The invention discloses a
coal slime flotation dosage
dynamic prediction method and
system based on multiple working conditions, and relates to the technical field of
mineral processing process control, and the method comprises the steps: based on
coal slime particle data, employing a density clustering
algorithm to carry out preliminary dynamic division on real-time working conditions, generating initial working condition labels, and through a
time sequence correlation analysis and dynamic correction mechanism, carrying out
dynamic prediction on the real-time working conditions; optimizing the initial working condition
label to generate a multi-working condition
label; a flotation
cell simulation model is constructed, advanced
simulation is conducted according to the real-time working condition operation parameters, the
clean coal ash value and the
recovery rate in a future time window are predicted, dynamic calibration is conducted, and a chemical adding amount adjusting instruction is generated; and executing the dosage adjusting instruction, evaluating the
coal slime flotation efficiency through a multi-dimensional flotation dynamic evaluation method, and dynamically optimizing the dosage adjusting instruction according to an
evaluation result. A flotation
cell mechanism model is established, a predicted value is dynamically calibrated through parameter inversion and extended Kalman filtering, a dosing instruction is generated in combination with MPC, and advanced precise regulation and control of
clean coal quality and
recovery efficiency are achieved.