Freeway Ramp Metering Control Based on Neural Dynamic Optimization

Abstract

In this paper, we consider the optimal metering control problem for a local freeway ramp. The traffic model is formulated as one with stochastic and nonlinear properties. The control objective is to maintain the freeway operated at a desired traffic density and to diminish the queue length on the ramp as possible. Such a nonlinear stochastic optimal control problem is known hard to solve exactly, then we concentrate on providing approximate solution. Our approach consists of two steps: the first is to solve a finite-horizon optimal control problem based on neural dynamic optimization; the second is to adapt the finite-horizon optimal controller to work for the infinite-horizon case by using the recedinghorizon control mechanism. We choose the multilayer feedforward neural network as the approximating structure, and the stochastic steepest descent method for the training. Extensive simulation studies demonstrate the efficiency of our approach.

Multi-ANN Predictive Control of Citrus Peel Supercritical Extraction Temperature

“…The large-scale supercritical fluid extraction (SFE) system of citrus peel requires classification extraction to get variety of ingredients with high stability and accuracy temperature control, and has the system disturbance of flow rate and pressure, showing a strongly nonlinear, inertial and time-varying characteristic. At present, for such problems, the traditional cascade PID control is hard to overcome nonlinear and inertia [6]. Predictive control based on state variables [7,8], which rely on rigorous mathematical model, is not suitable for the model under time-varying conditions mismatch.…”