RESUMEN
After injection molding, plastic gears often exhibit surface defects, including those on end faces and tooth surfaces. These defects encompass a wide range of types and possess complex characteristics, which pose challenges for inspection. Current visual inspection systems for plastic gears suffer from limitations such as single-category defect inspection and low accuracy. There is an urgent industry need for a comprehensive and accurate method and system for inspecting defects on plastic gears, with improved inspection capability and higher accuracy. This paper presents an intelligent inspection algorithm network for plastic gear defects (PGD-net), which effectively captures subtle defect features at arbitrary locations on the surface compared to other models. An adaptive sample weighting method is proposed and integrated into an improved Focal-IoU loss function to address the issue of low inspection accuracy caused by imbalanced defect dataset distributions, thus enhancing the regression accuracy for difficult defect categories. CoordConv layers are incorporated into each inspection head to improve the model's generalization capability. Furthermore, a dataset of plastic gear surface defects comprising 16 types of defects is constructed, and our algorithm is trained and tested on this dataset. The PGD-net achieves a comprehensive mean average precision (mAP) value of 95.6% for the 16 defect types. Additionally, an online inspection system is developed based on the PGD-net algorithm, which can be integrated with plastic gear production lines to achieve online full inspection and automatic sorting of plastic gear defects. The entire system has been successfully applied in plastic gear production lines, conducting daily inspections of over 60,000 gears.
RESUMEN
An approach for improving and maintaining a consistent weld quality of the deposited material during the FFF printing process is proposed. The approach is based on the analysis of the printing process thermal cycle and the real-time nozzle temperature control. The mathematical model of the FFF printing process has been developed with the use of real-time control in the algorithm of numerical implementation. The successful solution of the thermal conductivity problem made it possible to determine segment-wise heating settings for use during the printing process, resulting in a high and stable quality of welding. Comparison of the results of modeling with other well-known mathematical models of the FFF printing process and experimental results showed the adequacy of the proposed model. A maximum deviation of 17.7% between the simulation results and the thermography data was observed. The proposed model was verified using rectangular 3D polylactide shapes printed with and without regulation of the power of the heat source according to the previously estimated settings. The overall quality of regulation, stability of the system, and the PI coefficients of the controller were evaluated using a simulated model of the control system. The results of the experiment fully correspond with the modeling results.
RESUMEN
In order to improve plant factories, an appropriate control system on fertilization is urgently required. An automatic management system to control nutrient concentration was constructed using a programmable logic controller (PLC) and ion selective electrodes (ISEs) of nitrate, phosphate, and potassium ion. The concentration of nutrient components in a culture solution was monitored using these ISEs. When the concentration of the nutrient components diminished to the threshold set as an optimum condition (0.1 - 2.0 mM), an appropriate amount of a concentrated solution of each nutrient component was added to the culture solution using solenoid valves connected with the PLC. The present cultivation system was simply constructed without any computers and pumps. Three kinds of automatic control systems simultaneously worked and did not influence each other.
RESUMEN
Our research was designed for on-line detection of multi-index in the concentration process of Ganmaoling granules by integration of near infrared spectroscopy and automatic control system. First, on-line detection system was set up in the concentration tank for Ganmaoling granules production. Spectra were scanned and values of chlorogenic acid, linarin, solid content and relative density were measured. Models of partial least squares regression were built and imported into near infrared workstation. By connecting the control system, real-time multi-index values were determined automatically in the concentration process. Results showed that correlation coefficients of chlorogenic acid, linarin, solid content and relative density models were 0.963, 0.989, 0.993 and 0.918, respectively. Relative standard errors of prediction were 3.71%, 4.28%, 4.17% and 0.24%, respectively, indicating a good performance and high accuracy of the models. Real-time data collection during the whole process was measured by the near infrared detecting system in the control system. In conclusion, the near infrared detection system is able to perform real-time automatic determination of multi-index in the concentration process of Ganmaoling granules with significant advantages.
RESUMEN
Transducer is applied to automatic control system extensively,but it causes certain electromagnetic interference to control circuit and electricity equipments around it and sometimes may result in wrong action.This paper carries on the experimental analysis and formulates some anti-disturbance measures.