K-means clustering power grid adaptability evaluation method based on Mahalanobis distance

The invention relates to a K-means clustering power grid adaptability evaluation method based on Mahalanobis distance, and the method comprises the following steps: selecting a to-be-built region of aline in a power system, and determining a to-be-built line data set; based on the historical load data, a Mahalanobis distance K-means clustering algorithm is utilized to perform clustering and per-unit processing on the daily load curve, the type set is W, then the weight vector M = [m1, m2,..., mw] of each type of load is determined, and the maximum value of load prediction in the planning period is utilized to multiply by the clustered per-unit daily load curve, thus to obtain a load prediction curve after clustering in a planning period; establishing a power grid adaptability evaluation model; solving an evaluation model; obtaining an economic adaptability index, a renewable energy utilization adaptability index, a power supply reliability adaptability index and a system branch load rate expected value of the