Study On The Method Of Diesel Engine Fault Diagnosis Based On Ant Clustering Algorithm

The diesel engine, as an important power machinery, is widely used in all fields of national economy. The diesel engine holds the extremely important status in ships as the main power source. The operating condition of diesel engine has a direct impact on the whole equipment. However, due to some characters of diesel, such as numerous components,complicated motion and poor working conditions, the diesel engine is easy to be faulty. Once a failure happening, the running state and the safe operation of the diesel engine will be affected. It would cause heavy economic damages, sometimes resulting in damaging diesel engine accessories straightly and causing casualties. On this condition,it makes a lot of sense to study the technology of diesel engine fault diagnosis.In the paper, fault diagnosis of diesel engine based on the ant colony algorithm is studied and it is also proved through software simulation and experiments.Based on the diesel structure,the diesel multi-cylinder is abstracted to ant swarm,then the paper proposes a novel diesel diagnosing method through comparing the cylinder swarm operation character, based on the ant colony algorithm. A GT-POWER model of diesel engine is established to simulate normal and abnormal working condition. Based on MATLAB, the simulation is clustered through ant colony algorithm. Based on this, the abnormal cylinder will be got, and it is proved that the method advanced in the paper is effective. Cluster-based global outlier factor(CBGOF) is introduced into diesel fault diagnosis research. Then the degradation degree of abnormal cylinder can be represented by calculating the CBGOF under the different faulty degree. Finally, the paper establishes a thorough test bed of diesel fault diagnosis,and shows the hardware devices and software systems. A man-machine interface of operation parameters acquisition is established through LabVIEW. Based on the experimental platform, the experimental program like misfire is developed, and the fault diagnosis experiment like misfire shows that the method advanced in the paper is reasonable and feasible. The paper aims at providing novel diesel fault diagnosis method under the condition of fault sample deficiency.