Case Study on Optimized Control of an Ultra-Clean, High-Purity Electronic-Grade Reagent Facility at a Certain Factory
Category:
Typical Cases
Ultra‑pure, high‑purity reagents are typically highly corrosive. To ensure that their purity meets specifications, the manufacturing process must be extremely stringent: raw materials require pre‑treatment, equipment and materials must be specially selected, and analytical instruments must offer high precision. Moreover, some measuring instruments cannot be installed because they would suffer corrosion and contaminate the product; consequently, many critical parameters cannot be measured online.
Zhejiang Bangye Technology Co., Ltd. has extensive experience in using VOA, constructing linear and nonlinear mechanistic or empirical models based on measurements of multiple process variables to predict critical parameters. VOA can be calibrated using reliable laboratory analytical data to mitigate model bias and the effects of unmeasurable disturbances.
By leveraging VOA online prediction to provide a basis for real-time critical variables, Zhejiang Bangye Technology has integrated VOA with APC control technology to design and develop an advanced control system for ultra‑clean, high‑purity reagents. After commissioning the APC system, fluctuations in key process parameters have been reduced by more than 35%, product purity and stability have improved, plant material consumption has declined, and operators’ workload has been significantly alleviated.
Figure 1 shows a comparative plot of the tower‑bottom temperature control performance; after implementation, the temperature fluctuations at the tower bottom were significantly reduced.
Advanced control not only ensures stable operation of individual distillation columns but also enables material‑balance control across multiple columns. As shown in Figure 2, when the load of a downstream column is adjusted, the controller promptly modulates the feed and product withdrawal rates of the upstream column, thereby maintaining material balance among the entire set of distillation columns. Given the lengthy process flow of the purification system, implementing advanced control facilitates coordinated material flows into and out of multiple columns, enabling rapid load adjustments and ensuring stable column operation—tasks that are difficult to achieve through manual control.
Figure 2 shows the effectiveness of production target control.
Figure 2 shows the effectiveness of production target control.
Ultra‑pure, high‑purity reagents are typically raw materials for downstream products, and fluctuations in their purity are critical to ensuring the quality of subsequent products. Statistical Process Control (SPC) is an effective quality‑management tool that enables monitoring of the manufacturing process, identifies and mitigates abnormal variations in process parameters, keeps production under control, and achieves quality‑control objectives.
Each point represents the operating condition for a specific time interval; the yellow lines indicate the upper and lower 3σ limits, while the red lines denote the upper and lower specification limits of the process parameters. Users can flexibly configure out-of-control criteria as needed to detect anomalies, trigger alerts, and guide corrective actions.
Bangye’s SPC software has been deployed in real-world manufacturing environments, delivering significant results. Overall, it boasts comprehensive functionality, flexible configuration, and robust reliability: (1) It is suitable for continuous or semi‑continuous processes, allows flexible configuration of start‑up and shutdown criteria, and includes built‑in statistics on start‑up and shutdown durations. (2) It supports online computation and updating of X‑bar charts, enabling real-time monitoring of equipment status. (3) It permits on‑line modification of certain parameters, such as control limits and specification limits. (4) It provides offline statistical analysis capabilities, including X‑bar charts, control charts, normal distribution plots, and process capability index (PCI) charts. (5) It supports online anomaly alarms for SPC. (6) It also enables the export of control charts, process capability charts, and other relevant reports.
Address：19F, Building S3, HeRui Science and Technology Park, No.475, Changhe Road, BInjiang District, Hangzhou, Zhejiang, China