{"responseHeader":{"status":0,"QTime":3,"params":{"q":"{!q.op=AND}id:\"103006\"","hl":"true","hl.simple.post":"","hl.fragsize":"5000","fq":"!embargo_tdt:[NOW TO *]","hl.fl":"ocr_t","hl.method":"unified","wt":"json","hl.simple.pre":""}},"response":{"numFound":1,"start":0,"docs":[{"file_name_t":"Yen-On_Representation.pdf","thumb_s":"/72/f7/72f7ba8fb5c5c5ca588626389cf227af55ceb40b.jpg","oldid_t":"compsci 11119","setname_s":"ir_computersa","restricted_i":0,"format_t":"application/pdf","modified_tdt":"2016-05-26T00:00:00Z","file_s":"/77/5d/775dbe0fbf1dd11507fbcb817dcb9cea81f8681b.pdf","title_t":"Page 45","ocr_t":"34 meanwhile errors may propagate. The second one is called the contouring method [9, 62]. This method totally abandons the previous mesh and rebuilds a new mesh by triangulation with the aid of contour curves. However it relies on the generation of contour curves and the selection of points on the contours for triangulation. An extension of this method is to examine the analysis result and use it as mesh grading heuristic. In particular, since the total strain energy can be used as a measurement of solution accuracy in a structure analysis, strain energy distribution can serve as a guidance for grading the new mesh. In other words, for areas with high strain energy variation, smaller (and hence more) elements should be in place, and vice versa. This approach is definitely more attractive so long as the measurement of accuracy, e.g., strain energy distribution, can be properly quantified. 3.2.3 Grid Synthesis Method In [50], Shephard, Gallagher, and Abel suggest a grid synthesis method which synthesizes the conventional isoparametric method with the variation of the strain energy distribution (SED). Recall that in the conventional isoparametric approach, the key nodes are those along the boundary whose positions determine the parameter values on which isoparametric lines are drawn from this boundary to the opposite side. Using a previously conducted analysis result this method repositions the key nodes such that the SED variation between each adjacent key node is roughly the same. To govern the spacing of key nodes along one domain boundary, the total energy variation is calculated by adding up all absolute energy differences between adjacent key nodes. The segment energy variation is then calculated by dividing the total energy variation by the number of segments where the number of segments is just the number of nodes minus one. A cubic spline with proper end conditions is interpolated through these nodes, and some number ( -15 times of number of nodes) of evenly-spaced data points are calculated","id":103006,"created_tdt":"2016-05-26T00:00:00Z","parent_i":103066,"_version_":1679953856817004546}]},"highlighting":{"103006":{"ocr_t":[]}}}