14 results about "Dijkstra's algorithm" patented technology

The application relates to the technical field of supply chain logistics scheduling, and particularly discloses a chemical supply chain intelligent logistics scheduling optimization system based on big data analysis, which belongs to the technical field of supply chain logistics scheduling. A data acquisition module is arranged in the system to obtain a multi-agent multi-scheduling cycle supply chain correlation data set, a supply and demand analysis module generates a supply and demand gap distribution map according to order demand and inventory state data, and a road condition evaluation module generates a road traffic cost matrix based on vehicle positioning and road traffic data. An improved Dijkstra algorithm with a node weight dynamic updating mechanism and an edge cost time-varying attenuation mechanism is adopted in a scheduling optimization module, joint path optimization is carried out on the two types of graph matrices, an optimal logistics scheduling scheme is generated, and a command issuing module generates and issues a vehicle and path planning instruction set according to the scheduling scheme, so that the dynamic intelligent logistics scheduling demand of the chemical supply chain is adapted.

The invention discloses a hierarchical strategy-planner hybrid architecture for audio-visual navigation (AV-Nav) of a body-equipped agent, and aims to solve the challenge of tight coupling of multi-modal reasoning and precise action control in a traditional end-to-end model. According to the high-level strategy, a Transform-based network is adopted, deep fusion is carried out on real-time visual and acoustic features through a token and self-attention mechanism, and a navigation target point on a 3 * 3 local grid is output. The local target is then passed to a low-level classical planner that calculates the shortest path on the internally maintained dynamic navigation map using the Dijkstra algorithm and determines the atomic action to be performed. The method is characterized in that a planner is integrated with an interactive trial and error obstacle avoidance mechanism: once a collision signal Ct is received from the environment, the planner can immediately remove a corresponding blocked path edge from a navigation map, so that a system is forced to automatically re-plan a collision-free path in the next time step, and the navigation efficiency and robustness are remarkably improved. In addition, the architecture may perform end-to-end training in conjunction with novel audio enhancement strategies, including interfering sound sources, mixing or spectrogram masks.

The invention relates to the technical field of coal mine tunneling management and control, and discloses a coal mine tunneling face labor organization intelligent management and control system and a handheld terminal, and the system comprises an online meeting module which is used for enabling a coal mine tunneling face operator to participate in a meeting before class online; the personnel underground commuting path self-adaptive planning module is used for calculating the equivalent length of a commuting path by adopting a Dijkstra algorithm and planning an optimal commuting path according to a calculation result; the shift-over module is used for inputting key information of an on-duty state, so that a shift-over team can conveniently check the input information before getting on duty; and the operation process safety management and control module is used for positioning an operator so as to grasp the position of the underground personnel in real time, determining and delimiting a danger area according to a danger source, and persuading the operator to leave when the operator approaches or approaches the danger area. According to the invention, meeting before shift, underground commuting lines, shift change and personnel safety management and control are realized through a digital intelligent technology, and meanwhile, parallel operation of partial processes is realized, so that commuting time is shortened, labor intensity of workers is reduced, and safety is improved.

The invention discloses a distributed power distribution management method of a power distribution cabinet, and relates to the technical field of power distribution management. The method comprises the following steps: constructing a standardized data set through a three-phase linear power flow algorithm based on power operation data of a power distribution cabinet; training a graph neural network model based on a comparative learning mechanism and a generative adversarial network through the standardized data set to obtain a first power distribution neural network model; constructing a federated learning architecture, carrying out federated learning on the first power distribution neural network model based on the federated learning architecture, obtaining a second power distribution neural network model, inputting the standardized data set, obtaining a power distribution prediction result, and formulating a power dispatching basic strategy; based on the power dispatching basic strategy, the substitution method, the improved Dijkstra algorithm and the improved ant colony algorithm, optimization is carried out, a power dispatching optimization strategy is obtained and executed, feedback data is collected in real time to construct an optimization data set, the second power distribution neural network model is updated, and finally a power dispatching final strategy is obtained.

This invention discloses a site selection and capacity optimization method for electric vehicle charging stations that considers user charging accessibility, belonging to the field of electric vehicle charging infrastructure planning technology. The method includes constructing a dynamic traffic network considering road impedance, planning the shortest path using an improved Dijkstra algorithm, predicting the spatiotemporal distribution of charging demand using Latin hypercube sampling and Monte Carlo simulation, innovatively proposing an electric vehicle user charging accessibility index based on a Gaussian two-step moving search method to quantify the matching degree between charging station layout and demand, establishing a dual-objective optimization model with the goals of minimizing annual comprehensive cost and maximizing charging accessibility, and solving for the Pareto optimal solution set using the NSGA-II algorithm, and finally ranking the candidate schemes using the entropy weight method and the TOPSIS method to determine the optimal scheme. This invention balances the economic benefits for operators and the convenience for users, achieving precise allocation of charging resources and providing a scientific basis for urban charging facility planning.

The application provides a visual route path control method and device, equipment and a storage medium. The method comprises the following steps: acquiring a routing map, confirming a starting routing node and a terminating routing node in the routing map, generating a candidate path planning task based on the starting routing node and the terminating routing node, sending the candidate path planning task to a controller, confirming a target segment list that needs to be modified in response to a candidate path set sent by the controller, confirming at least one routing node as a necessary node on the routing map, generating a modified path according to the Dijkstra algorithm based on the target segment list and the necessary node, modifying the weight value of the target segment list in the modified path, and generating a routing control strategy. By directly displaying each routing node on the routing map, a user can modify the necessary node of the segment list of the candidate path on the routing map as needed, and the modification method is simple, fast and less prone to errors.

The application discloses a kind of based on Internet of Things battery swap station monitoring method and system, it is related to electric power dispatching optimization and intelligent energy management technical field, collection electric power data and pre-processing, set up node, extract electric power feature, generate electric power data block using SHA-256 algorithm, carry out electric power data error detection, construct the block chain system based on threshold editable hash function, carry out editing log record and responsibility tracking;Build road network diagram, calculate the dynamic weight of each edge of road network diagram, calculate optimal electric power path using Dijkstra algorithm, with MILP model as optimization target, define three-layer constraint condition, build XGBoost prediction model to predict the load of battery swap station, adjust electric power distribution strategy, by using improved SHA-256 algorithm and constructing the block chain system based on threshold editable hash function, improve the security of electric power data, by combining road network diagram to calculate dynamic weight, using Dijkstra algorithm and adjusting electric power distribution strategy, enhance the intelligentization of monitoring system.

The invention discloses a construction method of a charging load prediction model, a load prediction method and related equipment, and relates to the technical field of load prediction, and the key points of the technical scheme are that based on the position information of electric vehicle charging stations, a Dijkstra algorithm is adopted to construct a dynamic adjacency matrix representing the association relationship between different electric vehicle charging stations; calculating a Spearman rank correlation coefficient between the historical meteorological data and the historical charging load data, and screening at least one meteorological factor from the historical meteorological data as a meteorological feature according to the Spearman rank correlation coefficient; determining date features based on the date data; and inputting the dynamic adjacency matrix, the meteorological features and the date features into a graph convolutional neural network-Transform model, defining a regularized mean square error loss function, training the graph convolutional neural network-Transform model by adopting an optimizer, and outputting a charging load prediction model when the training is finished. According to the invention, the problem of low load prediction precision when a plurality of charging stations are dispersedly coupled is solved.

This invention provides a path fitting method and a geographic information visualization system incorporating this method. During online billing, the Dijkstra algorithm combined with a multi-validation model is used to verify and optimize vehicle on-the-go data and nodes in the fitted path, removing gantry systems with incorrect traffic relationships or abnormal traffic conditions to ensure that the final fitted vehicle travel path more closely reflects reality. Simultaneously, by establishing a comprehensive multi-layered data and business management system, the accuracy of complex business operations is guaranteed, while reducing the system's operational load.

This application provides a method, apparatus, device, medium, and program product for obtaining the aortic centerline, comprising: preprocessing a CTA image based on a pre-trained convolutional neural network algorithm to obtain target segmentation prediction results and trunk segmentation prediction results; preprocessing the CTA image based on a pre-trained convolutional neural network algorithm to obtain target segmentation prediction results and trunk segmentation prediction results for the target aorta; obtaining a set of vessel vertices and a first set of endpoints for vessel segments in the target aorta based on the vascular skeleton line of the target segmentation prediction results; extracting the aortic sinus endpoints from the first set of endpoints based on the trunk segmentation prediction results, and determining a second set of endpoints; extracting a second shortest path set from the first shortest path set based on the Dijkstra algorithm and the set of vessel vertices, and determining the second shortest path set as the aortic centerline to improve the efficiency of obtaining the aortic centerline.

This invention discloses a remote collaborative operation system for multiple electric excavators based on the Internet of Things (IoT), comprising: a status processing module for acquiring operational status data and preprocessing it to form a status data stream; a path construction module for constructing control path data; a controlled differential prediction module for constructing an improved neural controlled differential equation and outputting prediction results; a work map update module for creating a work map and updating its status; a task charging orchestration module for determining the set of excavators requiring charging and the target charging node; a collaborative planning module for obtaining the path sequence and operation sequence of each excavator based on the Dijkstra algorithm; and a command feedback module for issuing commands to the corresponding excavators, executing the command packets, and feeding back the execution status. This invention achieves remote collaborative operation of multiple electric excavators by combining the improved neural controlled differential equation and the Dijkstra algorithm.

The invention relates to the technical field of disaster simulation surveying and mapping systems, in particular to a disaster situation-oriented rapid surveying and mapping simulation system, which is characterized in that a hidden Markov model is used for statistical modeling of a stage demarcation sequence, and a stage mean value and a duration ratio are used as statistical characteristics in a state transition matrix; through weight adjustment and change chain construction of a high-frequency transfer stage group, probability distribution self-adaptive updating of disaster intensity between stages is realized, continuity and statistical stability of a disaster evolution process are maintained, a Dijkstra algorithm is adopted in path calculation, and by superposing gradient, friction resistance and energy loss parameters between geographic nodes, the probability distribution self-adaptive updating of the disaster intensity between the stages is realized. A path cost matrix is formed, the spatial precision and energy diffusion rationality of disaster propagation direction deduction are improved, clustering division and grade calibration are performed on a fluctuation value and a distance mean value in a surveying and mapping grid system, multi-level risk spatial distribution is generated, and the level difference of high-risk nodes in space is quantitatively expressed.

The invention discloses a charging guiding method based on Internet of Vehicles data driving, which belongs to the field of Internet of Vehicles and comprises five steps of data acquisition, path optimization, reputation management, double-layer iteration and guiding execution. Aiming at the problems that the traditional static path planning (such as the Dijkstra algorithm) does not consider actual traffic factors and the path search efficiency is low, dynamic path optimization is realized by improving an A * algorithm: a known item G (i) introduces a traffic light mechanism, traffic light waiting time is taken as a node weight to be incorporated into actual cost calculation, and the path search efficiency is improved. The problem of time consumption exceeding estimation caused by traffic light waiting is avoided; the pre-estimation item H (i) is added into a return road penalty item, the actual preference of'little turning around and no return road 'of a vehicle owner is met, and invalid path search is reduced; and finally, an optimal path is preferentially selected through a decision function, roads having no influence on an optimal result are avoided, the path searching efficiency is remarkably improved, and the total consumed time of'road driving time + charging station waiting time 'of the user is shortened.

The invention discloses a network information security supervision method, and relates to the technical field of network information. The method comprises the steps of obtaining access data of a user in a preset time window and performing semantic feature extraction to obtain a semantic behavior feature chain; inputting the semantic behavior feature chain into a Markov model, and outputting to obtain a network behavior risk index; judging the semantic behavior feature chain based on the network behavior risk index to obtain an abnormal behavior feature chain; constructing a time sequence behavior topological graph based on the abnormal behavior feature chain; performing risk path search on the time sequence behavior topological graph through a heuristic Dijkstra algorithm to obtain a cross-domain attack chain; and carrying out feature extraction on the cross-domain attack chain to obtain cross-domain attack chain features, inputting the cross-domain attack chain features into the multi-attribute decision model, and outputting to obtain an optimal processing strategy, thereby realizing supervision on network information security.