The invention discloses a decision making method for robust unit commitment scheduling considering CVaR (Conditional Value at Risk), which includes the following steps: constructing a polyhedral uncertainty set from interval, time and space dimensions; measuring possible system risks by CVaR measurement; constructing a robust unit commitment optimization model considering the admissible wind powerinterval based on the polyhedral uncertainty set and a robust optimization method; and solving the model, and outputting the admissible wind power range, risk loss, unit start/stop status and wind abandonment and load shedding decisions of the optimized system. Compared with a method in which the boundary of the uncertainty set is given, the boundary of the constructed uncertainty set is obtainedby optimization, the conservativeness of the robust optimization model can be controlled by adjusting the uncertainty parameters, and the model is based on the current power system scheduling architecture, complies with the adaptive mechanism between unit commitment and robust optimization and can be directly used to solve the unit commitment problem of the current power system.