Research Article

Improved Hypercube Optimisation Search Algorithm for Optimisation of High Dimensional Functions

Table 1

The performances of the HOS and HOS+ algorithms using different dimensions.

HOSHOS+
Benchmark functionsDBestMeanStd. Dev.BestMeanStd. dev.

Sphere (F1)301.19E − 068.88E + 004.86E + 009.95E − 1372.63E − 514.29E − 51
601.30E − 025.92E + 014.00E + 017.79E − 966.39E − 396.19E − 39
901.20E − 011.49E + 027.49E + 011.70E − 812.19E − 323.59E − 32

Schwefel 2.22 (F2)303.00E − 012.67E + 011.41E + 011.19E − 951.98E − 273.19E − 27
606.50E − 011.01E + 024.40E + 013.21E − 723.88E − 226.39E − 22
904.75E + 001.67E + 027.66E + 012.49E − 595.72E − 186.91E − 18

Rotated hyper ellipsoid (F3)302.59E + 022.10E + 041.11E + 043.39E − 1342.69E − 472.03E − 47
604.08E + 022.77E + 051.59E + 055.25E − 1091.75E − 342.09E − 34
905.85E + 029.81E + 055.09E + 051.05E − 751.35E − 281.79E − 28

Ackley (F4)305.00E − 031.06E + 013.49E + 008.88E − 163.57E − 154.69E − 15
605.40E − 011.51E + 013.86E + 008.88E − 162.53E − 142.79E − 14
905.70E − 011.69E + 014.40E + 004.44E − 154.29E − 136.69E − 13

Griewank (F5)305.00E − 023.10E + 011.85E + 010.00E + 000.00E + 000.00E + 00
606.90E − 011.50E + 021.19E + 020.00E + 000.00E + 000.00E + 00
901.20E + 014.19E + 022.39E + 020.00E + 000.00E + 000.00E + 00

Hyper ellipsoid (F6) sphere (F1)305.10E + 012.61E + 024.59E + 011.55E − 1392.19E − 481.77E − 48
606.30E + 012.75E + 036.49E + 024.35E − 1133.39E − 363.59E − 36
906.90E + 011.04E + 042.45E + 034.89E − 792.99E − 293.96E − 29