Bearing Fault Diagnosis Method Based on Multidomain Heterogeneous Information Entropy Fusion and Model Self-Optimisation

Algorithm 1

The framework of optimising feature extractor.

	Input: the number of sparrows N
	Output: optimal fitness value F;Optimal value of feature extractor structure ybest;
(1	)initialization: initialize sparrow population via (21) and (22);
(2)	repeat
(3	) while k Population size do
(4)	Calculate fitness values of the each individual via (25);
(5)	Sort each individual according to its fitness value;
(6)	Random R2 value in the interval [0, 1];
(7)	for each producer i = 1 to (N  PD) do
(8)	Update the location of the producer via (7).
(9)	end for
(10)	for each producer j = (N  PD) + 1 to N do
(11)	Update the location of the scroungers via (8).
(12)	end for
(13)	Use the elite reverse strategy to reverse solution and update outstanding individual via (23) and (24);
(14)	Update F and ybest;
(15)	k = k + 1;
(16)	end while
(17)	until a fixed number of iterations
(18)	return F, ybest