19results about How to "Avoid local optimum problems" patented technology

The invention discloses a power grid safety enhanced intelligent scheduling method and device based on twinborn adversarial imitation learning and a storage medium, and belongs to the technical field of power scheduling. The method comprises the following steps: acquiring current real-time state information of a power grid system; inputting the real-time state information into a pre-trained scheduling strategy agent to generate a corresponding real-time scheduling action; and judging whether the real-time scheduling action in the current state is an expert-level action or not based on a state action data pair pool formed by the scheduling strategy agent in a training stage, and outputting a final scheduling strategy to a real environment according to a judgment result. Through real-time data analysis, intelligent decision and expert verification, the method can automatically adapt to the complexity and dynamic nature of the power grid, not only realizes efficient, safe and stable operation of the power grid, but also provides powerful support for intelligent management of the power grid.

The invention discloses a circuit breaker anti-tripping function test method and system, and belongs to the field of power system protection and control, and the test method specifically comprises the following steps: I, extracting circuit breaker data, and constructing an anti-tripping function map through a relation analysis technology; according to the method, the model optimization efficiency is improved, the local optimization problem is avoided, the model can make more accurate response to the state change of the circuit breaker under the actual operation condition, the flexibility and instantaneity of a test system are improved, the sample representativeness is optimized, the anti-tripping test effect is more reliable, the data processing cost is reduced, and the data processing time is shortened. According to the method, errors caused by human intervention are reduced, the intelligent level of a test process is improved, hysteresis of static assessment can be avoided, misjudgment caused by a single factor is reduced, the accuracy of health assessment is improved, the risk management and control capability of the system is enhanced, the possibility of fault occurrence is reduced, and the overall stability of the power system is improved.

The invention discloses a farmland soil heavy metal prediction method and system, and the method comprises the steps: carrying out the preprocessing of collected soil data, dividing a preprocessed data set, and obtaining a training set and a test set; initializing the extreme random tree regression algorithm model, and optimizing hyper-parameters of the extreme random tree regression algorithm model based on an ant colony algorithm; constructing a decision tree set based on an extreme random tree regression algorithm model; and training the decision trees in the decision tree set based on the training set, testing the performance of the decision trees through the test set, and evaluating the prediction precision based on the decision coefficient index to obtain an optimized extreme random tree regression algorithm model. According to the method, environmental and crop factors related to heavy metal pollution are effectively utilized, the adaptability of the model to complex environmental conditions is enhanced, the ant colony optimization limit tree algorithm can better capture the nonlinear relation between the soil heavy metal content and input variables, and the prediction precision and reliability are improved.

The invention provides an energy management method and device for a micro-grid group energy storage system, and the method comprises the steps: building an energy management model for the micro-grid group energy storage system; establishing an objective function and a constraint condition of the micro-grid group based on the cost of the energy storage system; and solving the established energy management model of the energy storage system of the micro-grid group by using an improved artificial hummingbird algorithm, and determining an optimal energy management strategy. According to the embodiment of the invention, the constructed micro-grid group energy storage system energy management model is solved by fusing the local escape operator and the improved artificial bee-bird algorithm based on the opposite learning technology, so that the solution space of the micro-grid group energy management model is better explored, the convergence speed is accelerated, and the problem that the solution falls into local optimum is solved.

This invention discloses a real-time rockfall detection method for open-pit mine slopes based on edge intelligence, belonging to the field of mine safety monitoring and early warning technology. The method includes: acquiring slope video to extract rockfall images, generating a dataset through manual annotation and region-adaptive dual-domain enhancement; constructing a DGN-YOLOv10 edge detection model, replacing the static convolutional layers of the backbone and neck network with dynamic convolutional modules DyConv, and replacing the C2f module with a lightweight composite module C2f_G_D; and using the Star Raven Optimization Algorithm (NOA) to dynamically and adaptively optimize the learning rate of the model to obtain the final model for rockfall detection. This invention suppresses environmental interference through region-adaptive dual-domain enhancement and the DyConv mechanism; it integrates the C2f_G_D module with the DFC attention mechanism to maintain the global receptive field while reducing the number of parameters and computational cost; and the NOA algorithm solves the local optimum problem under data scarcity.

The invention provides a multi-strategy improved distributed cross-domain cooperative task allocation optimization method, and belongs to the technical field of planning control. The method comprises the following steps: firstly, constructing a target allocation optimization model based on three types of target functions of comprehensive efficiency, time cost and resource utilization rate; a control distribution optimization model is constructed based on three objective functions of sensor detection capability, sending link connectivity and return link connectivity; and then combining the two models and corresponding constraint conditions to construct a cross-domain collaborative task allocation optimization model. And finally, carrying out distributed cross-domain collaborative task optimization allocation calculation by adopting a multi-strategy improved particle swarm optimization algorithm. According to the multi-strategy improved distributed cross-domain cooperative task allocation optimization method, the overall combat effectiveness is improved, the concealment and the collaboration among combat nodes are enhanced, powerful technical support is provided for cross-domain cooperative combat, and the task execution efficiency and the resource utilization benefit can be effectively improved.

This invention relates to the field of intelligent battery swapping recommendation technology, and discloses an intelligent battery swapping recommendation method and system based on historical battery data. The method includes: a vehicle central control system identifying the vehicle from a user-triggered battery swapping request signal to obtain parameter information of the current vehicle; querying historical battery matching data corresponding to the parameter information from a cloud database, and calculating the compatibility score between each battery model and the current vehicle; a battery swapping cabinet control system performing multi-objective collaborative recommendation decision-making with the compatibility score and the inventory batteries to obtain the optimal recommended battery identifier; and the battery swapping cabinet opening the corresponding compartment door according to the optimal recommended battery identifier to execute the battery swapping process and generate a battery swapping operation record. This invention improves the accuracy of battery compatibility assessment, can discover performance correlations between different battery models, and thus improves the automation and intelligence level of the entire battery swapping process.

The application discloses a high-voltage switch cabinet discharge cross-model dynamic positioning method, relates to the technical field of switch cabinet discharge, and comprises the following steps: acquiring historical multi-source monitoring data and local discharge fault records of a plurality of different models of switch cabinets, and generating a basic generalization model; deploying a multi-physical field sensor array on a target switch cabinet, and performing pretreatment and feature extraction; inputting the extracted features into the basic generalization model to generate an individualized diagnosis model; processing the features to acquire effective discharge pulses; generating a plurality of initial hypotheses about discharge source positions; for each initial hypothesis, simulating and calculating a theoretical waveform and a theoretical arrival time; calculating the comprehensive confidence of each initial hypothesis; and according to the comprehensive confidence, iteratively optimizing the initial hypotheses to determine the local discharge source positions. The application can deeply combine feature extraction and physical simulation, and through multi-dimensional matching and iterative optimization, realizes high-reliability automatic positioning of a discharge source in a complex switch cabinet scene.

The application relates to a large language model assisted cross-layer optical network routing optimization method. The method comprises the following steps: constructing the architecture of the cross-layer optical network, collecting network physical node composition, link resource parameters and service request characteristics; designing an auxiliary graph model, converting the cross-layer network routing optimization problem into a minimum cost path calculation problem in a virtual topology; building a large language model enabled auxiliary graph model weight optimization framework, and preliminarily generating the edge weight of the auxiliary graph model by the large language model; calculating the minimum cost path based on the initial edge weight of the auxiliary graph, evaluating the network performance index corresponding to the path, inputting the evaluation result into the large language model as feedback data, iteratively adjusting the edge weight until the performance index converges to the optimum; and completing the routing planning and wavelength allocation of all service requests in the auxiliary graph model according to the optimal edge weight. The application combines the large language model and the auxiliary graph model framework, improves the flexibility and multi-objective adaptation capability of the routing optimization, and realizes global optimization.

The application relates to the technical field of clean room air conditioner air outlet design, in particular to a clean room air conditioner air outlet optimization design method based on fluid dynamics; through construction of a multi-physical field coupling simulation model based on a fluid dynamics model, accurate prediction of a temperature field, a flow velocity field and a pollutant concentration field in the clean room is realized; the method takes Reynolds average NS equations and a discrete phase model as cores, combines a Standard k-epsilon turbulent flow model and a wall surface function method, effectively captures complex air flow organization and particle diffusion behavior in the clean room, provides a high-precision data basis for subsequent multi-target optimization, minimizes energy consumption under the premise of guaranteeing air quality, and has important engineering application value; and the technical problem that existing clean room air conditioner air outlet design methods mainly depend on experience and industry standards or focus on optimization of a single parameter and fail to fully consider the action of multiple factors is solved.

The present invention provides an intelligent car and an intelligent control method thereof, belonging to the technical field of elevator cars. The intelligent car comprises a car body, one end of which is provided with a switch slot, two air dispersion plates, the two air dispersion plates are fixedly connected to the switch slot, an exhaust fan, the exhaust fan is connected to the adjacent ends of the two air dispersion plates by rotating an electric roller, and two sets of driving mechanisms are arranged on the upper side of the car body. The exhaust fan can be driven to rotate between the two air dispersion plates by the rotation of the electric roller to perform ventilation treatment in the car body. At the same time, the rotation of the electric roller will also start the driving mechanism, and the sterilization and deodorization liquid such as spices in the spray pipe will be sprayed out through the diffusion nozzle through the driving mechanism. At the same time, the sprayed liquid is further diffused through the rotation of the exhaust fan, thereby achieving the effect of sterilization and deodorization, and improving the user experience.

This invention relates to a dual-modal flexible temperature-strain sensor and its fabrication method. The sensor achieves efficient detection of temperature and strain signals by combining direct ink writing and screen printing processes. The sensor features a unique design, comprising a rigid substrate and a flexible substrate, housing a thermal resistance temperature sensor and a piezoresistive strain sensor respectively, ensuring effective separation and accurate measurement of temperature and strain signals. The use of flexible materials and a FA-BP neural network algorithm based on an improved firefly algorithm enhances stability and measurement accuracy under varying environments. Furthermore, the sensor can still operate normally at high temperatures, outperforming common flexible thin-film materials on the market. This sensor demonstrates significant application potential in smart wearable devices, health monitoring, and industrial automation, providing a new solution for achieving high-precision and high-stability temperature and strain monitoring.

The invention discloses a distributed photovoltaic access distribution line protection method and device, and the method comprises the steps: collecting distribution line parameters, and building a multi-target distribution line protection fixed value setting model with the shortest protection fixed value time and the optimal sensitivity as targets; taking the multi-target distribution line protection setting value setting model as an improved hawk algorithm fitness function, and solving the model by using the improved hawk algorithm to obtain an optimal protection setting value; and the optimal protection constant value is embedded into the switch equipment, and distribution line relay protection is carried out. According to the method, the protection constant value adaptive to the current system can be automatically generated without knowing the fault voltage and current change characteristics of the system, and the method can be well suitable for the distributed photovoltaic access power distribution network. The reliability of the system is improved by improving the hawk algorithm, and the method has practical engineering significance. And a distribution line protection fixed value setting algorithm is embedded into the equipment, so that the integration level of the distribution equipment and the sensing capability of the distribution line are improved, and the operation reliability of the distribution network equipment is improved.

The invention provides an underwater unmanned platform path planning method based on adaptive step length deviation sampling, and relates to the technical field of path planning. The method comprises the following steps: firstly, constructing two bidirectional sampling trees, and selecting beacon points according to a current underwater unmanned platform path planning state; determining sampling points according to the selection mode of beacon points; after the sampling point is determined, firstly, a point closest to the sampling point is found in a sampling tree, and after the point extends to the sampling point, a new sampling point to be added to the sampling tree is obtained; judging whether the new sampling point is added to the sampling tree or not according to whether an obstacle exists between the new sampling point and a point closest to the original sampling point or not, deleting unnecessary sampling points from the currently constructed sampling tree, and selecting a new beacon point; repeating the above process until K times of sampling are executed to obtain two sampling trees; and finally, determining the shortest path from the starting point to the target point in the two sampling trees as the planned path of the underwater unmanned platform.

This invention belongs to the field of virtual reality technology, specifically relating to an XR task planning and instruction issuance method for brewing automation, comprising the following steps: Step 1: Collecting operator demonstration action sequences and voice process descriptions of the brewing production line in an extended reality environment to obtain a hierarchical task graph consisting of process layers and action layers; Step 2: Marking the execution devices corresponding to each action layer node in the hierarchical task graph as game participants, and the global coordination strategy includes the coordinated movement path scheme and rhythm allocation scheme of all execution devices; Step 3: Constructing a flexible Petri net model of the brewing production line based on the global coordination strategy, replacing the original strategy corresponding to the local subnet in the flexible Petri net model with the local replanning scheme, and continuing to issue instructions. This invention achieves formalized automatic mapping from process semantics to equipment actions, improving the programming efficiency, coordination quality, and anomaly response capability of the brewing production line.

The invention provides a method, device and equipment for determining a thickening CO2 flooding development scheme, a medium and a product, and relates to the technical field of oil and gas development. The method comprises the steps of obtaining a plurality of initial development schemes of a target oil reservoir, wherein the initial development schemes are obtained by permutation and combination of a plurality of preset values corresponding to a plurality of development characteristic parameters; inputting each initial development scheme into a pre-trained development dynamic prediction learning model to obtain development dynamic data corresponding to each initial development scheme; adopting a preset multi-target mathematical model to calculate a plurality of index values corresponding to each initial development scheme; determining a candidate development scheme set by adopting a multi-objective collaborative optimization algorithm according to the plurality of index values corresponding to each initial development scheme; and determining a thickening CO2 flooding target development scheme of the target oil reservoir based on the preset weight threshold values corresponding to the indexes. According to the method, through fusion of the preset multi-target mathematical model and the multi-target collaborative optimization algorithm, the optimization efficiency and precision of the thickening CO2 flooding development scheme are remarkably improved.

The invention discloses a refined raw material blanking method for buildings, relates to the technical field of building construction, and has the effects that data is acquired in real time through a data acquisition and modeling module, then the size and shape of a material are processed through a path generation module, and a surface image of the material is captured in real time; a geometric digital model of a material is generated after key features are extracted, then the geometric digital model of the material is combined with data collected by a sensor in real time, an initial path is generated through an algorithm, detailed optimization is performed by adopting specific parameters, and the accuracy of the path and the material utilization rate are ensured; and finally, the material utilization rate and the blanking precision are improved through an optimization algorithm, and after an initial path is generated, the system dynamically adjusts the path according to sensor data fed back by the real-time monitoring module.

The invention discloses a student score online monitoring method and system based on improved FA algorithm optimization SVR, and relates to the technical field of student score monitoring, and the method comprises the following steps: 1, carrying out the normalization processing of initial sample data; 2, setting a support vector regression model and improved firefly algorithm initial parameters; 3, initializing a population; 4, decoding operation is executed, and the fitness is calculated; 5, updating the positions of the fireflies; 6, updating the optimal solution; and 7, adding one to the number of iterations, and returning to the step 4. According to the method, model parameters are optimized by introducing the improved firefly algorithm, the prediction accuracy of the model is improved, compared with a traditional method, the improved firefly algorithm simulates glitters and mutual attraction behaviors of fireflies, better solutions are continuously searched in a parameter space, finally, a parameter combination enabling the model prediction error to be minimum is found, and the prediction accuracy of the model is improved. Therefore, the prediction accuracy of the SVR model is improved.

This invention discloses an optimization method for a natural gas pressure energy system based on a multi-objective frost algorithm, relating to the field of energy utilization technology. The specific steps are as follows: A comprehensive optimization model is established for the natural gas pressure energy system from three aspects: system efficiency, economic benefits, and environmental benefits. Then, algorithm parameters are initialized and the population is uniformly initialized. Next, a neighborhood structure is constructed by calculating Euclidean distance, and the fitness of each individual is calculated using a weighted sum decomposition method. The quality of local solutions is optimized by combining a frost neighborhood perturbation mechanism. In the stage of generating new individuals, simulated binary crossover and polynomial mutation operators are used, and a hard frost puncture mechanism is introduced to avoid local optima. Subsequently, the population and solution set are updated, retaining non-dominated solutions and maintaining the convergence and diversity of the solution set. Finally, the optimal result is selected by comprehensively evaluating the optimized solution set using the analytic hierarchy process (AHP)-fuzzy decision method. This invention can optimize and improve the performance of a natural gas pressure energy system.