Model-based computational intelligence multi-objective optimization for gasoline direct injection engine calibration

Abstract

For modern engines, the number of adjustable variables is increasing considerably. With an increase in the number of degrees of freedom and the consequent increase in the complexity of the calibration process, traditional design of experiments–based engine calibration methods are reaching their limits. As a result, an automated engine calibration approach is desired. In this paper, a model-based computational intelligence multi-objective optimization approach for gasoline direct injection engine calibration is developed, which can optimize the engine’s indicated specific fuel consumption, indicated specific particulate matter by mass, and indicated specific particulate matter by number simultaneously, by intelligently adjusting the engine actuators’ settings through Strength Pareto Evolutionary Algorithm 2. A mean-value model of gasoline direct injection engine is developed in the author’s earlier work and used to predict the performance of indicated specific fuel consumption, indicated specific particulate matter by mass, and indicated specific particulate matter by number with given value of intake valves opening timing, exhaust valves closing timing, spark timing, injection timing, and rail pressure. Then a co-simulation platform is established for the introduced intelligence engine calibration approach in the given engine operating condition. The co-simulation study and experimental validation results suggest that the developed intelligence calibration approach can find the optimal gasoline direct injection engine actuators’ settings with acceptable accuracy in much less time, compared to the traditional approach.

Enhanced intelligent proportional-integral-like fuzzy knowledge–based controller using chaos-enhanced accelerated particle swarm optimization algorithm for transient calibration of air–fuel ratio control system

“…32 The output value of Sugeno-type fuzzy controller is transformed to a constant value directly by fuzzy rules without the need to calculate the weighted average of output membership functions, which allows the Sugeno-type fuzzy controller to perform more efficiently for realtime implementation. Figures 2 and 3 give the simplest membership functions for input and output signals of the Sugeno-type fuzzy controller which are defined over [21,1] First of all, input signals X 1 (k) and X 2 (k)are normalized by scaling them with K i and K p , respectively. Then, a set of fuzzy decision rules relating inputs to the output are constructed as Table 1 shows.…”

“…Global optimization algorithms are widely used to resolve nonlinear optimization problems, because they are able to find globally optimal results instead of resulting in multiple locally optimal results. [18][19][20][21] Metaheuristic algorithm is one of the most well-known global optimization methods. 22,23 And swarm intelligence algorithm, such as particle swarm optimization (PSO) algorithm, is one of the main categories of metaheuristic algorithm.…”

fSDE: efficient evolutionary optimisation for many-objective aero-engine calibration

“…A B Feng Wu wufeng_my@mail.nwpu.edu.cn working condition is often described by a number of parameters depending on the actual usage of engine, such as the speed. Traditional calibration process is achieved by manually changing the engine parameters, simulating its state under different working conditions, comparing its states on engine maps [23], and repeating the above steps until the states are closed to some desired states. Figure 1 illustrates the manual process of adjusting engine parameters.…”

“…The calibration process is usually time-consuming and highly relies on the human knowledge. Ma et al [22,23] proposed a multiobjective optimisation approach to automate the gasoline direct injection engine calibration process, in which five engine parameters were adjusted under four cases and three evaluation criteria were considered for each case. However, some of the engines, such as aero-engines, are controlled by dozens of parameters.…”

Multi-objective optimization of water injection in spark-ignition engines using the stochastic reactor model with tabulated chemistry

“…They concluded that the multi-objective optimization using neural networks is a promising approach to find optimum solutions for engine valve-timing. Ma et al 25 presented a multi-objective optimization strategy using a mean-value model of a gasoline direct injection engine to predict fuel consumption and soot emissions together with the Strength Pareto Evolutionary Algorithm (SPEA). Their results showed an improvement of 3.2% of the indicated specific fuel consumption (ISFC) with the optimized engine control parameters.…”