Research Article

The Improved Least Square Support Vector Machine Based on Wolf Pack Algorithm and Data Inconsistency Rate for Cost Prediction of Substation Projects

Table 2

The forecasting values and actual values of the test set (unit: yuan/kV·A).

Serial numberActual valueBPNNSVMLSSVMILSSVMWPA-DIR-ILSSVM

67271.7286.4287.1284.6280.5268.6
68262.6251.3273.1255.3269.6264.2
69253.5270.0239.6264.0259.6255.9
70244.5225.6254.5259.5239.7247.2
71235.4251.4245.9246.5243.1236.7
72326.3335.6335.2341.1335.7330.5
73217.2208.3201.8203.6211.6215.6
74208.1219.0216.5218.0213.9209.5
75199.1210.2195.2208.6204.9200.6
76390.0376.9395.6398.5381.6387.6
77280.9296.1294.8269.6285.6280.5
78285.1302.8269.6296.9295.0288.0
79289.4271.9302.1303.2280.2293.1
80293.7310.5281.0285.6302.8291.3
81297.9313.7312.3311.1307.6298.1
82402.2412.5389.5385.6413.6405.9
83306.4325.3321.6322.6303.7309.7
84310.7298.7327.0292.7320.2309.1
85274.9290.3287.2288.6283.2276.7
86319.2338.1302.9323.5307.5318.6
87283.4300.1295.7270.6293.2283.6
88369.5350.2385.9387.1380.9372.0