88 results about "Constraint algorithm" patented technology

The invention relates to the technical field of production task decision making, in particular to a furniture production task automatic matching method and system based on intelligent manufacturing. And extracting personalized demand information corresponding to the production task, converting the personalized demand information into standard design parameters, and generating a customized task. The production task is disassembled into a plurality of sub-tasks, and each sub-task has independent attributes including process requirements, resource requirements and time requirements; and performing one-to-one mapping on the customized tasks and the subtasks to generate an execution plan. Performing resource matching by using a constraint algorithm based on the execution plan and the existing production resource information to obtain a production plan; and obtaining an optimal scheduling plan by using a constraint algorithm according to the production plan. And based on the real-time production data and the burst information, adjusting the optimal scheduling plan through the data analysis model to obtain a final scheduling plan. The method can cope with emergencies or real-time data changes, and manual decisions are reduced.

The invention provides a semi-supervised non-negative matrix factorization multi-view clustering method based on double constraints, which is realized by adopting a double-constraint algorithm, the double-constraint algorithm comprises a point constraint propagation algorithm and a pairwise constraint propagation algorithm, and specifically, when limited label information is obtained, the point constraint propagation algorithm and the pairwise constraint propagation algorithm are combined to form a semi-supervised non-negative matrix factorization multi-view clustering algorithm. According to the double-constraint algorithm, not only is label regularization constructed by using a point constraint propagation algorithm, but also multi-view hypergraph regularization is constructed by using a hypergraph pairwise constraint propagation algorithm, so that the detection accuracy of the model on multi-view clustering is improved. According to the method, the performance of multi-view clustering can be improved by utilizing the label information and the paired supervision information.

The invention discloses a semi-supervised multi-view clustering method based on double-constraint concept decomposition, which relates to the technical field of machine learning, combines point constraint with pairwise constraint, constructs a double-constraint algorithm based on prior label information of a complementary supervision mechanism, and realizes maximum utilization and progressive propagation of limited supervision information. And based on a concept decomposition framework, breaking through the limitation of a traditional method on non-negativity of data, processing complex data distribution in combination with a kernel method, obtaining a consensus matrix of multi-view data fusion as low-dimensional representation of views, performing clustering by adopting a k-means algorithm, and outputting a final category division result. According to the method, effective processing of complex multi-view data is realized, finally, the performance of a multi-view clustering task is comprehensively improved, and the bottlenecks in clustering precision, data adaptability and supervision information utilization efficiency in the prior art are broken through.

According to the emergency safety management GIS big data vector slicing service map, the vector data are thinned and simplified under each level by using the optimized global constraint algorithm, and the simplified and thinned vector data are cut and sliced and are published in a service form. The style of the tile is designed, and the client renders and visualizes the requested vector tile according to the specified style. And designing a cutting thought, and cutting point, line and surface elements in the vector data. And organizing and storing the generated tiles by using a quadtree index. A cache strategy of a client is designed, existing WMS and WMTS service standards are expanded, a vector tile map style is designed, and vector tiles are published in a service form. According to the method, a prototype system is created, static and dynamic slicing services of the vector map are realized, loading and rendering are carried out at a client, and particularly, when the emergency safety management quantity is huge and the quality is poor, the map precision and efficiency advantages are obvious.

The invention relates to the technical field of power grid digital checking, in particular to a checking method of a power grid digital design model, which comprises the following steps: S1, acquiring an initial parameter set of a to-be-checked power grid model; s2, performing data standardization processing on the initial parameter set; s3, constructing a three-dimensional topology verification model, and identifying abnormal nodes by using a spatial topology constraint algorithm; s4, mapping the real-time operation data to the three-dimensional topology verification model, and generating a dynamic parameter comparison matrix; s5, executing multi-dimensional check analysis, and generating a multi-dimensional difference report; s6, performing parameter replacement or topology reconstruction on the abnormal nodes, and outputting a final check result; according to the method, various power grid parameters are standardized in a unified manner, the accurate three-dimensional topology verification model is constructed, and automatic identification and quantitative correction of abnormal nodes are realized in combination with real-time data, so that the verification accuracy and operation reliability of the power grid model are remarkably improved.

The invention relates to the technical field of education and artificial intelligence, in particular to an intelligent teaching assistance system based on guided thinking, which comprises a heuristic guided teaching module which adopts a heuristic dialogue mechanism to replace direct answer provision and sets a constraint algorithm Constraint (T) = {1, if Count (Q in T) = 1 for'only one question is proposed each time '; the method comprises the following steps of: (1) calculating a formula (Q < in > T > = 0, if Count (Q < in > T > is not equal to 1), guiding students to autonomously think solutions, and automatically triggering knowledge summarization according to a completion signal detection algorithm CompletionScore = sigma (i = 1 to m) alphai * Match (keywordi, response) in answers of the students. According to the method, the understanding state of a student is automatically identified through a heuristic dialogue algorithm, personalized teaching content creation and dynamic teaching strategy adjustment are performed by utilizing the generation capability of a large model, and accurate course related knowledge retrieval is provided by combining a retrieval enhancement generation RAG technology, so that the limitation of directly giving answers to a traditional AI teaching system is broken through, and the teaching efficiency is improved. And real guided teaching is realized.

The invention provides a source load storage access optimization method and device based on user and power grid interaction external characteristics. The method comprises the steps of obtaining power grid user load data; extracting a plurality of multi-element load typical day scenes from the power grid user load data; carrying out weight calculation according to the user interaction external characteristics of each typical day scene; obtaining a user load portrait according to the user interaction external characteristics and the corresponding weights; constructing a source load storage double-layer collaborative planning model according to the user load portrait; and optimizing the source-load-storage double-layer collaborative planning model in combination with a normalized normal constraint algorithm, and generating a source-load-storage configuration scheme according to the optimized source-load-storage double-layer collaborative planning model.

The invention provides an unmanned cluster battle decision-making method and device. The method comprises the following steps: acquiring current state information and own resource information of a plurality of enemy target individuals in a cluster battle; based on the current state information of the plurality of enemy target individuals, estimating the state information of the plurality of enemy target individuals in the next time period by using an interactive multi-model filtering algorithm; inputting the state information of the plurality of enemy target individuals in the next time period and the resource information of the enemy target individuals into a constructed multi-target optimization model under a multi-constraint condition; solving the multi-target optimization model under the multi-constraint condition based on a multi-target optimization algorithm to obtain output target distribution information; and generating an own strike decision in different decision spaces according to the target distribution information. By adopting the scheme, the prediction of the enemy target individual is realized only by using the information observed in the confrontation, and the attack decision is obtained by comprehensively considering various constraint conditions in the battle, so that the method is more suitable for an actual confrontation scene.

The invention discloses a muscle-driven fruit fly larva-imitating intelligent body modeling and hierarchical motion control method, which comprises the following steps of: modeling a fruit fly larva multi-body-segment all-soft body based on a spring mass point method, modeling a muscle tension calculation node based on a muscle length-speed-tension nonlinear model, modeling a static fluid skeleton based on a volume constraint algorithm, and performing hierarchical motion control on the static fluid skeleton. And forming a drosophila larva imitating intelligent body under the simulation modeling framework. The method comprises the following steps: constructing a muscle bottom layer controller based on a von-mises cyclic distribution central mode generator for a fruit fly larva imitating intelligent body, and compressing a control dimension from a muscle number to a bottom layer controller parameter number; a reinforcement learning method is adopted to optimize a high-level strategy network for controlling parameters of a bottom-layer controller, so that a high-dimensional muscle system can drive a drosophila larva imitating intelligent body to generate multi-mode wriggling movement under the coordination of the hierarchical movement control method. The invention provides a novel structural design and a novel control algorithm, which are beneficial for promoting the application of the soft robot in the fields of industry, medical treatment and the like.

The invention discloses a digital human-driven key point generation method and device and a medium, and the method comprises the steps: extracting a static spatial domain feature and a dynamic time domain feature of a standardized target voice through a constructed spatial domain-time domain double-flow feature decoupling model; according to a cross-modal dynamic attention fusion mechanism, performing dynamic weighted fusion on the static spatial domain features and the dynamic time domain features to obtain fusion features; according to a biomechanical constraint algorithm, trajectory operation is carried out on the fusion features, and a key point motion trajectory conforming to biomechanical constraints is generated; and through a joint loss function, calibrating the key point motion trail and the key point trail of the standardized target video, and generating a final face key point. According to the scheme, the synchronization precision of the digital human lip shape is improved by constructing a space-time domain double-flow feature decoupling architecture; the non-physical motion is eliminated through a biomechanical constraint algorithm, the track smoothness is improved, the phoneme and lip-shaped semantic association ability is enhanced through cross-modal fusion, and the generalization of roles is ensured.

The invention discloses a fault constraint-based Dal-UNet stratum curved surface reconstruction method, which is applied to the field of information processing and the field of oil-gas exploration, and aims to solve the problems that an existing deep learning reconstruction algorithm is based on a supervised learning mode, a large amount of data information and manually marked training labels are needed for pre-training a network, the required economic and time cost is relatively high, and the efficiency is relatively high. The problems that in geological research with relatively rare data, an algorithm based on supervised deep learning is prone to generating an over-fitting phenomenon and is more difficult to be directly applied to a stratum curved surface reconstruction task of a complex geological structure containing a fault in the prior art are solved. According to the method, implicit features are captured from limited and sparse stratum curved surface data by using the Dal-UNet network, and a fault-containing complex stratum curved surface is reconstructed in combination with a fault constraint algorithm, so that the real landform is effectively restored.

The invention provides a park integrated energy optimization scheduling method and system, and relates to the technical field of energy scheduling, and the method comprises the steps: obtaining the state information of a park integrated energy system; establishing an energy scheduling model according to the state information; determining a constraint condition and a target function of the energy scheduling model; solving the energy scheduling model through an epsilon-constraint algorithm under the constraint of a constraint condition by taking the minimization of the objective function as a target, and generating a scheduling solution set; processing the scheduling solution set in a fuzzy decision mode to generate an optimal scheduling solution; and taking the optimal scheduling solution as a scheduling scheme, and carrying out optimal scheduling on the park integrated energy system. The energy utilization efficiency can be improved, and the energy safety and flexibility of the park are enhanced.

The invention discloses an elderly electroencephalogram time sequence diffusion staging method based on intergenerational transfer learning, and belongs to the crossing field of biomedical engineering and machine learning. In order to solve the problems of poor adaptability of weak features and strong noise of the elderly electroencephalogram, dependence on large-scale labeled data and the like, an elderly specificity time sequence diffusion model is constructed. In the preprocessing stage, an improved ICA algorithm is adopted to separate myoelectricity artifacts, and multi-band features are extracted; the diffusion model is fused into an elderly exclusive noise library and physiological priori, denoising is performed through a time sequence U-Net inverse process, and N3 period weak delta wave features are enhanced in combination with an attention mechanism; a common disease constraint algorithm is introduced to construct a common disease-sleep correlation tensor, and a loss function is optimized. The small sample enhancement module generates pseudo samples conforming to elderly features, and semi-supervised training reduces labeling dependence; and stage conversion advanced prediction is realized by combining stage reasoning with a BiLSTM classification head and dynamic confidence correction. According to the method, the electroencephalogram staging accuracy of the old people is improved by 18%-22%, the accuracy in a small sample scene reaches 82%, the dynamic conversion prediction error is smaller than or equal to 10 seconds, and the method is suitable for home sleep monitoring and clinical auxiliary diagnosis of the old people.

This invention discloses a method, system, and device for measuring the orientation of nonwoven fibers, belonging to the field of nonwoven fabrics. The method binarizes a color image of the nonwoven fabric to extract the fiber skeleton. Based on this, a T-shaped corner detection algorithm is used to determine the intersection points of the fiber skeleton. A fiber length constraint algorithm is used to limit the length of the skeleton segments, and a fiber projection algorithm is used to calculate the projected length of the skeleton segments, thereby determining the orientation of each length-constrained skeleton segment. The T-shaped corner detection algorithm used in this invention can accurately handle fiber skeleton intersection points, the fiber length constraint algorithm can effectively avoid interference from long fibers on orientation measurement, and the fiber projection algorithm can accurately calculate the projected length of the fibers. These three algorithms together make the measurement of nonwoven fiber orientation more accurate and efficient, and are applicable to various types of nonwoven materials.

The application relates to the field of equipment operation state monitoring, in particular to a medical equipment operation state monitoring and adjusting system based on big data, which comprises a multi-source acquisition module, a federal preprocessing module, an equipment multi-feature extraction module, an analysis module, an optimal adjustment strategy module and a database, the multi-source acquisition module acquires operation parameters and generates adversarial samples; the federal preprocessing module embeds a fairness constraint algorithm to output an initial balanced data set; the equipment multi-feature extraction module extracts frequency domain and space-time features and integrates the features into a time feature sequence; the analysis module enables a random greedy algorithm according to a dynamic decision, inputs a processed feature subset into a bidirectional LSTM model to output equipment states; and the optimal adjustment strategy module combines Bayesian network probability reasoning and deterministic rules of a rule engine to generate an optimal adjustment strategy, so that the data robustness and fairness are improved, accurate equipment state determination and optimal strategy generation are realized, and the safe operation of medical equipment is ensured.

The invention relates to the technical field of detection robots, in particular to a consistency-constrained detection robot task allocation method, which comprises the following steps: determining to-be-detected tasks and the number of detection robots; the to-be-detected tasks and the number of the detection robots are input into a task allocation model, task allocation results of the detection robots are obtained, and the task allocation model is constructed based on a consistency constraint algorithm; and driving the detection robots to perform detection work based on the task allocation result, updating the number of the to-be-detected tasks and the number of the detection robots in real time in the detection process, continuously inputting the to-be-detected tasks and the number of the detection robots into the task allocation model, and updating the task allocation result until the detection is finished. The detection robot task allocation method based on the consistency beam is more practical after the tasks are allocated, allocation of the detection tasks can be dynamically adjusted in the detection process, and resource waste is greatly reduced.

This application discloses a method, apparatus, and storage medium for printing imposition positioning, which improves imposition efficiency and accuracy while reducing the operational difficulty of imposition. The method includes: obtaining first imposition layout parameters for a selected printing press model from a limited dataset based on the selected model and task requirements; establishing a two-dimensional reference coordinate system based on the first imposition layout parameters, with the vertex of the printing plate of the selected model as the starting point; calculating the first reference position of a single row of code values ​​in the two-dimensional reference coordinate system according to a dynamic constraint algorithm; mapping the first reference position to a multi-row positioning network based on a topology expansion matrix; verifying whether the multi-row positioning network conforms to preset layout rules; if so, performing adaptive error compensation on the multi-row positioning network and outputting the final positioning result.

Network risk control neural rule update method and device based on feature pattern consistency constraint. The method includes: S1. Determine the neural rules to be updated, mainly including the neural rules in the deep image classifier based on the convolutional neural network; S2. Randomly strengthen the input data of the neural rules within a well-constrained range; S3. Run the neural rules and calculate the loss based on the feature pattern consistency constraint algorithm and the classification result; S4. Optimize the loss through backpropagation to achieve the update of the neural rules. The present invention can complete the update of neural rules for network risk control on the premise of introducing a small additional computational overhead, provides a neural rule update method with better network risk control effect, and can more effectively improve the security and robustness of the target network; provides a flexible neural rule update method for network risk control that can be implemented in a plug-and-play manner at any usage stage of the network, including the initial stage and the ex-post stage.

The application provides a park comprehensive energy optimization scheduling method and system, and relates to the technical field of energy scheduling. The method comprises the following steps: obtaining state information of a park comprehensive energy system; establishing an energy scheduling model according to the state information; determining constraint conditions and an objective function of the energy scheduling model; under the constraint of the constraint conditions, solving the energy scheduling model by an epsilon-constraint algorithm to generate a scheduling solution set, processing the scheduling solution set by a fuzzy decision method to generate an optimal scheduling solution, and taking the optimal scheduling solution as a scheduling scheme to optimize the park comprehensive energy system. The application can improve energy utilization efficiency and enhance park energy security and flexibility.

The invention provides a micro-grid energy distribution optimization method based on a genetic algorithm, and belongs to the technical field of micro-grid energy optimization based on computer data processing. The method comprises the following steps: firstly, collecting operation parameters of a micro-grid, then determining the priority and the operation parameters of discharging of a photovoltaic unit, a diesel generator and an energy storage unit, determining a power constraint condition, and taking power instructions of the energy storage unit, the photovoltaic unit and the diesel generator in every N time periods at T moments as an initial particle population of a gene random generation genetic algorithm; dynamically adjusting the generated initialized particle swarm by using a charge and discharge constraint algorithm, and finally, carrying out iterative calculation on the particle swarm generated after adjustment through a genetic algorithm and the charge and discharge constraint algorithm to find an optimal solution; according to the method, it is ensured that particles participating in fitness calculation and iteration each time meet constraint conditions, and the time for finding an optimal solution is shortened. Meanwhile, the algorithm is adjusted by setting the constraint condition meeting the optimization target, and approaching to the optimal solution is accelerated.

The present invention belongs to the technical field related to structural topology optimization design, and discloses an isogeometric topology optimization design method and application considering stress constraints, including the following steps: (1) An improved stress vanishing constraint function is used to replace the stress constraint condition in the isogeometric topology optimization design model of the structure to be optimized, so as to obtain an improved topology optimization design model; (2) The augmented Lagrangian function method is used to construct an approximate model of the improved topology optimization design model, and the sensitivity information of the approximate model is deduced and calculated; (3) Based on the sensitivity information, the unconstrained MMA algorithm is used to realize the iterative update of the design variables of the approximate model. After every N l times of design variable updates, the Lagrange multiplier is updated once based on the KKT condition until the optimization design result of the approximate model converges, and then the current approximate model is used for the isogeometric topology optimization design of the structure. The present invention realizes the high-efficiency and high-precision control of the stress states of each local region of the structure.

The invention relates to the field of equipment operation state monitoring, in particular to a medical equipment operation state monitoring and adjusting system based on big data, which comprises a multi-source acquisition module, a federal preprocessing module, an equipment multi-feature extraction module, an equipment multi-feature analysis module, an optimal adjustment strategy module and a database, and is characterized in that the multi-source acquisition module acquires operation parameters and generates an adversarial sample; the federal preprocessing module is embedded into a fairness constraint algorithm to output an initial balanced data set; the equipment multi-feature extraction module extracts frequency domain and time-space features and integrates the frequency domain and time-space features into a time feature sequence; the analysis module starts a random greedy algorithm according to a dynamic decision, and inputs a processed feature subset into a bidirectional LSTM model to output an equipment state; the optimal adjustment strategy module is combined with Bayesian network probabilistic reasoning and rule engine deterministic rules to generate an optimal adjustment strategy, the robustness and fairness of data are improved, accurate judgment of the equipment state and generation of the optimal strategy are achieved, and safe operation of medical equipment is guaranteed.

The application discloses a semi-supervised multi-view clustering method based on double-constraint concept decomposition, relates to the technical field of machine learning, combines point constraints with pair constraints, constructs a double-constraint algorithm based on prior label information of a complementary supervision mechanism, realizes maximum utilization and progressive propagation of limited supervision information, breaks through the limitation of data non-negativity of a traditional method by relying on a concept decomposition framework, processes complex data distribution by combining a kernel method, obtains a consensus matrix of multi-view data fusion as low-dimensional representation of a view, adopts a k-means algorithm for clustering, and outputs a final class division result. The application realizes effective processing of complex multi-view data, finally comprehensively improves the performance of a multi-view clustering task, and breaks through the bottleneck of the prior art in clustering precision, data adaptability and supervision information utilization efficiency.

The invention discloses a general-inductance integrated waveform power distribution method for the Internet of Vehicles, and belongs to the technical field of Internet of Vehicles and general-inductance integration. Comprising the steps of constructing a communication and sensing integrated system architecture based on the Internet of Vehicles, designing a power distribution scheme based on a communication data rate index and a sensing Cramer-Rao bound, designing a constraint algorithm of multi-objective optimization, solving a communication and sensing performance boundary to obtain an optimal distribution strategy and the like. Aiming at the advantage that a delay-Doppler domain modulation technology has high robustness in a high-speed vehicle network scene, the invention provides an Internet of Vehicles communication and inductance integrated waveform power distribution method under limited power budget, and meanwhile, the communication downlink sum rate and speed sensing Cramer-Rao bound performance is improved; and an efficient and reliable sensing means is provided for intelligent traffic cooperative control.

The invention belongs to the technical field of shoreline identification, and provides a shoreline identification and optimization method based on a point cloud neural network in an intelligent ship. Point cloud data are collected through the laser radar and are not affected by illumination conditions. The problem that a visual algorithm is limited by weather and illumination is solved, and the method adapts to dynamic environments such as foggy days and tidal changes. The point cloud density is reduced through voxel grid filtering, the data processing amount is reduced, the Edge-Aware Point Net network is improved, characterization learning of dense features does not need to be carried out, the lightweight design of the network structure is achieved, the end-to-end processing delay time is shortened, the ship berthing real-time requirement is met, and the real-time bottleneck is overcome. Differentiation features of a natural shoreline and an artificial shoreline are adaptively processed through a double-path U-net full-convolution integral segment network, and the problem that a traditional method is poor in adaptability to different types of shorelines is solved; the normal vector consistency constraint is introduced into the dynamic constraint K-means algorithm, and the fitting precision of clustering to the complex shoreline form is improved.

The invention discloses an anti-forgetting large language model continuous learning method and system, and the method comprises the steps: selecting samples which can represent the universal capability of a model or samples generated through an initial large language model, and forming a universal anti-forgetting sample set; inputting each sample in the universal anti-forgetting sample set into the initial large language model, and calculating a hidden state value of each sample; counting distribution of all hidden state value sets, and calculating related statistical values; calculating a positive activation threshold and a negative activation threshold for threshold edge loss calculation according to the related statistical value; and fusing the universal anti-forgetting sample set with an original continuous learning data set, and performing continuous learning of the initial large language model by adopting a hidden state threshold constraint algorithm. According to the large language model continuous pre-training and fine tuning method based on universal anti-forgetting sample hidden state threshold constraint optimization, downstream new tasks can be effectively learned in an anti-forgetting mode, and the universal capacity of the model is kept.

The present invention relates to a method for classifying and evaluating oocyte toxicity based on machine learning, comprising: obtaining an oocyte image; processing the oocyte image through a neural network segmentation model of a U-shaped structure to obtain key areas in the oocyte image; wherein the key areas include polar body areas, oocyte cytoplasm areas, oocyte outer ring areas and oocyte inner ring areas; calculating key features of oocytes in oocyte images based on key areas and mathematical constraint algorithms; wherein the key features are determined based on expert experience knowledge; extracting local features of oocytes in oocyte images based on a residual neural network model; mapping key features and local features to the same feature domain based on an adaptive attention network; and determining the category of oocytes in oocyte images based on features of the same feature domain. In the method, the entire process is automated and integrated, greatly improving analysis efficiency and accuracy.

The invention discloses a scientific investigation ship multi-support system integrated design method, and relates to the technical field of overall ships. The scientific investigation ship multi-support system integrated design method comprises two main parts, namely a spatial constraint algorithm suitable for scientific investigation ship support system integration and a hydraulic power source integrated design method. According to the scientific investigation ship multi-support system integrated design method, a multi-parameter spatial constraint algorithm is adopted in scientific investigation ship application for the first time, the efficiency and effectiveness of integrated design can be effectively improved, and it is ensured that multiple types of scientific investigation support systems are reasonably arranged and cooperatively run in a limited space; a hydraulic power source is integrally designed for a specific scientific investigation support system of a scientific research ship, pressure is flexibly supplied to subsystems according to working conditions, the guarantee requirements of various working conditions of a complex multi-support system can be effectively met, the system complexity is reduced, and the overall resource allocation of the scientific investigation ship is optimized.

The present invention relates to the field of detection robot technology, and in particular to a detection robot task allocation method based on consistency constraints, comprising: determining the tasks to be detected and the number of detection robots; inputting the tasks to be detected and the number of detection robots into a task allocation model, and obtaining the task allocation results of the detection robots, wherein the task allocation model is constructed based on a consistency constraint algorithm; driving the detection robot to perform detection work based on the task allocation results, and updating the tasks to be detected and the number of detection robots in real time during the detection process, continuously inputting them into the task allocation model, and updating the task allocation results until the detection is completed. The detection robot task allocation method based on consistency bundles of the present invention is more practical after allocating tasks, and the allocation of detection tasks can be dynamically adjusted during the detection process, greatly reducing resource waste.

The present invention discloses a method for generating a multi-objective non-inferior solution set for inter-basin water transfer based on a convex set feasible domain. The method comprises the following steps: collecting hydrological data from the study area, selecting a primary objective function from pre-constructed objective functions, constructing a plane set covering the entire spatial angle, and searching for boundary points of the target space corresponding to other objective functions; pre-processing the boundary points of the target space, and constructing a convex set boundary range of the target space using a spatial boundary convex hull algorithm; generating an ε-constrained threshold combination within the convex set boundary range of the target space; and for each ε-constrained threshold combination, converting the multi-objective optimization problem into multiple single-objective optimization sub-problems, solving each of these sub-problems using a sparse optimization interior point method, to generate a non-inferior solution set for the inter-basin water transfer scheme. The present invention effectively reduces the possibility of generating infeasible solutions and duplicate solutions in traditional ε-constrained algorithms, and improves the efficiency and quality of generating non-inferior solution sets.