Hypercortisolism is associated with tiredness and fatigue during the day and disturbed sleep at night. Our goal is to employ a wearable brain machine interface architecture to regulate one's energy in hypercortisolism. First, we present a state-space model to infer a hidden cognitive energy-related state from one's Cortisol secretion patterns. Particularly, we consider circadian upper and lower bound envelope curves on Cortisol levels, and timings of hypothalamic pulsatile activity underlying Cortisol secretion as observations. We then use Bayesian filtering to estimate the hidden cognitive energy-related state. Finally, we close the loop using a knowledge-based control approach. In a simulation study based on experimental data, we illustrate the feasibility of designing a wearable brain machine interface architecture for energy regulation in hypercortisolism. In this architecture, we infer one's cognitive energy-related state seamlessly rather than monitoring the brain activity directly and close the loop using fuzzy control. This simulation study is a first step towards the ultimate goal of managing hypercortisolism in real-world situations.