29results about How to "Improve optimization accuracy" patented technology

The invention provides a process window optimization method and device, a medium, a program product and a terminal, and solves the problems of low light intensity contrast ratio, limited process window and high finished product defect rate of weak point and dead pixel areas through an objective function optimization and numerical programming solution method. And establishing a reasonable optimization model by taking the sum of squares of light intensity logarithmic slopes of weak and dead pixel areas as an optimization target and combining an edge section moving range, an edge position error and mutual influence factors. And meanwhile, a plurality of matrix methods such as a sparse matrix and a dense matrix are introduced, so that the optimization efficiency and precision are improved, the accurate optimization of the mask pattern is realized, the process window is expanded, the defect rate is reduced, the process stability and adaptability are enhanced, and the production efficiency and the product quality are remarkably improved.

The invention relates to a three-dimensional channel structure design method of a micro-channel heat exchange module and a related device thereof. The method comprises the following steps: inputting a heat conduction coefficient and a convective heat transfer coefficient of a micro-channel heat exchange module into a Rayleigh-Ritz coupling model for kinetic analysis to obtain channel structure parameter data; performing temperature field and pressure field acquisition on the channel structure parameter data, and performing time sequence feature extraction through a double-space-time automatic encoder to obtain a fluid dynamic feature data matrix; inputting the fluid dynamic characteristic data matrix into a recursive window Bayesian network, and calculating a heat exchange efficiency weight coefficient, a fluid resistance weight coefficient and a heat transfer uniformity weight coefficient to obtain a channel optimization objective function; and according to the channel optimization objective function, carrying out global iterative calculation on the channel main body structure by utilizing a genetic algorithm, and carrying out local optimization on the shunting and converging structure by adopting numerical simulation to obtain a three-dimensional channel structure optimization scheme. According to the method, the optimization precision of the channel structure parameters is effectively improved.

The invention belongs to the technical field of array antenna synthesis, and particularly relates to a thin cloth rectangular planar array antenna synthesis method, a program, equipment and a storage medium. According to the method, a thin cloth rectangular planar array antenna comprehensive method is converted into an optimization problem of an array element position in a quadrant, an improved sparrow algorithm is adopted, Kent mapping population initialization is introduced, and diversity of an initialized population is enhanced; the idea of a sine and cosine algorithm is fused in a position updating strategy of a discoverer, and a nonlinear sine learning factor is introduced to enhance all search capabilities; and finally, a crisscross strategy is introduced into a sparrow algorithm, and the optimization precision is improved by improving a search mode of a supervisor.

The invention relates to a central air conditioning system collaborative optimization control method based on a master-slave game. The method is suitable for the technical field of building energy saving and intelligent building control. According to the technical scheme, on the basis of the system structure of the central air conditioning system, an overall optimization control task is set and comprises an overall objective function and energy conservation constraints of a central air conditioning water system and a central air conditioning air system, and the overall objective function aims at minimizing the total operation energy consumption of the central air conditioning system while maintaining the thermal comfort degree of end users; an overall optimization control task of the central air conditioning system is decomposed into a series of subsystem-level main optimization tasks; according to the equipment composition of each subsystem, the subsystem-level master optimization task is decomposed into a series of equipment-level slave optimization tasks; and based on the main optimization task of the subsystem level and the slave optimization task of the equipment level, performing game optimization of the main-slave level by using a multi-level alternating direction multiplier method.

The invention discloses a process optimization method, device, equipment and system, and relates to the field of process optimization. The method comprises the following steps: acquiring input data of a user, wherein the input data comprises process boundary conditions; generating an orthogonal test scheme through a preset orthogonal test design method based on the input data, wherein the orthogonal test scheme comprises a plurality of orthogonal test samples; determining an alternative parameter combination based on the plurality of orthogonal test samples and a prediction model, wherein the prediction model is constructed based on a random forest regression model; and determining the alternative parameter combination as an initial solution of a simulated annealing algorithm, and obtaining an optimized parameter combination through the simulated annealing algorithm, wherein the optimized parameter combination is used for process optimization. According to the method, the optimized parameter combination can be obtained and used for process optimization.

The present disclosure provides a communication channel optimization method, the method comprising: obtaining the channel availability rate and channel usage information of some or all channels in the working frequency band within at least one cycle; determining the busy period of the service according to the channel usage information of the said part or all channels obtained within at least one cycle; determining the optimal channel of a certain busy period according to the channel availability rate and channel usage information of the said part or all channels obtained within a certain busy period of at least one cycle. The present disclosure also provides a communication channel optimization device, an electronic device, and a computer-readable storage medium.

The invention provides a language model training method and device, equipment and a storage medium, and the method comprises the steps: determining a first matrix, a second matrix and a third matrix according to a pre-training matrix of a language model; determining a first scaling vector of the second matrix and a second scaling vector of the third matrix; in response to the condition that the norm of the second matrix is greater than or equal to the norm of the third matrix, determining the second matrix and the second scaling vector as a target training matrix; obtaining a training text and an actual reply result of the training text; inputting the training text into the target training matrix to obtain a prediction reply result; determining a result error according to the predicted reply result and the actual reply result; and training the target training matrix based on the result error to obtain an optimized language model. Through the singular value decomposition algorithm, the training parameters of the language model are reduced, and the complexity and the calculation cost of the model are reduced.

The invention discloses a lightweight semantic segmentation model remote sensing sample optimization method, which comprises the following steps of S1, collecting public remote sensing semantic segmentation sample data in the Internet, sequentially performing data enhancement on the sample, constructing a lightweight semantic segmentation FCT network model, constructing a Beta hybrid model, establishing a label correction refinement model and optimizing the sample; s2, putting the high-quality samples selected in the step S1 into a sample database, and further screening and processing the samples according to requirements of an incremental learning technology on the samples; and S3, carrying out continuous training on the basis of the pre-training model by utilizing the data set screened out in the step S2. According to the method, rapid training of the model can be completed under the condition that a large number of samples exist, the purposes of correcting and refining the samples and removing redundant samples are achieved at the same time, and the precision of the model is not reduced while training resource consumption is reduced.

The invention belongs to the field of parameter optimization algorithms, discloses a meta-heuristic parameter optimization algorithm based on variation chain search, and particularly relates to a meta-heuristic evolutionary parameter optimization algorithm for variation chain search based on individual characteristics of sample points and overall generality of a sample point set. The method comprises the following specific processing steps: (1) hydrological model parameter determination and hierarchical division; (2) collecting an initial sample group; (3) extracting the optimal evolution direction (acceleration) of the individual; (4) optimizing subgroup division; (5) generating an optimization speed vector; (6) determining a search direction; and (7) convergence conditions. According to the method, on the basis of the similarity and anisotropy between the sample points in the hydrological model parameter optimization process, the sample points are updated and iterated through similarity analysis and a spatial clustering method, and a new thought is provided for the optimization method of the hydrological model.

The invention discloses an energy management balance scheduling optimization method and system based on deep learning, and the method comprises the steps: collecting multi-source feature data, carrying out the standardization and preprocessing, and obtaining structured multi-dimensional time series data; constructing a node relation graph by using an improved dynamic graph attention model, and generating a global attention matrix; node features are updated according to the global attention matrix, optimized features are formed, and a scheduling strategy is generated; optimizing the scheduling strategy by adopting an improved Harlisia eagle algorithm to obtain an optimized scheduling scheme; a node operation strategy is adjusted according to the optimization scheme, and power and load distribution is completed; and a final operation scheme is generated and dynamically corrected. According to the method, the improved dynamic graph attention model is constructed, and the Harris eagle optimization algorithm combining a boundary escape mechanism and a dynamic energy regulation and control mechanism is combined, so that intelligent modeling of multi-source data of an energy network, multi-target optimization of a scheduling strategy and dynamic balance scheduling of energy operation are realized.

This invention discloses an optimization design method for S-curve nozzles used in aero-engines that considers structural deformation, belonging to the field of aero-engines. The method selects key geometric parameters of the S-curve nozzle as characteristic parameters, chooses the structural deformation value at a characteristic location and two aerodynamic performance parameters as objective functions, then conducts orthogonal experimental design and constructs a response surface model based on the characteristic parameters and objective functions. After each optimization step, a cyclical evaluation is performed to obtain the final optimization result. This invention solves the problem of high aerodynamic performance loss caused by large structural deformation in S-curve nozzles used in aero-engines in existing technologies.

This invention discloses a software code optimization method and system based on big data. The method includes the following steps: constructing a code optimization knowledge base; acquiring the target code to be optimized and its corresponding estimated runtime data features; parsing the target code, extracting its structural features, and fusing the structural features with the estimated runtime data features to generate a target code fingerprint; performing similarity matching between the target code fingerprint and historical code fingerprints in the code optimization knowledge base to determine target historical code fingerprints that match the target code fingerprint; and determining a target optimization strategy for the target code based on the optimization strategy corresponding to the target historical code fingerprint. This invention can improve development efficiency, reducing the performance tuning work of senior architects, which can take hours or even days, to automated recommendations at the minute level, greatly improving software iteration efficiency.

This invention relates to a hierarchical optimization method for positive and negative pulse charging of lithium batteries based on deep reinforcement learning (DRL) and model predictive control (MPC), belonging to the field of battery management. This method establishes a long-term decision layer based on deep reinforcement learning and a short-term optimization layer based on model predictive control. The long-term decision layer learns the mapping relationship between positive and negative pulse charging actions and battery states across the entire process of positive and negative pulse charging, forming a long-term feasible region describing the overall charging trend, which is then passed to the short-term optimization layer in the form of a policy sequence. Near the long-term feasible region, the short-term optimization layer optimizes the positive and negative pulse charging actions on a short-term scale based on battery safety constraints and the current state, achieving optimal real-time control. This invention combines the efficient search capability of DRL in the nonlinear policy space with the rolling optimization mechanism of MPC under electrochemical constraints, realizing optimized control of the positive and negative pulse charging process of lithium batteries.

This invention discloses a method and system for coordinated scheduling of two-way arbitrage and demand response by energy storage aggregators. Through an innovative business model, it enables energy storage aggregators to achieve two-way arbitrage in both wholesale and retail markets, as well as multiple value acquisitions through collaborative demand response with industrial and commercial parks. This invention constructs a mixed-integer linear programming optimization model with the objective of maximizing the aggregator's total day-ahead revenue, accurately characterizing the aggregator's trading behavior and energy storage operation constraints in a multi-market environment. Addressing the nonlinear effective response capacity segmentation calculation and evaluation mechanism in demand response rules, this invention proposes a piecewise linearization method based on auxiliary binary variables and the Big-M method, accurately transforming complex nonlinear constraints into standard linear constraints. This method can effectively coordinate the optimal allocation of energy storage among multiple revenue sources, including wholesale market price arbitrage, retail market revenue sharing, and demand response compensation, significantly improving the economic benefits of energy storage aggregators and providing a scientific decision support tool for the commercial operation of energy storage aggregators.

This invention provides an analytical method based on CFD-DEM-PBM coupled simulation, an ultra-high nickel cathode precursor, a ternary cathode material, and a preparation method. The method includes: constructing a CFD-DEM bidirectional coupled simulation model for a reactor with a volume ≥1000 L; calculating the macroscopic flow field using RANS equations combined with a turbulence model; transferring the data to the DEM module via an Eulerian-Lagrange framework to calculate the trajectory of discrete particles in the flow field, inter-particle collision energy, and collision frequency; feeding back the calculated particle volume fraction to the CFD; introducing a population equilibrium model to select a stirrer configuration and rotation speed combination that maximizes the particle orientation index, minimizes collision energy below the critical collision energy threshold, and narrows the Span value over time, thus confirming the stirrer type and optimal rotation speed. The preparation method of this invention is simple, significantly shortens the preparation cycle of the ultra-high nickel cathode precursor, and has broad application prospects.

The invention discloses a femtosecond laser refractive operation parameter optimization method and device, a medium and a product, and relates to the field of refractive operation parameter optimization. Refractive operation parameters such as the diameter of an optical area and the thickness of a lens substrate are automatically designed according to parameters such as the diopter, the pupil diameter, the cornea center thickness and the predicted cornea cutting thickness of a patient; when the pupil diameter of a patient is large, the diameter of the optical area is increased as much as possible on the premise that the cornea biomechanical property after an operation is guaranteed. When the degree of the patient is high or low, all parameters are comprehensively considered to optimize the diameter of the optical zone. According to the method, corneal refraction operation parameters such as the optical area of the patient and the thickness of the lens substrate can be automatically optimized, automatic personalized design can be achieved, the postoperative visual quality and corneal biomechanical characteristics of the patient are comprehensively considered, the optimization accuracy is improved, the working intensity of doctors is reduced, and the working efficiency is improved.

The invention discloses a wing profile optimization method based on an improved firefly algorithm, belongs to the technical field of wing profile design, and solves the technical problems that multi-objective and nonlinear optimization conditions are difficult to process, the global optimization capability is insufficient and the design period is long in the traditional wing design process. The solution sequentially comprises the following steps of firefly algorithm improvement, airfoil data parameterization, parallel processing, constraint processing, external penalty function method, airfoil optimization, grid generation and flow field calculation. According to the method, the firefly algorithm is improved through the Matlab platform, the geometric shape of the airfoil is parameterized through the CST parameterization method, the lift-drag ratio of the airfoil serves as an optimization target, the airfoil is optimized through the improved firefly algorithm, the lift-drag ratio of the airfoil is improved, and then the aerodynamic performance of the airfoil is improved. The method has the advantages that the design is reasonable, the optimization precision is relatively high, and the designed wing section has high aerodynamic performance.

The present invention discloses a method for planning an offshore wind power distributed power source based on a hybrid improved seagull algorithm, comprising: optimizing the basic seagull algorithm to obtain a hybrid improved seagull algorithm; initializing the parameters of the hybrid improved seagull algorithm; using the initialized hybrid improved seagull algorithm to solve the upper optimization model, obtain the distributed power site selection and capacity information and input it into the lower optimization model for solving, and obtain the distributed power planning information; returning the distributed power planning information to the upper optimization model; calculating whether the total cost meets the conditions, if so, outputting the distributed power planning information, if not, repeating the above steps. The present invention takes demand response into consideration during the planning process, effectively improves the amount of distributed energy consumption, reduces network losses, reduces annual comprehensive costs, realizes economical and efficient, safe and stable operation of the system, and has higher optimization accuracy and faster optimization speed in solving the problem of double-layer planning of offshore wind power distributed power sources.

The invention relates to the technical field of recycled asphalt mixture design, and discloses a recycled asphalt mixture design method and device, computer equipment and a storage medium, and the method comprises the following steps: constructing a grading modulus data set; determining a grading optimization domain; constructing a collaborative optimization mechanism; the design model is based on a target modulus, a collaborative optimization mechanism is combined, individuals of an adaptive target are iteratively screened in a grading optimization domain according to a gradient strategy of first modulus primary screening and then cost-working condition double-constraint fine screening, and an optimal grading result with the target modulus reaching the standard, the cost being controllable and the working condition being adaptive is output. The device, the computer equipment and the storage medium all correspond to the method. According to the method, the optimization precision of the high-aging RAP and the grading fluctuation aggregate is improved through the collaborative optimization mechanism of the grading modulus data set and the adaptive reclaimed material characteristics, and the gradient screening strategy is combined, so that it is guaranteed that the target modulus reaches the standard, cost controllability and working condition adaptation are considered, the trial preparation cost is reduced, and the reclaimed material utilization rate and the engineering practicability of the design scheme are improved.

The embodiment of the application provides a parameter optimization method for an outer contour surface of a door glass and an electronic device, the method comprising: obtaining initial control point position information, reference point position information and a target gap range, wherein the reference point position information is used to represent a space reference position corresponding to the outer contour surface of the door glass, and the target gap range is used to represent a gap constraint condition corresponding to the reference point position information; constructing an initial outer contour surface corresponding to the door glass based on the initial control point position information; performing evaluation processing on the initial outer contour surface by using the reference point position information and the target gap range to generate a modeling and engineering evaluation score; and performing iterative optimization on the initial control point position information based on the modeling and engineering evaluation score to obtain target control point position information, wherein the target control point position information is used to construct a target outer contour surface corresponding to the door glass. The application solves the technical problems of long development cycle and low optimization precision of related door glass large surface optimization methods.

The present application mainly relates to the technical field of automobile engineering. The present application discloses a kind of automobile wire harness full life cycle cost optimization method, the method includes: the multi-source data of integration component selection, design, assembly and logistics stage is generated cost base data table, breaks the data island of each link;Identify the optimizable item of each cost optimization dimension, generate cost reduction proposal scheme;Filter target cost reduction proposal scheme and combine, calculate the cost saving amount of each combination scheme, select the optimal cost reduction scheme;The optimal scheme is output to the execution end, and receive execution result data for iterative optimization.This scheme realizes the full life cycle cost optimization through the combination optimization and closed-loop feedback of global perspective, ensures the synergy of each link cost measure, avoids the overall cost rise caused by local optimization, and finally realizes the full life cycle cost optimization.

The present disclosure relates to the field of image processing technology, and specifically to an image optimization method and device, a computer-readable storage medium, and an electronic device, wherein the method comprises: performing spatial domain filtering on the N-th frame image to obtain a first filtered image; obtaining a second filtered image after performing spatial domain filtering on the N-1-th frame image and a reference optimized image after optimizing the N-1-th frame image, wherein N is a positive integer greater than or equal to 2; determining a differential image according to the second filtered image and the first filtered image, and determining the state information of each pixel in the first filtered image according to the differential image; fusing the first filtered image and the reference optimized image according to the state information of each pixel to obtain a fused image; determining a weight image according to the differential image, and fusing the first filtered image and the fused image according to the weight image to obtain a target optimized image. The technical solution of the embodiment of the present disclosure improves the accuracy of image optimization and reduces the amount of calculation during image optimization.

The invention discloses a graphic rendering performance optimization method and system for a high-dimensional configuration space, and belongs to the technical field of graphic rendering. According to the method, firstly, configuration parameters of a rendering system are optimized and modeled into a multi-objective optimization problem, and an initial population is generated; evaluating a performance index value of each individual by using a performance prediction model, and forming a target function vector in combination with a penalty function; a tournament selection strategy based on crowding distance and ranking is adopted in the evolution process, and filial generations are generated through crossover and variation; then, by updating ideal points, calculating convergence indexes and distribution indexes and combining angular point solutions to screen individuals, a next-generation population is constructed; and continuously iterating until ending, and finally obtaining a non-dominated solution set as a recommended configuration scheme for graphic rendering performance optimization. The method is suitable for various rendering engines and application scenes, and the operation stability and the user experience of the mobile terminal graphic rendering system can be remarkably improved.

The present invention discloses a DNA storage coding optimization method based on a barnacle algorithm with weights and a hybrid mutation strategy, which specifically comprises the following steps: constructing an optimal DNA coding sequence that meets constraints, firstly constructing a certain number of DNA sequences as an initial set, and evaluating and ranking the fitness of the DNA sequences in the initial set. Secondly, using the obtained initial set, the barnacle algorithm based on weights and a hybrid mutation strategy is used to perform continuous iterative search and optimization to obtain a DNA sequence with a higher fitness. Then, it is determined whether the DNA sequence meets the constraints, and if so, it is added to the DNA coding set. Finally, the optimal DNA coding set is output. This method can search for a relatively optimal number of DNA coding sets.

The embodiment of the invention relates to the field of new energy battery lamination processes, and discloses a lamination model parameter adaptive optimization method, device and equipment and a storage medium, and the method comprises the steps: randomly generating a plurality of particles in a preset parameter space according to a lamination model of a battery lamination machine, and respectively representing a group of lamination model parameters; according to the lamination operation process executed by each group of lamination model parameters, the fitness value of each group of corresponding lamination model parameters is obtained according to the offset of the swing rod of the battery lamination machine; defining the distance and direction of each particle during each iteration movement, obtaining updated lamination model parameters during each iteration, and calculating an updated fitness value; and judging whether a preset convergence condition is met or not according to the updated fitness value, if so, outputting a corresponding lamination model parameter, and if not, continuing iteration. The problems that the iteration time is long and the optimization precision is low due to the fact that polling is carried out by adopting a fixed step length in an existing lamination model parameter optimization method can be solved.

The application provides a mechanical arm trajectory optimization method based on model memory data mapping and related equipment, and relates to the technical field of robot control. The application uses a visual language model to quickly identify and position the object to be worked on in the current work scene, quickly understand the expected work task issued by the user for the target mechanical arm, and then combine the model memory data mapping technology, the relationship key point constraint technology and the diffusion strategy technology in the mechanical arm trajectory optimization process. The expected mechanical arm trajectory required for the target mechanical arm to execute the expected work task on the object to be worked on in the current work scene is quickly and accurately optimized, so as to effectively improve the trajectory optimization precision, reduce the trajectory optimization time consumption and the trajectory optimization difficulty, improve the trajectory optimization real-time performance, achieve the synchronous improvement effect of the trajectory optimization real-time performance and the trajectory optimization precision, and ensure that the corresponding mechanical arm can more flexibly adapt to different work requirements of different objects.

The invention belongs to the technical field of wind power generation. The invention provides a wind power plant non-uniform layout micro-siting optimization method and system. The method comprises the following steps: determining a site boundary of a wind power plant according to a sensitive area range and a to-be-arranged position; a parallelogram is used as a grid shape arrangement of a minimum unit, and the maximum generating capacity of the whole wind power plant is used as an optimization target; and performing iterative optimization valuing on the origin coordinate, the long side azimuth angle, the included angle between the long side and the short side, the distance between the starting point and the short side adjacent point, the distance between the starting point and the long side adjacent point, the short side distance diffusion coefficient and the long side distance diffusion coefficient of the coordinate system to obtain an optimized wind turbine generator regular arrangement scheme. According to the method, optimization parameters of eight degrees of freedom are adopted, so that the optimization algorithm has excellent adaptive capacity, the optimization precision is improved, and the method can be suitable for micro-siting optimization of wind power plants on flat terrains and on the sea.

The invention discloses a high-voltage circuit breaker closing spring fatigue degree on-line monitoring method based on an energy storage action model, and belongs to the technical field of high-voltage circuit breaker monitoring, and the method specifically comprises the following steps: S1, constructing an energy storage action model; s2, acquiring a voltage value and a current value of an energy storage motor and an angle displacement signal of a crank of the operating mechanism; S3, determining a fitness function; s4, solving the energy storage action model by using a dogvessel colony algorithm and a fitness function to obtain the closing spring stiffness K; and S5, calculating the fatigue degree of the current closing spring according to the closing spring stiffness K. According to the online monitoring method for the fatigue degree of the closing spring of the high-voltage circuit breaker based on the energy storage action model, the energy storage action model including parameters such as crank angular displacement, closing spring rigidity, energy storage motor voltage and energy storage motor current is constructed, solving is carried out by adopting a dogvessel colony algorithm, and accurate online measurement of the closing spring rigidity is realized.

The application belongs to the technical field of aircraft engine oil supply and aircraft tank liquid sloshing, and particularly relates to a spatial position optimization layout method of an aircraft engine oil suction port, steps S1: obtaining the maximum mass of fuel consumed by the engine per unit time, determining the cross-sectional area characteristic parameter of the engine oil suction port, determining the minimum distance between the oil suction port and the nearby structure wall plate, and obtaining a first feasible domain in which the oil suction port and the nearby structure wall plate maintain a minimum distance or above; step S2: obtaining multiple working conditions that affect the fuel movement of the tank, obtaining the limit position of the oil surface trajectory under each working condition through simulation, and obtaining the common part of the limit positions of the oil surface trajectory under all working conditions as a second feasible domain; step S3: obtaining the common part of the first feasible domain and the second feasible domain as the final selection area of the best position of the engine oil suction port, and when the first feasible domain and the second feasible domain have no common part, modifying the tank parameters and returning to step 1.