Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

1351results about How to "Improve search efficiency" patented technology

Big data-based energy storage power station battery health state diagnosis method and system

The invention provides an energy storage power station battery health state diagnosis method and system based on big data, and relates to the technical field of energy storage batteries, and the method comprises the steps: collecting multi-mode operation data of an energy storage power station battery, carrying out homomorphic encryption to obtain ciphertext data, carrying out analysis and calculation, extracting explicit and implicit correlation features, and obtaining fusion time sequence data; based on fusion time sequence data, extracting multi-path quantum entanglement features, combining with a self-evolution quantum attention mechanism, dynamically adjusting the phase and amplitude of quantum bits, generating diagnosis feature enhancement representation, realizing entanglement storage through a quantum circuit, and combining with a gating unit to predict a health state sequence; and determining a battery health prediction result, and obtaining a battery health state diagnosis result by combining quantum state superposition and quantum parallelism, searching a diagnosis strategy combination space on line, embedding a multi-order causal relationship tensor model, modeling causal relationship strength between battery performance indexes, determining a battery health degradation index, generating a causal diagnosis map and obtaining the battery health state diagnosis result.
Owner:JIANGSU ZHONGJUN ENERGY TECH CO LTD

Traffic flow prediction optimization method based on intelligent optimization algorithm

The invention relates to a traffic flow prediction optimization method based on an intelligent optimization algorithm, which is applied to the technical field of traffic flow prediction. The system comprises a multi-modal data acquisition and fusion module, a hybrid deep learning prediction module, an intelligent parameter optimization module, an integrated learning module, a multi-scale error correction module, a prediction result analysis and visualization module, a distributed learning and privacy protection module and the like. Wherein the hybrid deep learning prediction module adopts an attention mechanism to fuse models such as CNN, Transform and LSTM, and introduces an adaptive error correction mechanism, so that the prediction precision is remarkably improved; the intelligent parameter optimization module accelerates the model search efficiency by using technologies such as quantum computing and swarm intelligence optimization; the integrated learning module and the adversarial training strategy enhance the robustness and generalization of the model; and the multi-scale error correction module adopts a time sequence decomposition and self-adaptive online learning method to improve the accuracy of long-term and short-term prediction.
Owner:JIANGXI VANDT COLLEGE OF COMM

Intelligent search engine system based on Simbert algorithm

The invention relates to the technical field of intelligent search engines, in particular to an intelligent search engine system based on a Simbert algorithm, which can improve semantic comprehension ability, process complex query scenes, realize personalized search experience and improve search efficiency and accuracy. The system comprises a data collecting and preprocessing module for collecting massive text data, preprocessing the collected text data and constructing a corpus; the index construction module is used for coding the preprocessed corpus through a Simbert algorithm to generate a semantic vector and storing the semantic vector in a search engine; the query processing module is used for acquiring a query request input to a search engine by a user, and encoding the query request through a Simbert algorithm to generate a query vector; and the similarity calculation module is used for matching the query vector of the user with the semantic vector stored in the database, calculating the similarity between the query vector and the semantic vector, and obtaining the similarity score of the query request.
Owner:DIGITAL JIANGXI TECH CO LTD

Comprehensive energy system short-term multi-element load prediction method fusing deep learning combination model

The invention discloses an integrated energy system short-term multivariate load prediction method fusing a deep learning combination model, and the method comprises the steps: obtaining the multivariate load historical data and meteorological data of an integrated energy system, and obtaining a correlation factor which is greatly affected by the multivariate load; dividing the multi-element load historical data into two parts, namely a training set and a test set, and performing maximum and minimum normalization processing; extracting multi-level spatial features of the multi-element load historical data by using a series-parallel superposed multi-scale convolutional neural network; load historical data time sequence features are extracted by using a bidirectional gating cycle unit neural network, and a multi-head attention mechanism is introduced to extract important information of a load prediction task, so that a multi-scale convolution bidirectional gating cycle unit neural network fused with the attention mechanism is designed; and a self-adaptive multi-strategy grey wolf optimization algorithm is utilized to realize optimization parameter search of the bidirectional gating circulation unit neural network, an evaluation index is adopted to carry out error analysis on a prediction result, an optimal parameter is obtained, and accurate load prediction is realized.
Owner:JIANGSU UNIV OF TECH

Particle size distribution measuring method and system based on machine learning and light scattering method

The invention relates to the field of particle size distribution measurement, in particular to a particle size distribution measurement method and system based on machine learning and a light scattering method.The method comprises the steps that a to-be-measured particle sample is irradiated through a light source, and meanwhile light scattering intensity data of different scattering angles and different wavelengths are collected through a multi-angle scattered light detector; constructing a multi-modal deep learning model, and forming an integrated learning architecture in combination with a support vector machine model and a random forest model; inputting the preprocessed multi-wavelength light scattering intensity data into the trained integrated learning architecture to obtain a preliminary particle size distribution result of the to-be-detected particle sample; based on the improved Mie scattering theory, a high-precision theoretical model considering the particle shape, the refractive index, the surface roughness and the internal structure is constructed; an improved Bayesian optimization algorithm is adopted, and the initial particle size distribution result is optimized and adjusted by minimizing the error between the theoretical light scattering intensity and the actually collected light scattering intensity, so that a corrected particle size distribution result is obtained.
Owner:CHANGCHUN UNIV OF SCI & TECH

Dynamic language test case automatic generation method based on large language model

The invention discloses a method for automatically generating a dynamic language test case based on a large language model, which comprises the following steps of: firstly, in a preprocessing stage, arranging related information of a to-be-tested program into prompts, analyzing a parameter type of the to-be-tested program by the large language model according to the prompts, and generating code information required by a test through an abstract syntax tree; in the test case generation stage, a dynamic multi-objective optimization algorithm is adopted, test cases are continuously crossed and mutated through branch coverage feedback, in the partial evolution process, test data which cannot be generated by conventional variation are generated by utilizing a large language model, and finally, the screened excellent test cases are organized into a complete test suite. In a program repair phase, the test suite is run to detect potential errors, the errors are analyzed and iteratively repaired using a large language model, and an executable and error-free test suite is generated. According to the method, the test coverage rate of the dynamic language is remarkably improved, and the achievement of a coverage target is accelerated.
Owner:HANGZHOU DIANZI UNIV

Sleep quality evaluation method based on millimeter wave radar and AI monitoring sensor

The invention discloses a sleep quality evaluation method based on a millimeter wave radar and an AI monitoring sensor, and belongs to the technical field of sleep quality evaluation, and the method comprises the steps: collecting annotation data through the millimeter wave radar and the AI monitoring sensor module; performing sample generation based on a conditional generative adversarial network of locality preserving projection to realize data expansion; inputting the expanded data into a feature extraction model to train the feature extraction model; inputting the data after feature extraction into a feature dimension reduction model to train the feature dimension reduction model; inputting the data subjected to feature dimension reduction into a classifier to train a classifier model; and processing new sleep monitoring data by utilizing the trained feature extraction model, the data dimension reduction model and the classifier model. According to the method, the newly input multi-source sleep data can be effectively processed, real-time classification and evaluation are carried out, the classification accuracy and efficiency are improved, a scientific sleep quality basis is provided for a user, and the sleep quality can be managed and improved.
Owner:ZHEJIANG JOYCHINE IOT TECH CO LTD

Tree component optimization method based on React framework and fusing virtualization rendering and increment rendering

The invention belongs to the technical field of front-end development, and particularly relates to a React framework-based tree component optimization method fusing virtualization rendering and increment rendering. Comprising the steps of S1, data preprocessing; S2, dynamic virtualization rendering: dynamically calculating a node index range in a visible area according to a real-time scrolling position of a user viewport, only rendering nodes in the range, and realizing seamless lazy loading of data through scrolling event monitoring; s3, intelligent incremental updating: identifying changed nodes through a state change comparison algorithm, and executing local re-rendering only for the changed nodes based on a component life cycle control mechanism of a React framework; and S4, lazy loading cooperative control. The method aims at solving the problems in the aspects of performance, memory management, user experience and the like when a tree component in a React ecological system processes large-scale data, and is widely applied to various Web application program development scenes needing to display and operate a large-scale tree data structure.
Owner:HANGZHOU ARTECH

Wind, light and water resource element simulation method and device for optimizing key parameters of confluence mode

The invention discloses a wind, light and water resource element simulation method and device for optimizing key parameters of a confluence mode. The method comprises the following steps: acquiring a river network database and gridded hydrotopographic data, and interpolating the gridded hydrotopographic data to a refined confluence grid with a target resolution to obtain interpolated elevation data; according to the altitude data, the flow direction data and the cumulative flow after interpolation, determining vector data of a river network center line and a water collection area boundary, segmenting the river network according to the Strahler river level, and determining river reach of different Strahler levels; inputting the vector data of the central line of the river network and the boundary of the water collecting area into a pre-established river network confluence model, and simulating by adopting different Manning coefficients for river reaches with different Strahler levels to obtain simulated flow; and through a dynamic dimension search algorithm, the Manning coefficients of the river reaches with different Strahler levels are optimized respectively, so that the objective function is minimum.
Owner:THREE GORGES GROUP IND DEVELOPMENT (BEIJING) CO LTD +1

Robot autonomous learning and adaptive trajectory planning method oriented to man-machine cooperation

The invention provides a robot autonomous learning and adaptive trajectory planning method oriented to man-machine cooperation aiming at a mechanical arm simulating a human teaching sample to obtain a motion skill and generalizing the motion skill to an adaptive obstacle avoidance behavior of a new situation, and specifically, a dynamic motion primitive algorithm is introduced to carry out learning and generalization of a motion trajectory of the mechanical arm. Based on the initial state and the search area of a generalized trajectory improved particle swarm algorithm, rapid search of a local obstacle avoidance trajectory is achieved, and mechanical arm motion trajectory planning with an autonomous obstacle avoidance function is comprehensively achieved. The method has the advantages that the method has high consistency with an original generalization track and wider adaptability, and the path searching efficiency is remarkably improved; a virtual simulation experiment and a physical experiment verify that the method has good stability and practicability.
Owner:CHAOHU UNIV

Path planning method based on improved bidirectional A star search algorithm

The invention relates to a path planning method based on an improved bidirectional A star search algorithm, relates to the technical field of intelligent vehicles, and solves the problem of low search efficiency caused by bloated search nodes or special topographic reasons in an existing path search scheme. The path planning method based on the improved bidirectional A star search algorithm comprises the following steps: acquiring an obstacle grid map of a target scene; determining a motion path from a starting point position to a target position in the barrier grid map based on an optimized bidirectional A * search algorithm; wherein in the optimized bidirectional A * search algorithm, the forward heuristic function value of the child node is a heuristic function value from the child node to the child node with the minimum reverse total cost at present, and the reverse heuristic function value of the child node is a heuristic function value from the child node to the child node with the minimum forward total cost at present.
Owner:BEIJING INST OF TECH

Design and optimization method for multilayer resonance composite sound absorption structure of high-capacity high-frequency transformer

According to the design and optimization method for the multi-layer resonance composite sound absorption structure of the high-capacity high-frequency transformer, the sound absorption performance of different composite sound absorption structures is compared and analyzed, and the optimal composite sound absorption structure is determined; a porous sound absorption material with the best sound absorption performance is determined according to the determined best composite sound absorption structure; establishing an electromagnetic field-structural force field-sound field three-dimensional simulation solution model of the high-frequency transformer, analyzing noise characteristics of the high-frequency transformer without a sound absorption device, and determining an analysis frequency; determining an optimization variable and a target function of the composite sound absorption structure, and obtaining sound field simulation results of the sound absorption device under different structure parameters by adopting a method of combining center combination test design and finite element simulation; constructing a radial basis function neural network model to obtain the influence degree of different structure parameters on the sound pressure level of the high-frequency transformer; and on the basis of a sparrow search optimization algorithm fusing sine and cosine and Cauchy variation, obtaining optimal structural parameters of the composite sound absorption device. According to the method, the influence of noise on the surrounding environment and power equipment can be reduced.
Owner:国网黑龙江省电力有限公司绥化供电公司

Data query method and device, readable medium, electronic equipment and program product

The invention relates to a data query method and device, a readable medium, electronic equipment and a program product. The method comprises the steps of obtaining a target statement input by a user in a natural language; querying in a first database according to the target statement through a machine learning model to obtain target data table information, and generating a structured query language statement for data query according to the target statement and the target data table information; and executing the structured query language statement to obtain a data query result, and displaying the data query result to the user. Therefore, the data can be automatically searched based on the target statement input by the user through the machine learning model, so that the data searching process is simplified, and the data searching efficiency is improved. Besides, the user can take the natural language as the query statement, and does not need to generate a structured query language statement for querying the data in advance, so that the data query difficulty and complexity can be reduced, and the data query efficiency is further improved.
Owner:BEIJING YOUZHUJU NETWORK TECH CO LTD

Homogeneous real estate space intelligent search method and system based on space partition

The invention discloses a homogeneous real estate space intelligent search method and system based on space partition, and relates to the technical field of space data processing and intelligent search. According to the homogeneous real estate space intelligent search method and system based on space partition, the data of the target real estate area is obtained and preprocessed, so that the accuracy and integrity of the data are ensured; dividing the pre-processed area into a plurality of space subareas through a geographic information system, extracting homogeneous features through a clustering algorithm, and analyzing a peripheral matching adaptation degree index of each subarea in combination with geographic position preference of a user to form a homogeneous house resource set; and searching and matching the homogeneous house resource set through a collaborative filtering algorithm, generating a user recommended house resource list, calculating a matching degree index of house resources in each partition, setting a dynamic feedback mechanism, and sorting the recommended house resource list in real time, so as to improve the user satisfaction. Personalized housing resource recommendation is provided for the user, and the convenience and accuracy of real estate transaction are improved.
Owner:PEKING UNIV SHENZHEN GRADUATE SCHOOL +1

Unmanned aerial vehicle path planning method based on improved sparrow algorithm

The invention provides an unmanned aerial vehicle three-dimensional path planning method based on an improved sparrow algorithm, and the method comprises the steps: carrying out the initialization of a population through fusing Tent chaotic mapping and Sine chaotic mapping, thereby improving the diversity of the population, and improving the search range of the population; secondly, introducing a butterfly optimization algorithm, updating the position of a discoverer in a sparrow search algorithm, and enhancing the search capability of the algorithm for a globally optimal solution; then, position updating is carried out on the participant by using adaptive T distribution, the adaptability and flexibility of the algorithm are increased, K-means clustering is introduced to select the alert, and the path planning capability of the unmanned aerial vehicle in a complex environment is enhanced; finally, in order to prevent the algorithm from falling into local optimum, optimal individual Gaussian disturbance variation is introduced, and the algorithm is prevented from falling into local optimum in the search process. A simulation experiment result shows that compared with a traditional sparrow algorithm, the improved sparrow algorithm can find a safe and feasible path, and has relatively stable optimization capacity.
Owner:CHANGCHUN UNIV OF TECH

Subway pedestrian anomaly detection method and system based on improved yolov8 network and clip model

The invention provides a subway pedestrian anomaly detection method and system based on an improved yov8 network and a clip model, and relates to the technical field of subway pedestrian detection, and the method comprises the steps: obtaining a video frame in a subway scene, and carrying out the preprocessing of the video frame; inputting the preprocessed frame data into an improved yolov8 neural network model to obtain an abnormal behavior target detection frame, and performing target candidate tracking on the target detection frame by using a target tracking algorithm to obtain a candidate image of a target area of a specific behavior; the method comprises the following steps: inputting a candidate image containing a target area into a clip model, inputting a set detection text into the clip model, mapping the detection text and the candidate image of the target area, performing text-image similarity matching, and optimizing a similarity matching threshold by adopting a Q-Learning reinforcement learning immune algorithm to obtain a similarity matching result; and the abnormal behavior detection accuracy is improved.
Owner:JINAN RAILWAY TRANSPORT GRP CO LTD

Evolutionary neural architecture searching method and system based on similarity agent assistance

The invention provides an evolutionary neural architecture searching method and system based on similarity agent assistance, and the method comprises the steps: 1, initializing an architecture population, and selecting an architecture with the best performance as an initial reference architecture; 2, constructing a graph neural network variant as a feature extractor; 3, constructing an agent model, and training the agent model through a joint loss function; 4, the high-potential architecture is reserved according to the fitness value, authenticity performance evaluation is carried out on the high-potential architecture, and an evaluation result is added into the training set; combining the current population with the high-performance architecture predicted and screened by the proxy model, and updating the population through an environment selection strategy; and step 5, repeating the step 3 and the step 4 until the population performance converges, and finally outputting a global optimal architecture. According to the method, the high-performance neural network architecture can be quickly searched under limited computing resources, and an efficient and intelligent novel solution is provided for automatic design and optimization of the neural network.
Owner:NANJING UNIV OF INFORMATION SCI & TECH

Navigation method of intelligent navigation robot based on multi-sensor fusion

The invention belongs to the technical field of dung beetle optimization methods, discloses a robot path planning system and method based on a dung beetle optimization algorithm, and realizes comprehensive perception and accurate modeling of a complex three-dimensional environment by introducing multi-sensor fusion and an advanced environment modeling technology. The robot can dynamically update the environment model according to environment data collected in real time, and sensitivity and adaptability to environment changes are kept. Meanwhile, a self-adaptive navigation algorithm is adopted, so that the robot can autonomously adjust a navigation strategy and path planning according to task requirements and environment changes. The robot can flexibly cope with various challenges in complex and dynamic environments due to the high self-adaptive capacity, and the reliability and the intelligent level of the robot in practical application are remarkably improved.
Owner:NAVAL UNIV OF ENG PLA

Searching method and device, electronic equipment and storage medium

The embodiment of the invention provides a search method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a search request of a user, the mode of the search request being a text mode or an image mode; performing feature extraction on the search request through a multi-modal large model to obtain a to-be-searched feature vector; determining a target scene search area based on the vector similarity between the to-be-searched feature vector and the scene type vector; based on an index structure of the to-be-searched feature vector in the target scene search area, searching a target bottom library feature vector from the bottom library feature vectors of the scene search area; based on the target bottom library feature vectors, target bottom library data are searched and obtained, the target data are returned to the user, and each bottom library feature vector corresponds to one piece of bottom library data. According to the method, the search request organization time of the user is shortened, only the base library data of the corresponding scene is searched in the target scene search area, the search data volume is reduced, and then the search efficiency of the scene event is improved.
Owner:SHENZHEN INTELLIFUSION TECHNOLOGIES CO LTD +1

Mobile robot path planning method and system

The invention relates to a mobile robot path planning method and system, and the method comprises the steps: determining a multi-neighborhood mixed search mechanism for the mixed use of eight neighborhoods and new sixteen neighborhoods according to whether there is an obstacle in the eight neighborhoods of a current node; deleting invalid nodes according to the positions of the current node and the target node; a bidirectional search strategy is adopted, a dynamic weight coefficient is added to adjust the heuristic function, and a planned path is obtained; and a redundant inflection point deletion strategy is adopted for the planned path, useless inflection points are removed, and an optimal path is obtained. By improving a neighborhood search mechanism, the search efficiency of a path planning algorithm is improved; through a node optimization strategy, invalid nodes in the algorithm searching process are removed, and the path planning algorithm searching time is shortened; by introducing a two-way search mechanism, the algorithm search efficiency is improved, meanwhile, the dynamic weight factor is added to adjust the algorithm search speed, search of invalid regions is reduced, and therefore the search efficiency is improved.
Owner:GUANGZHOU INST OF ADVANCED TECH CHINESE ACAD OF SCI

Knowledge question and answer method and device, readable medium, electronic equipment and program product

The invention relates to a knowledge question answering method and device, a readable medium, electronic equipment and a program product. The method comprises the steps of obtaining a target question input by a user in a natural language; searching in a knowledge base according to the target question through a machine learning model to obtain a target knowledge document used for answering the target question, and determining a target answer for the target question based on the target knowledge document, the knowledge base being used for storing a business knowledge document; and displaying the target answer to the user. Therefore, the question input by the user can be understood through the machine learning model, the knowledge document is automatically searched and sorted based on the question input by the user, and the answer to the question is directly displayed to the user. According to the method, the accuracy and searching efficiency of searching the knowledge document in the specific field by the user can be improved, the answer of the question can be directly obtained, and the efficiency of understanding business knowledge in the specific field by the user is improved.
Owner:BEIJING YOUZHUJU NETWORK TECH CO LTD

Dynamic operator acceleration tuning method combining meta learning and reinforcement learning

A dynamic operator accelerated tuning method combining meta learning and reinforcement learning comprises the following steps: firstly, constructing a generalization search space based on operator scheduling primitives, extracting tensor program structure features, and establishing a dynamic operator cost model; secondly, intelligent agent parameters are quickly adjusted by applying a meta-learning optimization algorithm, and search is accelerated by utilizing a reinforcement learning state encoder and a meta-gradient descent algorithm; and further, an Autotuning technology is combined with a deep reinforcement learning architecture to carry out operator optimization strategy search. Designing a reward function, and feeding back and guiding a search behavior according to code performance. Through a control variable method and a quantitative analysis method, a calculation and memory access performance model is constructed, operator performance key factors are analyzed, and limit performance under different hardware conditions is evaluated. And finally, selecting an optimal operator to generate a code, and converting the code into a machine code of the target hardware platform. By designing a high-quality generalized search space and performance limit analysis method, the efficiency and precision of operator tuning are remarkably improved, and optimization support is provided for deployment of a deep learning system.
Owner:NAT SUPERCOMPUTING SHENZHEN CENT (SHENZHEN CLOUD COMPUTING CENT)

Industrial internet equipment fault prediction method and system based on digital twinning

The invention discloses an industrial internet equipment fault prediction method and system based on digital twinning, and belongs to the field of industrial internet. Operation data and data volume processing data of scene equipment related to industrial internet-of-things abnormal behaviors are acquired for equipment fault primary prediction; and finally, performing final prediction on the fault equipment based on the operation data of the primary prediction fault equipment and the equipment fault primary prediction, and finally displaying fault equipment information obtained by final prediction on the digital twinborn model, and reminding maintenance personnel to maintain the equipment. According to the method, the abnormal behavior before the fault occurs is accurately judged based on the digital twinborn model, and the equipment causing the abnormal behavior is searched through the parameter correlation degree and the time-space correlation between the equipment and the behavior in the abnormal behavior analysis process, so that the searching efficiency of the fault equipment is improved, and the fault diagnosis accuracy is improved. And the security of the industrial internet scene is further improved.
Owner:NAT IND INFORMATION SECURITY DEV RES CENT

Fault recovery method and device of power distribution network and electronic equipment

The invention discloses a fault recovery method and device of a power distribution network and electronic equipment. The power distribution network fault recovery method comprises the following steps: determining a load node, a power generation node and a fault area in a fault power distribution network; the fault power distribution network is divided into a plurality of island areas based on the fault areas and the power generation nodes, and the load nodes in each island area supply power through the power generation nodes in the island area; calculating the power consumption demand of each load node during the fault period of the power distribution network, and determining a target function of a fault recovery strategy through the power consumption demand of each load node; and by taking the target function as a fitness function of a binary particle swarm algorithm, optimizing the fault recovery strategy through the binary particle swarm algorithm to obtain a target fault recovery strategy, and executing the target fault recovery strategy on the fault power distribution network. According to the invention, the problem of long time consumption of power restoration of the power distribution network in related technologies is solved.
Owner:STATE GRID BEIJING ELECTRIC POWER CO +2

Plant protection unmanned aerial vehicle path planning method based on improved A*

PendingCN119063730AReduce the number of inflection pointsreduce flight timeData processing applicationsInternal combustion piston enginesThree-dimensional spaceSimulation
The invention discloses a plant protection unmanned aerial vehicle path planning method based on improved A *. The method is applied to unmanned aerial vehicle field operation of Xinjiang and weapon groups. Firstly, a three-dimensional space diagonal distance is adopted as a heuristic function, the characteristic that the heuristic function needs to be weakened as a target point distance of an A * algorithm becomes closer is considered, a dynamic weight factor is designed, and a target path node of the unmanned aerial vehicle is selected more reasonably. Secondly, a target path is optimized through a redundant node deletion rule, redundant node coordinates are removed, and the number of inflection points in the flight process of the unmanned aerial vehicle is reduced; and finally, smoothing the adjusted flight line by using a cubic B-spline curve to obtain an optimal unmanned aerial vehicle path planning result. Compared with a traditional unmanned aerial vehicle route planning scheme, the optimal route can be effectively selected, meanwhile, the target searching time is saved, the flight route length is reduced, and the method can adapt to the characteristics of vast regions and diverse terrains of Xinjiang and weapon groups.
Owner:SHIHEZI UNIVERSITY

Electronic file intelligent management method and system based on AI

The invention provides an electronic file intelligent management method and system based on AI. The method belongs to the technical field of information, and comprises the following steps: carrying out multilevel feature extraction on file content based on a convolutional neural network and a recurrent neural network, and training an adaptive classification model based on the extracted features; a user portrait is constructed based on a deep neural network according to historical operation records and behavioral habits of a user, a reinforcement learning mechanism is introduced, and a recommendation strategy is dynamically adjusted according to the user portrait and a current context. Through the convolutional neural network and the recurrent neural network, multi-level features can be extracted from multi-modal files such as texts, images and audios, internal information of the files is fully mined, and the file classification accuracy is improved.
Owner:ZHEJIANG STARSINO INFORMATION TECH

RRT-Connect mechanical arm path planning method fused with beetle beetle algorithm

The invention relates to an RRT-Connect mechanical arm path planning method fused with a beetle antennae algorithm, and belongs to the technical field of mechanical arm path planning. The method comprises the following steps: taking a midpoint between a given starting point and a target point as a third node, and if an obstacle exists at the midpoint between the starting point and the target point, selecting one or more unit distance points on a perpendicular bisector of a connecting line of the starting point and the target point to translate upwards or downwards from the midpoint of the connecting line; the RRT-Connect algorithm fused with the beetle antennae can grow four trees from a starting point to a third node, from the third node to the starting point, from the third node to a target point, and from the target point to the third node. And meanwhile, in combination with the beetle antennae algorithm, the odor concentration sensing capability of the left and right antennas of the beetle antennae algorithm is fused into the RRT-Connect algorithm, so that the number of expanded redundant points is reduced, the path length is shortened, and the search efficiency is improved.
Owner:CHONGQING UNIV OF POSTS & TELECOMM

Long, medium and short term nested cascade hydropower station forecast scheduling system

The invention provides a long-medium-short-term nested cascade hydropower station forecasting and dispatching system. The system comprises a long-medium-short-term nested power generation optimal dispatching module and a reservoir inflow forecasting module. The long-medium-short-term nested power generation optimization scheduling module comprises a cascade hydropower station long-term optimization scheduling module, a cascade hydropower station medium-term optimization scheduling module, a cascade hydropower station short-term optimization scheduling module, a cascade hydropower station in-plant economic operation module and a scheduling model heuristic optimization algorithm module; the reservoir inflow forecasting module comprises a monthly inflow forecasting module based on a self-made probability distribution model, a daily inflow LSTM forecasting module based on multi-dimensional feature data driving and an hourly inflow forecasting module. An economic operation module is embedded, operation of cascade hydropower stations is optimized, and efficient management of time and space scales is achieved; a novel algorithm is adopted to correct an initial flow value, water balance and final water level constraint are met, and optimization efficiency is improved; the water inflow is predicted through data driving, model calculation is simplified, and the forecast scheduling efficiency is improved.
Owner:POWERCHINA HUADONG ENG CORP LTD +1

Generative adversarial network architecture search method and system based on hybrid convolution operation

The invention provides a generative adversarial network architecture search method and system based on hybrid convolution operation, and the method comprises the steps: firstly designing a wide search space, then constructing a super-net integrating all candidate operations, enabling the training process of the super-net to be completely differentiable through a continuous relaxation method, and enabling the training process of the super-net to be completely differentiable; therefore, efficient training is carried out by using a gradient descent method, and finally, the optimal operation on each position is selected to obtain the optimal architecture. Different from a traditional architecture search method, the method introduces a novel hybrid convolution operation into a search space, and optimizes a search process by integrating a partial channel attention mechanism. According to the generative adversarial network architecture search based on hybrid convolution operation, the long-distance information capturing capability and architecture search speed of the network are enhanced, and compared with a traditional search method, the method has higher efficiency and better performance, and the optimal neural network architecture suitable for tasks can be found more quickly.
Owner:NANJING UNIV OF INFORMATION SCI & TECH

Ground unmanned equipment cluster path planning and formation control method

The invention discloses a ground unmanned equipment cluster path planning and formation control method. The method comprises the following steps: step 1, constructing a grid map: constructing the grid map of an operation area of a ground unmanned equipment cluster according to map information; 2, completing path planning by adopting a bidirectional A * algorithm: simultaneously searching from a starting point and an ending point, estimating the cost from a current node to a target node through an evaluation function, and completing path planning; 3, a dynamic step length adjustment mechanism: dynamically adjusting the search step length according to the environmental complexity and the state of the unmanned equipment; 4, performing formation control based on the main path: performing formation control by adopting a Leader-Follower mode; and 5, generating a cluster path: generating a final action path by using the path planning and the formation control. According to the invention, the efficiency and adaptability of the multi-ground unmanned equipment system are improved, and the stability and safety of formation control are improved.
Owner:杭州智元研究院有限公司