{"id":"arxiv/2606.26446","url":"https://arxiv.org/abs/2606.26446","tags":["math.OC","eess.SY"],"score":0,"title":"Input Convex Neural Network as a Surrogate in Stability-Constrained Optimization for IBR-dominated Power Systems","domain":"math.OC","source":"arxiv","authors":["Wangkun Xu","Hongyang Jia","Yi Wang","Ning Zhang","Fei Teng"],"reviews":0,"subject":"math.OC","summary":"Input convex neural networks (ICNNs) are increasingly used as surrogates for stability indices and embedded as constraints in power-system optimization. This letter clarifies two recurring formulation limitations that can negate ICNN convexity benefits: (i) applying generic Big-$M$ mixed-integer reformulations introduces auxiliary binaries that are unnecessary for enforcing ICNN sublevel constraints; and (ii) reversing the stability inequality transforms a convex sublevel set into a generally nonconvex superlevel set, invalidating global-convergence guarantees of cut-based methods. After…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-06-24","reactions":{}}