17 results about "Component Load" patented technology

A combined rock mass strength prediction and floor strength mapping method based on a PCA-KAN fusion neural network comprises the following steps: preparing combined rock mass test pieces with different layer sequences, thickness ratios and interface states, and obtaining multi-dimensional mechanical parameters through a compression mechanical test; carrying out dimension reduction processing by adopting a principal component analysis method, constructing a comprehensive strength index according to a principal component load coefficient, and forming a dimension reduction feature vector; constructing an Attn-KAN model, and adjusting hyper-parameters by adopting a Bayesian optimization algorithm; the performance and interpretability of the model are ensured through layered K-fold cross validation and SHAP analysis; the method comprises the following steps: carrying out a compression mechanical test on a field rock core sample to obtain multi-dimensional mechanical parameters, carrying out dimension reduction processing by adopting a principal component analysis method, and predicting by taking a principal component feature vector as input data to obtain a comprehensive strength predicted value; and adopting a Kriging interpolation method to output an intensity distribution and risk grade partition map. According to the method, accurate prediction of the strength of the combined rock mass can be realized, and spatial visual mapping of the strength of the coal seam floor and the risk level can be realized.

The application provides a sensitive component life detection method and system, and relates to the field of sensitive component life detection.The method comprises the following steps: recording the factory UTC time of a sensitive component and initial performance data at the factory; based on a preset timing monitoring period, acquiring the current UTC time and average performance data of the sensitive component in the timing monitoring period at the end of each timing monitoring period; adjusting the weight coefficient corresponding to different timing monitoring periods according to environmental data and sensitive component load parameters; performing weighted summation on the average performance data according to the weight coefficient to obtain a total integral value; when it is detected that the total integral value is greater than or equal to a preset integral threshold value or the variable value of the average performance data exceeds a preset amplitude threshold value, triggering alarm information that the life has reached the limit. By implementing the method, the performance change of the sensitive component under different environmental and load conditions can be dynamically responded, and the life end of the sensitive component can be more accurately predicted.

The invention belongs to the technical field of electric power safe operation, and particularly relates to a medium and long term load prediction method based on data decomposition and multi-model fusion, which comprises the following steps: processing missing values and abnormal values of historical load data of a power grid to obtain a historical load sequence of the power grid; decomposing the processed historical load sequence of the power grid into a trend component, a season component and a residual component according to a data decomposition means, and allocating weights to the three components; based on the generated three weighted components, constructing an independent CNN-LSTM prediction model for each component to obtain a load prediction sequence of each component; and performing linear addition based on the obtained component prediction results to obtain a final load prediction value. The method not only considers the overall characteristics of the load sequence, but also establishes a special prediction model for the time sequence characteristics of different components, and significantly improves the precision and reliability of load prediction.

The invention provides a customer scene whole vehicle wheel force load spectrum acquisition method, which comprises the steps of installing a wheel six-component device on a target vehicle, installing a strain sensor on the surface of a chassis part of the vehicle, and simultaneously acquiring wheel force load spectrum data and chassis part load spectrum data of the vehicle on a test field road; determining associated chassis parts equivalent to the wheel force load frequency in frequency response in the Fx, Fy and Fz directions through frequency response analysis; establishing a fitting equation of the wheel force load and the associated chassis part load; performing backstepping on test field wheel force load data by using the associated chassis part load data and the fitting equation, and calculating an equivalent value Z; and if the equivalent value Z is in the calibration interval, applying the verified fitting equation to the customer scene chassis part load data, and carrying out backstepping to obtain a customer scene wheel force load spectrum.

The application provides a 3D model loading method, device, equipment and medium, and relates to the technical field of image data processing. The method comprises: performing spatial grid division on a 3D model to obtain a plurality of components; determining multi-dimensional data of the components, wherein the multi-dimensional data at least comprises semantic units; obtaining a current component loading queue of the 3D model and initial weight values of to-be-loaded components in the component loading queue; determining a scene to which the 3D model belongs, calculating fine-tuning weight values of the to-be-loaded components in the scene based on the semantic units of the to-be-loaded components, wherein the fine-tuning weight values are used to measure the importance of the to-be-loaded components in the scene; calculating target weight values according to the initial weight values and the fine-tuning weight values; sorting the to-be-loaded components in the component loading queue according to the target weight values, and loading the to-be-loaded components according to the sorting result. The application can improve the loading speed of the 3D model.

The invention belongs to the technical field of overall design of aircrafts, and particularly relates to a low-resistance modeling design method of an unmanned helicopter. Comprising the following steps: 1, selecting a control point set of longitudinal and transverse contours of an unmanned helicopter body, respectively establishing longitudinal and transverse contour parameterized control curves passing through the control points, and determining initial values of tension proportion parameters of the control points on the longitudinal and transverse curves; 2, constructing a corresponding parameterized control curve as an internal load space control optimization model, and solving the internal load space control optimization model by adopting a genetic algorithm on the basis of an optimization target with the minimum windward area and by taking the equipment and part loading space, the mounting gap and the process manufacturing as constraint conditions, so as to obtain a control curve of a typical section of a fuselage; and 3, importing the fuselage three-dimensional model into a fluid dynamic tool to calculate aerodynamic resistance, obtaining a fuselage sample aerodynamic data set, constructing an agent model for optimization, and obtaining an optimal unmanned helicopter model.

The application belongs to the technical field of ground test of aviation aircraft, and relates to a multi-component load working condition synchronous implementation test design method and device, which comprises the following steps: S1, determining whether the component load working condition can be trimmed; S2, when the component load working condition cannot be trimmed, determining whether there is another component load working condition capable of stress decoupling; S3, when there are two component load working conditions capable of stress decoupling, determining the principal direction and magnitude of each component load working condition; S4, when the principal directions are opposite, determining the load difference of the synchronous implementation of the two component load working conditions according to the magnitude; and S5, when the load difference is less than a preset load threshold, determining that the two component load working conditions can be synchronously implemented, forming a combined working condition, and trimming the load of other components of the aircraft according to the load difference of the combined working condition. The application can significantly reduce the magnitude of the trimmed load and reduce the difficulty of test load design.

The invention provides a 3D model loading method and device, equipment and a medium, and relates to the technical field of image data processing. The method comprises the following steps: carrying out space grid division on a 3D model to obtain a plurality of components; determining respective multi-dimensional data of the components, the multi-dimensional data at least comprising semantic units; obtaining a current component loading queue of the 3D model and respective initial weight values of to-be-loaded components in the component loading queue; a scene to which the 3D model belongs is determined, a fine tuning weight value of the to-be-loaded component in the scene is calculated based on the semantic unit of the to-be-loaded component, and the fine tuning weight value is used for measuring the importance degree of the to-be-loaded component in the scene; calculating a target weight value according to the initial weight value and the fine adjustment weight value; and sorting the to-be-loaded components in the component loading queue according to the target weight value, and loading the to-be-loaded components according to a sorting result. According to the invention, the loading speed of the 3D model can be improved.

The invention relates to the technical field of component loading optimization, and discloses a dynamic component loading optimization method, which comprises the following steps: collecting and identifying current real-time operation data to obtain a corresponding prediction demand category and an actual operation state; setting a plurality of component loading nodes based on the predicted demand category, generating a first component loading strategy according to all the component loading nodes, performing analogue simulation according to the first component loading strategy, and calculating a simulation demand coefficient and a simulation state coefficient according to a simulation result; calculating a simulation application coefficient of the first component loading strategy according to the simulation demand coefficient and the simulation state coefficient, and judging whether the first component loading strategy is optimized or not, if not, issuing a loading instruction according to the first component loading strategy, and if yes, generating a second component loading strategy, the component loading strategy is dynamically adjusted according to the requirement during operation, the user requirement is met, stable operation is achieved, and meanwhile the system response speed is increased.

The present invention provides an evaluation method for spectral standardization, including step 1, calculating a score matrix: from the host spectrum X m Decompose the principal component load matrix P m And the principal component score matrix T m , and then through the slave spectrum X′ t and P m , calculate the principal component score matrix T′ of the slave machine t ; Step 2, calculate the principal component score error rate: through T m and T′ t The principal component score error rate (PCSER) is calculated; the quality of spectral standardization is ultimately judged by the size of the PCSER value; the smaller the PCSER value, the better the spectral standardization. The spectral standardization evaluation method of the present invention can evaluate the differences between spectra based on the correction model established by the partial least squares method, improve the similarity of spectra, and thus realize model sharing between different instruments. In addition, compared with traditional evaluation methods, the evaluation method of the present invention does not require the complete execution of a complete set of model prediction work, thus saving a lot of time and cost.

The invention discloses a power transmission and transformation project cost prediction method and system based on data mining, and relates to the technical field of project cost prediction, and the method comprises the steps: collecting multi-dimensional data covering project attributes, environmental factors and historical cost, and building a project database; performing missing value filling and preprocessing on the data, and calculating a correlation coefficient matrix; determining the number of principal components based on principal component analysis, calculating a principal component load matrix, and extracting a candidate key field set; and carrying out partial correlation analysis, cross validation and regression significance test on the candidate fields and the project cost data, identifying key factors which have significant influence on the cost, constructing a project cost prediction model based on the key factors, and carrying out scientific prediction on the cost of a new project. According to the method, the problems of high dependence and low data utilization rate of traditional experience estimation are solved, systematic identification and cost prediction of key factors of the cost are realized, and support is provided for investment decision and cost control of power grid construction.

The application discloses a random multi-parameter load spectrum compiling method based on principal component analysis, and utilizes the principal component analysis method to convert multi-parameter random load into several independent load histories, i.e., multi-parameter principal component load histories, by taking the multi-parameter measured load spectrum of a component as basic compiling data, and then performs peak-valley value extraction processing to obtain the peak-valley value sequence and non-peak-valley value point set of the principal component load, performs rain flow cycle counting to obtain the rain flow cycle matrix of the principal component load history, reconstructs the load history from the rain flow cycle matrix, inserts the non-peak-valley value points randomly to obtain the principal component load reconstructed history, and linearly solves the principal component load history to obtain a new multi-parameter random load spectrum. The application fully considers the multi-parameter load correlation and multi-axis damage information in the random multi-parameter load spectrum compiling process, and provides a basis for fatigue damage analysis of practical complex mechanical components and multi-axis fatigue test.

The invention provides a sine load equivalence method of a multi-degree-of-freedom load system. The method comprises the following steps: calculating a principal stress and a principal stress angle according to a strain signal of a target part of a chassis; the multi-component load signals serve as a first group of load signals, and the multi-component load signals are recombined to obtain an in-phase recombined load component and a reverse-phase recombined load component to serve as a second group of load signals; identifying a frequency response function corresponding to each recombined load component in the second group of load signals according to the principal stress, and calculating a time domain response caused by the frequency response function on the chassis target part and a contribution degree to the principal stress; selecting the phase of the recombination load component with the maximum contribution degree as the phase of the equivalent sine load; and if the principal stress angle of the target part of the chassis and the principal stress angle under the action of the multi-component load signal meet a preset similar condition under the action of the equivalent sine load, setting a durability test target. According to the method, the problem that coupling among six components of an original complex road spectrum is difficult to effectively decompose in the prior art is solved.

The application provides a kind of retired component load adaptive mechanical arm multi-target trajectory control method.The application includes: real-time acquisition of mechanical arm multi-class operation data, identification of load mass and centroid and determination of load mutation;According to the change of load, update the kinematic constraint boundary and multi-objective optimization weight, re-plan the smooth transition trajectory;Construct a dynamic model and estimate the lumped disturbance in real time through a nonlinear disturbance observer;Fusion of reference trajectory, model parameters and disturbance estimate, output torque command using compound control strategy, realize mechanical arm closed-loop disassembly operation.The application can effectively suppress load mutation disturbance, improve the stability and control accuracy of mechanical arm operation.

This application relates to an intelligent fault diagnosis and management system and method for rail transit control consoles. The method includes: constructing a fault prediction probability model and an adaptive dynamic diagnostic threshold adjustment model; constructing a causal intensity quantification model based on fault features corresponding to the fault prediction probability; inputting potential root causes, fault features, and root cause contribution; and outputting a causal intensity score to classify fault root causes and obtain fault level coefficients; and constructing a dynamic parameter adaptive adjustment model based on the fault level coefficients, outputting component dynamic load parameters for dynamic adaptation of component loads. This invention supports zero-sample / small-sample fault detection, solving the problem of scarce fault samples. It adopts a fusion framework of unsupervised contrastive learning and cross-domain transfer learning, achieving anomaly identification without relying on a large number of real fault samples. By combining digital twins to generate high-fidelity virtual fault samples and optimizing their weights, it fundamentally solves the industry problem of few fault samples and difficulty in modeling in rail transit.

The invention provides a kiloton six-component load sensor calibration device which comprises a supporting assembly, the supporting assembly comprises a bottom plate, first supporting frames, second supporting frames and a supporting plate, and the first supporting frames are fixedly installed at the two ends of the top of the bottom plate respectively. According to the kiloton-level six-component load sensor calibration device provided by the invention, through mutual cooperation of the supporting assembly, the loading assembly, the loading tool, the guiding assembly, the moving assembly and other structures, a worker only needs to place a kiloton-level six-component load sensor needing to be calibrated on the top of the moving platform; the kiloton-level six-component load sensor is loaded on the loading tool, then the moving platform is pushed until the kiloton-level six-component load sensor is moved to the bottom of the loading tool, then the loading tool is pushed through the loading assembly, the kiloton-level six-component load sensor can be calibrated, other transfer equipment is not needed for cooperative operation, the operation process is simplified, and the working efficiency is improved.