Improved GSO Optimized ESN Soft-Sensor Model of Flotation Process Based on Multisource Heterogeneous Information Fusion

Table 2

Predictive data set of the soft-sensor model.

Serial number	Color features		Texture features				Process variables and boundary conditions					Predicted variables
	Saturation	Brightness	𝑓1	𝑓2	⋯	𝑓14	Feed grade (%)	Feed flow rate (m^3/h)	Feed concentration (%)	Feed granularity (%)	Pharmacy flow rate (L/min)	Concentrate grade (%)	Tailings recovery rate (%)
1	0.077	0.633	0.107	1.154	⋯	0.268	62.76	329	35	90	15.5	70.51	97.7
2	0.061	0.602	0.131	1.168	⋯	0.472	63.67	297	35	90	11.5	69.74	97.2
3	0.059	0.591	0.147	1.359	⋯	0.502	65.07	285	37	92	11.3	69.69	97.0
4	0.038	0.464	0.103	2.786	⋯	0.618	65.48	214	36	95	7.5	68.98	93.5
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
300	0.034	0.415	0.076	2.125	⋯	0.584	65.9	310	36	96	5.5	67.29	90.2