Research on Optimization Method of Combination Weigher Weighing Module
In industries such as agriculture, combination weigher and food, the traditional single-scale weighing has simple structure and single function, which cannot realize personalized quantitative weighing. The appearance of the combination weigher speeds up the process of quantitative weighing and improves production efficiency. However, during the working process of the equipment, it has been accompanied by actions such as the falling impact of the weight and the closing of the hopper switch. The weighing signal contains noise. In order to solve the noise problem, the weighing module has been partially optimized in the industry. Among them, high-digit AD conversion chips are mostly used in hardware, and a high-precision voltage regulator chip is connected to the reference voltage pin, or a higher frequency MCU chip is used. These methods increase the cost of module production and are not conducive to production and promotion. In software filtering, the de-extreme average method is mostly used to process the weighted data, and since this method is based on the noisy signal, the calculated weight will inevitably be biased. Therefore, it is necessary to optimize the hardware at low cost, and use more complex algorithms to filter out the noise to ensure that the data involved in the calculation is the useful data after the noise is removed. At present, the commonly used signal denoising methods are mostly analysis algorithms such as Fourier transform and Kalman filtering. However, the above algorithms are more effective in filtering stationary signals. When the signal fluctuates greatly, the functions of traditional denoising algorithms are difficult to match. [4]. The moment the heavy object hits, the weight value changes suddenly in a very short time, and the Fourier transform must decompose a lot of sinusoidal waveforms to fit. The effect is not good, and the time is relatively long. Therefore, a filtering algorithm that can filter out dynamic noise is required: Xu Xinyin et al. [5] used the QRD-LSL adaptive filter algorithm to control the vibration effect of the overall structure, but could not control the order of the filter to deal with different environments, and could not filter the noise signal generated by the temperature drift and zero drift of the hardware circuit. . Et al. proposed a filtering algorithm based on wavelet filtering, which can effectively filter the noise of the weighing system, but the specific implementation and application still need to be studied. Zhang Xiliang et al. [7] used discrete wavelet transform to process dynamic weighing signals, and compared the filtering effects of fast Fourier transform and discrete wavelet transform, and determined the effectiveness of discrete wavelet transform for filtering processing. The wavelet transform method of multi-head combination weigher is a time-frequency analysis method in which both the time domain and the frequency domain window form can be changed. It has higher time resolution and lower frequency resolution in the high frequency stage, and is especially suitable for processing noisy signals. Medium and high frequency noise. In this paper, aiming at the weighing accuracy of industrial combination weighers, a new combination weigher weighing module based on wavelet denoising method is developed, and the weighing performance of this module is tested and analyzed on site.