Optimum positioning of newly installing seismographs along highway lines is discussed. First, we introduce a method to estimate the spatial distribution of earthquake intensity under the condition that several observation data of earthquake intensity are provided at given sensor locations. We also define a spatial index which indicates the estimation error level of the spatial distribution of earthquake intensity of which value becomes zero at the sensor locations. The optimum sensor location problems can be formulated as a minimization problem of this index value which is a function of the sensor locations and the prior information of the special distribution of earthquake intensity in the target area. The design parameters are locations of the sensors. Genetic Algorithm (GA) is used for the optimization. After a brief introduction of real coded GA, optimum sensor locations along highway lines are determined as to minimize the spatial index value by using this GA code.
[1]
R. E. Kalman,et al.
A New Approach to Linear Filtering and Prediction Problems
,
2002
.
[2]
Shigenobu Kobayashi,et al.
Theoretical Analysis of the Unimodal Normal Distiibution Crossover for Real-coded Genetic Algorithms
,
1999
.
[3]
Tomoyuki Hiroyasu,et al.
Distributed genetic algorithms with a new sharing approach in multiobjective optimization problems
,
1999,
Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).
[4]
A. Journel,et al.
Kriging in terms of projections
,
1977
.
[5]
Noel A Cressie,et al.
Statistics for Spatial Data.
,
1992
.
[6]
Masaru Hoshiya,et al.
IDENTIFICATION OF CONDITIONAL STOCHASTIC GAUSSIAN FIELD
,
1996
.