Academic Journal of Science and Technology 
ISSN: 2771-3032 | Vol. 3, No. 2, 2022 

 

144 

Analysis	of	Microterrain	Data	Based	on	GIS	and	Applied	
to	Power	Grid	Layout	
Nai Wei, Wei Wang 

Wuhan University, Wuhan, 430072, China 

 

Abstract: With the rapid development of economy and electronic information industry, the demand for electricity is getting 
higher. As a basic project, the reasonable point layout of power grid is becoming more and more important. Topographic elements 
are the key factors affecting power grid layout. However, the actual terrain is complex and changeable, so it is urgent to quantify 
all kinds of topographic factors to provide reference for power grid layout. Based on three-dimensional GIS technology, this 
paper provides a terrain factor algorithm with controllable variables for power grid layout project, obtains terrain factors such as 
slope, aspect, slope position, underlying surface type and surface roughness with different accuracy according to the actual terrain, 
and outputs the data format in accordance with the specification to provide decision support for power grid layout. At the same 
time, this study can also provide a reference for other decision-making works that rely on topographic elements. 

Keywords: Three-dimensional GIS, Tiny terrain. 
 

1. Introduction 
Terrain is an important factor affecting environmental 

change, hydrological process, biological distribution, 
geomorphic features, etc., and the spatial distribution 
characteristics of terrain attributes are important indicators to 
describe these spatial change processes[1].The micro-
topography formed by the combination of regional large 
terrain and local topographic terrain topography has a 
significant impact on local meteorological factors, and even 
leads to the special enhancement of some meteorological 
factors (such as ice and wind).Therefore, micro-topography is 
directly related to the selection of grid layout points and the 
maintenance of poles and towers. For example, the formation 
of debris flow and landslide is closely related to the slope and 
surface roughness, and the slope aspect and elevation will 
affect the sunlight irradiation intensity, thus affecting the ice-
covered melting speed of transmission lines in winter 
[2].Therefore, digitization and visualization of complex 
micro-terrain factors can provide reference for power grid 
layout and effectively improve the quality of electricity 
consumption. The early power line projects in China are 
mainly based on the geographical data of manual field survey 
to design power grid layout scheme. However, this way has 
the disadvantages of high cost, long cycle and large error. 
With the continuous integration of computer science and 
cartography, geographic information system has shown a 
good application prospect in the electric power industry. By 
using GIS spatial analysis method based on DEM data to 
calculate topographic, geomorphic and hydrological 

information, the problem of tower selection and maintenance 
in power grid layout can be solved. In this paper, Guangdong 
Province is taken as the case area to design and develop the 
power grid layout GIS. The micro-terrain factor algorithm 
module is added into the power grid layout GIS to realize the 
calculation of slope, aspect and slope position based on DEM 
data, and obtain the terrain factor data to provide strong 
support for power grid layout. 

2. Research Data and Methodology 

2.1. Overview of Study Area 
The terrain of Guangdong Province is complex and diverse. 

The terrain is high in the north and low in the south. It 
gradually descends from the mountains in the north to the 
coastal areas in the south, forming a landform pattern 
dominated by mountains in the north, hills in the middle, and 
plain platforms in the south. Therefore, the deployment and 
maintenance of power grids in Guangdong Province are faced 
with great difficulties [3]. 

2.2. Data Sources and Preprocessing 
The data in this study include DEM data of 30m×30m 

resolution, soil type data, and Excel spreadsheet data 
containing information about the height of surrounding trees. 
First of all, Excel data were associated with "bar segment" as 
the key field to obtain the surface roughness factor data set, 
including rod segment, coordinate information and the 
vertical distance between the log and log of the traverse. The 
research data of this paper are shown in table 1. 

 
Table 1. Description of research data 

Contains the fields Contains the fields data format 
DEM elevation shp 

Soil type soil type shp 
Hidden Danger File - List 
of tall plant hidden danger 

bar segment, tree species, horizontal distance between traverse 
and log, vertical distance between traverse and tree 

excel 

The province line 
bar segment, Vertical distance of rod coordinates and wires to 

trees 
excel 

 
 



 

145 

 

2.3. The research methods 
2.3.1. Slope aspect calculation 

Slope refers to the ratio of vertical height and horizontal 
distance of slope surface, usually including two parts: slope 
and slope aspect. Slope refers to the maximum ratio of slope 
height change; Slope aspect is the direction reflecting the 
maximum value of slope surface change ratio[4].According 
to the actual needs of topographical features of Guangdong 

province and the grid layout, is proposed in this paper in view 
of the Guangdong power grid layout the gradient of slope to 
the classification standards (table 2 and table 3), because of 
ArcGIS can't direct output grade level and tilt, 
recommendations for layout grid, in order to solve the 
problem, based on DEM data computing slope and aspect to 
the information and classification, directly output the 
information needed for power grid layout, without secondary 
processing of ArcGIS data: 

 
Table 2. Slope classification 

Slope Grade Slope Degree Slope /(°) Suitable for power grid deployment 
Grade 1 flat 0~5 Yes 
Grade 2 gentle 6~15 Yes 
Grade 3 normal 16~25 Not recommended if it is not necessary 
Grade 4 steep 26~35 No 
Grade 5 Sharp 36~40 No 
Grade 6 dangerous 41~45 No 
Grade 7 Risk Above 46 No 

 
Table 3. Aspect classification 

Aspect classification Aspect/(°) Whether the power grid is properly laid 
shady 0~45 No 

Half-shady 45~135 Yes 
Sunny 135~225 Yes 

Half-sunny 225~315 Yes 
shady 315~360 No 

 
In this study, the fitting surface method was used to solve 

the slope aspect under the micro-terrain, as shown in Fig. 1. 
Based on the 3x3 window, the local function of the quadric 
surface was constructed for calculation. 

 
e5 e2 e6 
e1 e e3 
e8 e4 e7 

Figure 1. 3×3 Moving window 
 
Through the surface fitting algorithm to get slope in the 

direction of X-axis (𝑠𝑙𝑜𝑝𝑒 ) and Y-axis (𝑠𝑙𝑜𝑝𝑒 ): 
 

𝑠𝑙𝑜𝑝𝑒
）

         (1) 

 

𝑠𝑙𝑜𝑝𝑒
e 2e 𝑒 e 2𝑒 𝑒 ）

8𝑐𝑒𝑙𝑙𝑠𝑖𝑧𝑒
 

 
Calculate the slope and aspect of the center point e: 
 
 

slope arctan 𝑠𝑙𝑜𝑝𝑒 𝑠𝑙𝑜𝑝𝑒        (2) 
 

𝑎𝑠𝑝𝑒𝑐𝑡 270° arctan
𝑠𝑙𝑜𝑝𝑒
𝑠𝑙𝑜𝑝𝑒

90°𝑠𝑙𝑜𝑝𝑒 /|𝑠𝑙𝑜𝑝𝑒 | 

cellsize is the resolution of the grid, and e  is the elevation 
value of each grid dot. 

After the slope aspect value is calculated by the above 
formula, it can be classified according to the classification 
table, and the slope aspect data required for grid layout can be 
output. 

 

2.3.2. Calculate slope positions 
Aspect refers to the geomorphic position of the slope 

surface. For example, on positive or negative terrain. The core 
work of slope position extraction is the division of 
geomorphic parts and geomorphic types [5].In this study, the 
fuzzy inference method of slope position gradient information 
is used to calculate the slope position information. According 
to the typical positions of various slope positions in space, the 
similarity between other positions and typical positions is 
calculated, so as to systematically and quantitatively describe 
the spatial gradient information of slope position. The 
gradient information of important slope position types such as 
ridge, slope shoulder, back slope, slope foot and valley can be 
reasonably described [6]. 

The algorithm mainly involves four terrain parameters: 
slope, profile curvature, horizontal curvature and other local 
terrain attributes, and a regional terrain feature RPI.RPI is 
defined as the sum of the shortest Euclidean distances from 
the point to the valley and the shortest Euclidean distances 
from the point to the ridge and the valley. The RPI value 
calculated based on ridge and valley can better reflect the 
relative position information of the terrain. Profile curvature 
is a measure of the change rate of ground elevation along the 
direction of maximum slope of the ground slope, and 
horizontal curvature is the curvature value of the curve at a 
point on the terrain surface obtained by cutting the terrain 
surface along the horizontal direction, that is, the curvature 
degree of the ground contour at the point. Since profile 
curvature and horizontal curvature have similar signs and 
functions, and horizontal curvature can effectively avoid the 
unreasonable extreme value of contour curvature, horizontal 
curvature is used to replace contour curvature in this study. 
Douglas-Peuker algorithm was used to extract feature points 
for ridge, and different elevation thresholds could be set based 



 

146 

on different study areas as the lower limit of ridge existence, 
so as to eliminate unreasonable ridge extraction results. 

Douglas-Peuker algorithm was used to extract feature 
points for ridge, and different elevation thresholds could be 
set based on different study areas as the lower limit of ridge 
existence, so as to eliminate unreasonable ridge extraction 
results. 

The algorithm principle of Douglas-Peuker method is as 
follows (Fig. 2) : 

Connect the first and last points of the curve to be processed 
with a straight line, find the distance between all intermediate 
points and the straight line, and find the maximum distance 
value dmax, and compare dmax with threshold: 

If dmax < threshold, all the middle points on this curve are 
dropped; 

If dmax≥threshold, the curve is divided into two parts by 
taking this point as the boundary, and the above process is 
repeated for these two parts of the curve until all points are 
processed. 

Due to the coexistence of multiple landforms in 
Guangdong Province in this study area, the lower limit of 
ridge elevation was set as 950 meters, and a relatively obvious 
ridge line was extracted. 

 

 
Figure 2. Feature points extracted by Douglas-Peuker 

method 
 
The valley line adopts O'Callaghan algorithm, which is also 

known as D8 algorithm. Its basic idea is to use a moving 3x3 
window to traverse the DEM data from the upper left corner 
of the input image to calculate the distance weight drop p 
between the center pixel and other eight neighboring raster 
cells. Assume that the position of the center pixel of the raster 
window is (i, j).Its elevation value is hi, and the elevation 
value of its domain grid is hk. k starts from east and is taken 
from 1 to 8 clockwise: 

 

       𝑝  
ℎ ℎ     𝑘 1,3,5,7

√
    𝑘 2,4,6,8           (3) 

 
Then take the maximum value of P, mark the k value of the 

cell where max of p is located, and finally get the flow 
direction dir of the center pixel: 

 
𝑑𝑖𝑟 2                (4) 

 
DEM elevation value and flow direction value of one of the 

operation Windows are taken, as shown in Fig. 3 [7]: 
 

 
Figure 3. DEM elevation value and stream direction 

coding value 
 
D8 algorithm is used to calculate each pixel of the upstream 

catchment area, the catchment area is greater than the given 
threshold pixel identification as the valleys, the algorithm can 
produce continuous, the reasonable channel, can set different 
threshold values according to different research areas, 
generate realistic topography of river network as a result, in 
this study, the Guangdong province has complex and varied 
water networks. The threshold of the upstream catchment area 
was set as 2km², and a relatively obvious valley line was 
extracted. 

Based on the deterministic classification algorithm of slope 
position [6], and the limitation of RPI value, rules can be 
generated to automatically extract the typical positions of 
various slope positions in the region (table 4). These rules can 
be applied to extract the typical positions of various slope 
positions, and the rules can also be set according to the 
geomorphological conditions of different regions. 

 
Table 4. Parameter Settings for typical positions of slope position extraction 

 ridge Slope shoulder Back slope Slope foot valleys 
RPI ≥0.99 [0.9, 0.95] [0.5, 0.7] [0.2, 0.3] ≤0.1 

Section curvature(×10-3m-1) ≥1 ≥1 [-0.1, 0.1] ≤-1 [-0.5, 0.5] 
Horizontal curvature(×10-3m-1)   [-0.1, 0.1] ≥0  

elevation ＞950m     
slope(°) ≤1  ≥6  ≤1 

 

2.3.3. Surface roughness and surface type algorithms 
Surface roughness reflects the undulation of the land 

surface and is the main influencing factor of many land 
surface processes[8].There are two kinds of understanding of 
ground roughness. One is from the perspective of 
aerodynamics. Due to the uneven surface or the geometric 
shape of ground objects, the position where wind speed is 0 
is not at the surface, but at a certain height from the surface, 
which is defined as ground roughness, also known as 
aerodynamic roughness. Another kind of understanding is 
mainly from the perspective of topography, and defines the 
degree of ground roughness as roughness, also known as 

surface microtopography [9].In this paper, the surface 
roughness mainly refers to the degree of uneven ground. 

The extraction of surface roughness factor is mainly based 
on the corresponding vertical distance data between wires and 
trees provided by Guangdong Electric Power Bureau. The 
surface roughness factor data is divided by a 30m×30m grid, 
and the mean value of the vertical distance between wires and 
vegetation in each grid is the surface roughness of the grid. At 
the same time, according to the corresponding relationship 
between the surface roughness and different vegetation types 
provided by Guangdong Electric Power Bureau (table 5), the 
surface type of each grid can be obtained. 

 



 

147 

Table 5. Correspondence between surface roughness and surface type 
ground type Corresponding surface roughness (m) 
Paddy field 0 

The dry land 1 
forestland 15 

bush 8 
Land with few trees 10 

Other woodland 10 
High coverage grassland 0.6 

Medium coverage grassland 0.4 
Low coverage grassland 0.2 

grass 0 
lakes 0 

Reservoir pits 0 
Permanent glacial snow 0 

Tidal flats 0 
beaches 0 

Urban land 30 
Rural settlement area 10 

Other construction land 10 
sand 0.2 

Gobi area 0.2 
Saline-alkali land 0.5 

marsh 0.5 
Bare land 0.2 

Bare rock stony ground 0.2 
other 0.1 

 

3. Results and Analysis 
Based on the 30mx30m DEM data of Guangdong Province, 

the quadric surface fitting algorithm is used to obtain the 
classification data of slope and aspect, the ArcGIS module is 
used to calculate the profile curvature and horizontal 
curvature, and the ridges and valleys are calculated according 
to Douglas-Peuker algorithm and O'Callaghan algorithm. 
And then you get RPI. Based on table 4, the slope, curvature, 

RPI value and elevation value were judged to obtain the slope 
position data. Based on the surface roughness factor 
preprocessing results and DEM data partitioning, the surface 
roughness data are obtained. The underlying surface data 
were obtained based on "Soil Type Data of Guangdong 
Province". Then, the data are output in accordance with the 
standard format. Flowchart of system algorithm 
implementation is showed in Fig. 4. 

 

 
Figure 4. Flowchart of system algorithm implementation 



 

148 

 
The output result is shown in Fig. 5-Fig. 6: 

 

 
Figure 5. Microterrain data output 

 
 

 
Figure 6. Slope position data output 

 
 
In order to improve the efficiency of data output, a 512×512 

one-dimensional array of BigGrid was established to read 
data. The numbering method was consistent with the grid 

numbering rule of each image, and the raster values were read 
from the lower left corner, line by line. Generate the data 
format required by the grid (table 6) and store it in the 
database: 

 
Table 6. Microtopographic information extraction results 

Serial 
number 

Central 
latitude 

Center 
longitude 

Altitu
de 

Aspe
ct 

Slop
e 

Slope 
position 

Underlyi
ng 

Surfa
ce 

Roughne
ss 

004614-01 20.38 109.87 0 -1 0 0 255 127 127 



 

149 

 

4. Conclusion and Prospect 
Topographic elements play an indispensable role in all 

kinds of construction work. How to extract different 
topographic factors flexibly according to the needs of the 
project is very important. This paper provides an algorithm 
for extracting terrain factors such as slope, aspect, position, 
surface roughness, surface type and underlying surface type 
based on DEM data and other types of data provided by the 
project. A variety of standards and variable parameters are 
provided in the algorithm to make reasonable Settings 
according to different projects. Finally, the data of different 
terrain factors are stored in the database according to the 
prescribed format, which is convenient for the staff to access 
at any time, and provides an important reference basis for the 
maintenance and planning of the power grid tower. 

However, there are not only six types of micro-terrain 
factors involved in this study, but also different micro-terrain 
factors required in different projects. This study will continue 
to deepen the flexible setting and on-demand extraction of 
micro-terrain factors, with the ultimate goal of including all 
the terrain factors in the system. In addition, the parameters 
and classification standards of various micro-terrain factors 
can be flexibly set according to different landform types, 
providing construction workers with more convenient micro-
terrain factor data as a reference for project decision-making. 

References 
[1] Reddy, G. P. O. (2018). Geographic Information System: 

Principles and Applications.  

[2] LI Longji,XI X.,LI Zhijian,WANG Xiaoguang,WEN Qingfeng, 
LI Qiran, ZHOU Kai,LIU Yong, YAO Juntao (2020). 
"Analysis and Prediction of Micro-meteorological Parameters 
for Power Transmission Lines in Micro-terrain Environment." 

[3] GUO Xin-chun, W. G.-q. (2019). "A GIS-based Method for 
Identifying and Extracting Micro-topography Factors." 

[4] XU Jia-hao, Z. W., HUANG Xiao-ning, YANG Cheng-shun 
(2014). "Transmission Line Icing Prediction Model Under 
Micro-Meteorological Conditions." 

[5] QIANG Xiao-huan, Y. C.-a. (2005). "Research of Algorithms 
and Accuracy for Calculating Slope Based on Digital Elevation 
Model."  

[6] Paul A. Longley, M. F. G., David J. Maguire, David W. Rhind 
(2005). Geographical Information Systems: Principles, 
Techniques, Management and Applications, 2nd Edition, 
Abridged 

[7] QIN Cheng-zhi, Z. A.-x., SHI Xun, LI Bao-lin, PEI Tao, 
ZHOU Cheng-hu (2007). "Fuzzy inference of spatial gradation 
of slope position." 

[8] GAO Qi, F. J.-f., HE Hui-xin, KONG Wei-hua, ZHOU Yu-ke 
(2017). "Experimental Comparison and Analysis of Multi-core 
Parallelization of D8 Algorithm under Different Data Partition 
Methods." 

[9] Shanshan, J. C. F. H. W. (2012). "Review of Land Surface 
Roughness Parameterization Study."