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

925 results about "Simulated annealing" patented technology

Simulated annealing (SA) is a probabilistic technique for approximating the global optimum of a given function. Specifically, it is a metaheuristic to approximate global optimization in a large search space for an optimization problem. It is often used when the search space is discrete (e.g., the traveling salesman problem). For problems where finding an approximate global optimum is more important than finding a precise local optimum in a fixed amount of time, simulated annealing may be preferable to alternatives such as gradient descent.

Smart park full-life-cycle management system and method based on digital twinning and Internet of Things

The invention discloses a smart park full life cycle management system and method based on digital twinning and Internet of Things, and relates to the technical field of smart park management, and the system comprises a sensing edge module, a data governance module, an intelligent analysis module, a life cycle module and a twinning modeling module. According to the invention, multi-protocol access and edge computing capability are supported, and the data transmission efficiency and stability are greatly improved; the intelligent analysis module outputs an accurate analysis result by constructing a multi-class feature matrix and deep multi-task joint modeling mechanism, and provides data support and model guidance for dynamic management and intelligent decision making of the park; the life cycle module integrates a Kepler optimization algorithm and a multi-agent reinforcement learning and simulated annealing algorithm, establishes a collaborative optimization mechanism, realizes combination of global search and local fine tuning of resource scheduling, and effectively optimizes energy consumption, response time, space utilization and safety risks; and the twin modeling module constructs a park three-dimensional model, so that the interactivity and operability of the system are improved.
Owner:SUQIAN NANYOU DIGITAL ECONOMY IND RES INST +1

Urban functional area gridding identification and classification method based on multi-source spatio-temporal data

The invention provides an urban functional area gridding identification and classification method based on multi-source spatio-temporal data, and relates to the technical field of area identification, and the method comprises the steps: carrying out the dynamic grid division of an urban area through employing a self-adaptive grid division algorithm, obtaining the multi-source spatio-temporal data of a gridding area, carrying out the space-time alignment, and generating a multi-dimensional time sequence feature set; extracting multi-dimensional feature vectors of the grid units through a multi-modal feature fusion algorithm, acquiring long-range time sequence features by using a bidirectional gating recursion unit, calculating spatial correlation features based on a multi-head attention mechanism, and inputting the two types of features into a feature fusion network with jump connection to generate enhanced spatial-temporal features; and an optimization strategy is adopted, a functional area clustering center point is determined through density peak value clustering in sequence to obtain an initial division result, the initial result is optimized by combining spectral clustering of spatial adjacency constraints, and the boundary of the functional area is optimized by using a simulated annealing algorithm, so that accurate identification and classification of the urban functional area are realized.
Owner:XINHUA FUSION MEDIA TECH DEV (BEIJING) CO LTD

Liquid cooling server safety management system and method

The invention relates to the technical field of liquid cooling servers, and discloses a liquid cooling server safety management system and method, and the method employs a multi-mode sensor network to synchronously collect the thermodynamic parameters of a liquid cooling system at 100 Hz, and constructs a three-dimensional thermal field digital twinborn model after the processing of an extended Kalman filtering algorithm. A distributed cooling strategy is designed based on a federated learning framework, and each node locally trains a thermal dynamic prediction model and is optimized by a central aggregator. A dynamic control instruction is generated by using a near-end strategy optimization algorithm, and cooling liquid flow distribution is optimized in combination with a quantum derivative simulated annealing algorithm. And designing a dual-threshold phase change control mechanism, establishing a block chain log to ensure traceability and tamper resistance of the instruction, and realizing closed-loop feedback control through a CAN bus. The system and the method can accurately monitor and intelligently control the liquid cooling system, improve the heat dissipation efficiency, reduce the power consumption, and guarantee the data safety and the system stability.
Owner:百信信息技术有限公司 +1

Factor multiple optimization-based flood risk dynamic zoning method and system

The invention discloses a factor multiple optimization-based flood risk dynamic zoning method and system, and the method comprises the steps: constructing an initial flood risk assessment index set, carrying out the screening of the initial flood risk assessment index set, and calculating the frequency value of each index; constructing a comprehensive discrimination index value, and further screening to obtain a key flood risk assessment index set; constructing a risk assessment model and a key flood risk assessment index weight optimization model, and adopting a genetic-local search hybrid algorithm of simulated annealing optimization to calculate and obtain an optimal weight combination scheme; and calculating the similarity between the scene of each time point of the research area and a historical typical flood event, selecting an optimal weight combination scheme, and inputting the optimal weight combination scheme into the risk assessment model for calculation to obtain a flood risk dynamic division of the research area. According to the method, scientific and reasonable dynamic risk assessment can be provided, effective decision support can be provided for related decision makers, and the flood control toughness and the extreme environment coping capacity of the drainage basin area can be improved.
Owner:NANJING HYDRAULIC RES INST

Intelligent construction method based on data driving

The invention discloses an intelligent construction method based on data driving, and relates to the field of intelligent construction, and the method comprises the steps: collecting multi-source heterogeneous data of an intelligent construction project; performing feature selection on the acquired multi-source heterogeneous data by adopting a recursive feature elimination (RFE) algorithm to obtain a feature subset reflecting project progress, quality and safety; according to the feature subset, a multi-target model used for predicting project progress, quality and safety is constructed through a support vector machine (SVM) algorithm; a non-dominated sorting genetic algorithm NSGA-II is adopted to solve the multi-target model, a simulated annealing strategy is introduced in the iterative search process of the NSGA-II to jump out of a local optimal solution, and an optimized Pareto optimal solution set is output; and selecting an optimal parameter combination according to the Pareto optimal solution set, and adjusting the BIM model of the current intelligent construction project by using the optimal parameter combination. In view of low multi-objective optimization efficiency in intelligent construction project management in the prior art, the optimal model parameters considering different management objectives are obtained, and the efficiency is improved.
Owner:ZHUHAI HUAZHANG ENG MANAGEMENT CONSULTING CO LTD

Laser flight processing track planning method and device for complex curved surface

The invention relates to the technical field of intelligent laser processing path planning, and discloses a laser flight processing path planning method and device for a complex curved surface, and the method comprises the steps: obtaining the geometric continuity features of the curved surface, carrying out the region division and boundary feature extraction, obtaining the geometric features of the boundary points of each region, and obtaining the geometric features of the boundary points of each region; performing track generation based on boundary geometric features to obtain an initial processing track path, then performing heat accumulation analysis and track screening, identifying a track section needing to be adjusted, executing multi-target path scheduling optimization, and obtaining adjusted track execution feature data; processing sequence optimization is carried out based on the feature data, and an optimized processing sequence is obtained; performing thermal input analysis and parameter correction according to the processing sequence to obtain an updated processing parameter set; and finally, performing global coordination based on a simulated annealing algorithm, generating a final processing track set, completing track optimization and instruction set generation, and outputting a complete processing scheme. According to the method, the problem of insufficient track fitting degree can be solved.
Owner:SHENZHEN ZHIDING AUTOMATION TECH CO LTD

A method for scheduling optical imaging satellite missions based on a hybrid heuristic memetic evolutionary algorithm

The invention provides an optical imaging satellite task scheduling method based on a hybrid heuristic memetic evolution algorithm, solves the problem of how to accurately and efficiently obtain satellite images, belongs to the engineering field, and is based on a constructed satellite data set, a task data set, a satellite dynamics model, a task decision model and constraint conditions, takes maximizing the satellite task completion benefit value as an objective function, and constructs a satellite scheduling decision model; constructs a hybrid heuristic memetic evolution algorithm: obtains an initial observation path solution set through clustering processing and a constraint check model; inputs the initial solution into the satellite scheduling decision model to obtain a total benefit value of the initial solution; adopts a neighborhood search algorithm to obtain a feasible solution and a total benefit value of the initial solution, and obtains the quality of the feasible solution from the total benefit value difference; combines a simulated annealing algorithm and an ant colony evolution algorithm to search, and obtains an optimal solution for optical imaging satellite task scheduling after a preset stop condition is met; the invention realizes accurate and efficient acquisition of satellite images.
Owner:PLA PEOPLES LIBERATION ARMY OF CHINA STRATEGIC SUPPORT FORCE AEROSPACE ENG UNIV

Aircraft flight path optimization method and system based on inertial navigation

The invention provides an aircraft flight path optimization method and system based on inertial navigation, and relates to the technical field of flight control, and the method comprises the steps: generating an initial path according to a flight task and a limiting condition; acquiring original acceleration and angular velocity data; the attitude estimation precision is improved based on a quaternion updating algorithm in combination with Kalman filtering and an adaptive noise model; position updating is carried out by combining the acceleration and attitude information; dynamically adjusting and correcting parameters by using a machine learning algorithm; path optimization: defining an objective function which comprehensively considers path length, time, fuel consumption and safety, and adopting a particle swarm optimization algorithm and a simulated annealing algorithm to carry out optimization; evaluating and correcting the path, evaluating the quality of the path in real time, and correcting according to the track deviation degree, the safety and the fuel efficiency. According to the method, through combination of various advanced algorithms and models, the optimization effect of the flight path is remarkably improved, and efficient, safe and economical flight of the aircraft in a complex and dynamic environment is ensured.
Owner:XIAN SANHANG INTELLIGENT INFORMATION TECHNOLOGY CO LTD

Orthogonal fractured reservoir frequency variation AVO inversion and parameter prediction method

The invention discloses an orthogonal fractured reservoir frequency variation AVO inversion and parameter prediction method, and relates to the technical field of exploration geophysics, and the method comprises the steps: deducing and reconstructing a fluid saturation horizontal-vertical orthogonal fractured medium rock physical model based on a multi-group fracture-pore medium Chapman model, realizing fluid saturation rock physical modeling containing two groups of orthogonal fracture media; based on an anisotropic reflectivity algorithm and a fluid saturation horizontal-vertical orthogonal fractured medium rock physical model, realizing frequency variation AVO forward modeling of the fractured reservoir, and calculating a frequency variation AVO reflection coefficient and a synthetic seismic record of the fractured reservoir under the interface reflection model; a frequency-varying AVO inversion theory based on a Bayesian framework is constructed by taking a forward modeling process of the frequency-varying AVO as a drive, and the frequency-varying AVO inversion based on the Bayesian inversion framework is realized and reservoir parameter prediction is carried out by utilizing pre-stack seismic AVO data and introducing a simulated annealing-particle swarm algorithm in an inversion process. The method provides a new method and a new means for fine description and prediction of the fractured reservoir.
Owner:INST OF GEOPHYSICAL & GEOCHEMICAL EXPLORATION CHINESE ACAD OF GEOLOGICAL SCI

Ship shore-end intelligent management method, system and equipment and storage medium

The invention discloses a ship shore-end intelligent management method, system and device and a storage medium, and the method comprises the steps: carrying out the ship position prediction based on an electronic chart, the current navigation state data of a ship, and the current environment data, and obtaining the information of a to-be-docked ship according to the future position information of the ship, constructing a port operation knowledge graph based on the information of the ships to arrive at the port, the equipment data of the ships approaching and leaving the port, the historical tug operation data and the port data, and predicting the tug workload demand based on the port operation knowledge graph; taking tug operation data as a state space, taking a tug scheduling strategy and a cargo stacking layout adjustment strategy as an action space, taking maximization of wharf operation efficiency as a reward target, and adopting a reinforcement learning method and a simulated annealing method to carry out wharf operation scheduling optimization to obtain an optimal wharf operation scheduling strategy. According to the method, the port working efficiency is improved, manual interference is not needed when the wharf operation scheduling strategy is obtained, and autonomous decision-making can be completely achieved.
Owner:XIAMEN UNIV OF TECH

Communication system for dividing VLAN (Virtual Local Area Network) into network cards based on high-speed signal connection switch

The invention relates to the technical field of virtual local area networks, in particular to a communication system for dividing a virtual local area network (VLAN) into network cards based on a high-speed signal connection switch. The system comprises a network state monitoring module, a flow prediction module, a resource allocation decision module, a flow control execution module, a behavior pattern analysis module, a VLAN configuration adjustment module, a resource allocation optimization module and a network configuration adaptive adjustment module. According to the method, the software-defined network technology and the time sequence analysis are combined, the flow prediction is carried out by using the anomaly detection, the long-short-term memory network and the convolutional neural network, the bandwidth, the priority and the routing are adjusted through the optimization theory and reinforcement learning, the network flexibility is enhanced, and the strategy deployment efficiency is improved through the software-defined network technology; the recurrent neural network and the long short-term memory network deepen behavior recognition, the VLAN configuration is optimized by using the minimum cut maximum flow algorithm, the resource utilization rate is improved, the network adaptive adjustment adopts the genetic algorithm and simulated annealing, the routing is optimized, the QoS strategy is automatically adapted, and the adaptive capability is enhanced.
Owner:BEIJING ANTAI DIANTONG SCI & TECH

Filling body multi-age strength prediction and material inversion model construction method

The invention discloses a filling body multi-age strength prediction and material inversion model construction method, and belongs to the technical field of metal mine paste filling, and the method comprises the steps: carrying out the preprocessing operation according to the characteristic variables of a material ratio parameter and a tailing characteristic parameter, and obtaining a characteristic variable data set; performing data division on the characteristic variable data set according to a preset proportion in a preset multi-age period; constructing a multi-age strength prediction model, and training the model by using the training set to obtain an unconfined compressive strength prediction result in a preset multi-age; verifying the constructed multi-age strength prediction model by using a test set by adopting a Shapley addition explanation method; and adding a simulated annealing algorithm to the verified multi-age strength prediction model to obtain a reverse optimization model. The method has higher generalization ability and universality, and the filling design period is greatly shortened by combining model prediction of the multi-age strength prediction model with the simulated annealing optimization algorithm.
Owner:KUNMING UNIV OF SCI & TECH

Charging pile energy consumption scheduling method based on load prediction model

The invention provides a charging pile energy consumption scheduling method based on a load prediction model, and the method comprises the steps: carrying out the data collection and preprocessing, obtaining multi-source data, such as charging, equipment, environment and electricity price, from a charging station management system, and cleaning abnormal values; and then using grey correlation analysis to determine weights of factors influencing energy consumption, constructing a grey prediction model, and combining preprocessing and weight distribution data to predict future charging pile load demands. And the control center receives a load prediction result and vehicle reservation information, preliminarily plans a charging distribution scheme according to the real-time state of the charging pile, and optimizes scheduling by taking the minimization of the total charging cost and consideration of the user satisfaction and the stability of the power system as targets through a simulated annealing algorithm. During operation, an actual state and an external environment are continuously monitored, a charging plan is rapidly adjusted when deviation occurs, finally, vehicle charging bills of all departments are generated according to an execution scheme, and system performance indexes are regularly evaluated to realize continuous optimization. The method can effectively reduce energy consumption cost and guarantee efficient and stable operation of the charging station.
Owner:HUBEI INT LOGISTICS AIRPORT CO LTD

Dynamic equipment management service system and method for deep learning

The invention discloses a dynamic equipment management service system and method for deep learning. The method comprises the following steps: S1, collecting real-time operation data of equipment and preprocessing the real-time operation data; s2, constructing a long-short-term memory network model, and outputting comprehensive early warning indexes of the equipment; s3, constructing an optimization problem based on the comprehensive early warning index of the equipment, and setting an objective function and constraint conditions; s4, performing global search on the optimization problem by adopting an improved dragon fly algorithm, and generating a plurality of resource scheduling and maintenance candidate schemes; s5, performing local optimization on the candidate scheme by using a simulated annealing algorithm, and determining an optimal resource scheduling and maintenance decision scheme; and S6, feeding back the real-time prediction result and the optimal decision result to form closed-loop control. Through fusion of deep learning prediction, the improved dragon fly algorithm and the simulated annealing algorithm, real-time closed-loop control of accurate early warning of the equipment state and dynamic resource scheduling decision is realized, so that the failure rate is effectively reduced, and the equipment management efficiency is improved.
Owner:上海济士智能科技有限公司

Unmanned aerial vehicle route planning method based on simulated annealing-crown porcupine optimization algorithm

The invention discloses an unmanned aerial vehicle route planning method based on a simulated annealing-crown porcupine optimization algorithm, is applied to the technical field of unmanned aerial vehicle task planning, and aims to solve the problems of low efficiency and low planning accuracy when solving a route planning problem in the prior art. The method comprises the following steps: firstly, based on the actual demand and distance of a flight mission, establishing an unmanned aerial vehicle route planning model containing a range constraint; an initial parameter set of an improved crown porcupine algorithm is defined, a population is initialized through a Latin hypercube sampling strategy, and the diversity and coverage range of the population are ensured. Then, simulating the foraging behavior of the crown porcupines to construct an optimization model, and simulating the route search of the unmanned aerial vehicle in a variable environment; and finally, an optimal route planning sequence is obtained through iterative solution, and efficient task planning of the unmanned aerial vehicle under the limitation of the voyage is realized. According to the method, the global search capability and the solving speed of route planning are remarkably improved, and the method is suitable for the optimal route planning problem of the unmanned aerial vehicle in a complex environment.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Power system unit commitment model solving method based on improved simulated annealing algorithm

The invention discloses a power system unit commitment model solving method based on an improved simulated annealing algorithm, and the method comprises the steps: carrying out the optimization solving of a model objective function through employing the improved simulated annealing algorithm on the basis of a standard simulated annealing algorithm, and carrying out the optimization solving of a model objective function through adaptive temperature control and a multi-starting point strategy. And the global search capability and the convergence speed are improved when the unit commitment optimization problem of the power system is solved. Through the improvement of the invention, the algorithm can more effectively avoid falling into local optimum during solving, and the solving process is accelerated.
Owner:NANTONG UNIV

Dangerous goods transportation charging method based on risk assessment

The invention discloses a dangerous goods transportation charging method based on risk assessment, and relates to the technical field of dangerous goods transportation charging, and the method comprises the following steps: collecting the data of traffic flow, accident occurrence frequency, environmental factors and the like of each road segment in a dangerous goods transportation network, and calculating a dangerous goods transportation charging result based on a population coverage risk model and an accident expected value calculation method; and evaluating the potential risk level of each road section. Through multi-dimensional data fusion and an advanced algorithm, real-time assessment and path optimization of transportation network risks are realized. A population coverage risk model and an accident expected value calculation method are adopted, high-risk road sections are accurately recognized, transportation tasks are reasonably distributed, and safety is improved. Meanwhile, the genetic algorithm is utilized to optimize the transportation path, balance the cost and risk, reduce the transportation cost and guarantee the safety. Global path distribution is optimized through a simulated annealing algorithm, excessive load of high-risk road sections is avoided, task distribution is dynamically adjusted according to risk prediction, reasonable resource allocation is ensured, and transportation efficiency and safety are improved.
Owner:UNIV OF SCI & TECH OF CHINA

Traffic shaping method and system based on deep learning and antagonism strategy

The invention discloses a traffic shaping method and system based on deep learning and an antagonism strategy. The shaping method comprises the following steps: monitoring network traffic in real time through a traffic monitor, capturing information of traffic data, carrying out data preprocessing, carrying out traffic feature extraction through the preprocessed traffic data, and training a DNN classifier; jointly inputting the trained DNN classifier and traffic data into a patch generation module, preliminarily generating an antagonistic patch according to a dual simulated annealing algorithm added with a multi-objective optimization algorithm, and optimizing the antagonistic patch through a self-adaptive adjustment strategy and a bandwidth overhead adjustment strategy; and patch injection: injecting the optimized antagonistic patch into the original flow data through an injection controller, and shaping the original flow data. According to the invention, the problem that the confrontation capability and the privacy security protection capability of the Internet of Things equipment when facing equipment identification and traffic analysis attacks need to be enhanced is solved.
Owner:BOSHI FENGYUN (HUNAN) INFORMATION TECH CO LTD

Positioning method and system based on RSSI fingerprint database and generalized regression neural network

The invention discloses a positioning method and system based on an RSSI fingerprint database and a generalized regression neural network, and relates to the technical field of indoor positioning. The method comprises the following steps: acquiring RSSI values and corresponding position information of a plurality of sampling points in a target space, and constructing a fingerprint database based on the RSSI values and the position information; establishing a generalized regression neural network prediction model based on the fingerprint database, and globally optimizing model parameters of the generalized regression neural network prediction model by adopting a quantum particle swarm algorithm to obtain preliminary model parameters; performing local optimization on the preliminary model parameters by adopting a simulated annealing algorithm to obtain target model parameters, and training based on the target model parameters to obtain a target generalized regression neural network prediction model; and collecting a real-time RSSI value of a to-be-positioned target, inputting the real-time RSSI value into the target generalized regression neural network prediction model, and outputting a target position coordinate of the to-be-positioned target. By implementing the technical scheme provided by the invention, parameter optimization can be prevented from falling into a locally optimal solution, and the positioning precision is improved.
Owner:WUXI ZHENYUAN TECH CO LTD

Highway bridge intelligent maintenance decision-making system and decision-making method

The invention provides a highway bridge intelligent maintenance decision-making system and decision-making method. The system comprises a data acquisition module, a processing and storage module, a bridge condition evaluation module, a maintenance decision making module, an implementation and management module, and a monitoring and feedback module. The data acquisition module collects various types of data of bridges, environments, traffic and the like through multiple means such as fiber bragg gratings and vibration sensors. And the processing and storage module cleans the data by using a DBSCAN algorithm, processes the classified data by one-hot coding, and constructs distributed database storage based on a block chain. And the evaluation module adopts an Isomap algorithm to evaluate technical conditions and predict diseases. And the decision module generates a scheme by using a particle swarm optimization-simulated annealing algorithm, and analyzes and optimizes by using an analytic hierarchy process-grey clustering. And maintenance implementation is managed by means of a BIM model. The monitoring module collects data in real time and feeds back an adjustment strategy. The system can acquire data in real time through multiple sensors, accurately process data, accurately evaluate diseases, optimize maintenance decisions, dynamically adjust strategies, guarantee bridge safety and prolong the service life.
Owner:GUIZHOU QIANTONG ENG TECH CO LTD

Steel bar corrosion electrochemical parameter inversion method based on LSTM time sequence prediction

The invention provides a reinforcement corrosion electrochemical parameter inversion method based on LSTM (Long Short Term Memory) time sequence prediction, which comprises the following steps: S1, acquiring electrochemical time sequence data in a reinforcement corrosion process through an electrochemical workstation to form an original reinforcement corrosion electrochemical time sequence data set; s2, preprocessing is carried out to obtain a training set, a verification set, a test set and normalization coefficients of all parameters; s3, constructing and training an LSTM time sequence prediction model; s4, constructing and calibrating a steel bar corrosion electrochemical parameter forward modeling model; and S5, constructing an inversion framework fusing a particle swarm optimization algorithm, a simulated annealing algorithm and an Adam optimization algorithm, forming closed-loop cooperation by the particle swarm optimization algorithm, the simulated annealing algorithm and the Adam optimization algorithm so as to minimize an error between a target electrochemical response parameter and a theoretical electrochemical response parameter, and outputting an inversion result. According to the method, through organic combination of time sequence prediction and multi-algorithm cooperation, the problems that a traditional inversion method is low in precision and poor in stability are solved, and a reliable technical means is provided for reinforced concrete structure health monitoring.
Owner:SOUTHWEST JIAOTONG UNIV

Test box temperature intelligent regulation and control method

The invention relates to the field of temperature regulation and control, in particular to an intelligent regulation and control method and system for the temperature of a test box, and the method comprises the steps: calculating an optimal disturbance range in a simulated annealing algorithm, and obtaining the temperature in the test box in a next time period through the simulated annealing algorithm; the calculation method of the optimal disturbance range comprises the steps that a data set is obtained, the data set comprises multiple pieces of temperature data in the current time period, the dispersion degree of the data in the data set is calculated, the disturbance range optimization factor in the current time period is calculated, and the disturbance range optimization factor is in positive correlation with the dispersion degree. And taking a product of the disturbance range optimization factor and a preset disturbance range as an optimal disturbance range. The method has the effect of improving the calculation efficiency and the control precision of the simulated annealing algorithm.
Owner:WUHAN CLIMATE EQUIP

Comprehensive energy system planning optimization method and computing device

The invention discloses an integrated energy system planning optimization method and computing equipment, and the method comprises the steps: obtaining various equipment parameters, load demand data and energy prices in an integrated energy system, taking the lowest annual total cost of the integrated energy system as a first target, and taking the highest integrated energy efficiency as a second target; the total annual cost comprises the annual investment cost, the energy purchase cost, the annual operation and maintenance cost, the stepped carbon transaction cost and the wind curtailment and light curtailment penalty cost, and the comprehensive energy efficiency is used for representing the ratio of the total energy supply amount to the total energy consumption amount of each type of equipment in the comprehensive energy system; and solving the planning optimization model by using a simulated annealing chaos particle swarm complementary algorithm to obtain an optimal solution, and determining the optimal capacity configuration of each type of equipment in the integrated energy system based on the optimal solution. According to the method, the optimal capacity configuration of each type of equipment in the integrated energy system can be reasonably planned on the premise of meeting the load requirement of the system.
Owner:NORTH CHINA ELECTRIC POWER UNIV

Machine learning optimization method and system based on QUBO model and quantum annealing

The invention relates to the technical field of machine learning optimization, in particular to a machine learning optimization method based on a QUBO model and quantum annealing, and the method comprises the following steps: firstly, discretizing continuous parameters of the machine learning model into binary variables, and constructing a QUBO objective function; then optimizing the QUBO model by using a quantum annealing algorithm; and finally, solving a globally optimal solution by adjusting annealing parameters, wherein the annealing parameters comprise the initial temperature, the cooling coefficient and the number of iterations. The QUBO model can discretize continuous variables (including parameters such as weights and offsets) in models such as AR, SVM and CNN in machine learning into binary variables, so that a nonlinear relation is better processed, and the calculation complexity in the training process is reduced. Through the QUBO model, the regularization item can be better controlled, and the problem of overfitting is avoided. Meanwhile, the QUBO model can be solved in parallel through quantum calculation or a simulated annealing algorithm, and the calculation efficiency under large-scale data is remarkably improved.
Owner:GUANGZHOU UNIVERSITY

Method and device for determining parameters of cohesion unit

The invention provides a parameter determination method and device of a cohesion unit, and relates to the technical field of rock engineering. The method comprises the following steps: constructing a numerical model of a rock to be measured, which comprises a rock matrix unit and an embedded cohesion unit; applying a preset load to the to-be-tested rock for experimental testing to obtain a first stress-strain curve and a first peak point in the first stress-strain curve; constructing an objective function of a simulated annealing algorithm based on a stress error and a strain error between the first peak point and the second peak point; the second peak point is obtained by performing numerical simulation on the numerical model; and iteratively updating the current parameter value of the cohesion unit by adopting a simulated annealing algorithm and an objective function thereof to obtain an optimal parameter value. The parameters of the cohesion unit are automatically optimized and calibrated based on the simulated annealing algorithm, the calibration efficiency of the parameters of the cohesion unit in rock simulation can be remarkably improved, a large amount of time and manpower resources are saved on the basis of guaranteeing the parameter calibration accuracy, and the numerical simulation result is more reliable.
Owner:CHINA COAL RES INST

Grounding grid corrosion diagnosis method and system based on electric network theory and intelligent algorithm

The invention discloses a grounding grid corrosion diagnosis method and system based on an electric network theory and an intelligent algorithm. The method comprises the following steps: establishing a fault diagnosis equation of a grounding grid by using the electric network theory; iteratively solving the fault diagnosis equation by adopting a particle swarm simulated annealing fusion algorithm to obtain a globally optimal solution of the resistance of each branch of the grounding grid; and comparing the global optimal solution of the resistance of each branch of the grounding grid with the corresponding nominal resistance, and identifying the corrosion condition of each branch of the grounding grid. According to the method, the fault diagnosis equation is established by using the electrical network theory, the optimization model about the node voltage is constructed, and iterative solution is performed by using the particle swarm simulated annealing fusion algorithm, so that each corrosion branch of the grounding grid can be quickly and accurately identified, and technical support and technical reference are provided for grounding grid corrosion diagnosis research.
Owner:STATE GRID SICHUAN ELECTRIC POWER CORP ELECTRIC POWER RES INST

Wing design optimization method based on agent-assisted multi-initial-point simulated annealing

The invention discloses a wing design optimization method based on agent-assisted multi-initial-point simulated annealing, and belongs to the technical field of optimization design. Comprising the steps of 1, initializing algorithm parameters and a training data set; 2, constructing an agent model; 3, executing multi-initial-point parallel simulated annealing, and generating a batch of candidate new solution sets; 4, executing a double-elite active learning strategy based on the new solution set, screening the most potential sample to carry out real evaluation, and updating the agent model; 5, cooling the temperature and reducing the step length; 6, judging whether the cumulative evaluation times of the expensive objective function reach the set maximum evaluation times or not; if not, returning to the step 2; if yes, optimization is stopped; and finally, traversing the training data set, and selecting a sample point with the minimum real objective function value as a global optimal solution. According to the method, a multi-initial-point parallel simulated annealing search mechanism and a double-elite active learning strategy are combined, and the global optimal solution is quickly approached under the limited simulation times.
Owner:DALIAN UNIV OF TECH +1

Oil reservoir well pattern streamline fault avoidance calculation method

The invention belongs to the field of streamline calculation in oil reservoir development, and particularly relates to an oil reservoir well pattern streamline fault avoidance calculation method. The method comprises the following steps: (1) establishing a three-dimensional oil reservoir geologic model by using seismic and geological exploration data, and obtaining a gridded oil reservoir model by adopting a Delaunay triangulation technology; (2) streamline simulation is carried out based on a seepage equation of underground fluid, and streamline distribution of an oil reservoir well pattern is generated in the gridded oil reservoir model; (3) judging the contact risk of the streamline and the fault, and calculating an evasion path of the streamline; (4) well position optimization adjustment is conducted through a simulated annealing algorithm so as to ensure that the streamline and the fault do not intersect; (5) performing streamline verification, and if the verification is passed, outputting a streamline result; and if the streamline verification is not passed, returning to the step (4). According to the method, the streamline in the oil reservoir is simulated and optimized, the fault area is avoided in the well pattern layout, and efficient development of the oil reservoir is achieved.
Owner:QINGDAO UNIV OF SCI & TECH +1

Aviation composite material hot press molding workpiece loading method based on hybrid optimization algorithm

The invention provides an aviation composite material hot-press forming workpiece loading method based on a hybrid optimization algorithm. The method comprises the steps that workpiece demand information and autoclave platform information in composite material hot-press forming manufacturing are obtained; minimizing the opening times of the autoclave as a target function; establishing a two-dimensional multi-layer platform hot press molding workpiece loading model, defining space constraints, distribution constraints, resource constraints and type constraints among workpieces, and incorporating the constraints into model constraint conditions; a GA-SP is adopted to optimize a workpiece loading sequence and a rotation state, an optimal layout is generated through chromosome coding, decoding, selection, intersection and mutation operations, a hybrid optimization algorithm is constructed in combination with a neighborhood search strategy, efficient arrangement of workpieces on different platforms is realized in combination with a simulated annealing criterion, and an optimized loading scheme is output. The production efficiency of the hot press molding procedure can be improved, the manufacturing cost is reduced, and efficient technical support is provided for intelligent manufacturing of aviation composite materials.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS

Heterogeneous computing power resource allocation optimization method for deep reinforcement learning model training

The invention discloses a heterogeneous computing power resource configuration optimization method for deep reinforcement learning model training, and the method comprises the steps: constructing a heterogeneous graph for describing the features of a module and a data flow relation for a core module of a model-based Muzero series algorithm, mapping the features of nodes and edges to a same hidden space through a node-edge encoder based on multi-head attention, and carrying out the optimization of the heterogeneous computing power resource configuration. And predicting algorithm training time under different configurations by using the summarized global features. And performing configuration optimization according to the prediction training time output by the model by using a simulated annealing algorithm. And allocating CPU resources, GPU resources and memory resources according to the optimized configuration, and carrying out a training task. And calculating a resource equalization factor according to a training result, evaluating a resource configuration scheme, and continuously monitoring and optimizing. According to the method, the optimal balance between the performance and the resource consumption in the model training process is realized, the efficiency and the effect of model training are remarkably improved, and the technical difficulty of resource configuration optimization in a complex scene is reduced.
Owner:HANGZHOU DIANZI UNIV