Available Transfer Capability screening considering transient stability by Support Vector Machine

In the electricity power markets, available transfer capability (ATC) is an important index. Conventional ATC calculation takes much time because it repeats stability analysis many times. This paper proposes a neural network based technique applied to the ATC screening. Using this method no stability analysis is necessary at least in its first stage, and approximate ATC can be evaluated at high speed. Accurate ATC is expected to be obtained using the result of the proposed preliminary screening, which saves time greatly. This paper chooses so called support vector machine as the calculation tool which generalizes better than other neural networks. A numerical example is shown to demonstrate the validity of the proposed method.