An adaptive neural network for diesel engines testing

An adaptive neural network for fuzzy inference system (ANFIS) for testing internal combustion engines (ICE) is presented. The proposed ANFIS automatically obtains fuzzy rules for controlling ICE at the testing. Number of layers and theirs neurons was designed. The ANFIS was developed using MathWorks MatLab.Experimental results indicated that ANFIS has satisfactory precision for engine testing.
ADAPTIVE NEURAL NETWORKS FOR DIESEL ENGINE TESTING
The article presents the application of a fuzzy neural network for testing internal combustion engines. During the testing of an internal combustion engine, this network automatically generates the associated fuzzy inference rules. The number of layers and the corresponding number of neurons in each layer were selected experimentally. The network was implemented based on the Matlab package.
The conducted experiments demonstrated satisfactory network performance.
1. INTRODUCTION
Expansion of production of automobiles, tractors and theirs growing role at modern society’s satisfaction of needs brings to persistent upgrade of engines.
Rated engine power, efficiency, toxicity and other properties such as reliability and operating life is assessed from engine test at the stand. At the present time all new produced engines is testing at the stand. The essence, volume and number of tests are defined by their purpose. Engine testing is regulated by state standard specifications.
Tests are the final stage of difficult process of creation and modernization of internal combustion engines. So, all new created, modernized and serial internal combustion engines are putted to various kinds of tests. (…)

An adaptive neural network for diesel engines testing – PDF file for download