Abhishek Khanna edited Introduction.tex  about 9 years ago

Commit id: 0c1916797404b90f6505dfabe043bc6c6873c052

deletions | additions      

       

\section{Introduction}  Edward tufte’s work in visualization laid down basic principles of effective visualization as foundation to all visual design[1][2][3]. In mid 90s, The authors of [4] specifies three key to effective visual representation i.e. (i) Natural encoding of information; (ii) Task specific graphics; (iii) No gratuitous graphics. There were proposed researches to improve one line diagrams into visually informative, Dr. Thomas Overbye and his collaborators making significant contribution for improvement in one line diagram [5]. The contour plot is further enhanced to visualize variety of data, such as power flows in transmission lines[6], location marginal prices[7], contingency analysis[8], third dimension to visualize various data[9]; but still they developed the visualization with foundation as one line diagram.  Major drawback in each of the foregoing works is that the nodal positions remain arbitrary and exhibit no meaningful “natural encoding”. Some of the proposed researches explain the algorithm for bus positioning but none of these methods locate position nodes in a plot which is electrically meaningful. Visual presentation of any two nodes with a size variance are likely to be perceived as change in their values and if plotted in close proximity then it’s perceived as in the same group or cluster [10]. This paper illustrates different methods and explains the effects of graph layout on human sociogram perception. Random positioning of nodes to layout a simplified one line diagram may generate a wrong impression like low centrality node as a center node or an isolated sub network as a well-connected network. This paper redesigns power system network with a new methodology giving a new dimension by scaling the branch which corresponds to its admittance. The scaling and redesign of power system network uses two major layout which are discussed in the paper and are as follows: “Force Directed layout algorithm and Edge weighted spring embedded layout” which can generate electrically meaningful layout of power system network [11][12][13]. In [11][13], This paper proposes a fast convergence with high quality assurance of layout significantly. In [12], this paper introduces an edge length proportional to weight on the graph edges. This idea of creating “Electrical Lung” and using cytoscape for visualization is coined by Dr. Paul Cuffe. Our motivation to employ graph theory approach to power system is to understand system vulnerability to attack or component failure due to catastrophic events [14] [15], extensive work has been done in this area of research [16] to evaluate vulnerability via graph theory, using centrality measure to find critical nodes for energy transmission [17], topological investigation of power grid networks [18] and to evaluate the complex in large power system networks [19]. But one of the researches was based on the electrical distances to study the vulnerability of power grid[20]. The author of [18] discusses electrical distances to remodel electrical grid and use complex graph theory to analyze the electrical network where paper uses degree, centrality and other measure to evaluate the electrical grid. Vulnerability of electrical network is important as few of the researches shows the power law equation is followed [21].Unfortunately, these results are scattered in time and space. Modelling of electrical network as per the graph theory principle is a perplexing thought as it’s not evaluating an electrical network as per electrical network characteristics. In[22], a remarkable contribution was seen in biochemistry journal where they used resistance distances as a mean to restructure fullerene molecule. They used the electrical analogy to design the C_60 fullerene complex.  Complexity in electrical grid system is due to physics behind the network which will not align with network theory. Electrical distances in power system was were  enlightened in a research by Paul hines Hines and his collaborators. Electrical proximity methodology must be used to divide a power system into a meaningful zone, to utilize spectral clustering and embedding techniques to partition and visualize power system [23] where authors of [24] used electrical parameters to find electrical betweenness of a network and assessed the power grid on it.   A number of potential electrical distances measures and methodologies are being used in previous researches; The author of [25] used electrical distance and topological analysis and proposed the voltage control to divide the geographical areas and time constants. This concept allows the structural controllability and observability of proximity of the power system. The recent commendable contribution in work of power system visualization by Dr. Paul Hines, Dr. S. Blumsach  and his collaborators [26] [27] has laid down the grounds of research into power system network with electrical distances.   This paper is organized as follows; Section 2 describes the methodology employed and their usage in electrical grid visualization. Section3 presents several results from CytoscapeV3.2.0 of 2D visualization of variety of test case networks. The difference and the advantages of the proposed visualization with respect to conventional visualization are discussed in detail. Section 4 draws relevant conclusions. Finally, Appendix A briefly introduces software tools used for the simulations.