DCS Alarm Management

There are more process variations, and the alarm values can be revised through such changes

This post was last edited by ZhiGE on 2018-12-7 at 17:08. 1. Alarms need to be properly designed and implemented; by zone (different work sections/postings) and by level (H\M\L), the proportions should be 80%, 15%, and 5% respectively； Classification: process alarms, system alarms, operation records. All control systems have these functions; the key is to implement them properly ； 2. Alarm management: Statistics on alarms, including the total number of alarms per day, the number of repeated alarms, flashing alarms, and long-term alarms. An acceptable number of alerts is one alert per person per shift every ten minutes, with a maximum of 2 alerts; for an 8-hour shift, this corresponds to handling 48–96 alerts ； Due to process fluctuations, improperly set alarm thresholds, instrument failures, and other reasons, the number of alarms is often much higher than the actual frequency of incidents, causing valid alarms to be overwhelmed by invalid ones. 3. Alarm changes: Adjusting the set values due to unreasonable settings, or temporary shutdown of equipment due to failures; 4. Good guidance documents on alarms, such as Annex EEMUA191 and ANSI-ISA-18.2, which provide requirements for the design, management, and implementation of alarm systems ； A good alarm standard is one that can prompt effective alarms in a timely manner, without generating too many of them. It is recommended to analyze the top 10 alarm occurrences per day in order to gradually reduce the number of invalid alarms. An alarm management system can be considered, which can automatically perform alarm analysis.

Sort through them, set the important ones as alerts, and remove those that are completely unimportant. If the critical parameters cannot return to normal immediately, cancel them temporarily or set them to a higher value at any time, and keep a record of it.

Alarms related to the equipment should be resolved promptly. For alarms related to the manufacturing process, an application to change the alarm thresholds should be submitted each time the process sheet is modified; this way, the number of alarms can be reduced gradually

It’s the system after it’s been put into operation; I’m starting to sort things out now. There are still some devices that start and stop intermittently, as well as those with sequential control functions, and it’s difficult to handle the proper alarm responses for them

Why bother with so many gods? It’s either an alarm due to a malfunction in the instrumentation equipment or it’s the operator’s fault; if they find it too much work, they will naturally ask to have it canceled.

Management approach for alarms: 1. Data analysis – Conduct alarm analysis using large amounts of data to identify the top 10 alarms, and address the most critical ones first based on the 28-principle approach； 2. Alarm analysis – Conducts analysis of alarms based on the information displayed by them; it is possible to identify alarms such as oscillations, stale alarms, and flickering alarms, and by considering different types of alarms, the causes behind them can be analyzed ； 3. Alarm sorting – Use filtering, dead zones, and other methods to eliminate alarms ； 4. Alarm classification and grading – Establish level 3 and level 4 alarms as well as different types of alarms, and subsequently set alarm priorities ； 5. Advanced alarm management – Further configure and optimize alarms; do not simply aim to reduce the number of alarms, as it may be necessary to increase the number of alarm points, since the purpose of alarms is to ensure smooth production ； 6. The device features automatic control and smooth operation, ensuring stable, efficient, and high-quality performance. As long as the operational procedures remain within the defined process parameters and alarm limits, there will be fewer alarms. To achieve this, it is necessary to have a good alarm management system, as well as a team that has the right strategies and the ability to implement them. The alarm systems provided by Beijing Jialian Youkong’s Baojing Wang can help solve the above problems www.baojingwang.com. You can contact them by phone at 13911105995.

1. First, conduct alarm statistical analysis. Classification of process alarms, equipment alarms, and system alarms. 2. Evaluate the Rationality of process alarms; cancel unnecessary alarms and adjust unreasonable alarm values. For batch production, conditional alarms should be implemented (alarms that are activated under certain conditions). This requires a detailed modification plan from the manufacturing process; careful work yields high quality results. 3. Equipment alarms: Determine whether it is a genuine equipment alarm; handle any faults. Normal start-up and shutdown of the pump do not require alarms. 4. System alarms are categorized into network and hardware issues; it is necessary to determine exactly what the problem is, as an incident of this kind is serious. Redundant networks cannot be switched over; issues such as controller problems, card problems, and problems with power supplies/fans/lights need to be addressed promptly.

Whether it’s processes, systems, or mechanical and electrical equipment, there’s no need to worry about too many alarms. All that’s required is proper system maintenance. Based on the actual needs on site, in the directory of historical alarm records, my approach has been to retain historical data for one month; all data from more than a month ago is copied and stored on disk. It is then archived using a CD burner, allowing for organized management of all process operation records and historical alarm data since the start of testing. It doesn’t matter how much.