210results about How to "Convenient research" patented technology

The invention discloses an active learning-based electric power knowledge graph construction method and system, and relates to the technical field of electric power market transaction data analysis, and the method comprises the steps: collecting data, cleaning the collected data, generating word prediction task data of an electric power transaction scene, and training an electric power field word embedding model; generating an entity relationship training sample, inputting a word embedding model, performing entity relationship label prediction through an entity relationship joint extraction model, extracting a high-quality sample based on active learning to perform part-of-speech tagging, and training the model based on a tagged sample; and applying the trained entity relationship joint extraction model to power transaction data to construct a knowledge graph, performing manual correction and entity unification, and storing the knowledge graph into a Neo4j database. According to the method, active learning and deep learning are combined, so that the sample labeling cost is effectively reduced, and meanwhile, the model performance is ensured.

The invention provides a vascularized organ-like chip based on seal microstructure patterning and a preparation method thereof, the chip comprises a micro-channel layer, the upper surface of the micro-channel layer is provided with a gel perfusion channel, a central tissue chamber and culture medium flow channels at two sides which are connected, and tumor spheres or colon cells are inoculated in the central tissue chamber; the liquid storage pool layer is arranged on the upper surface of the micro-channel layer and corresponds to the inlet and outlet of the gel filling channel, the inlet and outlet of the central tissue chamber and the inlet and outlet of the culture medium runner, the liquid storage pool layer is provided with a plurality of liquid storage pools and a plurality of through holes, and the liquid storage pools contain culture solutions to provide nutrients for cultured tissues; and the seal module is arranged above the central tissue chamber, the seal module is provided with a tumor seal and a colon seal, and the tumor seal and the colon seal are respectively provided with a microstructure three-dimensional module and are respectively used for simulating the culture environment of tumor spheres or colon cells. According to the invention, in-vitro construction of the vascularized tumor model and the colon model is realized, and a guide is provided for construction of the vascularized organ-like model with specific organ three-dimensional microstructure characteristics.

The invention provides an experiment control device for researching an ejection system. The experiment control device comprises a high-pressure power source pipeline, a multi-stage ejector pipeline and a control and data acquisition system, the high-pressure power source pipeline is used for providing high-pressure gas for the multi-stage ejector pipeline, so that each stage of ejector in the multi-stage ejector pipeline can work normally; the multi-stage ejector pipeline comprises a first-stage ejector pipeline to an nth-stage ejector pipeline, and all the stages of ejector pipelines are sequentially connected in a cascade form; and the control and data acquisition system is used for controlling the multi-stage ejector pipeline and acquiring data. According to the experimental control device for studying the injection system, the performance of various ejector systems can be studied, so that the ejector systems meet the design target and adapt to various variable working conditions, and the injection performance is improved.

The invention relates to the technical field of capacitor testing, and provides an aluminum electrolytic capacitor testing and sorting machine which comprises a sorting rotary table, a rack and a plurality of U-shaped grooves, one U-shaped groove is used for feeding, the other U-shaped grooves are used for testing, a through groove is formed in the inner side face of each U-shaped groove, a placing plate penetrates through the through grooves, and the placing plates are arranged in the through grooves. A camera is arranged above the U-shaped groove used for feeding, mounting plates are arranged above the U-shaped groove used for testing, the mounting plates are used for mounting capacitor testing devices, each mounting plate is connected with a pair of electrode plates, the surface of the placing plate is rotatably connected with a movable circular plate, a second motor is arranged on the rack, and the movable circular plate is rotatably connected with the movable circular plate. An output shaft of the second motor is fixedly connected with an electromagnet, and a material collecting box is arranged below the U-shaped groove for testing; and the camera is used for collecting an image of the capacitor and identifying the positive electrode and the negative electrode of the capacitor, so that the positive electrode and the negative electrode on the capacitor correspond to the positions of the electrode plates, accurate alignment of the pins and the electrode plates is realized, and high efficiency and convenience are realized.

The invention discloses a satellite ground staring attitude control multi-software joint simulation system and a construction method thereof, and belongs to the technical field of computer simulation, and the construction method comprises the steps: 1, constructing an ADAMS-based microsatellite ontology model; 2, constructing a satellite orbit model and an attitude sensor model based on STK; step 3, constructing a torque execution mechanism model based on MATLAB (matrix laboratory) and Simulink; 4, designing an attitude determination algorithm and a control algorithm of a ground staring task simulation model capable of combining a microsatellite body model, a satellite orbit model, an attitude sensor model and a torque execution mechanism model; according to the method, ADAMS, STK and MATLAB are combined to simulate a satellite body, an attitude sensor, a torque execution mechanism and a running orbit, the attitude control task of the satellite is simulated, and the method has the advantages of being good in simulation effect and visual and intuitive in simulation process.

The invention discloses application of a mineralized carbon sequestration strain, the mineralized carbon sequestration strain is classified and named as Halomonas sulfidaeresis. ZN2 #, the mineralized carbon sequestration strain is preserved in the China General Microbiological Culture Collection Center (CGMCC), and the preservation number is CGMCC No.32033. The invention further discloses a preparation method of the mineralized carbon sequestration strain. The mineralization carbon sequestration strain can metabolize to generate urease and carbonic anhydrase at the same time, the two enzymes are subjected to a dual mineralization reaction in combination with a microorganism-induced calcium carbonate precipitation technology, the crystal form of a mineralization product is improved, the mineralization strength is improved, meanwhile, carbonic anhydrase can absorb carbon dioxide in the environment, high-strength soil reinforcement is achieved, meanwhile, carbon dioxide emission is reduced, and the soil quality is improved. Therefore, the dual effects of biomineralization remediation and improvement and carbon sequestration and emission reduction of the soil are achieved, and sustainable development of a soil bioremediation and improvement technology is achieved.

The invention relates to a single cell data integration method based on self-supervised learning, and the method comprises the steps: integrating single cell data from different researches through a single cell data integration algorithm scCorrector based on a research specific normalization layer, and mapping the single cell data to a public space; aligning a plurality of types of single cell data sets by adopting a single cell data integration algorithm scInterpreter which is based on knowledge regularization and is interpretable; integrating and annotating multi-species single cell data through a multi-species single cell data integration and annotation algorithm CAMEX based on a heterogeneous graph neural network; spatial transcriptional expressions are inferred from histological images using an algorithm stInfer based on a weighted K-nearest neighbor model. Compared with the prior art, the method has the advantage that adverse effects on downstream tasks caused by sequencing technology differences are relieved. In addition, gene expression is inferred through a dyed image, so that the research cost of a space transcriptome is greatly reduced, and convenience is provided for exploring a tissue microenvironment.

The invention discloses a mass spectrum-based maternal metabolite structure labeling method, which comprises the following steps of: inputting a structure SMILES character string of a maternal metabolite, and outputting a metabolite structure and a reaction type of the metabolite structure; calculating the mass-to-charge ratios of all output metabolites in different adduct ion forms, and performing correlation matching and annotation on the mass-to-charge ratios and a pre-constructed metabolite peak table to obtain candidate metabolites and candidate metabolite adduct ion forms; performing multi-dimensional scoring on the obtained candidate metabolites; screening the metabolite list according to a set filtering rule in combination with scores of multiple dimensions; and outputting annotation information and structure information of the metabolites in the peak table to a file, wherein the annotation information comprises a reaction type, an adduct ion form, a multi-dimensional score and a structure. The invention further discloses a labeling system and application for implementing the labeling method.

The invention discloses a test platform and a test method for simulating lunar soil penetration in a low-gravity environment based on bottom friction, and belongs to the technical field of low-gravity simulation devices. Simulated lunar soil is placed in the simulation box; the simulation box is arranged on the support adjusting unit; wherein the inclination angle of the simulation box is adjusted by adjusting the length of the support adjusting unit, so that the gravity of the moon is simulated; the monitoring unit penetrates through the simulation box and then penetrates into the simulated lunar soil, and the penetration resistance in the simulated lunar soil is monitored. The low-gravity environment of the moon can be simulated, and the low-gravity environments of other planets can also be simulated by applying the method principle provided by the invention; meanwhile, the method provided by the invention can be used for calculating the inclination angle alpha of the corresponding test bed according to different lunar soil or star soil materials (the friction coefficients mu of different materials are different).

The invention provides an application of atractylis lactone as a beta-glucuronidase inhibitor. Experiments find that atractylenolide can act on active sites of beta-glucuronidase to generate various bonding effects and play an inhibition role, and atractylenolide shows a dose-dependent relationship on activity inhibition of beta-glucuronidase. Furthermore, the atractylenolide can be used as a beta-glucuronidase inhibitor for preventing and treating related diseases (such as drug-induced diarrhea, diabetes mellitus, calculus and the like), and has very good research and application values.

The invention belongs to the technical field of marine surveying and mapping, and discloses an airborne LiDAR sounding wave correction method and system. The method comprises the following steps: calculating wave parameters according to an Elfouhail spectrum model; using a KD-Tree accelerated K-nearest neighbor search method to determine a search neighborhood according to wave parameters so as to realize search of sea surface points in the neighborhood; and according to the efficient search result of the sea surface points in the neighborhood, constructing a multi-scale irregular TIN array sea surface model, calculating the normal vector of each beam of sea surface laser points with seabed reflection echoes, and performing ray tracing to correct wave errors to obtain accurate seabed point coordinates. According to the invention, seabed sounding errors caused by path deflection when laser enters water from air due to sea surface waves can be reduced, the sounding precision of the ALB system is improved, and powerful support is provided for ocean resource development, ocean environment protection and the like.

The invention belongs to the technical field of soil sampling, and particularly relates to a grain planting soil detecting and sampling device which comprises a base, a supporting column is fixedly arranged on one side of the base, a top plate is fixedly arranged on the supporting column, a driving mechanism and a stratified sampling mechanism are arranged on the top plate, and a soil sample storage mechanism is arranged on the base. The stratified sampling mechanism comprises a sampling drill rod assembly and a pressure difference sampling assembly, the pressure difference sampling assembly is installed on the top plate, and the sampling drill rod assembly is located below the top plate and connected with the pressure difference sampling assembly; by combining the pressure difference principle with the longitudinally-arranged temporary storage cavities, soil of different depths can be collected in a layered mode in one-time sampling operation, after the soil is collected, the soil of different depths is sprayed into the corresponding sample receiving plates to be stored according to the pressure difference principle again, sample mixing is effectively avoided, accurate layered sampling is achieved, and the sampling efficiency is improved. And the collected soil sample is kept in an original state.

The invention discloses an electromagnetic lock running-in and friction torque automatic test system and method, and relates to the field of electromagnetic lock testing, and the system comprises a test bench which comprises a fixed base, an electromagnetic lock, a first coupling, a torque and rotating speed sensor, a second coupling and a servo motor. An electromagnetic lock fixing assembly and a servo motor fixing assembly are fixed on the fixed base, a stator part of the servo motor is fixedly connected to the servo motor fixing assembly, and a stator part of the electromagnetic lock is fixedly connected to the electromagnetic lock fixing assembly; a rotor part of the electromagnetic lock is fixedly connected with the torque and rotating speed sensor through a first coupler, and the torque and rotating speed sensor is fixedly connected with an output shaft of the servo motor through a second coupler. The problem that an existing electromagnetic lock friction torque testing system cannot meet the requirements for accuracy and efficiency of electromagnetic lock friction torque measurement is solved.

The invention belongs to the technical field of proton exchange membrane preparation methods, and relates to a method for preparing a proton exchange membrane through response surface optimization. According to the method, a single-factor extraction test is carried out by taking the mass fraction of a membrane casting solution containing liquid crystal performance, the pulling speed and the drying temperature as influence factors, the influence of the mass fraction of the membrane casting solution containing liquid crystal performance, the pulling speed and the drying temperature on the membrane thickness of the proton exchange membrane is determined, and a response surface method test is designed; the optimal process conditions for preparing the proton exchange membrane are determined as follows: the mass fraction of the membrane casting solution containing the liquid crystal performance is 2.08%, the pulling speed is 82.23 mm / min, the drying temperature is 60.30 DEG C, and the actual measured average value of the thickness of the proton exchange membrane obtained under the conditions for three times is 33 + / -1 [mu] m. The prepared proton exchange membrane has high proton conductivity, and the method is provided for improving the proton conductivity of the proton exchange membrane.

The invention discloses a large-scale new energy rapid power control transient calculation method and system, and relates to the technical field of large power grid transient calculation, and the method comprises the steps: modifying the power control instruction values of new energy power generation equipment in batches, and carrying out the pre-acceleration of a power outer ring, so as to achieve the purpose of actually changing the output value of the new energy power generation equipment. According to the specific implementation method, a new energy control plant-station set is configured, and power instructions of new energy power generation equipment in the plant-station set are uniformly increased or decreased according to a certain proportion, so that power instruction modification of a large number of equipment is realized; after the power instruction is modified, the difference value between the power instruction and the current actual power output value is calculated, integral operation is carried out on the difference value, an integral result serves as input of an original power outer ring, and the purpose of quickening response to the power instruction is achieved.

The invention provides a traditional Chinese medicine text key information intelligent capturing method and system. The method comprises the following steps: receiving a traditional Chinese medicine text data set, and preprocessing each traditional Chinese medicine text data in the traditional Chinese medicine text data set; performing word feature extraction on each traditional Chinese medicine text data by using a word vector model to obtain a first word data set, and performing semantic analysis on each word data in the first word data set to obtain a simplified second word data set; inputting the second word data set and the traditional Chinese medicine knowledge graph into a pre-constructed deep learning model, wherein the deep learning model outputs a plurality of pieces of prescription information; and repeating the above process, completing the processing of all traditional Chinese medicine text data, and carrying out integration processing on the obtained prescription information. According to the scheme, the key information can be efficiently and accurately extracted from the traditional Chinese medicine text.

The invention discloses a preparation method of a hole transport layer, a solar cell and an electric device thereof. The preparation method of the hole transport layer of the solar cell comprises the following steps: 1, sputtering a layer of NiOx film on the surface of conductive glass by adopting a magnetron sputtering method; 2, putting the NiOx thin film substrate into a mixed solution of water and hydrogen peroxide for ultrasonic cleaning, and then cleaning with ethanol to obtain a modified NiOx thin film; 3, the modified NiOx film substrate is put into a mixed solution of SAMs and ethyl alcohol to be subjected to ultrasonic cleaning, then ethyl alcohol is used for cleaning, and a modified NiOx film is obtained; and 4, drying the modified NiOx thin film substrate, eliminating the residual solvent, heating and annealing to obtain the hole transport layer. The hole transport layer prepared by the invention is more matched with the energy level of the perovskite layer, and hole transport is facilitated; and non-radiative recombination at the interface can be inhibited, and the extraction of current carriers is facilitated. The performance of the perovskite solar cell is effectively improved, and processing of large-area devices is facilitated.

The invention provides a method for realizing automatic extraction of compound information in magnetic resonance mass spectrometry, which belongs to the field of quality control of natural product extracts and comprises the steps of original data information conversion, ion information self-adaptive integration, compound chromatogram construction and compound chromatogram outflow characteristic information extraction. Firstly, originally collected signals are converted into data in a NetCDF format, information of all components in each mass spectrum is reserved, and then high-precision mass spectra, retention time and ion abundance information of all ions in a sample are obtained; then, according to the information of the ion fengdu and the high-precision mass spectrum, through a DBSCAN clustering algorithm, obtaining potential valuable characteristic ion center points, and clustering all ions; constructing a chromatogram by the ions in each class according to a retention time sequence; a dynamic window is combined with a continuous smoothing strategy to extract a compound chromatographic peak in each chromatogram, information such as compound retention time, high-precision m / z and peak area is obtained, and finally extraction of compound information in a sample is achieved.

The invention discloses a method for evaluating combined toxicity of nano plastic and tamoxifen based on caenorhabditis elegans and a method for resisting drug toxicity caused by the nano plastic, a sample to be detected is an NGM solid culture medium of a mixture of the nano plastic and tamoxifen, and the caenorhabditis elegans is inoculated into the sample to be detected. The toxic effect of the joint exposure of the nano plastic and the tamoxifen on nematode movement, nerve, development and reproduction under the actual environment condition can be quickly and accurately evaluated, and a basis is provided for evaluation of human health risks. The invention provides a combined toxicity detection method which is suitable for laboratories, simple and convenient to operate and high in applicability by fully considering the problems that the toxicity is changed due to the chemical effect generated by nano-plastics and medicines and the in-vivo conditions are complicated and are difficult to observe, and provides a combined toxicity detection method which is suitable for laboratories and can be used for detecting the toxicity by supplementing N-acetylcysteine through diet. The effect of N-acetylcysteine in relieving tamoxifen toxicity caused by nano-plastics is further evaluated.

The invention discloses a preparation method and application of a diatomic catalyst based on a covalent organic framework, the prepared diatomic catalyst based on the covalent organic framework has a clear structure, the bimetal active center distance is clear and adjustable, and research on the synergistic effect of the diatomic catalyst is facilitated. The invention also provides an excellent catalyst for electrochemical preparation of hydrogen peroxide, and the catalyst has the advantages of stable structure, high specific surface area, high repeatability, simple preparation, and hydrogen peroxide selectivity of more than 90%.

The invention discloses a method for directly inducing generation of organoids from IPS and application, and belongs to the field of cell engineering. The invention provides a method for directly and rapidly generating a target organ based on human IPS, and the lung organ, the placenta organ and the blood vessel organ are obtained. According to the method, the organoid represented by three germ layers is induced, so that the method is suitable for constructing various organoid. Compared with the prior art, the method has the advantages that step-by-step induction is not needed, one-step construction is successful, the operation is simple, the construction time is greatly shortened, the process is repeatable, long-term stable culture and passage are realized under in-vitro conditions, and the freezing survival rate is high; the method has the advantages that the method is simple in operation, the cultured organoid is good in homogeneity, RNA-seq results show that high expression of the blood vessel related genes appears in the organoid, and the maximum length of the organoid can reach 400 micrometers, so that the culture method can spontaneously induce expression of part of the blood vessel related genes to support growth of the organoid. The invention provides an important tool for researching pathogenesis of diseases and subsequent drug development.

The invention discloses a novel microroot canal root phenotype detection method based on water-soluble holing, which comprises a water-soluble holing mechanism, the water-soluble holing mechanism sprays water and holes in plant root soil, and the water-soluble holing mechanism comprises a water tank, a stacking groove plate, a communicating hole, a water inlet and outlet hole, a sealing cover, a water conveying pipe, a water pressure regulating valve, a water pump, a water spraying head, a connecting hole and a water spraying hole. The water conveying pipe is wound on the outer portion of the slurry suction pipe, the water spraying head is connected with the slurry suction short pipe, the water spraying head sprays water to punch holes in soil at the roots of plants, the slurry suction short pipe sucks away slurry, then the transparent root pipe is placed in the holes, and the transparent root pipe is placed in the holes. Shooting is carried out after the shooting position of the macro camera is adjusted, so that plant root system images can be conveniently collected, after collection is completed, the transparent root canal is taken out, the hole is filled with slurry in the slurry storage tank again, the plant root system is buried, and damage to the plant root system is reduced.

The invention discloses a solvent-removing lossless precise transfer method for a two-dimensional material. The method comprises the following steps: carrying out surface functionalization on a silicon wafer by using HMDS (hexamethyldisilazane); carrying out spin coating on the functionalized silicon wafer by adopting a PMMA anisole solution, and drying to form a PMMA supporting film; picking up the PMMA support film by using a heat release adhesive PDMS; pDMS / PMMA is transferred to the surface of a two-dimensional material / first substrate needing to be transferred, PDMS and PMMA are separated after heating, and PMMA covers the surface of the two-dimensional material; the PMMA / two-dimensional material is separated from the first substrate through etching of the etching solution; fishing up the PMMA / two-dimensional material separated from the first substrate by using PDMS; transferring the PMMA / two-dimensional material to a second substrate; and washing away PMMA with acetone, and blow-drying with nitrogen to obtain the product. The two-dimensional material can be cleanly, efficiently and accurately transferred, and the intolerance of the two-dimensional material to an organic solvent and the pollution of the organic solvent can be avoided.

The invention provides a current and voltage characteristic generation method and device of a transient arc, which can be used in the technical field of power system relay protection, and the method comprises the steps: constructing an arc model through employing an external electromotive force source, an internal resistor, and a plasma with a preset radius, length and transient temperature; generating an energy conservation model and a current model according to the arc model; converting the energy conservation model to obtain a direct current arc temperature model; and utilizing the direct current arc temperature model, the current model and a pre-generated conductivity regression model to generate current and voltage characteristics of the arc. According to the method and device for generating the current and voltage characteristics of the transient arc, the non-transient stable change process of the temperature and the voltage drop of the arc column is obtained, the change process of the arc column is described more accurately, and research on the grounding fault arc resistance of a power distribution network is facilitated.

The invention belongs to the technical field of graphene material preparation, and particularly relates to a universal covalent functionalized graphene electrolysis preparation device and a use method thereof. A reactant graphite paper electrode of the device is mounted in an electrode mounting hole in an electrolytic tank cover of a four-channel electrolytic reaction tank mechanism through a conductive clamp mechanism, the graphite paper conductive clamp mechanism is clamped at the upper end of the graphite paper electrode, and the graphite paper electrode extends into an inner cavity of an electrolytic tank main body; four groups of liquid conveying systems of the working liquid conveying mechanism are communicated with four liquid injection and drainage ports of the four-channel electrolytic reaction tank mechanism through working liquid pipelines, and the four-channel electrolytic reaction tank mechanism, the graphite paper electrode conductive clamp mechanism, the working liquid conveying mechanism and the electric control mechanism are integrally mounted on the shell and the mounting integration mechanism. According to the invention, automatic control parts are effectively combined, so that the covalent functionalized graphene electrochemical preparation method is automatically realized through program control.

The invention discloses a device and method for measuring the photothermal effect of a two-dimensional material, and the device comprises an optical bread board, the optical bread board is provided with a light heating module, an electric heating module and a measurement module, the measurement module is provided with a to-be-measured material, and is used for supporting the to-be-measured material and obtaining the temperature of the to-be-measured material; the electric heating module is used for heating a to-be-tested material and obtaining the temperature in the heating process; the light heating module is used for illuminating a to-be-tested material and acquiring light power during illumination; the measurement principle advantage of the measurement device is remarkable, the Newton's cooling law is used as the core calculation principle, a physical model for calculating the photothermal conversion efficiency is simplified, the measurement device can calculate the photothermal effect of the two-dimensional material more visually and accurately, the measurement cost can be effectively reduced, and the measurement operation can be simplified.

The invention discloses a method for efficiently producing branched human milk oligosaccharide lacto-N-neohexaose (lacto-N-neohexaose, LNnH, Gal beta 1, 4GlcNacbeta 1, 3 (Gal beta 1, 4GlcNacbeta 1, 6) Gal beta 1 and 4Glc), a recombinant strain and application of the recombinant strain. The invention further discloses a method for efficiently producing the branched human milk oligosaccharide lacto-N-neohexaose, LNnH, Gal beta 1, 4GlcNacbeta 1, 3 (Gal beta 1, 4GlcNacbeta 1, 6) Gal beta 1 and 4Glc. The whole-cell synthesis method of the LNnH is low in cost, high in efficiency and beneficial to industrial production. Besides, as the core structure of the branched human milk oligosaccharide, the method provided by the invention is beneficial to making up the blank of the biosynthesis technology of the branched human milk oligosaccharide and promoting the research and development of the branched human milk oligosaccharide.

The present invention relates to the field of sampling technology, and discloses a ginger root sampling device for root disease research, comprising a base and a mounting platform arranged above the base, a limiting plate 1 is fixed on a side of the mounting platform away from the base, a limiting plate 2 sliding on the mounting platform is arranged on one side of the limiting plate 1, the limiting plate 1 is provided with an opening, a cutting blade is arranged on the side of the limiting plate 1 close to the limiting plate 2, a clamp for fixing the ginger root is arranged on the mounting platform, and a sampling assembly is arranged on the base. The present invention can quickly sample the ginger root, is easy to use, has high safety performance, can adaptively regulate the sampling device, can sample different positions of the ginger root, thereby facilitating the planting personnel to study the diseases of the ginger root, and has high working performance.

The utility model discloses a dynamic balancer for teaching display, which comprises a transparent container extending left and right, a dynamic ball is arranged in the transparent container, and the transparent container is arranged on a bottom plate through a left connecting assembly and a right connecting assembly which are arranged at an interval left and right. And the left connecting assembly and the right connecting assembly can realize up-down movement of the left end and the right end of the transparent container, so that the dynamic ball rolls left and right in the transparent container. Left-right movement of the dynamic small ball in the transparent container is achieved through the left connecting assembly and the right connecting assembly, the overall structure is simple, and establishment of a subsequent model and research and implementation of a control algorithm are facilitated. Due to the fact that the transparent container is supported by the left connecting assembly and the right connecting assembly, the transparent container has two supporting points, the dynamic small ball can be static when the transparent container does not work, and overall machining is convenient.

The invention discloses a regression prediction method and device for the biological activity of polypeptide, and the method comprises the steps: collecting the biological activity numerical information of polypeptide, and constructing a polypeptide biological activity data set; the method comprises the following steps: performing feature extraction on polypeptides in a polypeptide biological activity data set by using an Evolutionary Scale Modeling-2 protein language model, and dividing the polypeptides into a training set and a test set; constructing a long short-term memory network model, and optimizing hyper-parameters of the long short-term memory network model through a bionic meta-heuristic algorithm; training the hyper-parameter optimized long-short-term memory network model by adopting a training set to obtain a polypeptide biological activity regression prediction model; evaluating the biological activity regression prediction model by adopting the test set; and substituting to-be-detected polypeptide sequences into the biological activity regression prediction model in batches to obtain the biological activity value of the polypeptide. According to the polypeptide biological activity regression prediction method and device, the specific biological activity numerical value information of the polypeptide can be rapidly and accurately predicted, and meanwhile the high fitting degree and the low root-mean-square error are achieved.