CN-122017477-A - Monitoring analysis method for rapidly and accurately diagnosing titanium alloy ion nitriding arc discharge

Abstract

The invention provides a monitoring analysis method for rapidly and accurately diagnosing titanium alloy ion nitriding arc discharge, and belongs to the technical field of titanium alloy surface strengthening equipment. The arc pre-diagnosis method comprises the steps of collecting electric parameters of arc discharge in ion nitriding, establishing an arc discharge electric parameter database, constructing an arc prediction mixed architecture model, wherein the arc prediction mixed architecture model can identify abnormal modes of the electric parameters related to the arc discharge, at least comprises a convolutional neural network layer, a memory network layer and an attention mechanism layer, training the arc prediction mixed architecture model by using the electric parameter database, and carrying out arc discharge prediction by using the trained arc prediction mixed architecture model in the operation process of the titanium alloy high-frequency ion nitriding furnace. The method can realize the advanced prediction of arc discharge, provide advanced early warning for the inhibition of subsequent arc discharge, and ensure the operation stability of the titanium alloy high-frequency ion nitriding furnace and the surface quality of titanium alloy parts.

Inventors

• YANG XINGKUAN
• ZHEN GUANGCHUAN
• DU ZHIWEI
• PAN BILIN
• SHAO PIYAN
• WU SHAOLIANG

Assignees

• 中国铁道科学研究院集团有限公司金属及化学研究所
• 中国铁道科学研究院集团有限公司

Dates

Publication Date

20260512

Application Date

20251219

Claims (8)

1. 1. A monitoring and analyzing method for rapidly and accurately diagnosing titanium alloy ion nitriding arc discharge is characterized by comprising the following steps of: collecting electric parameters of arc discharge in ion nitriding and establishing an electric parameter database of arc discharge; Constructing an arc prediction hybrid architecture model capable of identifying abnormal patterns of electrical parameters related to arc discharge, the arc prediction hybrid architecture model comprising at least a convolutional neural network layer, a memory network layer and an attention mechanism layer; Training the arc prediction hybrid architecture model using the electrical parameter database; And in the running process of the titanium alloy high-frequency ion nitriding furnace, performing arc discharge prediction by using the trained arc prediction mixed architecture model.
2. 2. The method of monitoring and analyzing for rapid and accurate diagnosis of ion nitriding arc discharge of titanium alloy according to claim 1, wherein the electrical parameters include at least voltage, current density, ion temperature distribution and/or arc morphology.
3. 3. The method for monitoring and analyzing the rapid and accurate diagnosis of the titanium alloy ion nitriding arc discharge according to claim 1, wherein the convolutional neural network layer carries out convolutional operation on the electrical parameters and extracts local spatial features of the electrical parameters, and a feature map sequence is output.
4. 4. A method of monitoring and analyzing for rapid and accurate diagnosis of titanium alloy ion nitriding arc discharge according to claim 3, wherein the electrical parameter is voltage, and the shape and/or duration of the high frequency spike of the voltage is used as the local spatial feature.
5. 5. The method for monitoring and analyzing the rapid and accurate diagnosis of the titanium alloy ion nitriding arc discharge according to claim 4, wherein the preprocessed voltage is converted into a two-dimensional time-frequency diagram, the two-dimensional video diagram is input into the convolutional neural network layer, and the convolutional neural network layer extracts local spatial features of the voltage through a ReLU activation function.
6. 6. A method of monitoring and analyzing for rapid and accurate diagnosis of titanium alloy ion nitriding arc discharge as set forth in claim 3, wherein said local spatial features are input to said memory network layer, said memory network receiving timing dependencies capturing said local spatial features and outputting hidden sequence states.
7. 7. The method of monitoring and analyzing for rapid and accurate diagnosis of titanium alloy ion nitriding arc discharge according to claim 6, wherein hidden sequence states are input into the attention mechanism layer to increase key feature weights associated with arc discharge.
8. 8. A monitoring and analyzing method for rapidly and accurately diagnosing titanium alloy ion nitriding arc discharge according to claim 1, In the running process of the titanium alloy high-frequency ion nitriding furnace, the electric parameters of the titanium alloy high-frequency ion nitriding furnace are monitored in real time; inputting the electric parameters obtained by monitoring into the trained arc light prediction hybrid architecture model; The trained arc prediction hybrid architecture model identifies abnormal patterns associated with an arc discharge and calculates the probability of an arc discharge occurring.

Description

Monitoring analysis method for rapidly and accurately diagnosing titanium alloy ion nitriding arc discharge Technical Field The invention relates to the technical field of titanium alloy surface strengthening equipment, in particular to a monitoring and analyzing method for rapidly and accurately diagnosing titanium alloy ion nitriding arc discharge. Background The titanium alloy is widely applied to the high-end manufacturing fields of aerospace, medical equipment, new energy sources and the like, and the fields have extremely high requirements on the surface quality and performance of parts. The high-frequency ion nitriding furnace for titanium alloy is a core device for strengthening the surface of titanium alloy, nitrogen ions are permeated into the surface layer of titanium alloy through plasmas generated by a high-frequency pulse power supply to form a high-hardness and high-wear-resistance nitriding layer (such as TiN and Ti 2 N), and the wear resistance, corrosion resistance and fatigue strength of the titanium alloy are obviously improved, so that the titanium alloy meets the requirements of high-end manufacturing. Arc discharge (also known as "arcing") is the most common interference problem during operation of titanium alloy high frequency ion nitriding furnaces, and is a critical challenge affecting equipment stability, workpiece quality and production efficiency, with great harm to equipment and processes. Arc discharge is a phenomenon of intense discharge of plasma under abnormal conditions, and can generate instant high temperature and strong electromagnetic pulse up to thousands of degrees celsius. If the inhibition is not performed in time, the workpiece electrode and the furnace body cathode electrode are ablated, insulating parts such as ceramic sleeves and sealing elements are damaged, and even safety accidents such as furnace body short circuit and fire disaster are caused. Oil stains on the surface of a workpiece, point discharge, abrupt change of gas pressure and the like belong to abnormal conditions which cause arc discharge. Meanwhile, the stability of plasma is damaged by arc discharge, so that the concentration of nitrogen ions fluctuates, and the thickness of the nitriding layer is uneven and the hardness gradient is abnormal. For titanium alloy precision parts, the uniformity of the nitriding layer directly determines the fatigue life and biocompatibility of the titanium alloy precision parts. Thus, diagnosis and suppression of arc discharge is required throughout equipment operation and process optimization. Arc diagnostic suppression systems are typically used to reduce the occurrence of arc discharge. The arc diagnosis suppression system realizes effective protection of the core components of the equipment by detecting the arc in real time and rapidly triggering an arc extinguishing mechanism. Arc detection is a precondition for arc extinction, and the core objective is to rapidly capture an arc signal and accurately distinguish between "normal glow" and "abnormal arc". The existing arc light detection technology is mainly used for judging through acousto-optic and electric distinction between glow discharge and arc discharge. Current mainstream technologies for arc detection include ultraviolet imaging sensors, acoustic sensors, and electrical parameter monitors. The ultraviolet imaging sensor achieves high sensitivity detection of the arc (response time <1 ms) by capturing ultraviolet radiation of the arc (wavelength 185-400 nm). The acoustic sensor achieves spatial localization (error <5 mm) by detecting the "snap" sound of the arc (frequency 20-100 kHz). The electric parameter monitor judges the occurrence of arc light by collecting voltage and current fluctuation (such as voltage dip and current surge) of the power supply. The core goal of rapid arc extinction is to rapidly terminate the discharge after the arc light occurs and repair the damage to the surface of the workpiece. The current mainstream technology for rapid arc extinction includes a high-frequency pulse power technology. The high-frequency pulse power supply can inhibit the sustainable development of arc light by rapidly switching on and off current. The existing arc extinguishing technology mainly comprises the modes of series connection of resistors with large resistance, current cut-off negative feedback arc extinguishing, thyristor bypass arc extinguishing, ionic switch arc extinguishing, pulse power supply arc extinguishing and the like. Existing arc diagnostic suppression systems have a significant drawback in that considerable energy is concentrated at a very small point on the workpiece surface within a few microseconds to tens of microseconds of arc detection and eventual extinction. With respect to active metals such as titanium alloys, even very brief strong arcs may leave microscopic lesions or structural changes on the surface, affecting the quality of the titanium alloy. In view of this, the