Benefit Analysis of Low-Carbon Policy Mix Innovation Based on Consumer Perspective in Smart City
Algorithm 1
GA-BP genetic algorithm.
Input: the training set {xs, ys, zs}, the test set {xd, yd, zd}. The maximum iteration value is M, the size population value is N, the crossover probability value is θ, and the mutation probability value is β. The number of neuron nodes value is Nd.
Output: predicting the carbon emissions of private cars {y}
(1)
Define population information as a structure.
(2)
While iteration do
(3)
For i = 1, 2, …, N do
(4)
Calculate individual fitness.
(5)
End for
(6)
Find the best chromosome s and best fitness b, solve the optimal initial threshold t and weight .
(7)
For i = 1, 2, …, M do
(8)
Select the individual i, cross chromosome, mutate heredity.
(9)
For j = 1, 2, …, N do
(10)
Find the minimum fitness chromosomes p, maximum fitness l, and their position.
(11)
End for
(12)
If b > new b, then
(13)
b = new b
(14)
End if
(15)
End for
(16)
Assign the optimal initial threshold weight to network prediction.
