Research On Processing Quality Inspection Of Disc Turbine Fir-Tree Slots

With the rapid growth of aircraft development,manufacturing efficiency is seen as an increasingly important topic.The engine is one of the most important parts of the production process.Low-pressure turbine engine disks are an important component of aircraft engines.There are a lot of slots in each turbine disk that have striking similarities to the fir-tree,known as fir-tree slot.There are dozens of fir-tree slots on some roulettes,and even more than 200.Processing fir-tree slot is a challenge for jet engine manufacturers.Modern jet engines can have up to 40 or more turbine disks.There are many difficulties in the process of cutting the eucalyptus trough on the disc,for example the geometry of the fir-tree slot is complex and the dimensional tolerances can be compared to the diameter of the human hair.At the same time,the workpiece material is very hard but easily damaged.After processing,the surface treatment must meet strict specifications.Since each disk has a plurality of available slots,the manufacturing method must be efficient and efficient.Productivity is an important factor in the total cost of jet engine production.For decades,jet engine manufacturers have been investigating this type of operation and are always looking for cutting processes that improve quality.Nowadays,with the advantages of high precision and low cost,Wire Electrical Discharge Machining(WEDM)has gradually replaced broaching as the best way to process fir-tree slot.However,in order to optimize the application of the Wire Electrical Discharge Machining process in the production of fir-tree slot,there are still important problems that need to be optimized and solved during the processing.One of the most advanced research topics is the on-line inspection of processing quality during the cutting process of the fir-tree slot.Traceability during WEDM is difficult to achieve.The commonly used machining quality inspections are generally performed by a coordinate measuring machine after machining.However,this method is very time consuming and causes an increase in cost.Nowadays,with the development of machine learning,this paper puts forward some related testing ideas and related experiments.It applies two kinds of machine learning methods,unsupervised learning and deep learning.Through Python programming,the experiment realizes the initial machining tolerance detection and makes the processing quality.It provide the possibility of inspection of processing quality during the cutting process of the fir-tree slot.