The Fault Diagnosis Of Argon Bottom Blowing System In Steel Making Process

Argon bottom blowing technology has become one of the most critical processes in refining process of LF and the key role to improve the quality of the products.However, due to the complex and bad working conditions on site, Argon bottom blowing system is very easy to produce unstable factors and some faults which directly influence the production rhythm of steelmaking process and quality of steel. Find faults and promptly eliminate them, to reduce loss of enterprises and improve product quality have an important significance.Aiming at the faults of vent brick jams and gas leak, the fault diagnosis mechanism model is established in conjunction with the relevant knowledge of the fluid mechanics on the basis of analyzing Argon bottom blowing system. The mechanism model describes the mathematical relationships between the open area of bottom blowing argon and the main test parameters in actual system. By analyzing the model, influential factors and ascertainment method of main parameters in the model are got.Owing to that many parameters can’t be got or can’t be accurate got, fault diagnosis method based on the mechanism model can’t be used in Argon bottom blowing system. One fault diagnosis method based on neural network has an extensive application, because it needs no precise mathematical model. Inlet pressure, outlet pressure, quantity of flow and valve opening are elected as the input, and seven kinds of fault state describing the open area of bottom blowing argon as the output. Then the fault diagnosis model based on BP neural network is designed and the suitable network structure and the training method are chosen. The simulation results show that accuracy of the fault diagnosis based on BP neural network is higher and it has obtained a better effect in the Argon bottom blowing system.