42results about How to "Have diversity" patented technology

The invention relates to a color mixing 3D printer capable of controlling the material mixing proportion and a color mixing regulation and control method, and the color mixing 3D printer comprises two conveying parts which are arranged on a printing head and used for conveying particle consumables of different colors; the sleeve part is arranged on the printing head, the sleeve part is arranged on the periphery of the lower end of the conveying part, and the sleeve part is used for allowing consumables to pass through; the discharging spray head is communicated with the lower end of the sleeve part, and the discharging spray head is used for extruding consumables; the 3D printing device has the beneficial effects that a wire extrusion structure and a hot melting structure in an existing 3D printing technology are replaced by the conveying part, so that consumables cannot be interrupted in the extrusion and conveying process, the forming effect of a printed object is better, and the smooth degree of the printed object is guaranteed; and two regulation and control plates which are driven by the driving part to move are further arranged, and the angles of the two regulation and control plates can be changed in the sleeve part so as to divide the area of a passing area of the hot melting consumable in the sleeve part.

The present application discloses a detection device and a reagent card for detecting a coagulation project, the detection device includes: a reagent storage module, including a multi-person reagent storage module and a single-person reagent storage module, the multi-person reagent storage module is used to store reagent bottles; the single-person reagent storage module includes at least one reagent card storage channel to store reagent cards, and the reagent card includes at least one reagent storage position to store a single-person detection reagent; a distribution module, used to inject the collected sample into the detection cup, and inject the detection reagent in the reagent bottle or the single-person detection reagent in the reagent card into the detection cup; an incubation detection module, used to incubate the sample in the detection cup and detect the sample in the detection cup to obtain a coagulation detection result. Through the above method, the detection device can meet the detection of single-person samples and avoid the waste of large bottles of reagents.

The invention discloses equipment for producing various building materials, which comprises a supporting base, outer side plates are arranged at the two ends of one side of the top of the supporting base, first rails are arranged at the tops of the outer side plates, first sliding blocks matched with the first rails are arranged on the first rails, a first mold supporting plate is transversely erected at the tops of the first sliding blocks, and a second mold supporting plate is arranged at the top of the first mold supporting plate. A first lower die is arranged at the top of the first die supporting plate, main belt wheels are arranged on the two sides of the upper portion of the inner side of the outer side plate on one side, a first secondary belt wheel and a second secondary belt wheel are arranged above and below the side edge of each main belt wheel, and the main belt wheels are connected with the first secondary belt wheels and the second secondary belt wheels through belts. A conveying shaft is connected to the middle of the main belt wheel on one side, and the output shaft penetrates through the outer side plate and is connected with a first driver installed at one end of the outer portion of the outer side plate. The method has the beneficial effects that the produced building materials are high in universality, low in manufacturing cost, high in production efficiency, low in rejection rate and high in utilization rate of raw materials.

The utility model discloses a low power consumption type electric heating device for an air conditioner, which relates to the technical field of electric heating and comprises two fixing rods, and the sides, close to each other, of the two fixing rods are respectively and fixedly connected with a first supporting rod, a second supporting rod and a socket rod body. According to the utility model, the sectional type electric heating pipes are arranged, and each section of electric heating pipe can be independently installed and used, so that the corresponding number of electric heating pipes can be installed according to the outdoor temperature, and therefore, when the outdoor temperature is low, a large amount of heat can be generated indoors for heating in a short time; due to the fact that the types of the electric heating pipes are different and the lengths of the electric heating pipes are different, the distance between the electric heating pipes and the electric heating plug can be increased or decreased through the lifting assembly, the electric heating pipes with different lengths can be installed on the electric heating plug for use, and diversity is achieved.

The invention relates to the technical field of computers, in particular to a code review method and device, electronic equipment and a storage medium, which are used for efficiently screening out high-quality and high-diversity training data so as to improve the accuracy of code review. The method comprises the following steps: acquiring a training sample set and a test sample set, and dividing the training sample set into a plurality of training sample subsets; the test sample set comprises test samples reaching expected quality; after a to-be-trained code review model is preliminarily trained based on the training sample set, the test influence degree of each training sample subset on the code review model on the test sample set is determined; for at least one target training sample subset of which the test influence degree meets a preset influence condition in the plurality of training sample subsets, performing clustering according to a sample gradient to obtain a plurality of clusters; and sampling is carried out in each cluster, and the code review model after preliminary training is trained again according to the obtained sampling sample set.

The present application relates to the field of science and education display, and particularly relates to a display control method of a multi-axle three-dimensional dynamic pseudo-motion phenomenon display device, which comprises the following steps: S1, setting up a three-dimensional dynamic pseudo-motion phenomenon display device; S2, representing the pseudo-motion phenomenon effect of the display device by a control function G; S3, establishing a target function library, selecting the best function from the target function library, and obtaining the optimal solution of each coefficient in the control function G by using the least square method; S4, determining the rotation angular velocity range of the control function G; S5, combining the control function G to establish an input parameter database, constructing a machine learning control algorithm, and thus self-adjusting the associated parameters affecting the control function G. The present application can present the best display effect for different people in different environments.

The invention discloses a multi-cluster air computing method and system based on a deep expansion graph neural network, and belongs to the technical field of air computing in a wireless network, and the method comprises the steps: firstly, defining an aggregation center node, a wireless device node and an edge for connecting the two nodes in a heterogeneous graph neural network; secondly, a multi-cluster control calculation optimization problem needing to be solved is decomposed into a neural network solving module of an unfolded structure; secondly, designing a feature extraction and feature embedding method in a neural network solving module; secondly, designing a heterogeneous graph neural network encoder to aggregate the features, and mapping the aggregated features into variables needing to be optimized through a designed multi-layer perceptron decoder; and finally, designing a parameter sharing strategy of the whole network and a learning method of the neural network, and training the neural network to obtain a universal model. The scheme provided by the invention is superior to an existing optimization scheme in the aspect of calculation rate, the solving speed is greatly improved, and the method has very strong generalization ability.

This invention discloses a phosphonothiolated binaphthol derivative and its synthetic method. Using readily available binaphthioonium salts as starting materials, secondary phosphine oxides as coupling agents, and a base as a promoter, a series of phosphonothiolated binaphthol derivatives are generated. This invention features simple raw material preparation, very mild synthetic reaction conditions, convenient experimental operation, high reaction efficiency, and diverse functional groups.

This invention discloses a method for synthesizing phosphine oxide derivatives based on [2.2] p-cycloaranes. Using thionium salts and disubstituted phosphine oxides as starting materials, Ni(COD)₂ as a catalyst, CuBr as a co-catalyst, PPh₃ as a ligand, and K₃PO₄ as a base, the reaction yields phosphine oxide derivatives. This method is novel, uses readily available raw materials, is simple to operate, has simple and mild reaction conditions, and has broad substrate applicability.

The invention discloses an air deflector assembly and an air conditioner. The air deflector assembly comprises a rotatable air deflector. The first rotating shaft and the second rotating shaft are arranged at the two opposite ends of the air guide plate in the rotating direction correspondingly, and the first rotating shaft and the second rotating shaft are rotationally connected with the air guide plate; the first rotating shaft rotates with the joint of the second rotating shaft and the air guide plate as the rotating center and drives the air guide plate to rotate synchronously, the second rotating shaft rotates with the joint of the first rotating shaft and the air guide plate as the rotating center and drives the air guide plate to rotate synchronously, and the first rotating shaft and the second rotating shaft rotate intermittently. The air guide plate assembly intermittently rotates through the first rotating shaft and the second rotating shaft, the rotating centers of the first rotating shaft and the second rotating shaft are different, the rotating angle range of the air guide plate is enlarged, and therefore the swing angle of the air guide plate can be adjusted in a large range, air guide of the air guide plate at various angles is achieved, and then horizontal air outlet, vertical air outlet and angle air outlet are achieved; the air outlet direction of the air conditioner is diversified, and user requirements are met.

This invention relates to a MIL-100(Fe)-derived ErFeO3 / α-Fe2O3 gas-sensitive material, its preparation method, and its application, belonging to the field of gas sensor and environmental monitoring technology. A method for preparing a MIL-100(Fe)-derived ErFeO3 / α-Fe2O3 gas-sensitive material involves: slowly adding FeCl3 solution dropwise to H3BTC solution at room temperature; after uniform mixing, maintaining the mixture at 150℃~210℃ for 20 hours in a reaction vessel to obtain an orange-red substance; washing and drying to obtain MIL-100(Fe); dispersing the obtained MIL-100(Fe) in methanol; dissolving Er(NO3)3·6H2O in the above solution and reacting for a period of time; washing, centrifuging, and drying; transferring the obtained product to a muffle furnace and calcining under an oxygen atmosphere to obtain the ErFeO3 / α-Fe2O3 gas-sensitive material. The method for preparing porous metal oxide-based gas-sensitive materials using MOF materials as precursors and rare earth element modification proposed in this invention has advantages such as versatility, controllability, and recyclability compared to traditional methods for preparing metal oxide-based gas-sensitive materials.

This invention belongs to the field of public building floor plan layout technology, specifically involving a method, equipment, and medium for generating a public building master plan and floor plan layout. It obtains land boundary line and functional requirement data, acquires volume parameters for each functional area, constructs volume blocks, and arranges them within the boundary line according to constraints and preset arrangement rules, generating multiple initial master plan layout schemes. The schemes are then graded and screened to obtain candidate schemes. Based on the functional area type, a space division template is retrieved and adapted to the volume block base area to generate sub-functional space layouts and boundary contours, generating a planar wall layout. Road edges are generated based on the outer contour lines of the volume blocks, and the master plan design drawing is generated in conjunction with the wall layout. This invention constructs a design method from planning conditions to master plan design drawings, based on internal space generation, possessing scalability and reusability, improving the generation efficiency and rationality of design schemes.

The utility model relates to the technical field of water supply system detection equipment, in particular to a novel water supply exhaust pressure measuring device which comprises a pressure measuring meter, a quick-opening ball valve is installed at the lower end of a pressure buffer pipe, an L-shaped connecting pipe is installed at the lower end of the quick-opening ball valve, and the quick-opening ball valve is connected with a three-way connector through the L-shaped connecting pipe. A connecting pipe is installed at one end of the three-way connector, an automatic exhaust valve is installed at the upper end of the connecting pipe, a corrugated hose is installed at the other end of the three-way connector, and the end, away from the three-way connector, of the corrugated hose is connected with a multifunctional conversion connector. The multifunctional adapter is adopted, diversified use scenes and hierarchical functions of the device are enriched, the pressure measuring process is more flexible and convenient, the efficiency of on-site judgment and problem solving is greatly improved, meanwhile, the specification of the multifunctional adapter is diversified, the multifunctional adapter can be rapidly matched with water supply pipelines of different specifications, and the practicability is high. And the application range of the device is obviously expanded.

The invention provides a face frame set, a multi-head lamp and a lighting system. The face frame set comprises a first face frame and / or a second face frame. The first face frame is in a rectangular frame shape, a first clamping groove is formed in the first side of the first face frame, a first elastic fastener is arranged on the second side of the first face frame, the first side of the first face frame is opposite to the second side of the first face frame, and the first elastic fastener can be connected with the first clamping groove of the other first face frame in a matched mode. Each second face frame is in a shape similar to a rectangular frame, the first side of each second face frame is provided with a second clamping groove, the second side of each second face frame is provided with a second elastic fastener, the first side of each second face frame is adjacent to the second side of each second face frame, and each second elastic fastener can be connected with the second clamping groove of another second face frame in a matched mode; when the face frame set comprises the first face frame and the second face frame, the first elastic fastener can be connected with the second clamping groove in a matched mode, and the second elastic fastener can be connected with the first clamping groove in a matched mode. The multi-head lamp and the lighting system are both provided with the face frame set, and the face frame set can be freely spliced and matched and is high in splicing firmness.

The invention relates to the technical field of oil well fracturing prediction, and solves the technical problems in the prior art that distribution of important features in fracturing data cannot be fully reserved generally through simple data segmentation, and low result accuracy in fracturing state prediction is easily caused. In particular to a fracturing state prediction system and method applying a machine learning intelligent decision-making system, and the method comprises the following steps: S1, obtaining original data of a gulongshale oil fracturing detection system, preprocessing the original data, and obtaining a unified table through an SQLJOIN method; the statistical feature extraction algorithm and the hierarchical clustering algorithm are used to construct a more excellent algorithm model, the accuracy and robustness of feature expression are improved by realizing feature extraction and clustering, the feature redundancy is reduced by combining statistical features and hierarchical clustering, the interpretability and prediction precision of a subsequent model are improved, and the prediction efficiency is improved. And meanwhile, the nonlinear relationship in the gulonium shale oil data is optimized, so that the subsequent steps are more accurate and smoother.

The utility model discloses a multimedia player, including base and vertically setting in the loudspeaker box body of base top, and the top of loudspeaker box body is electrically connected with display screen, one side of the inside of loudspeaker box body is provided with projection mounting plate, and the forward side of projection mounting plate is provided with the projection lens that can carry out image projection, and projection lens penetrates loudspeaker box body, be provided with light machine booster plate between projection mounting plate and projection lens, and light machine booster plate is located one side of projection mounting plate, one side of light machine booster plate is electrically set up with light machine drive board, and one side of light machine drive board is electrically set up with light machine, and light machine and projection lens carry out electric connection and projection lens and display screen carry out electric connection between, and then further through projection lens, the film that display screen shows is projected.

This utility model discloses an electromechanical installation pipeline device, relating to the field of pipeline installation technology. It includes a base plate and a moving mechanism disposed on the top of the base plate for adjusting the fixed position of the pipeline left and right. The outer walls at both ends of the base plate are provided with lifting mechanisms for adjusting the vertical position, and the top of the base plate is provided with a fixing mechanism for fixing wires or pipes. The moving mechanism includes a sliding rod. A horizontal groove is formed on the top outer wall of the base plate, and the sliding rod is installed in the horizontal groove. Sliding blocks are slidably inserted into both ends of the outer circumference of the sliding rod. The bottom of the outer circumference of the lower arc-shaped plate is fixedly connected to the top outer wall of the sliding blocks. With this utility model, when the pipeline is fixed, due to human error, the pipeline may not be in a horizontal state. When it is necessary to adjust the pipeline position, the upper and lower arc-shaped plates can be moved left and right by moving the sliding blocks, thus adjusting and moving the pipeline position, facilitating use by operators.

The present application belongs to the field of computer vision and artificial intelligence, and particularly relates to a differentiable random image data augmentation method and system, aiming at solving the problem that the prior art does not consider the difference between the features of the data sets, so that the augmented data set is less helpful for the artificial intelligence method to more accurately achieve the target computer vision task. The present application comprises: defining an augmentation search space and operator differentiable parameters of the image data to be augmented; constructing a spatial search neural network model and training; performing relaxation sampling of the augmented operator obtained by training and normal distribution sampling of the transformation amplitude of the augmented operator, and normalizing the sampling results; and augmenting the image data to be augmented based on the final augmented operator and the transformation amplitude obtained by normalization. The present application learns the data augmentation strategy according to the characteristics of the data set itself, realizes fast and customized data augmentation, and thus helps the artificial intelligence method to more accurately achieve the target computer vision task.

The application discloses a multi-instrument replacement handle structure and an electric scooter. The application can quickly switch between an OLED instrument and a digital tube instrument and be connected with an integrated wire box through the setting of replaceable instruments. The instrument screen can be switched according to the needs of customers, and the application has diversity and is relatively convenient. The wire harness of the integrated wire box is located in the inside of the tap structure, and the appearance is relatively beautiful. Moreover, the maintenance and replacement of the external instrument are relatively more convenient in the later period. The integrated wire box is placed in the tap structure, the internal wiring of the external instrument is realized, and the replaceable instrument can be replaced only by dismounting the instrument panel, so that the cost price and accessories do not need to be increased, and the accessories do not need to be re-opened, thereby reducing the cost.

The application discloses a steel surface defect image generation algorithm based on coordinate attention and multi-scale texture edge guidance, which comprises the following steps: S1, obtaining a steel surface defect image dataset, and performing a pretreatment operation on image data in the dataset; S2, based on the steel surface defect image dataset pretreated in step S1, preparing a diffusion model training dataset; S3, constructing a diffusion image enhancement model based on multi-scale texture-edge guidance, wherein the diffusion image enhancement model comprises a forward diffusion process, a noise prediction network and a reverse denoising generation process, and a multi-scale texture-edge structure enhancement module and a coordinate attention module are introduced into the noise prediction network; S4, based on the training dataset obtained in step S2, training the diffusion image enhancement model to obtain a noise prediction network for reverse denoising; S5, based on the trained noise prediction network, obtaining an enhanced steel surface defect image by using a structure-guided img2img defect generation mode; and S6, merging the enhanced steel surface defect image with the original steel surface defect image, constructing an expanded steel surface defect detection dataset, and performing experiments and evaluation on the enhancement effect. The application can enhance the defect edge structure expression while maintaining the continuity of the steel surface background texture, generate steel surface defect images with clear texture, reasonable structure and diversity, thereby effectively expanding the existing steel surface defect detection dataset and improving the training effect of the subsequent defect detection model.

The present application relates to a kind of human motion sequence synthesis method and system based on motion graph.The present application first extracts key frame from human motion sequence to construct motion graph, transition edge is added in motion graph, and the transition sub-sequence corresponding to transition edge is generated;Based on dynamic graph, human motion sequence completion task is designed, and a kind of transfer number related graph path search algorithm is used to find multiple feasible paths between start node and end node, so as to generate a set of diversified motion sequence completion results;For human motion sequence prediction task, the method of forward walking on motion graph starting from start node is used, and the method can return multiple possible end nodes, so that human motion prediction task is attributed to human motion completion task with some different end nodes.The present application can generate real and diverse human motion sequence, and has excellent generalization ability between different data.

The utility model discloses a multi-channel myoelectricity pressure conduction rectum electrode which comprises a rectum wall air bag assembly used for conducting myoelectricity pressure signal detection and transmission on a rectum wall and an anal canal air bag assembly used for conducting myoelectricity pressure signal detection and transmission on an anal canal area. The rectum wall air bag assembly and the anal tube air bag assembly are coaxially assembled according to the relative positions during detection, the anal tube air bag assembly is provided with a conductive part corresponding to the position of an anal tube part, and the rectum wall air bag assembly and the anal tube air bag assembly are each independently provided with an air guide pipe set used for inflating corresponding air bags to conduct pressure detection; the rectum electrode is provided with the rectum wall air bag assembly and the anal tube air bag assembly, can be used for pressure signal transmission between a rectum wall and diagnosis and treatment equipment and pressure signal transmission between an anal tube part and the diagnosis and treatment equipment at the same time, can support multi-dimensional evaluation and treatment of the rectum electrode, and enables diagnosis and treatment to have diversity and accuracy.

The invention relates to a remote sensing image building extraction method based on multi-scale interaction and cross decoding, and belongs to the technical field of remote sensing image information extraction. The method comprises the following steps: S1, acquiring a remote sensing image, labeling the remote sensing image, dividing the remote sensing image into a training set, a verification set and a test set, and performing image data enhancement on the image when the image is input; s2, constructing a building extraction network model MGCC based on multi-scale interaction and cross decoding, wherein the network mainly comprises a multi-scale interaction guide module MSAG and a cross-layer cross decoding module CLCC; s3, training an MGCC model: inputting the training set into the MGCC model, calculating a corresponding loss function, and then adopting an AdamW optimizer and a cosine restart strategy to iteratively train the MGCC model until the model converges; and S4, inputting the test set into the trained MGCC model to obtain an image only having building features. The accuracy of remote sensing image building extraction is improved.

A method, system, equipment, and readable storage medium for load allocation across multiple units in a hydropower station are disclosed. This invention relates to the field of hydropower generation. Under the constraint of the total active power output of the entire station, an initial population is constructed within a safe range satisfying a preset unit vibration zone. Each individual in the initial population includes a set of load allocation schemes. For each individual, under preset rigidity and objective constraints, Pareto dominance iteration is performed to obtain a Pareto optimal solution set. The objective constraints include minimizing the total station's water consumption for power generation, the lifespan loss of unit equipment, the unit vibration zone crossing time, and the grid peak-shaving command response speed. For each candidate solution in the Pareto optimal solution set, a comprehensive score is determined based on the objective parameter value and preset weight coefficients. Based on the comprehensive score, a target load allocation scheme for the multiple units of the hydropower station is determined from all candidate solutions. This application ensures the safe operation of unit equipment while responding in real-time to changes in renewable energy output and meeting grid peak-shaving requirements.

The outdoor heat exchanger comprises a heat exchanger body, the heat exchanger body is provided with a heat exchange flow path, the heat exchange flow path is provided with a first main inlet and outlet and a second main inlet and outlet and comprises a first flow path, a second flow path and a third flow path, and the first end communicates with the first main inlet and outlet; the fifth end is communicated with the second main inlet / outlet; the gas-liquid separator is provided with a gas outlet, a first inlet and outlet and a second inlet and outlet, the second end and the third end are communicated with the first inlet and outlet, the second inlet and outlet is communicated with the sixth end, and the gas outlet is communicated with the fourth end and the first end; the first control valve is connected in series among the second end, the third end and the first inlet and outlet, and the second control valve is connected in series between the first end and the fourth end. According to the outdoor heat exchanger, the switching effect of different numbers of flow paths when the outdoor heat exchanger is in the refrigerating mode and the heating mode is achieved, so that the performance of the outdoor heat exchanger under different working conditions is improved.

The application discloses a traditional Chinese medicine hot air pressure therapeutic instrument, relates to the technical field of medical equipment, and solves the problems that the air pressure therapeutic instrument in the prior art cannot be inflated in sequence and is single in function. The device comprises a binding belt, an air pressure assembly installed on the binding belt, a medicine storage bag and a heater, a gas conveying pipe connected to the air pressure assembly, and the heater is located behind the medicine storage bag. The air pressure assembly comprises a plurality of air bags installed at equal intervals on the binding belt, a gas conveying channel communicated with the bottoms of the air bags, an exhaust channel arranged at the right side of the air bag, a piston arranged between the air bag and the gas conveying channel, and a spring sleeved on the piston, and the lowermost gas conveying channel is communicated with the gas conveying pipe. The device has diversity, the heat generated by the heater can assist the Chinese herbal medicine components in the medicine storage bag to better penetrate into the deep layer of the skin of a patient, the air pressure assembly, the heater and the medicine storage bag are used in cooperation, blood circulation can be promoted through mechanical pressure, the drug penetration and the heat therapy can be used in cooperation, and comprehensive treatment is achieved.

The utility model discloses a novel liquefied petroleum gas recovery skid-mounted device which comprises a mixed hydrocarbon buffer tank, a methane and ethane removal tower, a liquefied petroleum gas tower, a liquefied petroleum gas tower top cooler, a liquefied petroleum gas tower top separator, a liquefied petroleum gas tower reflux pump, a stable light hydrocarbon cooler, a stable light hydrocarbon buffer tank and a stable light hydrocarbon loading pump, the device is integrated into one skid-mounted device, the occupied area of the skid-mounted device is obviously reduced, equipment transportation is facilitated, and meanwhile the field installation work amount is small; the device is simple in structure and convenient to construct and install on site, realizes recovery treatment of mixed hydrocarbon in a liquefied natural gas factory, produces fuel gas, liquefied petroleum gas and stable light hydrocarbon, ensures normal production of the liquefied natural gas factory, reduces potential safety hazards of the liquefied natural gas factory, and improves economic benefits of the liquefied natural gas factory.

An air inflation table for positioning and locking a wafer box comprises an air inflation table seat for bearing equipment, an air inflation table assembly arranged on the air inflation table seat and used for fixing the wafer box, and a plurality of air measurement control devices located in the air inflation table seat and used for detecting air data. The inflation pedestal is externally connected with a gas conveying device used for conveying gas to the inflation pedestal, and the gas measurement control device is externally connected with control equipment used for controlling the equipment to move. The air inflation table assembly comprises an air inflation table base plate used for bearing a wafer box, a gripper arranged in the air inflation table assembly, a set of fixing assemblies arranged on the air inflation table base plate and a set of positioning assemblies arranged on the air inflation table base plate. The wafer box is fixed and locked through the gripper arranged on the inflation table substrate, so that the wafer box is prevented from moving up and down along with the impact of gas during inflation.

This invention relates to a protein sequence design method, apparatus, and storage medium based on the Graphormer algorithm. The method includes: representing the protein structure as a graph, using individual amino acids as nodes and the connections between amino acids as edges; extracting initial edge features and initial node features of the protein; concatenating the initial node features of the protein with a random matrix, adjusting the dimensions through a linear layer, and adding positional encoding information to obtain node features, which are used as input to the GPD model; concatenating the initial edge features of the protein with a matrix and then passing them through two linear layers to obtain edge features, which are embedded into the attention matrix of the GPD model; constructing and training a GPD model for designing protein sequences with a fixed backbone; and designing protein sequences based on the GPD model, node features, and edge features. Compared with existing technologies, this invention has a higher sequence recovery rate and higher design sequence diversity.

The Pareto solution search device is configured to be provided with: an evaluation value acquisition unit (1) that acquires, from a simulator, an evaluation value for each Pareto solution candidate in an envelope analysis method, the simulator calculates evaluation values in an envelope analysis method of M (M is an integer of 2 or more) Pareto solution candidates in a plurality of objective functions corresponding to a multi-objective problem solved by a genetic algorithm using the envelope analysis method; and a distance calculation unit (2) that calculates the relative distance between the M Pareto solution candidates. Furthermore, the Pareto solution search device is provided with: a Pareto solution candidate selection unit (3) that selects G (G is an integer greater than or equal to 1 and less than M) Pareto solution candidates from the M Pareto solution candidates on the basis of the relative distance calculated by the distance calculation unit (2) and the evaluation value acquired by the evaluation value acquisition unit (1); and a Pareto solution candidate generation unit (4) that regenerates (M-G) Pareto solution candidates different from the G Pareto solution candidates, using any one or more Pareto solution candidates among the G Pareto solution candidates selected by the Pareto solution candidate selection unit (3).