61 results about "Distributed algorithm" patented technology

The invention discloses a fire-fighting spatial-temporal trajectory data processing and off-line data warehouse construction method based on a stream batch fusion architecture, and relates to the field of digital fire fighting, and the core steps of the method are as follows: after multi-source data are collected by an edge gateway, high throughput distribution and storage are realized by Kafka; real-time optimization such as track cross repair and offset correction is completed through Flink streaming calculation; performing offline batch processing on the historical data by the Spark to generate an analysis result; generating an optimal rescue path through a distributed Dijkstra algorithm in combination with road topology, real-time traffic and fire fighting truck parameters; the method comprises the following steps: realizing mixed storage by adopting ElasticSearch secondary index and HBase total storage; based on hot and cold data hierarchical management, intelligent applications such as fire-fighting commanding and dispatching, battle comment and review and the like are supported. Practice verifies that the method can shorten the fire control time of the main urban area by 60%, significantly improves the fire emergency response capability, and is suitable for large-scale fire track data scenes.

The invention belongs to the technical field of edge computing, particularly relates to a water conservancy gate remote control system and a method thereof, and aims to solve the problem that a centralized cloud center architecture adopted by existing traditional water conservancy gate remote control cannot realize effective consideration and unified optimization in the aspects of real-time performance, reliability, economical efficiency, intelligent cooperation and the like. According to the scheme, the system comprises a multi-source data acquisition module; the collaborative decision-making module is connected with an edge calculation module, the edge calculation module is connected with a data processing module, and the data processing module is connected with the multi-source data acquisition module; according to the method, a cloud, edge and end collaborative multi-stage edge computing architecture is constructed, and a distributed algorithm based on a market mechanism is introduced, so that unified optimization and fundamental improvement on multi-dimensional targets such as low-delay real-time control, network interruption autonomy, computing resource elastic sharing and AI model safety evolution are realized.

The invention discloses an emergency multi-unmanned aerial vehicle distributed service quality assurance method and system based on private flow sensing integrated design. The method comprises the following steps: establishing an equivalent Nakagami-channel model for a physical channel of a base station-unmanned aerial vehicle link in an emergency scene, establishing a signal model based on rate division multiple access, and calculating effective capacity; establishing a sensing signal model, and constructing an extended effective capacity of the fusion communication and sensing performance indexes; constructing a joint optimization problem; designing a dynamic weighted federated learning (DW-FL) algorithm, and solving a joint optimization problem; according to the method, the private flow of rate segmentation multiple access is designed into the sensing dual-function signal, and the federated learning DW-FL distributed algorithm for dynamic weighting based on the performance contribution degree is combined, so that the sensing resources of the multi-unmanned aerial vehicle network can be subjected to high-time-efficiency collaborative optimization and distribution in an emergency scene with limited resources, and the resource utilization rate of the multi-unmanned aerial vehicle network is improved. And the total effective capacity of the system is maximized while the dual service quality of communication delay and perception probability is ensured.

The invention relates to the technical field of data processing, and discloses a log data batch updating method and device, equipment and a medium, and the method comprises the steps: obtaining a full-amount historical log in a preset database in real time, converting the full-amount historical log into an event stream, and writing the event stream into preset message middleware, obtaining a partition ID of the total historical log in the message middleware, generating a global unique identifier bound with a partition according to the partition ID to obtain a distributed ID, generating a consumer ID associated with the user identity information according to the distributed ID by calling a pre-designed distributed algorithm, and sending the consumer ID to the message middleware. And carrying out lock-free parallel updating on the database according to the consumer ID based on connection pool preheating and asynchronous submission to obtain an updating result, carrying out consistency verification on the updating result to obtain a verification result, and confirming the updating result according to the verification result. And the batch updating efficiency of the log data is improved.

The invention relates to the technical field of power distribution networks of power systems, in particular to a power distribution network elasticity improvement method based on distributed dynamic recovery, which comprises the following steps: step 1, acquiring operation data of a power distribution network; 2, establishing a topology reconstruction and rolling optimization dual-stage recovery model, taking maximized load recovery as a target, considering network constraints and resource operation constraints, and converting a probability model into a deterministic mixed integer linear programming problem through opportunity constraint conversion; 3, decomposing the model into distributed sub-problems by adopting an alternating direction multiplier method, introducing a boundary variable compensation mechanism to enhance the robustness under communication interruption, and realizing a multi-period dynamic recovery decision based on rolling optimization; step 4; an integer variable processing method combining projection and relaxation iteration is designed, and an induced acceleration objective function is constructed, so that the solving efficiency of a mixed integer sub-problem is improved, and the convergence and calculation speed of a distributed algorithm are ensured.

The application belongs to the technical field of target tracking and fusion, and relates to a nonlinear filtering distributed target tracking method based on variational inference. Measurement noise parameters of a conventional nonlinear filtering distributed algorithm are assumed to be known, and the application extends the nonlinear filtering distributed algorithm based on variational inference to a single-target tracking scene under unknown measurement noise parameters. Under the distributed algorithm, weight likelihood parameters of each sensor to the target state tend to be consistent, so that the algorithm has strong robustness, small calculation amount, high flexibility and timeliness, and in combination with the variational inference method, the algorithm can achieve high-precision target tracking in a distributed multi-sensor scene under unknown measurement noise variance. The algorithm has close tracking precision to a distributed particle filtering algorithm with known measurement noise variance, and enhances practicability.

The invention discloses an offshore wind power plant distributed frequency supporting method and device based on a communication link, and the method comprises the steps: obtaining a grid-connected topological structure of a target offshore wind power plant, dividing wind turbine generators on the same feeder line into a group of wind turbine generator units, and building a rapid communication link between a DC converter station and the wind turbine generators; determining a frequency supporting capability coefficient of the wind power plant according to the rotor rotating speed of the wind turbine generator; when the frequency deviation exceeds a threshold value, the active power of the wind turbine generator in the frequency supporting stage is determined according to the frequency supporting capacity coefficient, the frequency deviation of the power grid frequency and a consistency control algorithm; and determining whether the wind power plant exits frequency support according to the wind power plant output power and the system frequency, and controlling the wind turbine generator to stably recover the rotating speed. Based on the communication link and the distributed algorithm, the direct control wind turbine generator of the flexible direct current converter station collaboratively participates in frequency support, the frequency support effect is effectively improved, the response speed is increased, and safe and stable operation of the wind turbine generator is guaranteed.

This invention belongs to the field of signal processing, and particularly relates to direction-of-arrival (DOA) estimation for electromagnetic and sonar signals. Specifically, it is a subspace-based DOA estimation method based on a distributed algorithm, which can be used for passive localization and target detection. The method includes the following steps: S1, establishing a distributed array model at the receiver; S2, calculating the signal subspace of the received signal; S3, calculating the weight w(θ); S4, calculating the spatial spectrum and performing a spatial domain search to find the DOA value. The proposed method achieves distributed solution of the algorithm, thus fundamentally avoiding the shortcomings of centralized algorithms. Simultaneously, it maintains good estimation performance, similar to that of centralized algorithms, and outperforms distributed MUSIC methods at low signal-to-noise ratios.

The application discloses a kind of film thickness intelligent control methods in electromagnetic shielding film preparation, it is related to shielding film preparation technical field, comprising the following steps: setting multiple film thickness sensors;Electrostatic neutralization is applied in coating area;Microclimate control device is set in the periphery of coating platform;Non-contact deformation detection system is set, and the deformation time series data of each coating head unit is collected;Real-time value of film thickness and deformation is predicted;Compensation value is introduced to compensate film thickness deviation and deformation deviation, and control module executes displacement or pressure control according to compensation value to each coating head, and the input information of prediction model is updated synchronously;Cloud platform uses distributed algorithm to coordinate each control node, and is fused by asynchronous scheduling and prediction model;The application effectively inhibits the edge thickening / thinning phenomenon caused by coating head deformation, tension gradient.

The invention discloses a shared energy storage interactive configuration and lease pricing method and equipment oriented to multiple energy-using main bodies in a park. The method comprises the following steps: establishing a shared battery energy storage configuration and park operation collaborative optimization model and an in-park energy consumption main body energy management model; adding power interaction constraint between the park operator and the energy consumption main body based on the constructed model, establishing a shared energy storage interactive configuration model, and performing iterative solution based on a distributed algorithm to obtain the optimal energy storage configuration capacity of the park operator and the energy storage capacity and duration rented by each energy consumption main body; and taking the product of the lease capacity and the lease duration of each subject of the park as a lease settlement amount, establishing a differential pricing constraint based on a quantity discount mechanism according to the settlement amount of each subject, constructing a price interaction negotiation model between the park operator and the energy consumption subject, and performing alternate iteration solution to obtain a final lease pricing. According to the invention, reasonable configuration of energy storage resources and leasing strategy optimization are realized, the energy consumption efficiency is improved, and the energy consumption cost is reduced.

The invention provides a hybrid distributed decentralized control direct current micro-grid secondary control method and system, and the method comprises the steps: constructing a direct current micro-grid system which comprises a plurality of converters, and each converter communicates with a group of neighbor converters through a low-bandwidth communication network; detecting the output voltage, current and power of each power converter and the voltage of a common direct current bus; calculating a droop coefficient correction term by using a distributed algorithm; calculating a voltage correction term by using a distributed algorithm; power sharing and voltage recovery. According to the invention, the technical problems of poor voltage and power regulation of the micro-grid, low system stability, low response speed and low precision are solved.

The application discloses a distributed decoupling optimization method for a comprehensive energy system considering carbon trading, and comprises the following steps: establishing a centralized optimization framework of an electrically coupled system, initializing global data and information of a power system and a gas system; constructing a centralized model of a regional comprehensive energy system, wherein an objective function of the model comprises the power system, the gas system and carbon emission cost, and corresponding operation constraints are added to the objective function; converting an original non-convex problem into a mixed integer second-order cone programming based on SOCP relaxation, and performing reactive power optimization by using an ADMM consistent distributed algorithm; introducing consensus variables to represent the synergistic effect between the power and natural gas networks based on the ADMM method of consistent variables, so as to solve a distributed optimization problem of the regional comprehensive energy system; and finally, a case analysis of a regional electric and gas comprehensive energy integrated system is performed, and the distributed decoupling optimization of the system is realized by using the ADMM method of consistency. The application also comprises a distributed decoupling optimization system for a comprehensive energy system considering carbon trading.

The invention discloses an equipment access intelligent management and control method and system based on a distributed algorithm, and the method comprises the steps: when to-be-accessed equipment initiates an access request, the to-be-accessed equipment is verified by adjacent edge nodes, and after the verification is passed, each edge node cooperatively elects a task coordination node, the task coordination node carries out capability comprehensive scoring on the multi-dimensional capability index of each node according to the node capability evaluation model, and determines a candidate access node set according to the score; the task coordination node obtains feedback information when each candidate access node actually executes a task, and performs adaptive adjustment on the importance weight of each capability index of each candidate access node according to the feedback information, so as to update the comprehensive capability score; and according to the updated capability comprehensive score and the load condition of each candidate access node, the task coordination node determines a target node which finally bears the access of the to-be-accessed device from the candidate access node set.

The application discloses an engine model parameter dynamic calibration optimization method, which comprises the following steps: obtaining a target parameter to be optimized, and determining a corresponding dynamic optimization problem according to the target parameter to be optimized and a working scene; calling an algorithm portfolio according to the dynamic optimization problem, and updating optimal solutions of each algorithm in the algorithm portfolio through optimization information interaction and sharing strategies among algorithms; determining optimization promotion corresponding to each algorithm according to the optimal solutions obtained by each algorithm, and distributing corresponding calculation resources to each algorithm according to the determined optimization promotion degree; optimizing the target parameter to be optimized through the distributed algorithm portfolio, and calibrating the optimized target parameter as a parameter meeting a performance index according to an actual output value and an expected output value of an aero-engine simulation model. The application applies the algorithm portfolio to the dynamic calibration problem of the engine parameter, and improves the robustness and stability of the parameter calibration.

The application discloses a power distribution network optimization partitioning method considering three-phase power imbalance, relates to the technical field of large-scale power distribution network partitioning, and comprises the following steps: according to local sensitivity calculation, combining line admittance parameters of a power distribution network, calculating a voltage sensitivity matrix between nodes, and deducing electrical distance; then taking the electrical distance as an edge weight of a power distribution network topology, combining the concept of modularity, and using a Louvain algorithm to obtain a preliminary partitioning result; then according to a partitioning number constraint, selecting a flow method or a hierarchical flow method to perform sub-region merging, and obtaining a non-overlapping partitioning result; finally, according to a distributed algorithm requirement, the non-overlapping partitioning result can be overlapped based on a maximum information interaction principle. The application not only solves the influence of three-phase imbalance on partitioning, but also speeds up the partitioning algorithm, and has more advantages in partitioning effect and partitioning stability.

Systems and methods are disclosed for operating a wireless audio network including a plurality of wireless microphone units (e.g., wireless delegate units) and a central access point having a mixer. The wireless microphone units may perform voice detection and level sensing, and make a preliminary gating decision. The central access point may make a final gating decision, determine the granting of wireless communications channels, and generate a final mixed audio output signal.

The application provides a distributed unmanned aerial vehicle cluster parameter estimation method based on adaptive data selection, a cross matching mechanism is established based on fusion data transmitted at the last time, and then a new data selection factor lambda is obtained k (t) generating a strategy and determining whether unmanned aerial vehicle k exchanges data with neighbor unmanned aerial vehicles based on the new lambda k (t) The application improves the estimation accuracy while reducing the calculation cost and communication cost and saving the energy consumption of the unmanned aerial vehicle cluster. The distributed ADS-DLMS algorithm is designed based on the cross matching mechanism, which not only improves the parameter estimation accuracy of the distributed unmanned aerial vehicle cluster, but also reduces the calculation cost and communication cost of the unmanned aerial vehicle node, and provides a new information fusion technical scheme for the distributed unmanned aerial vehicle cluster.

The invention belongs to the technical field of demand response strategies, and provides an industrial park hierarchical response method and system based on a federal alternating direction multiplier method. An industrial park hierarchical response model giving consideration to multi-subject benefits of load aggregators and industrial users and considering the technological process is constructed, the industrial park hierarchical response model comprises a load aggregator layer and an industrial user layer, and the load aggregator layer optimizes a user-level response scheme with the minimum cost as the target; the industrial user layer optimizes an equipment-level response scheme by taking the maximum additional income as a target; based on the industrial park hierarchical response model, constructing a federal alternating direction multiplier method distributed algorithm; and solving the hierarchical response model of the industrial park through a federal alternating direction multiplier method distributed algorithm, and dynamically updating the compensation price by the load aggregator layer and the industrial user layer until the compensation price is converged to an optimal solution. The sensitive data protectiveness and the calculation efficiency can be improved, and a demand response scheme considering the multi-subject benefits and the technological process is obtained.

This invention discloses a solution to the joint path and destination planning problem based on a distributed generalized Nash equilibrium algorithm. First, the joint path and destination planning problem is modeled, transforming it into a non-cooperative game model. This model includes the objective function of each electric vehicle, global coupling constraints, and local constraints. Second, pseudo-gradients are used to transform the game model into a VI problem, introducing edge-based consistency constraints and a heterogeneous step size mechanism. Based on fixed-point iteration and proximal gradient operator theory, a distributed solution algorithm under complete information is proposed. Then, a global estimate of the plans of other users is introduced, proposing a distributed solution algorithm under partial information. This invention avoids the construction of double random matrices through edge-based consistency constraints, and can maintain solution accuracy while meeting low computational requirements when more users participate in the game model.

This invention discloses a multi-timescale market bidding method and system for energy storage systems considering dynamic opportunity costs, relating to the fields of electricity market and energy storage optimization technology. Through a dynamic opportunity cost quantification model, the potential impact of current decisions on future returns is dynamically quantified as constraints. While avoiding short-sighted decisions, a multi-market time-series coupling mechanism is constructed using the capacity state transfer equation and cross-market power balance constraints. This ensures the physical feasibility and temporal consistency of the dynamic allocation of energy storage capacity across day-ahead, intraday, and real-time markets. Furthermore, a stochastic-robust hybrid optimization framework is proposed, integrating generative adversarial network scenario generation, conditional risk value, and robustness margin to unify the optimization of expected returns and extreme risks. The constraints are transformed into optimization problems using KKT conditions and the Big M method, and solved using the ADMM distributed algorithm, ensuring the model's feasibility and real-time performance in engineering practice.

The application discloses a kind of comprehensive energy system space-time coordination scheduling method and system considering cross-zone interaction, comprising: to the comprehensive energy system of existing cross-zone interaction, with minimizing the running cost of whole scheduling cycle as target, the wind power output value of day-ahead scheduling stage is predicted, and the typical random scene of day-ahead scheduling stage is obtained according to the generation and reduction of random scene;By predicting the real-time fluctuation value of wind power in real-time scheduling stage, determine the current output plan, with minimizing the running deviation as target, with typical random scene as reference, correct the current output plan, to obtain optimal operation strategy;To the region of cross-zone interaction in comprehensive energy system, by setting constraint condition, decouple to regional interaction line, and by introducing alternating direction multiplier method for distributed iterative calculation, to obtain the optimal scheduling strategy between regions. Coordinate the economy and accuracy of different time scales, while using distributed algorithm to complete the interaction optimization between regions.

The application discloses a power distribution line load prediction and optimal scheduling method and system, relates to the technical field of intelligent scheduling of power systems, and comprises the following steps: generating time-aligned multi-source fusion input data sets; constructing a hybrid time series load prediction model; in the model training process, introducing a weighted quantile loss function; constructing a joint probability distribution model of renewable energy output and demand response participation rate, sampling the joint probability distribution; constructing a rolling time domain power distribution network optimal scheduling model; using a two-layer hybrid strategy to process uncertainty, decomposing the optimization problem into multiple sub-problems and using an alternating direction multiplier method with an adaptive penalty parameter to solve the problem in a distributed manner. The application establishes a multi-objective optimal scheduling model in a rolling time domain, realizes dynamic closed-loop optimization, and through the introduction of a two-layer hybrid solving strategy and an ADMM distributed algorithm with an adaptive penalty parameter, ensures global consistency, and significantly improves the calculation efficiency and scalability.

The invention relates to a spatio-temporal information driven reinforcement learning autonomous exploration method, which combines viewpoint information and spatial scale information of a robot position, enables a robot to reasonably evaluate gains of different areas, and avoids gain overestimation. By introducing spatial scale information, the exploration decision of the robot can be optimized, and resource waste and low-efficiency exploration are reduced. According to the method, a distributed SAC (Soft Actor-Critic) algorithm is creatively applied to an autonomous exploration task, and a distributed value function and an exploration strategy are fused. The fusion method enables the robot to obtain stable strategy update in a complex and high-uncertainty environment, and ensures that the robot maintains stability and high efficiency in long-term training, thereby improving the overall performance of an exploration task.

The invention relates to a multi-microwave-source heating system cooperative control method based on an F-Raft distributed consensus algorithm. The method comprises the following steps: constructing a microwave heating model simulation platform; establishing a numerical model based on a physical heating model, and configuring a plurality of microwave sources as distributed agent nodes; node communication is maintained through a heartbeat triggering mechanism for dealing with severe communication conditions; leader election, log copying and state synchronization are realized through an F-Raft algorithm; and dynamically adjusting the power output time sequence of the microwave source according to the real-time temperature field data. According to the method, the core consensus problem is modularly decomposed into three independent parts to improve the simplicity of a distributed algorithm, the dependence on a traditional centralized controller is reduced, the uniformity of a temperature field can still be maintained under the condition of uncertain communication, and the heating energy efficiency and the system robustness are improved.

The invention discloses a fully distributed path planning method and device based on correlation equilibrium and a medium, belongs to the technical field of traffic system path planning, and solves the problems that the Nash equilibrium solution is not unique and the centralization strategy distribution dependence is high in the existing traffic path planning. Comprising the steps that a distributed optimization model is established, the objective function of the model is to minimize global system cost, and related equilibrium constraints are applied; solving the distributed optimization model based on a fully distributed algorithm, updating the vehicle path selection preference during each iteration, and taking the path selection preference corresponding to each vehicle as a final driving path when the algorithm converges; according to the method, a mapping relation between suggested path preference and system travel cost is established, an objective function is constructed according to the mapping relation, and a mathematical model of a traffic jam problem is established in combination with simplex constraints and related equilibrium conditions which are required to be met by the path preference; and distributed correlation equilibrium calculation and path selection are realized on the premise that a central coordination node is not needed.

The invention discloses a user association and resource allocation method based on fair perception in a full-decoupling network, and aims to solve the problems of asymmetric interference, resource coupling and lack of user fairness caused by independent operation of an uplink and a downlink in a 6G full-decoupling wireless access network. According to the invention, a double-layer mechanism fusing short-term and long-term fairness is designed; short-term space fairness is realized through a congestion pricing mechanism; long-term time fairness is realized through embedding a Jain fairness index; a joint optimization problem is converted into a distributed two-layer multi-arm tiger machine model, a distributed algorithm based on mean index exploration, adaptive utilization of intervals and interruption driving coordination is provided, and efficient learning and fairness convergence are achieved while logarithmic-level regret and equilibrium convergence are guaranteed. According to the method, the Jain fairness index can be stabilized at a high level, the outage probability is reduced, the fairness and efficiency are remarkably superior to those of an existing scheme, and the method is suitable for a heterogeneous deployment scene of a 6G full-decoupling wireless access network.

The application discloses a kind of distributed material transport distribution methods based on average monotonicity strategy, it is related to distributed decision-making technical field.The method includes: establishing agent material transport distribution model, integrating capacity constraint and time window constraint;Design the task insertion and removal mechanism based on average monotonicity strategy, the task coordination benefit is smoothly handled, to ensure that the bid sequence is not decreasing;Average monotonicity correction is introduced in distributed auction-consensus framework, realize task conflict resolution and distribution consistency.The application solves the problem that traditional distributed algorithm cannot converge and cyclic distribution in logistics scheduling and other non-submodular scenes, significantly improves the task allocation success rate and system stability, while maintaining excellent scheduling efficiency, suitable for large-scale, dynamic agent collaborative scheduling scene.

The invention provides a deep reinforcement learning automatic driving model optimization method, and belongs to the field of intelligent driving. Constructing a simulation environment and training a basic IL model according to the long-tail scene; performing large-scale data collection according to the simulation environment, and performing model optimization on the basic IL model through a distributed RL algorithm; according to the optimized RL model, candidate tracks are generated, through an environment interactive learning mechanism of reinforcement learning, autonomous exploration and trial and error in a simulation environment are achieved, and dependence on real driving data is reduced; and in combination with a dynamic reward function, the model is guided to actively learn a coping strategy of a long-tail scene, and the generalization bottleneck of traditional data driving is broken through. In a closed-loop training environment combined with traffic simulation, model output can be corrected in real time through online strategy optimization, the adaptability of the system to dynamic changes is improved, and deviation caused by closed-loop control feedback is adapted.

The application provides a kind of dynamic energy regulation low-power distributed sensor network, node layer is distributed with multiple sensor nodes, responsible for collecting environmental data respectively;Network layer is connected with node layer, and is provided with dynamic energy management module, can adjust node energy according to each sensor node energy and task priority;Still be provided with distributed collaborative calibration module, use distributed algorithm, by neighborhood node collaborative mechanism, compare each node and neighborhood node data, automatically adjust calibration parameter, then upload data to cloud platform layer.Cloud platform layer analyzes, stores and assists sensor node calibration received environmental data.The application realizes the efficient monitoring and real-time calibration of each sensor under wide area environment, and overcomes the energy consumption bottleneck, insufficient measurement accuracy and low communication efficiency and other problems in the prior art.The system realizes efficient, accurate, low-power wide area monitoring through the comprehensive application of low-power sensor technology, dynamic energy regulation technology, automatic calibration technology, low-power wireless communication technology and intelligent cloud adjustment mechanism.

The present application relates to the technical field of Internet of Vehicles, in particular to a kind of multi-layer collaborative inference method in unmanned aerial vehicle assisted Internet of Vehicles environment.The present application designs the generative artificial intelligence multi-layer collaborative inference algorithm based on semantic perception in unmanned aerial vehicle assisted Internet of Vehicles environment.In the Internet vehicle computing environment assisted by unmanned aerial vehicle, based on communication model, inference delay model, content quality model and energy consumption model, the generative artificial intelligence multi-layer collaborative inference optimization problem based on semantic perception in unmanned aerial vehicle assisted Internet of Vehicles environment is established, to realize efficient balance between delay, energy consumption and content quality.The ADMM distributed algorithm is used to solve the proposed joint optimization problem, each vehicle node only solves its own subtask, greatly saves solving resources and delay, and realizes the optimal balance of energy consumption, delay and content quality.