49results about How to "Effective calculation" patented technology

The invention discloses a container damage information detection counting method based on yolov9 and DeepSort, and the method comprises the steps: collecting the data of a container through a camera, inputting the obtained container data into a trained target detection model and a semantic segmentation model, converging the results of the two models, and transmitting the results to a related system for the next operation. The automatic detection of the damaged position and category information and the accurate statistics of the container throughput are completed. The invention provides a comprehensive and effective method which can be applied to improving the efficiency and quality of container management.

The invention belongs to the technical field of water resource management, and discloses a ConvLSTM-based regional groundwater supply capacity assessment method, which comprises the steps of calculating a regional SPEI index, constructing a space-time cube by using the SPEI index, and determining space-time distribution of a regional drought condition; constructing a groundwater flow numerical model according to the hydrological parameters of the groundwater in the arid region, and identifying the water supply status of the groundwater; hydrological data of an evaluation area are acquired, weights of indexes and parameters are determined by using an entropy evaluation method, and a training set, a verification set and a test set are made; constructing a ConvLSTM model, and training the ConvLSTM model by using the training set; and inputting the test set into the trained ConvLSTM model, and evaluating the groundwater supply capacity. According to the invention, the model is established by using regional groundwater spatio-temporal data to accurately evaluate the groundwater supply capacity, so that groundwater resources can be utilized more effectively under drought conditions.

The invention provides a session quality evaluation method, system and device and a medium, and relates to the technical field of data analysis, the method comprises the following steps: determining candidate keywords based on session information of a current session; determining the keyword weight of each candidate keyword based on the undirected graph, selecting a preset number of candidate keywords as keywords based on the values of the keyword weights, and determining the sentiment weight of each keyword according to the sentiment tendency and sentiment intensity corresponding to each keyword under each evaluation dimension; according to the dynamic change condition of the emotional tendency and / or emotional intensity corresponding to each keyword in the session information, dynamically adjusting the emotional weight of each keyword; and in combination with the occurrence frequency, the emotion weight and the keyword weight of each keyword, performing weighted summation on the score of each score dimension to obtain a session quality score of the current session. According to the method, the session quality can be accurately and quickly evaluated.

The invention provides a human voice similarity detection method and system and a storage medium, and the method comprises the steps: obtaining a to-be-detected audio, and carrying out the preprocessing of the to-be-detected audio; performing feature extraction on the preprocessed to-be-detected audio to obtain audio features, and inputting the audio features into a pre-trained acoustic model for phoneme matching to obtain a vowel phoneme sequence; determining formant data of the audio to be detected according to the timestamp of each vowel phoneme in the vowel phoneme sequence, and determining a formant value of each vowel phoneme according to the formant data; and according to the formant value, calculating a formant difference value between the same vowel phonemes in the different to-be-detected audios, and according to the formant difference value, determining the human voice similarity between the different to-be-detected audios. According to the embodiment of the invention, the human voice similarity between different to-be-detected audios can be effectively determined based on the formant difference value, and the accuracy of human voice similarity detection is improved.

The invention discloses an end-to-end object 6D attitude estimation method based on a beam adjustment method, and relates to the technical field of two-stage object 6D attitude estimation. Comprises: data acquisition; preprocessing the data; establishing a model; training the model; and testing the model. According to the invention, three-dimensional translation and three-dimensional rotation of the target object can be detected in a complex scene and in the presence of shielding.

The invention discloses a method for evaluating the ratio of a modified hydrophobic insulating material based on magnesium oxide doped epoxy resin, which comprises the following steps of: preparing two groups of insulating materials of epoxy resin which is not doped with magnesium oxide nano-particles and doped with magnesium oxide nano-particles, and respectively measuring the contact angle and the water absorption rate of two groups of samples at room temperature; respectively calculating hydrophobicity evaluation factors of the two groups of samples before aging by using the data; then performing high-temperature accelerated aging treatment on the two groups of materials, measuring the contact angle and the water absorption rate of an aged sample, calculating a hydrophobicity evaluation factor after aging, and comprehensively considering the hydrophobicity evaluation factor before aging and the hydrophobicity evaluation factor after aging to obtain a hydrophobicity comprehensive evaluation factor of the modified epoxy resin insulating material; and further evaluating the accuracy of the ratio of the modified hydrophobic insulating material with the magnesium oxide doped with the epoxy resin. According to the method, the hydrophobicity comprehensive evaluation factor is utilized to evaluate the hydrophobicity of the modified epoxy resin, the accuracy of different raw material ratios of the modified epoxy resin can be quickly judged, and a basis is provided for selecting a proper magnesium oxide doped modified epoxy resin raw material ratio.

The invention discloses an incremental model merging method and system oriented to a large language model, and belongs to the technical field of machine learning. The method comprises the following steps: calculating confusion parameters of an expert model through a task vector updated by the expert model and a gradient of a pre-training model on a calibration data set during first incremental model merging, and converting the confusion parameters into a parameter mask matrix; calculating the parameter importance of the expert model according to the task vector, the gradient and the parameter mask matrix, sampling to generate a sampling mask matrix, and scaling the task vector by using the sampling mask matrix to generate an incremental task vector; adding the incremental task vectors into parameters of a pre-training model to generate parameters of a merging model; and during subsequent incremental model merging, different new expert models are sequentially merged to the merged model obtained by previous merging by adopting the same method as the method for the first time. According to the method, expert models of different specific tasks can be sequentially combined in an incremental superposition manner, parameter conflicts and task conflicts are avoided, and the model combination performance is improved.

The invention relates to the technical field of tunnel engineering, in particular to a method for determining the influence of shield tunnel holing on mechanical characteristics of a main body structure. Comprising the following steps: S1, constructing a three-dimensional numerical calculation model of a shield tunnel; wherein the three-dimensional numerical calculation model is composed of a stratum model and a shield tunnel model arranged in the stratum model; s2, calculating the initial ground stress of the stratum model in the three-dimensional numerical calculation model; s3, calculating deformation and stress of a shield tunnel model without holes in the three-dimensional numerical calculation model under the action of surrounding stratum load; s4, calculating the structural mechanical characteristics of the shield tunnel model after holing in the three-dimensional numerical calculation model under the action of a peripheral external load; according to the method, the problem that the relation between the structural analysis result of the shield tunnel holing working condition and the structural safety risk cannot be determined by an existing method is solved, and the structural safety risk after shield tunnel holing can be known more visually.

The present application relates to a kind of intelligent algorithm of reheater brush coating area, by the calculation of the actual deviation temperature rise Δt sjpc Of superheated tube, reheater steam temperature rise deviation Δt, superheated tube wall temperature deviation Δt sdpc , superheated tube control temperature rise Δt kzpc , the temperature rise Δt jsws That needs to control reduction, establishes the correlation of spraying coefficient, realizes the effective calculation of brush coating area.The algorithm of the present application is based on the principle of heat conservation, since the temperature of working medium in reheater heating surface is equal to the average value of the tube wall temperature of this section heating surface, therefore, in the heating surface where superheating occurs frequently, brush coating can make the wall temperature of superheating area and the outlet steam temperature approach, so as to achieve the purpose of balancing the heat absorption of corresponding tube row.Through the calculation of brush coating area, it can effectively prevent the phenomena such as superheating of the last stage reheater and screen reheater tube wall, at the same time, save wear-resistant materials and improve economic efficiency.

The present invention relates to a method and system for calculating the trajectory of human flow based on a millimeter wave base station, and belongs to the field of communication technology. The method comprises: obtaining a target signal received by a millimeter wave base station; constructing a signal covariance matrix based on the target signal; inputting the signal covariance matrix into a pre-trained target convolutional neural network model, and outputting a target angle; wherein the target angle represents the angle between the target signal and the receiving antenna of the millimeter wave base station in the vertical direction; determining the horizontal distance between the millimeter wave base station and the mobile terminal based on the target signal and the target angle; calculating the trajectory of human flow corresponding to the mobile terminal based on the horizontal distance and the orientation angle; wherein the orientation angle is the direction angle of the receiving antenna corresponding to the mobile terminal with the millimeter wave base station as the center. This method has high measurement accuracy, is less affected by the environment, and does not capture personal images, and has wide applicability.

The invention provides a turbine blade radiation temperature measurement error correction method based on a reflected radiation model, and the method comprises the steps: building a reflected radiation model through considering the reflected radiation of a preceding-stage guide blade, an adjacent moving blade and a combustion chamber wall surface at the same time; calculating a radiation angle coefficient between turbine blades and a radiation angle coefficient between wall surfaces of a combustion chamber by using a surrounding channel integral method; temperature distribution of the surface of the turbine blade and the wall surface of the combustion chamber at different temperatures is obtained through comsol simulation; and designing a multiple reflection radiation formula, accurately calculating the reflection radiation of the hot end part, and obtaining a corrected radiation temperature measurement result through inversion. According to the method, reflection radiation of a front-stage guide vane, adjacent movable vanes and a combustion chamber wall surface is considered, a surrounding channel integral method is adopted, surface element shielding between blade surfaces is considered, and a turbine blade surface radiation angle coefficient is calculated; the wall surface of the combustion chamber is increased, the reflection radiation of the hot end component is considered, a multiple reflection radiation formula is designed, the accurate reflection radiation of the hot end component is obtained, the reflection radiation calculation error is reduced, and the turbine blade radiation temperature measurement precision is improved.

This invention provides a depth prior-guided multi-view stereo 3D reconstruction method. It acquires N original images of the target to be reconstructed from different perspectives; it iterates through the N original images, using each image as a reference image and the remaining N-1 images as source images, inputting them into a trained depth map generation network to obtain N depth maps; and then fuses these N depth maps to obtain a 3D point cloud of the target to be reconstructed. Based on its network architecture, the subsequent depth map generation module uses the depth maps output by the preceding depth map generation module to generate a depth prior. This depth prior guides the Transformer in the subsequent depth map generation module to complete cost volume aggregation, generating the final depth map, thereby improving the final quality of the depth map and point cloud.

The present invention discloses a traffic flow prediction model considering dynamic spatial relations, belonging to the field of intelligent transportation. The present invention does not directly calculate the dynamic spatial relations of the traffic network, but designs a method for efficiently and simply calculating the dynamic spatial relations of the traffic network. The method converts real-time traffic flow information into the dynamic spatial relations of the traffic network through a spatiotemporal conversion module, an autoencoder, and a graph normalization module, which is recorded as a time-varying spatial graph, which truly reflects the time-varying characteristics of the dynamic spatial relations of the traffic network. Then, the spatiotemporal features of the traffic flow data are extracted and fused according to the time-varying spatial graph and the stable spatial graph based on the node positions of the traffic network to obtain the spatiotemporal fusion features of the traffic flow, which reflects both the stable spatiotemporal features of the traffic flow and the dynamic spatiotemporal features of the traffic flow, thereby being able to more accurately realize traffic flow prediction.

The invention discloses an application method of a fire extinguishing agent, which belongs to the technical field of fire extinguishing agent application, and comprises the following steps: preparing the fire extinguishing agent, and checking the fire extinguishing agent; according to the method, the fire extinguishing agent preset points of the area are established, the amount of the fire extinguishing agent at each preset point can be effectively measured and calculated, the amount comprises the current amount and the reserve amount, it can be effectively guaranteed that enough fire extinguishing agent can be used during fire extinguishing, the established preset points can cover the fire extinguishing area in an all-around mode, and the fire extinguishing effect is improved. The fire extinguishing agent realizes dead-corner-free control of an ignition point, can control the ignition of the lithium ion battery through multiple experiments in the early stage, has a good application effect, is specially provided with a specific simulation process and an evaluation process, can perform a fire extinguishing simulation experiment according to the obtained information and the arranged fire extinguishing agent, and has a good application prospect. And the fire extinguishing result is simulated, and subsequent improvement and upgrading are performed according to the result, so that the actual application effect of the fire extinguishing agent is further ensured.

The present invention discloses a method and device for measuring the lateral slack length of a tire, and the measuring method comprises the following steps: step S1, mounting a tire to be measured on a six-component force test bench, and making the tire run on the six-component force test bench with a preset load and a preset speed; step S2, when the tire runs stably, performing a sinusoidal input with a preset amplitude on the tire, and changing the frequency of the sinusoidal input for multiple times, and obtaining the yaw angle and lateral force when the tire is running; step S3, calculating the transfer function of the lateral force and the yaw angle; step S4, fitting the curve of the transfer function; step S5, calculating the lateral slack length of the tire according to the curve of the transfer function. According to the method for measuring the lateral slack length of the tire of the present invention, the lateral slack length of the tire under the influence of the tire turning can be effectively calculated, avoiding the problem of inaccurate calculation results due to the influence of the turning angle coupling in the traditional method, and improving the measurement accuracy of the lateral slack length of the tire.

The embodiment of the invention provides a generator stator single-phase earth fault positioning method and system based on data driving. The method comprises the following steps: constructing a generator stator single-phase earth fault electromagnetic transient model; transition resistance of a fault point of the generator stator single-phase earth fault electromagnetic transient model is set, and a training set and a test set of fault data are generated; an MLP neural network model is constructed, and the MLP neural network model comprises an MLP fault positioning model, an MLP transition resistance calculation model and an MLP fault branch discrimination model; training an MLP neural network model according to the training set of the fault data; and inputting new data during a fault into the MLP neural network model, and outputting a fault position, transition resistance and a fault branch. According to the method, the fault position can be accurately positioned, the transition resistance is effectively calculated, and the fault branch is reliably judged.

The application provides a bypass valve element modeling method based on Flowmaster, which comprises the following steps: determining a bypass valve structure, which comprises a control cavity spring and a working cavity spring; obtaining displacement amounts of a control cavity valve and a working cavity valve when the bypass valve is in a working state; solving the displacement amounts to obtain a bypass valve opening degree; establishing a linear equation of a bypass valve interface element according to a general linear equation of a multi-interface element; obtaining the opening degree and pressure of the bypass valve according to physical and geometric characteristic parameters of the bypass valve, the valve interface linear equation and inlet and outlet geometric parameters; obtaining the opening degree and pressure of the bypass valve according to a characteristic relation curve of the opening degree and the difference between a valve control end and an inlet pressure, or a characteristic relation surface of the opening degree and the control end pressure and the inlet and outlet pressure difference, the bypass valve interface linear equation and the inlet and outlet geometric parameters; and carrying out bypass valve element opening degree test and flow characteristic test to verify the correctness of the bypass valve element.

The invention belongs to the field of warehouse goods allocation identification, and particularly relates to a warehouse goods allocation detection method capable of identifying a position angle, which comprises the following steps of: 1) carrying out camera angle and scale calibration on a camera on a transfer robot so as to convert image pixel coordinates into real scale coordinates; 2) obtaining an offset value between the image coordinate system and a transfer robot coordinate system, namely obtaining a corresponding relation between the image coordinate system and the transfer robot coordinate system, and further obtaining a cargo position under the transfer robot coordinate system; 3) according to the cargo position under the coordinate system of the transfer robot, obtaining a boundary frame of the cargo through an image detection algorithm, and processing a sub-image in the boundary frame to obtain a cargo angle; according to the method, the flexibility of warehouse management can be greatly improved when goods allocation detection is carried out.

The present invention relates to the field of tire manufacturing technology, specifically a comparative measurement method for quantifying the length of the inner contour cord of a full steel radial tire, marking the front and rear sides of the bottom center steel wire of the cross-section wire ring with a connecting line, grinding off the part of the carcass and the wrapping cloth wrapped around the bottom of the steel ring until the steel wire of the steel ring is completely exposed, drawing the extension line of the steel wire in the middle of the steel ring to the carcass, and cutting it off after stripping; calculating the length of the inner contour cord, measuring the length of the steel wire that is stripped and cut off, and recording it as a; construction theoretical length, recorded as b; construction flat width, recorded as B; recording the number of steel wires with the widest spacing of the steel ring, recorded as N; the size specification of the steel wire diameter is recorded as c, and the size specification of the wrapping cloth thickness is recorded as d; the number of wrapping cloth layers is recorded as n; construction theoretical length b=B-2*(N / 2*c+n*d); there is a difference between the construction theoretical design and the actual finished product=a-b. The technical method of the present invention is equivalent to the adjustment of the construction flat width value, and the method is feasible and convenient, which greatly improves the efficiency of solving such problems.

The invention discloses a non-excavation repair cost measuring and calculating method for a grouting pipeline after wall combination of a stainless steel lining. The method comprises the following steps: 1, establishing a cost subject division standard; the method specifically comprises the following steps: importing a file in which the expense subject division standard is stored into an expense standard, and reading listed information in the file; 2, analyzing the historical project file; the method specifically comprises the following steps: importing historical project files of various existing projects, and analyzing the historical project files including GUID codes, compiling time, project cost names, project cost unit prices, quantities, amounts and cost structural composition of project feature information; 3, optimizing standard data; the method specifically comprises the following steps of: de-duplicating and combining all item feature information with similar or same feature values; 4, converting the standard data; and 5, further applying the converted standard data to obtain the total cost. According to actual engineering characteristics and process characteristics, comparative analysis and characteristic matching with historical projects are carried out, and economic and technical indexes are effectively measured and calculated.

The invention provides a method, a device and equipment for positioning a single-phase earth fault point of a current collector, and a medium. The method comprises the following steps: acquiring a topological structure of a target current collection line and transient traveling wave signals of each monitoring area after a single-phase earth fault occurs; determining the arrival time of the transient traveling wave signal of each monitoring point according to the transient traveling wave signal of each monitoring point; determining a fault distance between any two regions according to the arrival time of the transient traveling wave signal of each monitoring point, and determining the actual length of each region according to the topological structure; constructing a fault area discrimination matrix based on the fault distance of any two areas, the actual length of each area and the topological structure; and according to the fault area discrimination matrix, positioning a single-phase earth fault point of the target current collection line. According to the method, the fault area discrimination matrix can be constructed through the actual length of each area, and fault point positioning is performed based on the fault area discrimination matrix, so that the error of fault point positioning is reduced, and the fault positioning precision requirement of the wind power plant current collection line is met.

The present invention discloses a supervised cross-modal retrieval method, device, equipment and medium, the method includes: extracting features from training sample data of image modality and text modality; mapping the extracted image data features and text data features to a common representation space; respectively calculating the loss of the label space, the loss within each modality and between different modalities in the common representation space, and the invariance loss between the image and text modalities, and adding different weights to obtain the loss function of the retrieval model; optimizing the parameters of the retrieval model by minimizing the loss function; mapping the target retrieval data to the common representation space using the optimized retrieval model, calculating the similarity between the target retrieval data and the data in the image and text data set, and obtaining the corresponding retrieval ranking result. In this way, the discriminability of different semantic data samples and the semantic information of the original data are retained, the correlation between cross-modal data can be calculated more effectively, and the retrieval accuracy is higher.

The invention relates to the technical field of logistics delivery, in particular to an intelligent logistics delivery method and system based on an electronic map, and the method comprises the steps: obtaining a to-be-delivered customer set, obtaining a to-be-delivered position set of the to-be-delivered customer set based on the electronic map of a delivery area, carrying out the position clustering of the to-be-delivered position set, and obtaining a to-be-delivered customer set; obtaining a to-be-delivered point set of the to-be-delivered position set; and obtaining a logistics distribution center of the delivery area, and constructing a delivery communication network of the delivery area based on the electronic map of the delivery area, the to-be-delivered point set and the logistics distribution center. According to the method, the positions to be delivered are clustered, delivery resources can be better integrated, unnecessary repeated routes in the delivery process are reduced, the overall efficiency of logistics delivery is improved, the weighted delivery communication network and the solution algorithm are utilized, the optimal delivery path can be effectively calculated, the transportation cost is reduced, and the delivery time is shortened.

The invention relates to a sandstone type uranium ore enrichment zone determination method and device, and a storage medium. The method comprises the steps of obtaining the content of thorium elements in energy spectrum logging; calculating the uranium element content according to the thorium element content; calculating a sedimentary environment factor; calculating a clay mineral relative content factor; calculating a shale content factor; calculating an organic matter content factor curve; ore control comprehensive evaluation parameters are calculated; and the uranium abnormal enrichment section is judged. According to the sandstone type uranium ore enrichment zone determination method and device and the storage medium provided by the embodiment of the invention, the ore body enrichment depth can be effectively calculated, so that the exploration cost is reduced, the exploration success rate is improved, and the method and device can be applied to new area exploration and have relatively high popularization and application values.

The present invention discloses a method for predicting urban spatial density under a self-influence mechanism. First, effective spatial information is extracted from the current spatial distribution status of urban buildings, spatial units are divided, and buildings are abstracted as spatial capacity particles. Then, the small-world model in complex network theory is used to calculate the interaction between capacity particles to obtain the capacity change trend of the capacity particles. The spatial filtering model is then used to calculate the overall capacity change of the spatial unit, thereby obtaining a prediction of the change in urban spatial density. The present invention constructs a mathematical model based on relevant theories of architecture, urban morphology, and complex network research, and can effectively predict the spatial density of cities that have been completed and entered a renewal and development stage.

The present application relates to a circuit for performing neural network calculations on a neural network, the circuit comprising: a systolic array comprising a plurality of cells; a weight extraction unit, for each of a plurality of neural network layers, the weight extraction unit being configured to: for the neural network layer, send a plurality of weight inputs to cells along a first dimension of the systolic array; and a plurality of weight sequencer units, each weight sequencer unit being coupled to a different cell along the first dimension of the systolic array, for each of the plurality of neural network layers, the plurality of weight sequencer units being configured to: for the neural network layer, transfer a plurality of weight inputs to cells along a second dimension of the systolic array in a plurality of clock cycles, and wherein each cell is configured to calculate a product of an activation input and a corresponding weight input using a multiplication circuit.

The application relates to the technical field of oil and gas exploration, and discloses a strike-slip fracture identification method and device based on a variety of seismic attribute fusion technologies. The method first acquires seismic data and geological background data of a target area, then performs seismic profile interpretation and calculates coherence, curvature, tensor thickness, symmetry illumination attribute and maximum likelihood attribute based on the geological background and the seismic data, and finally determines the preferred attributes according to the scale of the fracture and performs quantitative fusion to identify the strike-slip fracture zone. The method is suitable for fractures of different scales, different attributes are selected for quantitative fusion, the advantages of different characteristics are complemented, the identification precision and sensitivity of the fracture are improved, and the research work of researchers in the aspects of the fracture and well position is effectively assisted.

The invention relates to the technical field of flow cell stack design, and provides a flow cell stack structure packaging design method, which comprises the following steps of: dividing a flow cell stack model into a stack core area and a stack sealing area positioned on the periphery of the stack core area, and respectively calculating equivalent stiffness k stack core area and k stack sealing area by adopting # imgabs0 #; the overall equivalent stiffness k galvanic pile of the flow cell pile model is equal to k galvanic pile core area + k galvanic pile sealing area, the clamping force F galvanic pile core area borne by the galvanic pile core area and the clamping force F galvanic pile sealing area borne by the galvanic pile sealing area are calculated, and the clamping force or / and the parameters of the flow cell pile model are adjusted according to the sealing performance or / and the electrochemical performance of the flow cell pile. According to the flow cell stack structure packaging design method provided by the invention, the overall equivalent stiffness of the flow cell stack is regarded as the combination of the equivalent stiffness of each component, and the flow cell stack under the acting force of the clamping force is simplified into an equivalent stiffness mechanical model, so that the design efficiency can be improved.

The application discloses a hydraulic fracture geometric parameter nonlinear inversion method, which comprises the following steps: a strain analytical model is established according to preset mechanical parameters; a fracture parameter nonlinear inversion model is established according to the established strain analytical model and axial strain observation data obtained by a low-frequency DAS system under a neighboring well monitoring scene; a global optimal inversion is solved by using a nonlinear least square iteration algorithm according to preset constraint conditions and the fracture parameter nonlinear inversion model, and a fracture height, a fracture height asymmetric offset and a net pressure are obtained by inversion; and a maximum fracture width is calculated according to the obtained fracture height and net pressure.

The application relates to the field of robot control and discloses a live working robot load gravity compensation method, equipment and a storage medium, the method comprises the following steps: constructing a gravity compensation model according to the position relationship of each joint of a robot system; designing an installation angle optimization model with error sum of squares as a constraint index based on the installation angle of a force sensor; collecting multiple sets of mechanical arm pose and force data under experimental conditions; solving the installation angle optimization model by using the multiple sets of mechanical arm pose and force data under the experimental conditions based on an improved PSO method to obtain the optimal installation angle of the force sensor; installing the force sensor through the optimal installation angle, and performing gravity compensation on the mechanical arm end through the gravity compensation model under actual working conditions to obtain accurate external environment contact force. The application has the beneficial effects that all gravity compensation parameters can be calculated at high speed and effectively, and higher-precision and higher-efficiency gravity compensation is realized.