Adaptive control techniques can be applied to online gain tuning of tokamak thermokinetics. Here, a self-tuning control scheme is explored for both the plasma profile and power control. The distributed parameter system of the flux-surface-averaged one-dimensional transport equations is discretized by a nonlinear variational procedure. A finite-dimensional multiple-input/multiple-output control algorithm is derived using the linearized equations. A particular class of nonlinear three-parameter profiles is used for plasma density, temperature, and deuterium fraction profiles. Feedback gains are determined using a simplified minimum variance control law of self-tuning control. In the examples, normal multiple-output specifications for the plasma profile parameters for the density and power control are shown to be controllable by multiple-particle inputs alone.