Patents by Inventor Jinmiao Zhang
Jinmiao Zhang has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20220043174Abstract: A model-driven deep learning-based seismic super-resolution inversion method includes the following steps: 1) mapping each iteration of a model-driven alternating direction method of multipliers (ADMM) into each layer of a deep network, and learning proximal operators by using a data-driven method to complete the construction of a deep network ADMM-SRINet; 2) obtaining label data used to train the deep network ADMM-SRINet; 3) training the deep network ADMM-SRINet by using the obtained label data; and 4) inverting test data by using the deep network ADMM-SRINet trained at step 3). The method combines the advantages of a model-driven optimization method and a data-driven deep learning method, and therefore the network has the interpretability; and meanwhile, due to the addition of physical knowledge, the iterative deep learning method lowers requirements for a training set, and therefore an inversion result is more reliable.Type: ApplicationFiled: July 8, 2021Publication date: February 10, 2022Applicant: XI'AN JIAOTONG UNIVERSITYInventors: Jinghuai GAO, Hongling CHEN, Zhaoqi GAO, Chuang LI, Lijun MI, Jinmiao ZHANG, Qingzhen WANG
-
Patent number: 11226423Abstract: A model-driven deep learning-based seismic super-resolution inversion method includes the following steps: 1) mapping each iteration of a model-driven alternating direction method of multipliers (ADMM) into each layer of a deep network, and learning proximal operators by using a data-driven method to complete the construction of a deep network ADMM-SRINet; 2) obtaining label data used to train the deep network ADMM-SRINet; 3) training the deep network ADMM-SRINet by using the obtained label data; and 4) inverting test data by using the deep network ADMM-SRINet trained at step 3). The method combines the advantages of a model-driven optimization method and a data-driven deep learning method, and therefore the network has the interpretability; and meanwhile, due to the addition of physical knowledge, the iterative deep learning method lowers requirements for a training set, and therefore an inversion result is more reliable.Type: GrantFiled: July 8, 2021Date of Patent: January 18, 2022Assignee: XI'AN JIAOTONG UNIVERSITYInventors: Jinghuai Gao, Hongling Chen, Zhaoqi Gao, Chuang Li, Lijun Mi, Jinmiao Zhang, Qingzhen Wang
-
Patent number: 7516673Abstract: Structural stress in a fatigue-prone region of a structure is determined and analyzed by using: i) the nodal forces and displacement values in the fatigue-prone region, or ii) equilibrium equivalent simple stress states consistent with elementary structural mechanics in the fatigue-prone region. Of course, it is contemplated that combinations, equivalents, or variations of the recited bases may alternatively be employed.Type: GrantFiled: June 15, 2006Date of Patent: April 14, 2009Assignee: Battelle Memorial InstituteInventors: Pingsha Dong, Jinmiao Zhang, Jeong Kyun Hong
-
Publication number: 20060219026Abstract: Structural stress in a fatigue-prone region of a structure is determined and analyzed by using: i) the nodal forces and displacement values in the fatigue-prone region, or ii) equilibrium equivalent simple stress states consistent with elementary structural mechanics in the fatigue-prone region. Of course, it is contemplated that combinations, equivalents, or variations of the recited bases may alternatively be employed.Type: ApplicationFiled: June 15, 2006Publication date: October 5, 2006Applicant: BATTELLE MEMORIAL INSTITUTEInventors: Pingsha Dong, Jinmiao Zhang, Jeong Hong
-
Patent number: 7089124Abstract: Structural stress in a fatigue-prone region of a structure is determined by using the nodal forces and displacement values in the fatigue-prone region, or equilibrium-equivalent simple stress states consistent with elementary structural mechanics in the fatigue-prone region. The determination is substantially independent of mesh size and is particularly well-suited for applications where S-N curves are used in weld fatigue design and evaluation, where S represents nominal stress or stress range and N represents the number of cycles to failure.Type: GrantFiled: November 16, 2004Date of Patent: August 8, 2006Assignee: Battelle Memorial InstituteInventors: Pingsha Dong, Jinmiao Zhang, Jeong Kyun Hong
-
Patent number: 6901809Abstract: This need is met by the present invention wherein structural stress in a fatigue-prone region of a structure is determined by using the nodal forces and displacement values in the fatigue-prone region, or equilibrium-equivalent simple stress states consistent with elementary structural mechanics in the fatigue-prone region. The determination is substantially independent of mesh size and is particularly well-suited for applications where S-N curves are used in weld fatigue design and evaluation, where S represents nominal stress or stress range and N represents the number of cycles to failure. The present invention is directed to structural stress analysis through various combinations of modeling, calculation, and direct measurement schemes.Type: GrantFiled: November 16, 2001Date of Patent: June 7, 2005Assignee: Battelle Memorial InstituteInventors: Pingsha Dong, Jinmiao Zhang, Jeong Kyun Hong
-
Publication number: 20050071091Abstract: This need is met by the present invention wherein structural stress in a fatigue-prone region of a structure is determined by using the nodal forces and displacement values in the fatigue-prone region, or equilibrium-equivalent simple stress states consistent with elementary structural mechanics in the fatigue-prone region. The determination is substantially independent of mesh size and is particularly well-suited for applications where S-N curves are used in weld fatigue design and evaluation, where S represents nominal stress or stress range and N represents the number of cycles to failure. The present invention is directed to structural stress analysis through various combinations of modeling, calculation, and direct measurement schemes.Type: ApplicationFiled: November 16, 2004Publication date: March 31, 2005Inventors: Pingsha Dong, Jinmiao Zhang, Jeong Hong
-
Patent number: 6768974Abstract: A method for determining a model for a welding simulation, and the associated model. The method includes the steps of determining a history annihilation model of a material being welded, determining a strain hardening model of the material being welded, determining a three-dimensional virtual elements detection model of the material being welded, and incorporating the above models into a constitutive model for the welding simulation.Type: GrantFiled: October 10, 2000Date of Patent: July 27, 2004Assignee: Caterpillar IncInventors: Ashok Nanjundan, Pingsha Dong, Jinmiao Zhang, Frederick W. Brust, Yi Dong
-
Publication number: 20020112548Abstract: This need is met by the present invention wherein structural stress in a fatigue-prone region of a structure is determined by using the nodal forces and displacement values in the fatigue-prone region, or equilibrium-equivalent simple stress states consistent with elementary structural mechanics in the fatigue-prone region. The determination is substantially independent of mesh size and is particularly well-suited for applications where S-N curves are used in weld fatigue design and evaluation, where S represents nominal stress or stress range and N represents the number of cycles to failure. The present invention is directed to structural stress analysis through various combinations of modeling, calculation, and direct measurement schemes.Type: ApplicationFiled: November 16, 2001Publication date: August 22, 2002Inventors: Pingsha Dong, Jinmiao Zhang, Jeong Kyun Hong
-
Patent number: 6398102Abstract: A method for providing an analytical solution for a thermal history of a welding process having multiple weld passes. The method includes the steps of inputting a plurality of files and parameters, preprocessing information from the plurality of files and parameters to determine a set of conditions associated with the welding process, determining a region of influence of at least one heat source used in the welding process as a function of the set of conditions, determining a plurality of point heat source solutions within the region of influence, determining a temperature solution for each weld pass as a function of a superposition of the plurality of point heat source solutions, and determining the thermal history of the welding process as a function of the temperature solutions.Type: GrantFiled: August 29, 2000Date of Patent: June 4, 2002Assignee: Caterpillar Inc.Inventors: Zhenning Cao, Jinmiao Zhang, Frederick W. Brust, Ashok Nanjundan, Yi Dong
-
Patent number: 6324491Abstract: A method for determining a heat source model for a weld. The method includes the steps of determining a double elliptical distribution of the heat density of the weld, modifying the double elliptical distribution as a function of a profile geometry of the weld, and determining the heat source model as a function of the modified double elliptical distribution.Type: GrantFiled: July 21, 2000Date of Patent: November 27, 2001Assignee: Caterpillar Inc.Inventors: Jinmiao Zhang, Yi Dong