基于快速数字信号处理的电火花加工放电状态识别技术

发布时间：2018-10-10 20:09

【摘要】：电火花加工因其适用于任何难切削导电材料且无明显宏观作用力的优势而被广泛应用。但是电火花加工频率高、随机性强的特点严重影响了放电状态的识别精度,而电火花加工过程的伺服控制系统是以放电状态识别结果为依据,所以准确高效的放电状态识别系统是实现电火花高精度、高效率加工的基础。因此,对高精度、高效率的电火花加工放电状态识别方法的研究有重要意义。本文以数据采集卡和计算机为基础建立了间隙状态信号的高速采集系统,通过Labview软件实现数据采集卡与计算机的交互,从而实现电压信号和电流信号的同步采集,为电火花加工放电状态的识别奠定基础。针对电火花加工信号不稳定的特点,提出了一种基于小波变换的电火花加工放电状态识别方法。首先通过实验的方法确定最优分解层数,然后通过对影响小波变换结果的因素进行分析,确定根据脉宽的大小设置阈值的方法,从而利用阈值对开路状态、火花放电状态、稳定电弧状态、短路状态进行区分。将此方法与高速采集系统结合,实现放电状态的实时检测。对电火花加工放电状态的特性进行研究,分析不同放电状态伏安特性和频率特性的区别。同时对维持电压的形成和维持进行分析,并对火花放电状态与电弧放电状态作对比,从而为放电状态识别提供依据。由于实验条件会对放电状态特性产生影响,因此对不同实验参数下的频率特性进行研究。通过改变开路电压、脉宽、工件材料、电极材料等实验参数进行单因素分析,确定影响放电状态特性的参数。通过设置间隙电压阈值、间隙电流阈值和频谱阈值的方法,实现开路状态、火花放电状态、稳定电弧状态、短路状态以及脉冲间隔的准确有效识别。最后与高速采集系统相结合,通过实验验证此放电状态识别方法的可行性。
[Abstract]:EDM is widely used because it is suitable for any difficult cutting conductive material and has no obvious macroscopic force. However, the high frequency and strong randomness of EDM seriously affect the accuracy of EDM, and the servo control system of EDM is based on the result of EDM. Therefore, accurate and efficient discharge state recognition system is the basis of high precision and high efficiency machining of EDM. Therefore, it is of great significance to study the high precision and high efficiency EDM discharge state recognition method. On the basis of data acquisition card and computer, a high speed acquisition system of gap state signal is established in this paper. The interaction between data acquisition card and computer is realized by Labview software, and the synchronous acquisition of voltage signal and current signal is realized. It lays a foundation for the recognition of discharge state in EDM. In view of the unstable characteristics of EDM signals, a method of EDM discharge state recognition based on wavelet transform is proposed. First, the optimal decomposition layer number is determined by experimental method, then the method of setting threshold according to the magnitude of pulse width is determined by analyzing the factors that affect the result of wavelet transform, so that the open circuit state and spark discharge state can be set by the threshold value. The arc state is stabilized and the short circuit state is distinguished. The method is combined with high-speed acquisition system to realize real-time detection of discharge state. The characteristics of discharge state of EDM are studied, and the differences of volt-ampere characteristics and frequency characteristics of different discharge states are analyzed. At the same time, the formation and maintenance of the maintenance voltage are analyzed, and the spark discharge state is compared with the arc discharge state, which provides the basis for the recognition of the discharge state. Because the experimental conditions will affect the discharge state characteristics, the frequency characteristics under different experimental parameters are studied. By changing the open-circuit voltage, pulse width, workpiece material, electrode material and other experimental parameters, single factor analysis is carried out to determine the parameters that affect the discharge state characteristics. By setting gap voltage threshold, gap current threshold and spectrum threshold, accurate and effective recognition of open circuit state, spark discharge state, stable arc state, short circuit state and pulse interval is realized. Finally, the feasibility of this method is verified by experiments combined with high speed acquisition system.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TG661

【参考文献】