HYBRID TYPE PARTIAL AND FULL STEP QUADRATIC SOLVER FOR MODEL PREDICTIVE CONTROL OF DIESEL ENGINE AIR PATH FLOW, AND METHOD OF USE

摘要

PROBLEM TO BE SOLVED: To provide an alternative quadratic solver capable of reduction of a computational time and improvement of accuracy in model predictive control, and a method of use the same.SOLUTION: Model predictive control using a hybrid type quadratic solver for solving a linear feasibility problem corresponding to a nonlinear problem of a diesel engine, solves a convex, quadratic cost function having optimization variables and constraints and controls a plant on the basis of an output solution to optimize plant operation. Depending on whether a partial step (iterative) or a full step (direct) is attempted, a combination of iterative and direct calculations in a primal space is included. Further, primal and dual space array matrices are pre-computed and stored offline and are retrieved via use of a unique identifier associated with a specific active set for a set of constraints. Such hybrid type and/or offline calculations allow for reduction in computational power while still maintaining accuracy of solution results.SELECTED DRAWING: Figure 4