23results about How to "Optimize weight" patented technology

The invention discloses a motion data management method and system based on AI body measurement, and relates to the technical field of artificial intelligence and machine learning, and the method comprises the steps: collecting motion data of each user, and carrying out the preprocessing; dividing user intervals by using a K-means clustering algorithm, constructing a load prediction model by using a fuzzy BP neural network, and obtaining a final load prediction value by using an improved adaptive genetic algorithm; and constructing a regression model by using a QHD reinforcement learning algorithm, dynamically adjusting a discount factor, outputting a training load adjustment strategy, and adjusting the discount factor according to feedback. User training load intervals are accurately divided through a K-means clustering algorithm, distribution of personalized training loads is ensured, the load prediction precision is improved by using a fuzzy BP neural network and an IAGABP algorithm, the weight and bias of the network are optimized, a local optimal solution is avoided, the prediction accuracy is improved, a QHD reinforcement learning algorithm and a dynamic discount factor are combined, and the prediction accuracy is improved. And a strategy of dynamically adjusting the training load is realized.

The invention provides an emotion synchronous visual design system based on multi-modal biological feedback, and aims to dynamically adjust visual design elements by analyzing physiological signals of a user in real time, so that the emotion experience of the user is enhanced. According to the system, various biological feedback sensors such as heart rate, skin conductance and brain wave sensors are utilized, an advanced data processing algorithm is integrated, and the emotional state of a user is accurately captured and analyzed. And then, the system changes visual elements such as colors, shapes and layouts in real time through a dynamic visual design adjustment module so as to match and respond to emotional changes of the user. In addition, the system further comprises a real-time feedback and interaction module and a personalized learning and optimization module so as to provide personalized visual experience and continuously optimize user interaction. The system not only improves the interactivity and attraction of the user interface, but also deepens the emotional relation between the brand and the consumer, and has important commercial application value and market potential.

The invention discloses a respiratory rehabilitation training data management method and system based on artificial intelligence, and relates to the technical field of data processing, the respiratory rehabilitation training data management method based on artificial intelligence comprises the following steps: obtaining training data of a respiratory rehabilitation person, and preprocessing the training data to obtain a multi-source data set; constructing a teacher network based on the high-performance model, training the teacher network through the multi-source data set, constructing a student network, and training the student network based on the trained teacher network; optimizing the trained student network by using a differential evolution algorithm to obtain a student network optimized model; and training a reinforcement learning agent by using data generated by the model after student network optimization, and automatically adjusting rehabilitation training parameters of the respiratory rehabilitation personnel through the reinforcement learning agent. According to the invention, computing resource and storage requirements are reduced, data-driven decision is realized, and a personalized rehabilitation training scheme is provided.

The invention relates to an AGV path smooth planning method based on adaptive heuristic optimization under multiple constraints. Comprising the following steps: 1, constructing a grid map based on a Layou graph of an intelligent manufacturing workshop, endowing each grid unit with a specific steering constraint according to a constraint condition of the operation flow direction of AGV equipment, and generating a grid steering constraint table; 2, improving a path cost evaluation function, and introducing a reverse penalty term; 3, optimizing the weight of the heuristic function, and designing a self-adaptive heuristic weight factor; and 4, introducing a secondary Bezier curve to carry out path smoothing, and finally obtaining a smooth AGV path. According to the AGV equipment operation flow direction required by an intelligent manufacturing workshop, steering constraints are given to each grid, a grid steering constraint table is generated, a reverse penalty term is added to a path cost evaluation function of a traditional A * algorithm, and a self-adaptive heuristic weight factor is designed in a heuristic function; and introducing a quadratic Bezier curve to smooth the planned initial path, and planning a smooth optimal path to realize efficient and smooth planning of the AGV path.

The invention relates to the technical field of deep learning, in particular to a large model knowledge base question answering method and system with a timeliness map. The method comprises the following specific steps: constructing a graph database capable of updating data in time and regularly performing timeliness data cleaning; inputting contents in a natural language, and performing triple extraction through a natural language large model; obtaining a specific query result from a graph database and a vector library according to the obtained triple result; the natural language large model carries out reference answering according to the graph data and the query result of the vector library; the method has the advantages that the data in the knowledge graph has the capacity of updating along with time change, the expired data in the knowledge graph can be deleted in time, and the timeliness value of the graph data is improved.

The application discloses a mechanical drilling speed prediction method based on micro-inclination characteristics, comprising the following steps: acquiring real drilling wellbore logging while drilling and logging data as a first data set; preprocessing the first data set and extracting the best feature and the micro-inclination feature as input features; the preprocessing comprises data cleaning and correlation analysis; training a mechanical drilling speed prediction model by using the input features; and predicting the mechanical drilling speed of a target well by using the trained mechanical drilling speed prediction model. The micro-inclination feature introduced in the application can significantly improve the ROP prediction accuracy of the model. By introducing the micro-inclination feature in the MLP neural network, the model weight is optimized, and the prediction accuracy of the ROP is improved. By introducing the micro-inclination feature in the SVR model and selecting the optimal hyperparameter, the prediction effect is significantly improved. The method provides a new idea for improving drilling parameter prediction and control by using machine learning, and can be applied to the optimization of drilling speed and efficiency.

The present invention discloses a curtain fabric defect recognition method based on feature cube and fuzzy detection dual recognition, comprising the following steps: capturing curtain fabric images at different angles, and performing multi-angle synthesis to obtain a curtain fabric panoramic image; obtaining material features and texture features, and correlating them with exposure levels to obtain a curtain fabric feature cube; graying the panoramic image, enhancing it based on a deep learning-based adaptive image enhancement algorithm, and adjusting the enhancement parameters to obtain an enhanced curtain fabric panoramic image; filtering and segmenting the enhanced image to obtain a fabric dynamic segmentation map, and inputting it into a detection model to identify fabric and texture defects, and obtaining a first recognition result; optimizing the edge of the enhanced image, and inputting it into an edge detection model to identify edge defects, and obtaining a second recognition result. The present invention achieves the purpose of improving the accuracy, robustness, adaptability and dynamic adjustment capability of curtain fabric defect detection.

The invention provides a method and a device for evaluating the transient support capability of an energy storage power station to a power grid, and the method comprises the steps: calculating the weight of an index value after the dimensionalization processing through different weighting methods, and optimizing the weight of the index value after the dimensionalization processing, and the transient support capability of the energy storage power station on the power grid is evaluated according to the obtained combination weight of each evaluation index. It can be seen that different weights obtained by different weighting methods are comprehensively considered, the upper layer adopts at least two weighting methods to calculate the weight of the index value after the dimensionalization processing, and the lower layer optimizes the weight of the index value after the dimensionalization processing to obtain the combined weight of each evaluation index. That is to say, the combined weight of each evaluation index is obtained by adopting the double-layer combined weighting, the precision of the evaluation result is improved, and the reliability of the energy storage power station for evaluating the transient support capability of the power grid can be further improved.

The invention discloses a cargo transportation method and device, and relates to the technical field of warehouse logistics. A specific embodiment of the cargo transportation method comprises the following steps: in response to a received delivery request, obtaining corresponding order information and real-time warehouse information of a plurality of warehouses; inputting the order information and the real-time warehouse information into a preset warehouse recommendation model, and determining a target warehouse from the plurality of warehouses by using the warehouse recommendation model; and generating a transportation task according to the target warehouse, and transporting target goods corresponding to the delivery request to the target warehouse. According to the embodiment, the target warehouse more suitable for receiving the goods can be screened out from the multiple warehouses before the goods are transported, so that the number of times of temporarily modifying the target warehouse in the goods transportation process is reduced, and the goods transportation cost is reduced.

The invention discloses an electricity market operation effect evaluation method based on a sparrow search algorithm, and belongs to the technical field of electricity markets. In order to solve the problems that a traditional index system fusion technology is high in subjectivity, complex in calculation and simplified in relation and quantitative analysis cannot be conducted on the current market, a total evaluation layer, an evaluation layer and a specific description layer are designed through five first-level indexes including the market structure, the economical efficiency, the safety, the social welfare and the link friendliness; the indexes are fused from three dimensions of time, region and industry, the index system construction is completed, and the method is simple and effective. Various power market operation data of different types and different regions are effectively processed through a sparrow search algorithm; meanwhile, an index system is designed through three levels and three dimensions, the index system serves as the basis, first-level indexes serve as fitness functions, index weights are optimized, the defect that quantitative analysis cannot be conducted on the current market is overcome, and an artificial intelligence methodology and a tool box can be provided for decision making in the aspects of the electricity market and the like.

The invention discloses a supplier evaluation method based on an entropy weight method, and relates to the technical field of data processing. The method comprises the following steps: S1, determining an evaluation index for evaluating the supplier level based on the type of a supplied commodity; wherein the evaluation indexes comprise a quality level index, a delivery capability index, a price level index, a technical capability index and a backup service index; s2, collecting historical data of each evaluation index, and preprocessing the data; and S3, classifying the evaluation indexes, including qualitative evaluation indexes and quantitative evaluation indexes. According to the method, the entropy weight method is adopted to evaluate the suppliers, the entropy weight method can deeply reflect the distinguishing ability of the indexes and determine better weights, weighting is more objective, theoretical basis is provided, the credibility is higher, the entropy weight method is also adopted to process each evaluation index, the accuracy of data of each evaluation index is improved, and the evaluation efficiency is improved. The problem that an existing supplier evaluation method is not ideal in accuracy is solved.

The invention relates to a compression method capable of local reduction for industrial CT data, and the method comprises the steps: traversing a whole preset product, obtaining the pixel value of each pixel point in each section of the preset product, and obtaining the to-be-compressed three-dimensional data of the preset product; and inputting the to-be-compressed three-dimensional data into a neural network to obtain target three-dimensional data after the preset product is compressed. The method has the beneficial effects that when the local data is restored, the local information of the specified position can be directly restored from the compressed data, the requirement is quickly met, and the efficiency is improved.

This invention relates to the field of intelligent inspection and image analysis technology, and discloses an intelligent hazard investigation method and system based on image recognition. The method includes establishing a hazard knowledge network representing the correlation between hazard types, features, and causes. Hazard images are collected during inspections and input into the network. Hazard entities are extracted, and their features are matched with hazard features in the network. Network reasoning is used to obtain the associated hazard types and causes. Hazard probabilities are calculated based on the correlation strength of causes. Intelligent investigation tasks bound to locations are generated and output based on a comparison of the probability and a threshold to guide targeted investigations. Finally, the correlation relationships in the hazard knowledge network are optimized based on feedback from on-site verification of the tasks. This invention achieves intelligent hazard identification and cause reasoning through a knowledge network, and optimizes the network using a closed-loop business feedback mechanism, improving the intelligence depth of hazard investigation and the system's adaptability.

The invention discloses a visible light indoor position sensing system based on a biological inspiration optimization width learning network model. The method comprises the specific steps that firstly, a photoelectric detector collects illumination intensity at different positions in a 4m * 4m * 3m three-dimensional space, and a database is established in real time; inputting the training data set into a network model for model training; and finally, inputting the test set into the trained network model to obtain a corresponding positioning result, and sending the positioning result to a receiving end. According to the method, the biological heuristic algorithm and the width learning network are combined, so that high-dimensional and large-scale data challenges in the field of machine learning are overcome. Besides, aiming at the problem that the weight and bias of width learning input are randomly selected and can influence the test performance of the algorithm, the invention adopts a biological inspiration algorithm so as to effectively improve the positioning precision and stability.

The invention discloses a task scheduling method and device, a nonvolatile storage medium and electronic equipment. The method comprises the steps that influence factors related to a target scheduling task and scheduling factors corresponding to the influence factors are determined according to historical work order data; a target weight set is determined, the target weight set comprises target weights corresponding to all the scheduling factors, the target weight set is a weight set enabling the change rate of a cost function of the task scheduling model to be within a preset change rate range, and the cost function is composed of a loss function and a normalized item; the normalization item is positively correlated with the weight of each scheduling factor, and the value of the loss function is correlated with the weight of each scheduling factor; and processing the scheduling factor and the target weight set through a task scheduling model to obtain a scheduling scheme corresponding to the target scheduling task. The technical problem of low accuracy of a scheduling method caused by insufficient response of a traditional scheduling method to dynamic influence factors is solved.

The present invention relates to the technical field of bird-detecting radar data processing, and specifically to a bird-detecting radar data processing system and method optimized for small targets. The present invention includes a data acquisition module, an adaptive filtering module and a graph neural network processing module, wherein: the data acquisition module acquires radar data and expected data; the adaptive filtering module dynamically adjusts the step size and weight based on the current error during the filtering process by setting an initial step size and weight to generate a radar feature vector; the graph neural network processing module generates a target node according to the radar feature vector, defines edge weights through adjacent errors, calculates attention weights, and finally updates the radar feature vector of the target node. This scheme improves the accuracy and stability of small target signal extraction and solves the signal noise and parameter fixation problems in traditional methods.

The invention discloses an adaptive particle swarm optimization BP neural network air quality prediction method, and relates to the technical field of air quality prediction. According to the adaptive particle swarm optimization BP neural network air quality prediction method, data set division is carried out on screened feature data to form a training set and a test set, normalization processing is carried out, and then a neural network model is established; a screened training set is introduced into a neural network model to optimize threshold parameters in the neural network, an air quality prediction model is established to predict air quality data, the weight and the threshold of the BP neural network are optimized by using the advantages of high optimization capacity and high execution efficiency of the particle swarm optimization, and the prediction efficiency of the air quality data is improved. And meanwhile, the position of the particle swarm is adaptively changed, so that the global search capability of the algorithm is improved, the problem that the BP neural network is easy to fall into local optimum is effectively solved, and the prediction precision and speed of the BP neural network are improved.

The invention relates to a VVC fast CU division method based on ensemble learning, which belongs to the field of video coding and comprises the following steps of: initializing a given coding unit (CU), and starting a coding process from the deepest level; then, feature extraction is carried out; the feature attribute value of the current CU is calculated; and model prediction: loading a pre-trained integrated learning classifier (i = 0, 1, 2), carrying out division prediction on the current CU by using the pre-trained integrated learning classifier, and then carrying out decision processing. And once the coding of the current CU is completed, moving to the next CU in the list and repeating the steps. Through the series of steps, the algorithm can systematically perform division decision on each CU so as to optimize the video coding process. The method can be used for coding under VVC intra-frame coding configuration, the coding time of VVC is remarkably shortened under the condition that the coding efficiency and the coding quality are hardly reduced, and the method can be applied to video conferences and other application scenes with high requirements for coding real-time performance.

The invention discloses a multi-anomaly detection early warning method and system based on a large model, and relates to the technical field of biological sample storage automation, and the method comprises the steps: collecting and obtaining a frame sequence and shooting parameters; outputting a structured exception list; calculating a unified score and an index; outputting a gating level according to the configurable gating level set and the threshold interval and the decision function thereof; and the recheck closed loop is butted with the visual early warning packet. According to the method, various abnormities can be covered, the method can be expanded to any newly-added exception category through an exception plug-in mechanism, and compared with a scheme which only completes identification or post-event disposal, scene understanding, structured exception, unified gating and correction reinspection are solidified into a standardized interface and time sequence, empty grabbing, falling and sample damage are reduced, and the method is suitable for large-scale popularization and application. The door opening exposure time is shortened, the product stability is enhanced, and the technical problem that low-temperature sample grabbing abnormities are difficult to process in a unified mode is solved.

The invention provides a chip appearance defect identification method based on an improved random forest algorithm, and aims to meet the requirements of high precision and high efficiency on chip quality control in the semiconductor manufacturing industry. According to the method, feature selection and parameter configuration of the random forest algorithm are optimized, so that efficient and accurate detection of chip appearance defects is realized. The method comprises the following specific steps: firstly, preprocessing a chip image, including size standardization, graying and noise removal, so as to improve the image quality and ensure the effectiveness of subsequent analysis; thirdly, simplifying decision-making trees in the random forest by utilizing a mutual information technology, and removing redundant decision-making trees with relatively low classification performance, so as to improve the efficiency and accuracy of the model; then, optimizing the weight of the simplified decision tree by adopting a genetic algorithm, reducing the over-fitting risk, and enhancing the generalization ability of the model; furthermore, key parameters (including the number of trees, the maximum depth and the like) of the random forest are further optimized by using a grid search method, so that the overall performance of the model is improved; and finally, evaluating model performance under different parameter configurations by using cross validation so as to select an optimal configuration. The chip appearance defect detection accuracy and efficiency can be remarkably improved, and the method is suitable for an automatic detection system in the semiconductor manufacturing industry, and is particularly suitable for application scenes which have extremely high accuracy requirements and need quick response.

The present invention relates to the field of vehicle control technology, and provides a vehicle scheduled charging control method and system. The method is applied to an in-vehicle computing unit, and includes: determining the scheduled charging waiting time according to charging related information, charging gun connection time and the estimated charging time sent by the battery management system of the vehicle, wherein the charging related information includes basic charging parameters and vehicle use time, wherein the scheduled charging waiting time is the time starting from the charging gun connection time; and starting the timing in response to the determination of the scheduled charging waiting time, and when the timing reaches the scheduled charging waiting time, sending a charging permission signal to the vehicle control unit to enable the vehicle control unit to control the vehicle to charge. In the present invention, the in-vehicle computing unit completes the calculation of the scheduled charging waiting time and the scheduled charging timing, while the T‑box is only responsible for signal forwarding, which greatly reduces the amount of information processing by the T‑box.

This invention provides a transformer-based method for foodborne disease analysis, belonging to the field of data analysis and processing. The method includes acquiring foodborne disease case monitoring data, obtaining a standardized sample feature set through preprocessing, extracting features using an improved ReliefF algorithm, constructing a multi-dimensional feature matrix, training a basic prediction model based on a transformer architecture and an associative attention mechanism to obtain a foodborne disease analysis model, inputting the foodborne disease case data to be monitored, outputting the foodborne disease analysis results, and performing adaptive learning on the foodborne disease analysis model to complete iterative optimization. This invention solves the problems of low disease screening efficiency, insufficient predictive analysis accuracy, and limited functionality in existing technologies due to insufficient utilization of multi-dimensional case features and the lack of mechanisms to highlight key case features.

The application discloses an environment perception and adaptive cleaning algorithm field, and discloses an unmanned environmental sanitation vehicle adaptive cleaning method and device based on multi-sensor fusion, aiming at solving the technical problem of insufficient ground material, stain and obstacle recognition in a complex environment. The method comprises the following steps: acquiring multi-source sensor environment data and fusing the multi-source sensor environment data to obtain optimal pose state estimation; constructing an environment state matrix based on the optimal pose state estimation, and fusing visual features to obtain environment pollution structured information; using a deep Q network to learn and decide an adaptive cleaning strategy, and outputting a continuous control quantity; converting the control quantity into a driving signal to adjust a cleaning action; determining a cleaning coverage rate according to an actual pose and environment information, and planning a rescan path; constructing a reward function based on the coverage rate change and actual energy consumption, updating deep Q network weights, and generating an optimal cleaning strategy. The application can realize precise perception of the environment and intelligent and efficient cleaning function of the unmanned environmental sanitation vehicle.