive risk of the disaster chain. According to the method, disaster prediction of each stage is realized through neural network improvement, path optimization analysis and variance tracking change, and comprehensive risks are evaluated through multi-element coupling.

The invention relates to the technical field of electric power systems, and discloses a GFL-GFM parallel system fault ride-through control method, system, device and medium, according to the method, under the condition that a GFL-GFM parallel system breaks down, a magnetic flux coupling current limiter is used for achieving fault ride-through, and when the fault of the GFL-GFM parallel system recovers to be normal, fault ride-through is achieved. When the power angle fluctuation of the GFL system and the power angle fluctuation of the GFM system are both larger than a preset power angle fluctuation threshold value, a phase-locked loop parameter of the GFL system and a control parameter of the GFM system are set based on a Sigmoid function until the power angle fluctuation of the GFL system and the power angle fluctuation of the GFM system are both not larger than the preset power angle fluctuation threshold value, the equivalent inertia and damping characteristics of the GFL system and the GFM system are enhanced, and the damping performance of the GFL system and the GFM system is improved. And the whole process of the GFL-GFM parallel system from a fault transient state to fault recovery is controlled, and the robustness of fault ride-through control of the GFL-GFM parallel system is improved.

The invention relates to the technical field of PWM (Pulse Width Modulation), and discloses a method for optimizing double random PWM, which comprises the following steps: randomizing a carrier frequency fc and a pulse position to obtain double random PWM, performing parameter setting on random distribution of the carrier frequency and the pulse position by adopting a Sigmoid function, improving a genetic algorithm by utilizing a particle swarm algorithm, and optimizing the double random PWM. Optimizing a minimum value fmin of a random number of a carrier frequency or a pulse position, a maximum value fmax of the random number of the carrier frequency or the pulse position and a curve transverse adjustment parameter sigma of random number Sigmoid distribution in the double random PWM by adopting an improved genetic particle swarm algorithm; according to the method, harmonic energy is dispersed from frequency and phase dimensions to a wider range through double random PWM, the switching frequency and the harmonic amplitude at the integral multiple of the switching frequency are reduced, random numbers are distributed in a Sigmoid mode, and the carrier frequency and the pulse position can more easily avoid a sensitive area by means of the nonlinear probability density characteristic that samples are more on the two sides and less in the middle, so that the switching efficiency is improved. And EMI peaks caused by extreme parameters in traditional uniform distribution are reduced.

The invention belongs to the technical field related to deep learning, and discloses a training method and application of a three-dimensional flow field prediction system of an underwater vehicle, and the training method comprises the steps: calculating the mass center of a neighbor point set for each surface grid point of a vehicle model, constructing a covariance matrix between the neighbor point set and the centroid of the neighbor point set, and performing eigenvalue decomposition to obtain a normal vector of the point; calculating a normal vector included angle between the surface grid point and each point in the neighbor point set, taking the obtained maximum included angle as the geometric feature measurement of the point, and converting the geometric feature measurement into a weight through a Sigmoid function, thereby obtaining the sampling probability of each point; randomly extracting surface grid points from the original surface grid of the aircraft based on the obtained sampling probability to obtain a point cloud of a corresponding model; and training a point cloud neural network by using the sampled point cloud data to obtain the three-dimensional flow field prediction system of the underwater vehicle. Based on the method, the prediction precision and the flow field detail recovery capability can be improved while the calculation efficiency is maintained.

The invention discloses a numerical control machine tool thermal error prediction method based on dynamic physical information fusion, and belongs to the technical field of intelligent manufacturing and precision machining. The method creatively introduces a Bayesian dynamic weight adjustment mechanism and a multi-stage joint optimization strategy through hierarchical fusion of a physical mechanism and a data driving model, and specifically comprises the following steps: establishing a lightweight analysis model based on a thermal deformation mechanism; key temperature and displacement data are collected through a thermal characteristic test, and model parameters are fitted; constructing a fusion prediction model based on Gaussian process regression, taking a mechanism model as a mean value function, and combining global and local kernel function combinations to fit the time-varying characteristics of thermal errors; based on the distribution consistency of a KL divergence dynamic evaluation mechanism and data prediction, generating an adaptive weight factor through a Sigmoid function; a double-stage training strategy is adopted, kernel parameters are optimized through pre-training, dynamic weight adjustment is gradually introduced, and collaborative optimization of mechanisms and data is achieved. According to the method, the thermal error prediction precision (the root mean square error is less than or equal to 0.6 mu m) is remarkably improved while the physical interpretability is ensured, the adaptability of the model to multiple working conditions is enhanced through a dynamic weight mechanism, and the method is suitable for real-time monitoring and prediction of thermal deformation of a high-precision numerical control machine tool.

The invention discloses a device and method for improving synchronous tracking precision of a double-permanent-magnet synchronous motor motion platform servo system, and belongs to the technical field of numerical control. Synchronous control of the double permanent magnet synchronous motors is realized by using the preset performance cross coupling controller; estimating system lumped disturbance through a self-adaptive extended state observer, and inhibiting the system lumped disturbance by adopting a novel sliding mode maker; according to the novel sliding mode controller, an exponential function term and a power function term are added in the reaching law to enable the system state to converge rapidly, when the system state is close to a sliding mode surface, the exponential term in the reaching law is attenuated rapidly, the improved power function term plays a leading role, convergence stagnation caused by premature reduction of the reaching rate is avoided, the adaptability of the reaching rate is enhanced, and the control precision is improved. The arrival time is reduced. A sigmoid function is used for replacing a switching function, continuous control input is achieved, and high-frequency buffeting is weakened. And the high-performance Hall sensor and the incremental encoder are adopted, so that the signal acquisition precision is improved.

The invention relates to the technical field of network acceleration, and discloses an edge-end large language model reasoning acceleration method and accelerator, and the method comprises the following steps: reconstructing a calculation process of a decoding stage, and carrying out the deep fusion of a multi-head attention mechanism and the calculation operation of a feedforward network; the weight and key value data are stored in HBM, and the coefficient and the accumulated attention score are stored in DDR; for linear matrix calculation, a unified matrix calculation unit is used for executing multi-precision matrix operation; for nonlinear function calculation, a mathematical transformation and linear fitting method is adopted, a Softmax function is converted into operation with 2 as the bottom through a bottom conversion formula, and truncation and third-order linear fitting are conducted on a Sigmoid function; and constructing a key value screening algorithm based on the accumulated attention score, dynamically adjusting a key value storage position, maintaining a recent key value cache region and an important key value cache region in a limited cache space, and realizing key value efficient cache in long text reasoning.

The invention relates to the technical field of image data processing, in particular to an MBR water production turbidity detection method and system based on image processing, and the method comprises the steps: obtaining an original image of an MBR water production pipeline, and extracting a water body region by using a mask matrix; calculating the local contrast of the pixel points relative to the neighborhood, constructing a suppression weight based on the noise standard deviation of the image sensor, and weighting the local contrast to obtain scattering response intensity; obtaining the texture disorder degree based on the ratio of the geometric mean value to the arithmetic mean value of the feature values of the structure tensor; the particle confidence is obtained based on the scattering response intensity and the texture disorder degree, the weighted particle confidence is obtained through weighting of a Sigmoid function, and the average value of the particle confidence is calculated to serve as the comprehensive turbidity so as to evaluate the water production state. The scattering response intensity and the texture disorder degree are fused, a weight suppression and soft threshold mechanism is introduced, pipe wall scratches and thermal noise interference are reduced, and the detection accuracy is improved.

The invention provides a hybrid energy storage cooperative control method and apparatus. The method comprises the steps of obtaining an actual value, a rated value and a frequency change rate of a power grid frequency; dynamically adjusting the initial virtual inertia coefficient by adopting an improved Sigmoid function based on the frequency change rate, and generating a dynamic virtual inertia coefficient; dynamically adjusting the initial droop coefficient by using the same function according to the deviation between the actual value and the rated value of the power grid frequency to generate a dynamic droop coefficient; calculating a power reference value of electrochemical energy storage according to the dynamic virtual inertia coefficient and the dynamic droop coefficient; based on the difference value between the power reference value and the power demand of the power grid, determining the number of weight blocks needing to be put into gravity energy storage; and controlling the gravity energy storage unit to output corresponding power according to the determined input quantity of the weight blocks so as to realize the stability of the power grid frequency. The core of the invention lies in determining the input amount of the weight block through the power difference value, thereby regulating and controlling the gravity energy storage output, and guaranteeing the stable power grid frequency.

The invention discloses a transformer substation intelligent monitoring system based on a digital twinborn technology, and relates to the technical field of intelligent monitoring, which comprises the steps of collecting and preprocessing operation data, constructing a digital twinborn model, simulating and generating twinborn data, counting the probability density of the operation data and the twinborn data, and calculating information entropy by using an information entropy formula. Calculating mutual information by using a mutual information formula; calculating entropy change of the information entropy, calculating a dynamic entropy adjustment coefficient by using a Sigmoid function, calculating an abnormal index by using a weighted fusion method and combining the dynamic entropy adjustment coefficient, performing state monitoring on the transformer substation, and dividing early warning grades based on a monitoring result. According to the method, through multi-scale coupling modeling of information entropy analysis and mutual information estimation, the precision and the real-time performance of transformer substation operation state monitoring are improved, a Sigmoid function dynamic adjustment mechanism and a weighted fusion algorithm are introduced, classification judgment is carried out on the abnormal state from the angle of entropy change response, and the monitoring sensitivity and the response accuracy are improved.

The invention relates to the technical field of laser measurement and photoelectric detection, and discloses a laser boresight equipment high-precision calibration method and system based on region segmentation mapping, and the method comprises the steps: firstly collecting a standard grid frosted glass target plate image, extracting grid intersection point pixel coordinates through employing an SURF algorithm, and building a data set corresponding to physical coordinates; dividing the field of view into a plurality of sub-regions, and respectively resolving a perspective transformation matrix mapped from a pixel coordinate system to a physical coordinate system; secondly, collecting a light spot image, performing initial positioning by using a Canny operator and a least square method, fitting a sampling sequence through a Sigmoid function, and extracting edge sub-pixel coordinates in combination with a central difference method; and finally, calling a corresponding transformation matrix according to the sub-region where the light spot is located to obtain physical coordinates, and calculating the light spot jerk value, the maximum diameter and the included angle between the optical axis and the mechanical axis. According to the invention, image distortion is compensated through view field partition mapping, and the calibration precision and detection efficiency of the laser boresight equipment are improved in combination with a sub-pixel positioning technology.

The invention discloses an intelligent control type floating fan system for extreme sea conditions. The system is composed of a sensing layer, a risk identification layer, a control decision layer, an adaptive learning module and a structure response layer. The sensing layer obtains wind speed, wave height, attitude inclination angle and anchoring tension, and calculates disturbance indexes; the risk identification layer generates a risk index by using a Sigmoid function; the control decision-making layer outputs connecting rod compression, air bag release and thruster thrust instructions according to the control decision-making layer; and the adaptive learning updates the weight according to the action deviation. The structure response layer comprises a carbon fiber adjustable connecting rod, a multi-cavity air bag provided with an air storage tank and a high-speed valve, a multi-direction vector propeller and a modularized interlocking floating body. According to the system, rapid risk identification and linkage control are realized, and the steady state and risk avoiding capability of the platform are improved.

The present invention relates to device for estimating a probability of a health state of a patient, particularly a probability of an infection of the patient, the device comprising an input unit configured to obtain physiological data of the patient, the physiological data comprising information about a plurality of time series of physiological parameters of the patient, a processor configured to extract the plurality of time series from the physiological data and to apply a machine learning algorithm on the plurality of time series to estimate the probability of the health state of the patient, and an output unit configured to provide the probability of the health state, wherein the machine learning algorithm is based on a generalized additive model, in which results of an application of a learned non-linear univariate function on the parameter time series are summed up and the probability of the health state is obtained by applying the standard sigmoid function on said sum.

The invention discloses a damage identification method and device for an in-service steel wire rope type horizontal lifeline, and belongs to the technical field of high-altitude operation safety facilities. The method comprises the following steps: synchronously acquiring a magnetic flux leakage signal and a surface image of an in-service steel wire rope through magnetic flux leakage detection equipment and a high-definition camera which are carried on a steel wire rope inspection robot; preprocessing the acquired magnetic flux leakage signal, wherein the preprocessing comprises singular value elimination and trend term removal processing; carrying out de-noising processing on the pre-processed signal by adopting an improved wavelet threshold de-noising algorithm fused with a Sigmoid function; extracting a characteristic value for representing the damage of the steel wire rope, and performing normalization processing to form a characteristic vector; and inputting into a BP neural network identification model optimized by a genetic algorithm for identification, and outputting an assessment result of the damage type and positioning of the steel wire rope. The method can realize automatic and quantitative detection and accurate identification of internal and external damages of the steel wire rope.

The invention discloses an attention reminding algorithm for intelligent driving based on multiple modes, and relates to the technical field of intelligent driving and active safety, and the algorithm comprises the steps: firstly, collecting data in real time through a multi-mode sensor group, carrying out the alignment, and then calculating the quantization features, such as the fixation deviation degree, the steering wheel disturbance entropy and the heart rate variability; then, mapping scores by adopting an S-type function, analyzing a historical sequence by utilizing a long short-term memory network to generate a dynamic weight, and performing adjustment and weighted calculation on a single-mode score in combination with an environment complexity weight to obtain a comprehensive attention score; and finally, according to the continuous driving duration, calculating a linear decreasing dynamic reminding threshold value, executing dual logic judgment in combination with the environmental risk level, and triggering graded feedback through sound and light or vibration. According to the method, the problems of poor adaptability and high false alarm rate of traditional single-mode monitoring in a dynamic environment are effectively solved, the robustness of the system is improved, and adaptive evaluation and accurate early warning of driver distraction behaviors in a complex scene are realized.

The invention discloses a gas molecule smell prediction method and system based on an isotropic graph neural network, and relates to the technical field of tobacco production equipment.The method comprises the steps that an SMILES expression of gas molecules is converted into a molecular graph structure containing atomic features and edge features and three-dimensional space coordinates through DGL-LifeSci and RDKit toolkits; then inputting the data into an EGNN isotropic graph neural network, aggregating neighborhood information through a message passing mechanism, and synchronously updating node features and coordinate features; a plurality of EGNN layers are stacked for deep feature extraction, and high-dimensional node representation and final coordinate values are obtained; performing global pooling through an MLPReadout module to obtain a global molecular representation vector of a fixed dimension; and finally, mapping the vector to a 138-dimensional space through a full connection layer, processing through a Sigmoid function to obtain an odor probability vector, and finally judging and outputting a specific odor category through a threshold value.

The invention provides an equipment quality and supplier intelligent evaluation method based on a space-time coupling Bayesian model and a two-ticket graph database, and belongs to the technical field of intelligent databases. The two-ticket graph database is firstly constructed, and a station environment, equipment full life cycle maintenance and supplier data are integrated; obtaining a supplier initial score by using an entropy weight method and a Sigmoid function; constructing a space-time coupling Bayesian model by means of ST-LSTM and VAE, and clustering equipment health states; introducing an environment coefficient and a KL divergence gradient to correct a score, and separating environment responsibilities from supplier responsibilities; calculating an equipment target score by combining an aging model and cosine similarity; and aggregating scores through a graph attention network, considering quality fluctuation, and generating a supplier global score. According to the invention, dynamic, precise and traceable evaluation is realized, a basis is provided for power grid supply chain management and the like, and the method is suitable for supplier evaluation in electric power and other industrial fields.

The application provides an improved multi-agent deep reinforcement learning unloading method, relates to the technical field of Internet of Vehicles and edge computing, and solves the technical problems of computing overhead, energy consumption and time delay in the task unloading process.The technical scheme comprises the following steps: S1: task unloading problem modeling;S2: improved unloading method design and implementation.The task queue smoothing algorithm designed by the application can accurately reflect the task load of the server and the mobile terminal, and assist the deep learning algorithm to make more accurate and stable unloading decisions.The application designs a smoothing penalty mechanism of a Sigmoid function in a reward function, and provides continuous and accurate gradient guidance for agent learning.The application corrects the action output by the deep reinforcement learning algorithm in real time in the early training stage through the closed-loop feedback mechanism of the PI controller, and improves the time delay performance of unloading.