111 results about "Similarity criterion" patented technology

A seismic wave identification method based on deep learning comprises the following steps: step 1, acquiring seismic wave signals to obtain original data, and performing center interception and normalization processing to obtain unified signals; 2, decomposing a unified signal by using CEEMDAN to obtain a plurality of orders of intrinsic mode functions and residual components, carrying out coarse graining processing on each order of mode function according to a preset time scale parameter to divide into subsequences, constructing a similarity criterion through a fuzzy membership function, calculating fuzzy entropy values of the subsequences, and integrating the fuzzy entropy values into a feature matrix; and 4, inputting the feature matrix after the dimension reduction into a CTCM-1D-CNN model optimized by a tribe competition and member cooperation algorithm, predicting the type of an output seismic wave, and completing seismic wave identification. Therefore, the design has good seismic wave identification precision and stability.

The invention discloses a daily load curve clustering method based on a mixed distance improved FCM algorithm, and belongs to the technical field of data processing, and the method comprises the steps: collecting daily load data of a user at a sampling frequency of 5 minutes, carrying out the corresponding data preprocessing, analyzing the fluctuation characteristics of a daily load curve, and carrying out the clustering of the daily load curve; euclidean distance representing load curve power amplitude information and cosine distance representing load curve power fluctuation direction information are constructed respectively, and then an entropy weight method is adopted to construct a hybrid similarity distance (HSD) fusing the Euclidean distance and the cosine distance as a similarity criterion between daily power load curves; replacing the Euclidean distance in the FCM algorithm with the constructed HSD, then calculating a corresponding membership degree and a clustering center, iteratively updating the membership degree and the clustering center by using a target function based on an intra-class error weighted quadratic sum, and completing daily load curve clustering according to a preset maximum iteration number; and finally, determining the optimal clustering number and the corresponding optimal daily power load curve clustering result by using the sum of intra-class errors (SAE), thereby effectively solving the prominent problems of poor effect, low efficiency and the like of the traditional power load curve clustering method.

The invention discloses a landslide surge physical model test drag force prediction method based on AM-LSTM, and the method comprises the steps: constructing a three-dimensional hydrodynamic coupling physical model of landslide surge for a target landslide region and a hydraulic structure structure based on Froude similarity criterion, and carrying out the prediction of the drag force of the landslide surge. The method comprises the following steps: acquiring dynamic data of dam surface drag force by setting test groups with different landslide mass volumes and water depths, constructing a data set containing time sequence features, performing normalization preprocessing, and dividing the data set into a training set, a test set, a feature data set and a label data set; a long short-term memory neural network LSTM layer and an attention mechanism AM layer are introduced, data are input for training, a long short-term dependency relationship between the data is captured, dynamic prediction of drag force is realized, and result visualization is performed. According to the method, the towing force sequential evolution rule caused by landslide surge is captured through the LSTM model with the attention mechanism, dynamic prediction of the towing force is realized, and the prediction accuracy is improved.

The invention discloses a tunnel-pile interaction experiment method and device based on a 3D printing technology, and belongs to the technical field of civil engineering underground structure tests. The method comprises the steps that the geometric and mechanical similarity ratio is determined according to an engineering prototype, a tunnel lining model and a pile group model are prepared through the 3D printing technology, and controllable adjustment of the rigidity of the tunnel model is achieved by adjusting the type, the filling rate and the internal structure of a printing material; controllable adjustment of the rigidity of the pile foundation model is realized by adjusting the size of the magnetic field; the model is placed in a soil body groove, the shield tunnel excavation process is simulated, and deformation data of a tunnel and a pile foundation are collected through a PIV technology and a sensor. Accurate control over the rigidity of the tunnel and pile foundation model is achieved through 3D printing, the influence of shield tunnel construction on the pile group foundation can be simulated under the condition that the similarity criterion is met, and the device has the advantages of being adjustable in structure, high in experiment efficiency and the like and can be used for researching a dense urban area tunnel-pile deformation coupling mechanism and optimizing related engineering parameters.

The invention relates to the technical field of icing wind tunnel tests, in particular to a propeller icing wind tunnel test working condition similarity matching method and a storage medium. A first calculation model is established based on a forward ratio similarity criterion of a propeller and a first similarity criterion, and a second calculation model is established based on a plurality of set similarity criteria. Then, basic parameters of the target propeller are obtained, and the basic parameters comprise the size of the target propeller, the first angular speed, the first incoming flow wind speed and the scaling ratio of the icing wind tunnel test on the target propeller; and inputting the first angular velocity and the scaling proportion into a first calculation model to obtain a scaled second angular velocity. And finally, inputting the basic parameters and the second angular velocity into a second calculation model to obtain test working condition parameters of the target propeller in the icing wind tunnel test. Therefore, the design period of similar matching of the propeller icing wind tunnel test working conditions is shortened, and the quality of the scaled test working condition data is ensured.

The invention provides a bridge deck reduced scale model design method and experiment system in a plateau environment, and the design method comprises the steps: building a multi-factor coupling similarity criterion based on a similarity principle, physical parameters affecting the prototype structure performance of a bridge deck, and plateau environment parameters; determining a dynamic stress similarity ratio model based on a multi-factor coupling similarity criterion and the plateau environment parameters; based on the dynamic stress similarity ratio model and the section shape and size of the bridge deck prototype structure, performing scale design on the bridge deck prototype structure to obtain an initial scale model; and performing simulation experiment and simulation verification on the initial scale model based on a preset numerical simulation strategy and a preset plateau environment simulation strategy, and performing parameter optimization of the dynamic stress ratio similar model to design a target scale model. According to the scheme provided by the invention, experimental simulation of the bridge deck in the plateau environment can be efficiently and accurately carried out, and the experimental simulation cost is reduced.

There is provided a technique of computerized substituting an odor item of interest in a formula of a smell. The method comprises: generating for a given inventory of odor items a dataset informative, at least, of similarity of the pairs of odor items of interest with other odor in the inventory; processing data in the generated dataset to reveal one or more odor items meeting one or more similarity criteria, thereby generating one or more candidate odor items; and applying to the one or more generated candidate odor items an optimization algorithm to find, among them, a candidate odor item with the best fit to the formula of the smell. The resulting odor item is usable for substituting the odor item of interest in the formula of the smell. Similarity labels in the dataset can be provided with the help of a plurality of labeling functions applied to the pairs of odor items.

The invention relates to the technical field of dam foundation treatment, and discloses a low-liquid-limit clay dam foundation soil body excavation tedding thickness calculation method, which comprises the following steps: firstly, collecting an undisturbed soil sample and a disturbed soil sample, and obtaining physical and mechanical parameters through an indoor test; performing a light compaction test by using the disturbed soil sample, and calculating a first soil layer compaction coefficient according to the height change before and after compaction; an indoor dynamic compaction physical model meeting the similarity criterion is constructed based on the white Jinhan Pi theorem, the conversion relation between model compaction energy and on-site compaction energy is deduced through a geometric similarity coefficient, a time similarity coefficient and a quality similarity coefficient, and a second soil layer compaction coefficient is measured; and the maximum value of the two compaction coefficients is taken as a final calculation parameter, and the excavation tedding thickness is calculated in combination with the design thickness of the dam foundation bearing stratum. According to the method, the indoor test is used for replacing field trial compaction, so that the test cost is reduced, two measurement schemes of a compression modulus method and an accumulated compaction settlement method are provided, and the calculation result is more in line with engineering practice.

A set Kalman filtering turbulence model assimilation method based on full field similarity criterion is disclosed, which comprises the following steps: sensitivity analysis is performed on the coefficients of the candidate turbulence model to obtain sensitive coefficients and their sampling range for Latin hypercube sampling, i.e., sampling interval; model constant sample combinations are extracted from the sampling interval by using a directional Latin hypercube sampling method, and target flow field results are calculated; a data assimilation algorithm based on the set Kalman filtering algorithm is used for the target flow field results and the model constant sample combinations, and after obtaining the optimized model constant combination, the model constant combination is updated and verified in cycles to obtain the optimal model parameter combination; finally, the improved turbulence model is used to simulate the internal flow field of the fuel assembly in the core of the pressurized water reactor with high similarity. The present application can provide an objective evaluation standard for the full physical field similarity of the data assimilation method, significantly improve the convergence speed of the algorithm, strengthen the robustness of the algorithm, and reduce the calculation cost of the algorithm.

A system and method for protecting ranging by feedback of at least one channel metric. The method can include receiving, by a first modem, a first training signal from a second modem, generating, by the first modem, a first estimate of a channel metric based on the first training signal, receiving, by the first modem, a second estimate of the channel metric from the second modem, determining, by the first modem, that the first estimate of the channel metric and the second estimate of the channel metric do not meet a similarity criterion, and denying, by the first modem, access to a resource.

The invention provides a method for constructing a bearing test distortion model of an ice-resistant structure under the action of a dynamic ice load, and belongs to the field of test similar reduced scale model design under an ice breaking load of the ice-resistant structure. The problem that in the prior art, the strain rate of a fully-similar model is dissimilar due to the scale effect is solved; a traditional distortion model design method has the problem of dissimilar energy distribution. The method comprises the following steps: establishing a fully similar model from an ice-resistant structure prototype to a geometric and plate thickness equal ratio reduced scale; establishing a plate thickness distortion reduced scale model by using the same material on the basis of the geometric similar reduced scale model; preliminarily determining a collision contact position of the thickness distortion reduced scale model; a dynamic ice load similarity criterion is applied, correction factors are deduced through dimensional analysis, and ice body parameters and collision speed of a distortion model system are adjusted, so that the ice breaking capacity and dynamic structure response are similar to those of a prototype system. The device is mainly used in the field of polar region icebreaking.

The invention discloses a seaplane wave sliding load spectrum compilation method. The method comprises the steps of determining gravity center acceleration data of pool test data of each task section according to overall and structural parameters of a seaplane; comparing a gravity center acceleration time history curve representing the gravity center acceleration data with an airframe trim angle time history curve, and intercepting stable section data of each task section; performing power spectrum density analysis on acceleration overload / overload increment in the stable section data to obtain an acceleration characteristic frequency band; determining a cut-off frequency according to the acceleration characteristic frequency band, and filtering the stable section data; carrying out accumulated transcendental statistics on the filtered stable section data to obtain an accumulated transcendental curve of each task section; according to the time proportion of each task segment and the similarity criterion, obtaining an accumulative surpassing curve of each task segment of the real machine water surface operation; and forming a body wave sliding equivalent spectrum according to a damage equivalence criterion. The technical problem of compiling and verifying the water skiing fatigue load spectrum of the seaplane is solved.

A method is implemented for automatically labelling images at a computer system having one or more processors and memory. The computer system obtains a first image including an object, e.g., from a camera disposed at a physical environment. A reference model is applied to process the first image and generate a reference label, e.g., identifying the object in the first image. An image generative model is applied to generate a reference image based on the reference label. In accordance with a determination that the first image and the reference image satisfy a similarity criterion, the computer system labels the first image with the reference label. The first image that is labelled with the reference label is added to a corpus of training data to be used to generate a target model for autonomously monitoring the physical environment.

The invention discloses a method and system for processing ground surface settlement monitoring data, and relates to the technical field of tunnel monitoring data processing, and the method comprises the steps: collecting the ground surface settlement monitoring data of each measuring point, and carrying out the outlier recognition and processing; extracting a feature image of the monitoring data of each measuring point by adopting a Mel-frequency cepstrum coefficient feature extraction method, and converting the ground surface settlement monitoring data into an MFCC feature map; the MFCC feature map is input into a pre-trained MFCC-Transform model for data classification, and measuring point data affected by tunnel excavation construction are reserved; an improved ELM time sequence prediction algorithm is adopted to fill data missing values; and decomposing the monitoring data into a plurality of intrinsic mode functions (IMF) by adopting a VMD time sequence decomposition algorithm, and selecting the IMF with the highest similarity as the monitoring data after noise reduction in combination with a similarity criterion. According to the invention, the full-automatic processing of the ground surface settlement monitoring data is realized, and the reliability of data processing is ensured while the data processing speed is improved.

The application provides a low-rank parameter update denoising method and system for continuous learning, comprising: for any pre-training model, obtaining original update parameters by training on a current task using low-rank self-adaption; obtaining left singular vectors of the update parameters as candidate subspace basis vectors by singular value decomposition; defining a subspace similarity scoring function based on Grassmann manifold to calculate the alignment degree of the update subspace and the pre-training model base subspace; automatically screening and retaining key components according to the highest similarity criterion, eliminating noise interference components, and reconstructing the denoised update parameters; and updating the model in an incremental or staged manner using the denoised update parameters. The application effectively balances the plasticity and memory of old knowledge of the model on a new task by removing noise components in the low-rank self-adaptive training update, and significantly alleviates the catastrophic forgetting problem in continuous learning.

The invention relates to the technical field of physical similar simulation, and particularly provides a physical simulation method for a lunar lava tube structure forming process, which comprises the following steps of: mixing a thermoplastic material and a condensation additive to prepare a simulation melt meeting a basalt lava rheology and thermodynamics similarity criterion; the low gravity effect, the high vacuum environment and the radiation cooling condition in the lunar environment are reproduced through an inclined flow channel and a thermal vacuum cavity, and the injection rate, the flow channel inclination angle and the target environment temperature of the simulated melt are determined by accurately matching dimensionless parameters such as the Peclet number, the dimension temperature and the Rayleigh number; under the condition, controlled flow and surface solidification crusting of the melt are simulated, and finally a cavity structure model is formed, so that physical research on the lunar lava tube structure forming process is realized. According to the method, the lava flow dynamic crusting and cavity forming process in the extreme lunar environment can be accurately reproduced, and a quantitative experimental data basis is provided for formation mechanism research and engineering application of the lunar lava tube.

This invention relates to the field of physical similarity simulation technology, specifically providing a physical simulation method for the formation process of lunar lava tube structures. The method involves preparing a simulated melt that meets the rheological and thermodynamic similarity criteria of basaltic lava by mixing thermoplastic materials with condensation additives. The low-gravity effect, high-vacuum environment, and radiative cooling conditions under lunar conditions are reproduced using inclined flow channels and thermal vacuum cavities. By precisely matching dimensionless parameters such as Peckley number, dimensional temperature, and Rayleigh number, the injection rate, flow channel inclination angle, and target ambient temperature of the simulated melt are determined. Under these conditions, controlled melt flow and surface solidification crust formation are simulated, ultimately forming a cavity structure model. This allows for the physical study of the formation process of lunar lava tube structures. This invention can accurately reproduce the dynamic crust formation and cavity formation process of lava flow under extreme lunar conditions, providing a quantitative experimental data basis for the study of the formation mechanism of lunar lava tubes and their engineering applications.

The invention discloses a method for manufacturing a complete aeroelastic model of a special-shaped high-rise structure, and the method comprises the steps: S1, determining a geometric scale ratio based on prototype parameter extraction and modal analysis and test conditions; s2, determining a similarity ratio of the model according to a geometric scale ratio and a similarity criterion, and selecting a printing material; s3, according to a similarity criterion and an equivalent design method of section stiffness, obtaining a variational progressive equivalent section of the aeroelastic model along the height; s4, establishing a reduced-scale finite element model, comparing the frequency and vibration mode of the reduced-scale finite element model and the full-scale finite element model, and confirming that the reduced-scale finite element model meets a preset similar condition; s5, preparing an equivalent aeroelastic model through a 3D printing technology; and S6, fixing and aerodynamic configuration compensation are conducted on the aeroelastic model, and the complete aeroelastic model of the special-shaped high-rise structure is manufactured. According to the manufacturing method for the complete aeroelastic model of the special-shaped high-rise structure, efficient design and manufacturing of the aeroelastic wind tunnel test high-precision model are achieved.

This invention provides an automated digital printing sample adjustment method and system, including a sample adjustment client that imports a preset pattern file to be printed and performs similarity matching calculations with a local process database or a cloud process database on a cloud server. The local process data and the cloud process database store historical process scenario data examples. The sample adjustment client retrieves examples that meet the similarity criteria, corrects the data, and sends it to the digital printing machine terminal. The digital printing machine terminal drives the digital printing machine to print and saves the process scenario data that meets the finished product effect criteria to the local or cloud process database. The cloud server also communicates remotely with service engineers to manually set process scenario parameters. By storing and managing the printing difference parameters of all digital printing equipment, when a customer adjusts a sample, the system automatically matches and retrieves similar historical examples and makes corrections, improving the sampling efficiency of digital printing plants.

Provided are a computer program product, system, and method for forming a hypothesis set from sentences across documents representative of different stances taken across the documents. Sentences from the documents are clustered into a plurality of clusters. Sentences in a cluster of the clusters have stance scores with respect to other sentences in the cluster that satisfy a stance criteria. At least one similarity group of sentences is formed in the clusters having similarity scores satisfying a similarity criteria. Sentences are selected from the similarity groups in the clusters based on stance scores of the sentences in a similarity group. A hypothesis set is formed of the selected sentences in the similarity groups. Stance scores are determined of sentences in the documents with the sentences in the hypothesis set to determine stances of the documents with respect to the sentences in the hypothesis set.

The invention relates to the field of data processing, and provides a distribution comb contour determination method based on three-dimensional data acquisition, a medium and equipment. The method comprises the following steps: clustering an obtained contour curve of a specific area of the head of a user to generate a centroid-like curve, dividing the centroid-like curve into segmented centroid curves according to a preset interval, and screening segment curves with the similarity reaching the standard in a corresponding cluster as fitting candidate curves; and establishing a multi-objective optimization function containing curvature continuity, second-order derivability and anatomy adaptation degree, performing global optimal combination on the fitting candidate curves by using a dynamic programming algorithm to generate an initial contour point location sequence, and performing smooth optimization to form a contour curve of the distribution comb attached to the scalp side. According to the method, user head shape feature categories are divided through clustering analysis, an initial hash point set is generated by using a clustering center, the curve fitting degree is optimized in combination with a standardized smoothing algorithm, and finally a standard curvature template adaptive to a corresponding group is output.

This invention discloses a dynamic simulation device and a method for optimizing the molding quality of overflow forming of OLED display glass, belonging to the field of glass overflow forming process development technology. The device includes a support frame, an overflow pull-down mechanism, a simulated liquid heating furnace, a cooling mechanism, and a traction roller. The overflow pull-down mechanism is equipped with an overflow trough, which is provided with a near-end and far-end dual lifting mechanism, a far-end height adjustment component, and a material flow rate control valve. It can dynamically adjust the trough depth, far-end height, and inlet flow rate without stopping the machine for disassembly and assembly, completely replicating the entire process of overflow, cooling, and traction forming, and preparing a solid sheet. The method establishes a physical model based on the Reynolds number as a similarity criterion, and adjusts the inlet flow rate, trough depth, and far-end height in a closed loop according to the sheet thickness deviation to quickly correct the thickness uniformity. The parameters can be proportionally converted to the industrial prototype, improving process development efficiency, reducing costs, and compensating for the deviation between numerical simulation and actual production, providing reliable support for the optimization of OLED display glass overflow forming process.

The application provides an active control method of an electro-hydraulic servo system based on finite time control. The method dynamically restricts the error boundary through a time-varying obstacle Lyapunov function, combines a backstepping recursive control framework with a fractional order convergence term, significantly improves the convergence speed and anti-interference ability of the system, effectively suppresses the overshoot problem and chattering phenomenon of PID control, designs a feedforward compensation and an adaptive disturbance observation mechanism in view of the nonlinear characteristics and anti-interference problems of the electro-hydraulic servo system, and still can quickly recover stability under strong external disturbance, thereby ensuring the accuracy and robustness of the control force output, and based on the similarity criterion, a scale experiment platform is constructed to verify the engineering universality of the control strategy, and an efficient, reliable and easy-to-deploy solution is provided for the anti-seismic control of complex structures.

The disclosure relates to a method implemented by computer means for determining correspondences between features of a structure in a sequence of images, the method comprising: obtaining a dynamic set of key points associated with positions in a reference image; for each feature identified in a given image of the sequence: determining (131, 151, 171, 191, 192) a correspondence between the feature and a reference key point of a set of reference key points based on a similarity criterion; updating (210, 220, 310, 320, 330, 330, 331) the dynamic set of key points based on the determined correspondence and at least one predefined rule for assigning positions and associated values to the features; and dynamically excluding (211) key points from the dynamic set of key points based on their associated values exceeding a predefined threshold.

The present application relates to compressed air energy storage technology field, specifically disclose a kind of compressed air energy storage process energy loss measurement method based on force heat flow simulation test.The present application is under the guidance of force heat flow coupling similarity criterion dynamically collected cavern air pressure and temperature, and uses gas internal energy and heat change calculation formula to calculate the energy loss of gas storage stage, realize the energy loss of gas storage stage in simulation test process and analysis.The method of the present application not only greatly reduces the high capital investment and long time period of large-scale simulation test in traditional method, but also significantly improves the test efficiency and research feasibility, provides more scientific, accurate and reliable technical support for energy efficiency optimization of compressed air energy storage system.

Method for characterizing an object using an ultrasonic transducer, the method comprising the steps of: Acquiring (201) at a first instant, using the ultrasonic transducer, a first ultrasonic signal resulting from the reflection of an ultrasonic field emitted by the transducer on a region of interest including the object, Identifying (202) a point of interest in the first signal, Extracting (203) from the first signal a so-called reference portion including the point of interest, Acquiring (204) at a second instant, a second ultrasonic signal, Calculating (205) a similarity criterion between the reference portion and the second signal, Determining (206) the position of the point of interest in the second signal from the position in the second signal corresponding to the maximum of the similarity criterion and the relative position of the point of interest in the reference portion. Figure 2

In various examples, a technique for low-latency fusion and tracking of objects from multiple cameras is disclosed that includes determining a plurality of input objects and corresponding object characteristics, individual input objects and respective object characteristics being determined based at least on an image generated using a respective camera. The technique also includes identifying at least one subset of the plurality of input objects, the at least one subset corresponding to a respective physical object and comprising at least one input object that satisfies a similarity criterion. The technique further includes generating an output object associated with one or more smoothed object characteristics, the output object being generated based at least on two or more input objects included in the at least one subset of the plurality of input objects. The at least one subset corresponds to a physical object that is visible to two or more cameras.

Disclosed are methods, systems, and apparatus for monitoring delivered packages using video. A method includes obtaining a notification of a delivery of a package at a property that includes a first image depicting the package; obtaining a second image captured by a camera at the property; determining, using the first image and the second image, that the package has been delivered to the property; generating a package model that represents an appearance of the package; and tracking a location of the package for use in determining whether to provide an alert to a device about the location of the package. Determining that the package has been delivered to the property comprises: comparing the first image to a model of a scene of the property; determining that the first image satisfies similarity criteria for matching the model of the scene; and determining that the second image likely depicts the package.

The invention discloses an MT deep learning cross-scale inversion method and system based on a physical simulation criterion, and the method comprises the steps: deducing a similarity criterion followed by the physical simulation of an electromagnetic method based on the physical characteristics of the electromagnetic method, creatively introducing the similarity criterion into an MT deep learning inversion technology, and achieving the equivalent switching of data between a training space and a real space. The network trained by an MT data set with a certain frequency and scale is enabled to be suitable for MT data in other scenes in the same field, the generalization ability of the network is greatly enhanced, and the time of training the network is saved. Switching space physical parameters of a real space to a training space by using a similarity criterion, further converting MT forward modeling response data of the real space into forward modeling response data of the training space based on the switched space physical parameters, inputting the forward modeling response data into an MT inversion model based on a neural network to obtain an inversion result, and finally, utilizing the similarity criterion again to obtain an inversion result. And switching the obtained inversion result to a real space.

The invention discloses a self-flowing type water taking pipeline water taking metering method and system. A water taking head water taking pipeline related design scheme is obtained through investigation, parameters such as pipeline materials, pipeline diameter, elevation and pipeline arrangement are sorted, a simulation range is determined, a normal model is designed according to a gravity similarity criterion, and physical model design and manufacturing are carried out in a laboratory; a scaling scale is determined, a model test is carried out through the model, a flow-tracing ball conveying speed relation curve is drawn, and a flow-tracing ball conveying speed relation expression is deduced; the front-section water taking amount test and the tail-end water taking amount test of the water taking pipeline are carried out, and the leakage water amount and the leakage rate of the self-flowing type water taking pipeline are obtained through calculation; and the water taking amount of the self-flowing type water taking pipeline is calculated through parameters such as a water taking pipeline tail end metering facility and a water taking pipeline leakage rate. The water taking amount of the self-flowing type water taking pipeline can be efficiently and accurately calculated, and the metering problem under the condition of lacking of a water taking metering device is solved.