9 results about "Identifiability" patented technology

This invention belongs to the field of power load forecasting and data privacy protection technology, and discloses a joint power load privacy forecasting method applied to a system composed of a server and multiple local entities. The server generates group prototype statistical information, local entities calculate behavioral identifiability, profile uniqueness, and reconfigurability risks, determine segmented protection strategies, and execute peak-shifting silent replacement, feature masking, statistical pruning, differential privacy perturbation, and encrypted upload. The server securely aggregates sufficient statistics and solves for global forecasting model parameters, while local entities fuse the global model and their local private models to output load forecasting results. Privacy-preserving joint modeling based on multi-entity load data achieves load forecasting while reducing the risk of load data leakage.

The invention belongs to the technical field of bridge structure health monitoring, and discloses a suspension bridge structure system identification method based on observability and state estimation and a high-efficiency algorithm thereof. According to the characteristics of different components of the suspension bridge, an observability state equation of the suspension bridge is established, the identifiability problem of unit parameters is converted into the solvability problem of a mathematical equation, a Jacobian matrix is constructed in combination with a state estimation high-efficiency algorithm (OM-SE), explicit mapping between measurement data and state variables is achieved, and the measurement accuracy of the suspension bridge is improved. And the parameter identification precision and the calculation efficiency are improved. The overall rigidity matrix is established through the parabola main cable unit, the Euler-Bernoulli beam unit and the direct rigidity method, parameter inversion is performed by adopting a double-layer iteration process, the method is suitable for identifying the bending rigidity of the main beam and the axial rigidity of the cable of the large-span suspension bridge, and the method is verified through a reduced scale model test and an actual suspension bridge numerical value case, and the method is suitable for identifying the bending rigidity of the main beam and the axial rigidity of the cable of the large-span suspension bridge. The identification error is controlled within 5%, and the calculation efficiency is improved by more than 70%.

The invention discloses a foundation surface defect identification and evaluation method, system, equipment and medium based on monitoring data, and belongs to the technical field of hydraulic structure nondestructive testing and operation period health diagnose.The method comprises the steps that an interface model of a foundation surface is constructed, the foundation surface is modeled into a two-dimensional discontinuous interface, and an interface constitutive equation and a seepage model are explicitly introduced; collecting operation monitoring data, performing feature extraction on the operation monitoring data, and establishing bidirectional mapping from defect parameters to monitoring responses; and identifying foundation surface defect parameters through reverse learning, substituting an identification result back to the forward model, and outputting engineering evaluation indexes of foundation surface levels. According to the method, the identifiability of the defect position, the scale and the interface force-water property is remarkably improved, distortion caused by indirect inference only based on overall measurement is avoided, the robustness is kept under noise disturbance and working condition fluctuation, and the result has uncertainty definition and engineering interpretability.

The application provides a kind of RTCP-based numerical control machine tool rotary table geometric error identification method, it is related to numerical control machine tool rotary table geometric error analysis technical field.According to the structural characteristics of machine tool, based on the principle of multi-body kinematics and homogeneous transformation matrix method, the kinematics modeling of machine tool is carried out.The geometric error of prismatic shaft and the geometric error of rotary shaft can be uniformly described by the spatial pose deviation of tool center point, and the comprehensive identification is carried out under the same measurement constraint condition.By introducing the measurement relationship with fixed length constraint, the spatial displacement deviation of tool center point in multi-axis linkage process is converted into the measurable length variation, thereby a unified error measurement model is established.In the measurement process, by reasonably designing the multi-axis linkage trajectory and the measurement posture, the response characteristics of different error parameters in the measurement data are effectively distinguished, thereby the identifiability of each error parameter is improved.

A method for improving a system model includes selecting a continuous-time model that includes numerically constant model parameters and time-varying variables; selecting an objective function; selecting from the model parameters a set of model parameters whose numerically constant values are to be estimated; selecting a data acquisition system to supply data for model operation and for estimation of the numerically constant values of the set of model parameters. The data acquisition system includes sensors operable to collect data associated with at least one model parameter of the set. The method further includes assigning temporary values as the numerically constant values of the set of model parameters; generating, via a processor, synthetic data from the model using the temporary values; determining, via the processor, an unidentifiable model parameter of the set presenting an identifiability issue; and displaying, via a user interface in communication with the processor, the unidentifiable model parameter.

This enables data analysis using external language models while reducing the risk of data leakage. [Solution] The sensitive data analysis support device uses its own first language model to extract data from validation data that shows cases related to the clinical question, generates pseudo-data from the extracted data that is statistically similar to the extracted data and has reduced personal identifiability, generates a pseudo-analysis specification that is formatted for the pseudo-data based on the clinical question and validation data, inputs the pseudo-analysis specification and pseudo-data into an external second language model to obtain a draft analysis code for the pseudo-data based on the pseudo-analysis specification, and formats the draft source code for analysis of the extracted data based on the analysis specification.

The invention provides an equivalent load model parameter identification method and device, equipment, a medium and a product, and the method comprises the steps: constructing an equivalent load model considering frequency characteristics, and selecting noise-like identification data; the method comprises the following steps: identifying noise-like identification data of an equivalent load model, classifying each to-be-identified parameter according to the identification sensitivity, and carrying out pre-identification-parameter regularization identification-variable torque load inertia identification step-by-step identification on the equivalent load model parameter according to the noise-like identification data and the identifiability of each to-be-identified parameter to obtain an identification result of the equivalent load model parameter. The equivalent load model structure considering the frequency characteristics can effectively describe the voltage / frequency comprehensive dynamic characteristics of the electric equipment; through pre-identification, parameter regularization identification and variable torque load inertia identification step-by-step decoupling identification, stable parameter identification under multi-input data identification is realized, various mechanical characteristics of a real motor load are adapted, a stable and reliable identification result is obtained, and identifiability of multiple parameters of an equivalent load model under noise-like identification data is improved.

A method for improving a system model includes selecting a continuous-time model that includes numerically constant model parameters and time-varying variables; selecting an objective function; selecting from the model parameters a set of model parameters whose numerically constant values are to be estimated; selecting a data acquisition system to supply data for model operation and for estimation of the numerically constant values of the set of model parameters. The data acquisition system includes sensors operable to collect data associated with at least one model parameter of the set. The method further includes assigning temporary values as the numerically constant values of the set of model parameters; generating, via a processor, synthetic data from the model using the temporary values; determining, via the processor, an unidentifiable model parameter of the set presenting an identifiability issue; and displaying, via a user interface in communication with the processor, the unidentifiable model parameter.

The invention belongs to the field of geophysical inversion, and particularly discloses a shear wave velocity two-dimensional profile probability inversion method based on multi-source data fusion, and the method comprises the steps: introducing transverse constraint and drilling information, and achieving the joint probability inversion of a multi-measurement-point shear wave velocity profile; under a Bayesian framework, uniformly regarding multi-measurement-point frequency dispersion curve data, transverse constraint and drilling data as observation data items, establishing posterior probability distribution of multi-measurement-point shear wave velocity profile model parameters under three types of data constraints, and introducing various types of observation information step by step by adopting a sequential Bayesian updating strategy; and finally solving the Bayesian equation of each stage to obtain the posterior distribution of the shear wave velocity profile of the multiple measuring points. The method of the invention can effectively quantify the identification uncertainty of the shear wave velocity profile of multiple measuring points, enhance the continuity and physical consistency of adjacent profiles in the transverse direction, and improve the identifiability of the shear wave velocity profile.