1318results about "Chemical data mining" patented technology

This invention discloses a method for preparing a microcapsule-based photothermal self-healing anti-icing coating, belonging to the field of functional coating materials technology. The method includes: obtaining the core material and wall material composition parameters of the microcapsules, and collecting the polyacrylamide concentration and dopamine polymerization time; constructing a morphology-photothermal correlation model of the microcapsules to obtain the predicted photothermal conversion efficiency; performing multi-objective optimization on the core and wall material composition parameters with the target photothermal conversion efficiency as a constraint, and iteratively inverting to obtain the optimal microcapsule configuration; preparing primary microcapsules and composited with a dopamine photothermal modification layer on their surface to obtain a photothermal self-healing microcapsule additive; mixing the photothermal self-healing microcapsule additive with an epoxy resin-based coating at an adaptive mass ratio, coating and curing to obtain an anti-icing coating. This invention achieves accurate prediction of the photothermal conversion efficiency of microcapsules and intelligent optimization of composition parameters, obtaining an anti-icing coating with both high photothermal conversion efficiency and rapid self-healing capability.

This invention relates to the field of mechanical property monitoring technology for mine backfill bodies, specifically to a system and method for acquiring and analyzing tensile strength data of mine backfill bodies. The system comprises a data acquisition unit, a physical information coupling unit, a tensile strength inversion unit, and a spatiotemporal evolution mapping unit. This invention acquires the waveguide flight time and energy attenuation coefficient of the backfill body using an embedded guided wave sensor array. It uses the monotonically increasing law of hydration degree as a physical constraint to train the physical information coupling network, outputting a compliant hydration degree time series. Based on a hydration degree-tensile strength mapping relationship library, it inverts and fits to generate a strength growth curve, which is then loaded into a four-dimensional digital twin model to invert the internal tensile strength gradient field. It automatically identifies local abnormal regions where the growth rate is below the safety threshold, improving the physical rationality and accuracy of the inversion results, and realizing visualization of the spatiotemporal evolution of backfill body strength and early safety warning.

The application discloses a deep learning-based building outer wall sound insulation effect optimization method and system, relates to the technical field of building energy-saving sound insulation materials, and comprises the following steps: establishing a material-acoustic performance database; constructing and training a process influence correction model; constructing a joint prediction model comprising a generator, a discriminator and an auxiliary predictor which are coupled, and performing joint supervised training based on the material-acoustic performance database; obtaining initial material composition parameters of a target scene, and iteratively optimizing the initial material composition parameters based on the process influence correction model and the joint prediction model to obtain optimal material composition parameters. The application solves the technical problem that the existing building outer wall sound insulation design cannot predict the acoustic performance under different material compositions and process conditions, leading to great blindness in material selection and structure design, and the global optimal configuration of material composition cannot be achieved.

The application relates to the technical field of catalysts and discloses a catalyst process parameter optimization method based on a mathematical model, which comprises the following steps: obtaining model data and operation process parameters of a catalyst; establishing a mathematical model of the catalyst based on the model data; calibrating the mathematical model based on the operation process parameters to obtain a calibrated model; obtaining at least two groups of operation process parameters of the catalyst; simulating particle size distribution of the catalyst based on the calibrated model to obtain particle size distribution data corresponding to each group of operation process parameters; screening optimal particle size distribution data and optimal process parameters corresponding to the optimal particle size distribution data from all the particle size distribution data; and updating the operation process parameters to the optimal process parameters. The method provided by the embodiment of the application optimizes the operation process parameters to the optimal process parameters, so that the produced catalyst has optimal particle size distribution.

The present application belongs to the technical field of intelligent control and modeling of steel metallurgy process, and relates to a kind of converter decarburization reaction kinetics prediction model construction method and system, comprising: constructing converter end steel composition TSO prediction model;According to the real production data and the steel composition TSO prediction model, the TSO content prediction result of the converter end steel composition is obtained by prediction;Introducing control parameter to correct the preset decarburization reaction kinetics basic model;Each TSO content prediction result is used as end constraint, the parameter iterative optimization of decarburization reaction kinetics basic model is carried out, and based on the corrected decarburization reaction kinetics basic model and the optimal result of parameter, the optimized converter decarburization reaction kinetics prediction model is obtained.The present application realizes the adjustable dynamic modeling of converter decarburization process, can accurately capture the influence of control parameter change on converter smelting decarburization reaction, realizes the closed-loop iteration mechanism of prediction, feedback and correction, and guarantees the continuous leading of model performance.

The invention relates to an electrolytic hydrogen safe operation boundary construction and application method and system based on dynamic characteristic identification, and electronic equipment. The method comprises the following steps: acquiring actually measured dynamic response data of an electrochemical device under a dynamic excitation working condition; identifying real dynamic characteristic parameters through a system identification algorithm based on actually measured dynamic response data to replace idealized dynamic setting to construct a high-fidelity security model with a coupling sub-model; a high-fidelity security boundary is obtained based on high-fidelity security model simulation; operating parameters are collected in real time and compared with the high-fidelity safety boundary, and an instantaneous risk component is calculated; when the instantaneous risk component exceeds a safety control threshold, a power limit instruction is generated to perform closed-loop limitation on the input power. According to the method, the limitation of a traditional steady-state model is broken through, the transient out-of-limit risk can be accurately predicted and intervened, and the intrinsic safety of the electrolytic hydrogen system under the fluctuation working condition is greatly improved.

The application provides a comprehensive manufacturing method for performance optimization of water-based wear-resistant substitute plastic finishing oil, comprising: obtaining wear resistance performance data and flexibility data of a paint sample, removing noise through preprocessing to obtain a standardized data set; according to the standardized data set, using a support vector machine to classify the relationship between wear resistance performance and flexibility, and determining boundary conditions of performance conflicts; extracting a feature vector from the boundary conditions of performance conflicts, and obtaining a high conflict point set through Euclidean distance judgment of high conflict points; for the high conflict point set, using a genetic algorithm to generate and evolve a candidate formula design scheme; evaluating the wear resistance performance and flexibility of the candidate formula design scheme through simulation testing to obtain a comprehensive score vector; obtaining a material stability index from the preliminary optimization result, and determining a final formula design scheme through comparison with original paint data.

The present invention belongs to the technical field of automated steel production and discloses a method for predicting the composition of molten iron applied to an automated steel production process, wherein the method is: performing a sampling and an examination of molten iron in a process in which the molten iron is tapped from a torpedo tank into a ladle, and calculating a loss of composition of this molten iron, wherein the composition of the molten iron is calibrated before the molten iron is tapped into the ladle;Performing a sampling and analysis of the molten iron in a process where the molten iron is transferred from the ladle to a converter, and calculating any loss in the composition of this molten iron, wherein the composition of the molten iron is calibrated before the molten iron is transferred to the converter. The present invention is easy to use and implement, and can acquire more accurate and reliable real-time parameters for computational models at all levels; in its specific application, the present invention does not require the modernization of existing automated steelmaking equipment; and it can be applied to a wide range of automated steelmaking scenarios.

This application discloses an online assessment method and system for the biotoxicity of groundwater complex pollution, relating to the field of environmental monitoring technology. The system includes a multi-channel monitoring device, a data processing and calculation unit, and an early warning output unit. The method includes: acquiring time-series data of the movement trajectories of each fish in multiple independent monitoring channels; extracting multi-dimensional behavioral features from each monitoring channel; determining the baseline values ​​of the behavioral features of each monitoring channel at the current moment using a dynamic baseline model; comparing the multi-dimensional behavioral features with the baseline values ​​to calculate standardized feature offsets; fusing the standardized feature offsets of each monitoring channel to obtain a comprehensive feature vector, which is then input into a biotoxicity assessment model to output a comprehensive assessment result of the biotoxicity of the water body. This method intelligently learns the complex mapping relationship between complex pollution characteristics and biological responses, achieving more sensitive, accurate, and intelligent online monitoring and assessment of groundwater complex pollution.

The application provides a peak detection and qualitative method for full two-dimensional mass spectrum data based on a multi-modal model and related equipment, and comprises the following steps: performing supervised fine-tuning training on a pre-trained multi-modal large language model by using a mixed data set, performing reinforcement learning training on an initial peak detection and qualitative model by using a reinforcement learning fine-tuning data set, inputting preprocessed multi-modal input data into the peak detection and qualitative model to perform data extraction, and obtaining a corresponding data extraction result; by combining the multi-modal large language model and the reinforcement learning fine-tuning training, the method can accurately detect target peaks when processing full two-dimensional mass spectrum data, and can efficiently analyze compounds; the use of a low-rank synthetic data set helps to enhance the accuracy of model data processing; in the reinforcement learning fine-tuning stage, the structure generation strategy of the model in unknown compound analysis is optimized, so that the model can cope with complex situations of unknown compounds and optimize the structure generation capability.

The application provides a kind of based on artificial intelligence's fluid chemical reaction process prediction and optimization method and system, the method includes: step 1, the acquisition of fluid chemical reaction process multi-modal original data;Step 2, original data preprocessing;Step 3, construct coarse-grained mechanism model+data-driven residual model+physical constraint neural network's hierarchical hybrid modeling system;Step 4, online identification and dynamic updating;Step 5, based on hybrid model executes integrated multi-objective and strong constraint optimization control strategy;Step 6, the prediction result and optimization strategy output to field control system.The method of the application can accurately predict key process variables through hierarchical hybrid model;Online identification and updating realize process real-time optimization;Through constraint control, optimization under complex conditions is realized.The method of the application can significantly improve production efficiency, reduce energy consumption, improve target product yield, and ensure safe and stable operation of the reaction process.

A non-transitory computer-readable recording medium stores therein a training program that causes a computer to execute a process including first inputting a first image capturing a target compound to an encoder of an auto-encoder including a latent space that is isometric with respect to an input space, second inputting a latent variable output by the encoder and a typical compound model corresponding to a typical case of a three-dimensional structure of the target compound to a decoder of the auto-encoder, and updating parameters of the encoder and the decoder, based on a reconfiguration error between a second image reconfigured based on an output of the encoder and the first image.

This device for generating crystal structure information on the basis of an artificial intelligence model comprises: a data preprocessing unit that constructs an interleaved dataset by arranging text and images to cross each other; a learning unit that includes a visual encoder for generating image feature data from the images, a language model for generating text-encoded data from the text, and a projector for converting the image feature data and the text-encoded data into data in a shareable form, learns a relationship between the images and the text on the basis of the converted data, initializes the projector, and then performs, on the basis of the interleaved dataset, fine tuning for simultaneously training the language model and the projector to perform condition learning for generating information, which satisfies a specific condition, and filling learning in a method for filling a masked portion; and an information generation unit for generating at least one piece of crystal structure information on the basis of a prompt input.

The present application relates to the technical field of culture medium formula data processing, in particular to a culture medium formula intelligent iterative optimization method and system based on deep learning. The method comprises the following steps: obtaining a formula response feedback pair; converting formula component information to obtain a functional state node; obtaining a stage response improvement amount when only the main functional state changes, and a stage response improvement amount when the main functional state and the collaborative functional state change together; determining a component functional relationship edge according to the difference between the two improvement amounts; retaining the positive component functional relationship edge in the component functional relationship edge and determining a component functional collaborative response feature; calculating a stage response offset feature according to the response drop amount; calculating the executable degree of a candidate formula, and optimizing and adjusting all candidate formulas according to the executable degree. The present application can improve the stability, executability and iterative optimization efficiency of candidate formula recommendation.

This material designing device is provided with: a search unit (121) that sets a generation condition (21) for generating a three-dimensional structure (23) of a material, the generation condition (21) including information indicating at least one of the characteristic and structure of the material; a cross-sectional image generation unit (123) that generates a cross-sectional image (22) of the material on the basis of the generation condition (21) set by the search unit (121); and a three-dimensional structure generation unit (124) that generates the three-dimensional structure (23) of the material on the basis of the cross-sectional image (22) generated by the cross-sectional image generation unit (123).

The present disclosure relates to a method, system and control device for regulating the performance of carbon sequestration filling slurry, using industrial solid waste as raw material, combining with glauber's salt and alkali residue composite activator, establishing a mapping relationship model of performance index and process parameters, through pre-carbonization and intensified carbonization grading process, considering the pumpability of slurry and the strength of filling body; The system integrates raw material storage metering, mixing and slurry preparation, grading carbonization, conveying and filling, CO2 gas supply and regulation and control modules, realizes full-process closed-loop control relying on online monitoring and model dynamic correction, self-adapts to raw material and working condition fluctuations, can efficiently consume large quantities of industrial solid waste, improve CO2 mineralization and carbon sequestration efficiency, and is suitable for mine goaf cementation filling, and has multiple benefits of resource recycling, carbon emission reduction and engineering stability.

A computer-implemented method and system are provided. An analytical instrument support system receives a first spectra dataset associated with scans of one or more first samples, the one or more first samples including a target analyte having one or more known levels of a parameter. The analytical instrument support system receives a second spectra dataset associated with scans of one or more second samples. The analytical instrument support system determines one or more spectra arrays by combining (i) the first spectra dataset and (ii) the second spectra dataset. The analytical instrument support system determines a chemometric model for one or more levels of the parameter of the target analyte based on, at least, the one or more spectra arrays and the one or more known levels of the parameter.

The invention relates to a method for extracting discrete neutron components (derived from photonuclear reactions between photons and at least one chemical element to be detected) from a neutron spectrum derived from photonuclear reactions obtained by irradiating a material comprising the at least one chemical element to be detected with a source of photons, at least one energy of the photons of the source being greater than the photonuclear reaction threshold of the chemical element to be detected, which method uses a multilayer and multichannel neural network having an architecture with a convolution stage and a deconvolution stage. The method comprises a preliminary step of training the neural network by supervised learning and a prediction step using the trained neural network.

The application relates to a deep learning-based low-temperature brittleness prediction method for wood components. The method comprises the following steps: acquiring temperature and humidity time series data of an environment in which a target wood component is located and performing standardization preprocessing to generate a standardized environment data sequence; based on the standardized environment data sequence, a temperature and humidity coupling derived index is calculated, and a multi-dimensional feature sequence is generated in combination with original time series statistical features of the standardized environment data sequence; the multi-dimensional feature sequence is intercepted according to a first preset sliding time window to generate a plurality of feature matrices, and each feature matrix is assigned a label to generate a deep space-time sequence sample; the deep space-time sequence sample is input into a trained deep space-time prediction model to generate a continuous risk probability curve of the target wood component occurring brittle fracture in a preset future time period. The method can deeply integrate time series environment data, scientifically quantify the temperature and humidity coupling effect, and significantly improve the timeliness and accuracy of risk early warning.

A foundation model for performing molecular-level tasks by learning multimodal data in the form of one-dimensional text and two-dimensional graphs, and a system therefor, according to an embodiment of the present invention, enable various molecular-unit tasks such as chemical reaction prediction, molecular attribute prediction, and natural language description generation to be effectively processed through a single foundation model. In addition, it is possible to increase prediction accuracy of the model by maximizing utilization of two-dimensional molecular graph information, and automatically generate, on the basis of statistical sparsity, high-quality descriptive text that emphasizes core and distinctive features of each molecule.

The application relates to the technical field of blood drug concentration detection, and discloses a blood drug concentration analysis method based on mass spectrometric detection, which comprises the following steps: adopting microfluidic technology in combination with solid-phase extraction to pretreat a blood sample to obtain a concentrated solution, so as to remove interfering components and concentrate target drug molecules; according to the physicochemical properties of the target drug molecules, the concentrated solution is selected to flow through a biosensor or is directly introduced into a chromatographic unit, and the target drug molecules in the liquid phase are converted into gas-phase ions when flowing through the biosensor; chromatography-mass spectrometry is adopted to detect the gas-phase ions or the concentrated solution, so as to obtain mass spectrum data; a machine learning model is used to analyze the spectrum data, to identify the drug types and calculate the concentration of the target drug molecules, and to generate an analysis report. The method is suitable for target drugs with different physicochemical properties, simplifies the detection process, and improves the applicability and accuracy of detection.

A method of analysing a structure of a composition of matter in a sample comprising obtaining a data set comprising a plurality of spectra from the composition, from a first method of analysis dividing each of the spectra into a plurality of bins determining a control parameter or parameters indicative of synchronised fluctuations in signal intensity across some or all channels, resulting in universal correlation between said bins determining a partial covariance of different bins across the plurality of spectra using the control parameter to correct the correlation of intensity fluctuations between said bins.

This invention discloses a chlorophyll concentration profile inversion method based on a physical information neural network, belonging to the field of physical information machine learning technology. It is used for chlorophyll concentration profile inversion, including acquiring marine optical profile observation data to form observation sample data, and performing fractional standardization on the observation sample data; constructing and training a physical information neural network model, destandardizing the predicted data, and constructing a physical constraint loss function; inputting the data to be inverted into the trained physical information neural network model, and calculating the chlorophyll concentration profile using the model's predicted values. This invention constructs a physical information neural network model and introduces a physical constraint loss function to ensure that the inverted absorption coefficient, backscattering coefficient, and irradiance profile conform to the light transmission law in water, avoiding physical anomalies generated by purely data-driven methods, and achieving cross-domain inversion from apparent optical quantities to intrinsic optical quantities without the need for step-by-step calculations or intermediate parameter estimation.

The present application relates to a kind of federal graph clustering method for drug molecule discovery, belong to the cross technical field of artificial intelligence, distributed machine learning and biomedical informatics, the method includes: in client, based on local mode separation mechanism, original drug molecule graph is separated into the common invariant subgraph reflecting common efficacy mode and the personalized variable subgraph embodying mechanism specific modification;Common invariant subgraph and personalized variable subgraph are personalized knowledge fusion based on functional similarity in server end, and two-stage clustering is realized in client through molecule class discovery.The present application is particularly suitable for cross-institutional drug discovery, molecular property prediction, lead compound screening and other key links of medical research and development, aims at realizing efficient collaborative analysis under the premise of privacy protection, improves new drug research and development efficiency and safety.

The application discloses a kind of wastewater zero discharge process for gas combined cycle power station, it is related to combined cycle power station wastewater treatment technical field, including wastewater is coagulated and precipitated, oil removal and activated carbon adsorption by air floatation, remove suspended solids, calcium magnesium hardness and grease in wastewater;The pretreated effluent is sequentially subjected to ultrafiltration and softening bed softening;Two-stage reverse osmosis is used to concentrate softening water in stages;Electrodialysis concentrated water is transported to evaporation pond, and forced atomization evaporation is carried out by mechanical atomization evaporator, to realize wastewater zero discharge.The application effectively solves the fouling and pollution risk of subsequent membrane system, greatly improves the equipment operation stability and service life;Compared with evaporation crystallization process, energy consumption and operating cost are greatly reduced;Evaporation speed is increased by more than 10 times, while salt mist is effectively controlled from drifting, and the risk of secondary pollution is eliminated;The first and second reverse osmosis fresh water produced by the whole process can be used for circulating cooling system, to realize high wastewater reuse rate.