Figure 7: Energy reduction for increasing state weight matrices.
From: Energy scaling of targeted optimal control of complex networks
We construct a single model network using the static model52 with the parameters n=300, nd=n/4, γ=2.7 and kav=5. The energy scaling is examined for the general quadratic cost function. We compute η for different values of ζ such that the state weight matrix Q=ζIn. The values of η for ζ=0, 1 and 10 are η=13.46, 13.66 and 13.53, respectively. Each point is averaged over 50 iterations of target node sets. The simulations are performed with initial condition set to the origin and the final condition chosen randomly such that ||yf||=1. Error bars represent one s.d.
