Working condition division method based on piecewise linear regression

The invention belongs to the technical field of working condition division in the operation process of a power plant, and particularly relates to a working condition division method based on piecewise linear regression. Comprising the following steps: step 1, acquiring data from a sequence database; step 2, carrying out smoothing processing on the sequence Pr; 3, segmenting the data; 4, judging a data state; 5, screening and confirming working condition states; and step 6, writing out the working condition state. The method has the beneficial effects that the specific working condition occurrence time period is accurately counted in combination with parameter trend form change and engineer free parameter setting, and the working condition result generated in the operation process of the power plant is recorded. Through test analysis of the method, the condition identification method is utilized to judge that an identification result is completely consistent with a manual statistical result of an engineer, and