The invention discloses a
coal reservoir permeability
test optimization method and
system, and the method comprises the steps: carrying out the clustering analysis of
coal samples of different
coal orders through a K-means clustering and
random forest algorithm, and constructing a permeability test
database containing the permeability of different coal orders, the
fracture pressure, the pore characteristics and other parameters; on the basis of a CT scanning result of the coal sample, physical characteristics such as a pore structure and fracture distribution of the lump coal are analyzed, a representative position is selected for
sample preparation, and a
fracture pressure interval of the sample is pre-judged according to a
database; intelligently adjusting a pressurization strategy according to the pre-judgment interval, and switching to slow gradient pressurization when the pressure is close to the
fracture pressure of the coal sample; and according to the measured data points, abnormal points and
invalid data points in a fracture area are automatically identified and eliminated based on a
random forest algorithm, representative data points are optimally selected in a permeability test stable section and a characteristic inflection point interval, and finally data constraint and fitting correction are performed through a P50 probability method. The method can improve the efficiency and accuracy of the laboratory
coal permeability test.