Based on the structural characteristics of the double-differenced normal equation, a new method was proposed to resolve the ambiguity float solution through a selection of parameter weights to construct an appropriate...Based on the structural characteristics of the double-differenced normal equation, a new method was proposed to resolve the ambiguity float solution through a selection of parameter weights to construct an appropriate regularized matrix, and a singular decomposition method was used to generate regularization parameters. Numerical test results suggest that the regularized ambiguity float solution is more stable and reliable than the least-squares float solution. The mean square error matrix of the new method possesses a lower correlation than the variancecovariance matrix of the least-squares estimation. The size of the ambiguity search space is reduced and the search efficiency is improved. The success rate of the integer ambiguity searching process is improved significantly when the ambiguity resolution by using constraint equation method is used to determine the correct ambiguity integervector. The ambiguity resolution by using constraint equation method requires an initial input of the ambiguity float solution candidates which are obtained from the LAMBDA method in the new method. In addition, the observation time required to fix reliable integer ambiguities can he significantly reduced.展开更多
This work applies concepts of artificial neural networks to identify the parameters of a mathematical model based on phase fields for damage and fracture.Damage mechanics is the part of the continuum mechanics that mo...This work applies concepts of artificial neural networks to identify the parameters of a mathematical model based on phase fields for damage and fracture.Damage mechanics is the part of the continuum mechanics that models the effects of micro-defect formation using state variables at the macroscopic level.The equations that define the model are derived from fundamental laws of physics and provide important relationships among state variables.Simulations using the model considered in this work produce good qualitative and quantitative results,but many parameters must be adjusted to reproduce certain material behavior.The identification of model parameters is considered by solving an inverse problem that uses pseudo-experimental data to find the best values that fit the data.We apply physics informed neural network and combine some classical estimation methods to identify the material parameters that appear in the damage equation of the model.Our strategy consists of a neural network that acts as an approximating function of the damage evolution with output regularized using the residue of the differential equation.Three stages of optimization seek the best possible values for the neural network and the material parameters.The training alternates between the fitting of only the pseudo-experimental data or the total loss that includes the regularizing terms.We test the robustness of the method to noisy data and its generalization capabilities using a simple physical case for the damage model.This procedure deals better with noisy data in comparison with a more standard PDE-constrained optimization method,and it also provides good approximations of the material parameters and the evolution of damage.展开更多
The intensity non-stationarity is one of the most important features of earthquake records.Modeling of this feature is significant to the generation of artificial earthquake waves.Based on the theory of phase differen...The intensity non-stationarity is one of the most important features of earthquake records.Modeling of this feature is significant to the generation of artificial earthquake waves.Based on the theory of phase difference spectrum,an intensity non-stationary envelope function with log-normal form is proposed.Through a tremendous amount of earthquake records downloaded on Kik-net,a parameter fitting procedure using the genetic algorithm is conducted to obtain the value of model parameters under different magnitudes,epicenter distances and site conditions.A numerical example is presented to describe the procedure of generating fully non-stationary ground motions via spectral representation,and the mean EPSD(evolutionary power spectral density)of simulated waves is proved to agree well with the target EPSD.The results show that the proposed model is capable of describing the intensity non-stationary features of ground motions,and it can be used in structural anti-seismic analysis and ground motion simulation.展开更多
Through the analysis on the migratory diffusion process of atmospheric pollutants,we proposed to seek atmospheric pollutant source with surface soil sample of data.Based on Gaussian plume model and deposition model,at...Through the analysis on the migratory diffusion process of atmospheric pollutants,we proposed to seek atmospheric pollutant source with surface soil sample of data.Based on Gaussian plume model and deposition model,atmospheric pollutants distribution model was deduced,with which a schema matching source seeking model was established.The model was used to seek the pollutant source by using the arsenic data in the surface soil sample of a city.展开更多
文摘Based on the structural characteristics of the double-differenced normal equation, a new method was proposed to resolve the ambiguity float solution through a selection of parameter weights to construct an appropriate regularized matrix, and a singular decomposition method was used to generate regularization parameters. Numerical test results suggest that the regularized ambiguity float solution is more stable and reliable than the least-squares float solution. The mean square error matrix of the new method possesses a lower correlation than the variancecovariance matrix of the least-squares estimation. The size of the ambiguity search space is reduced and the search efficiency is improved. The success rate of the integer ambiguity searching process is improved significantly when the ambiguity resolution by using constraint equation method is used to determine the correct ambiguity integervector. The ambiguity resolution by using constraint equation method requires an initial input of the ambiguity float solution candidates which are obtained from the LAMBDA method in the new method. In addition, the observation time required to fix reliable integer ambiguities can he significantly reduced.
基金support of the National Council for Scientific and Technological Development(CNPq),grant numbers 164733/2017-5 and 310351/2019-7the University of Campinas(UNICAMP)。
文摘This work applies concepts of artificial neural networks to identify the parameters of a mathematical model based on phase fields for damage and fracture.Damage mechanics is the part of the continuum mechanics that models the effects of micro-defect formation using state variables at the macroscopic level.The equations that define the model are derived from fundamental laws of physics and provide important relationships among state variables.Simulations using the model considered in this work produce good qualitative and quantitative results,but many parameters must be adjusted to reproduce certain material behavior.The identification of model parameters is considered by solving an inverse problem that uses pseudo-experimental data to find the best values that fit the data.We apply physics informed neural network and combine some classical estimation methods to identify the material parameters that appear in the damage equation of the model.Our strategy consists of a neural network that acts as an approximating function of the damage evolution with output regularized using the residue of the differential equation.Three stages of optimization seek the best possible values for the neural network and the material parameters.The training alternates between the fitting of only the pseudo-experimental data or the total loss that includes the regularizing terms.We test the robustness of the method to noisy data and its generalization capabilities using a simple physical case for the damage model.This procedure deals better with noisy data in comparison with a more standard PDE-constrained optimization method,and it also provides good approximations of the material parameters and the evolution of damage.
基金supported by the National Key R&D Program of China(2017YFC0703600)the National Foundation of China(Grant Nos.51678301 and 51678302).
文摘The intensity non-stationarity is one of the most important features of earthquake records.Modeling of this feature is significant to the generation of artificial earthquake waves.Based on the theory of phase difference spectrum,an intensity non-stationary envelope function with log-normal form is proposed.Through a tremendous amount of earthquake records downloaded on Kik-net,a parameter fitting procedure using the genetic algorithm is conducted to obtain the value of model parameters under different magnitudes,epicenter distances and site conditions.A numerical example is presented to describe the procedure of generating fully non-stationary ground motions via spectral representation,and the mean EPSD(evolutionary power spectral density)of simulated waves is proved to agree well with the target EPSD.The results show that the proposed model is capable of describing the intensity non-stationary features of ground motions,and it can be used in structural anti-seismic analysis and ground motion simulation.
文摘Through the analysis on the migratory diffusion process of atmospheric pollutants,we proposed to seek atmospheric pollutant source with surface soil sample of data.Based on Gaussian plume model and deposition model,atmospheric pollutants distribution model was deduced,with which a schema matching source seeking model was established.The model was used to seek the pollutant source by using the arsenic data in the surface soil sample of a city.
