Porcellio scaber Algorithm with t-Distributed Elite Mutation for Global Optimization

Algorithm 1

TPSA algorithm.

	Set the objective function f(x), x = [x1,x2, …, xd]^T
	Set the algorithm parameters: population size N, elite mutation probability parameter Pt, the upper and lower boundaries of the weight allocation parameter λ, and step scale coefficient γ.
	Initialize the Porcellio scaber population (i = 1,2, …, N)
	f(x) determines the environmental condition Ex at x
	for k = 1 to maxiter
	Obtain the current optimal position:
	Randomly generate an exploration environment vector τ = [τ1, τ2, …, τd]^T
	Calculate the optimal environmental conditions min{Ex} and the worst environmental conditions max{Ex} at position xi^k+τ
	for i = 1 to n
	Calculate the action intensity p according to Equation (4)
	Update individual position Xi^k+1 according to Equation (3)
	if pt > rand
	Update the mutant Porcellio scaber position Xi^k+1 according to Equation (5)
	end if
	The individual fitness after position update was calculated
	end for
	if MinFitness(k + 1) > MinFitness(k)
	Reserved the optimal information of the previous generation to the next generation
	end if
	end for