Thread Content
Our plant has recently experienced sudden crashes of the DCS operating system – all computers showed a frozen state. Since this happened suddenly and there was no relevant information available, we handled it by shutting down the equipment urgently, and no accidents occurred. But what should we do if a similar situation arises again in the future? Our plants are of the typical pressure and catalytic cracking type. I hope those with more experience can offer some guidance. Thank you! :handshake
The prize is quite substantial. If this happens again, the only option is to perform an emergency shutdown on-site. There’s a topic here about what to do when the DCS screen goes black; many people have participated in the discussions. Go and take a look, then prepare a document based on the actual situation outlining how to handle similar issues in the future. http://bbs.hcbbs.com/thread-78909-1-2.html
There should be an immediate shutdown, as the device is out of control; it’s similar to a black screen situation, and there’s a dedicated thread for discussing this
Let me say it; I’m not sure if it’s correct. Has there been a disruption in the network communication between your control station and operation stations (such as a power outage), preventing the operation stations from receiving data? I wonder if you have noticed whether the control valves on site are moving. If the operation stations freeze while the control station is working fine, then all the control valves on site will remain at their original setting, and automatic control will continue as before. If the entire control station freezes, all control valves will not receive signals and will open or close completely. Some DCS companies believe that if only the operator station freezes, it will not have a significant impact on production in the short term; however, the cause must be identified as soon as possible, otherwise an emergency shutdown will be necessary.
Regarding the issues with DCS controllers, the design should include redundancy so that if one unit fails, another takes over automatically; therefore, such situations are rare. Check to see if both units in your cabinet have failed
A simultaneous crash like this is quite rare; it’s probably not a problem with the computer! It is recommended to find out the cause. Also, in such situations, the only option is to perform an emergency stop. :(
In our plant, the DCS crashed simultaneously; all control valves failed to operate, the temperature readings remained fixed, and it was not possible to switch to the backup system. The air pressure, power supply, and signal lines were all normal
There’s no other choice – emergency stop! This situation is extremely rare; the cause must be identified, otherwise it will happen again!
This has also happened in our company; generally, it’s due to power supply issues. It is recommended to use a voltage stabilizer and a USB protection device
I think it’s overloaded and needs an upgrade.
In the event of a sudden shutdown, those on site must remain at their posts, carefully monitor the instruments available there, and promptly report any data to the control room. The control room will then issue instructions for adjustments to be made on site; it is essential not to panic
In fact, a DCS crash is similar to a DCS screen going black, and the solutions are more or less the same.
Then it’s necessary to determine whether the cause of the crash is a problem with your computer hardware, or a conflict between the DCS system and the operating system. By the way, I was wondering if your DCS system is developed in-house, or do you use external professional systems such as those from Hilscher or Zhejiang University Control System?
The situation you mentioned could be due to one of the following: 1: There is an interruption in the communication between the system hardware and the display screen. In such cases, the display screen might not function, but it could still be possible to operate the system; production processes can continue as usual. In this situation, you need to call a technician to handle it. 2: There is a fault with the system hardware, such as problems with communication cards or controllers. Again, you should call a technician to diagnose the issue and restore normal operation of the system.
In situations like this... if it’s the control unit that has stopped working, there’s no need to think about anything else; an emergency shutdown should be carried out immediately. If it’s not the control station that’s faulty, then it depends on how quickly the operation station can return to normal operation. If it takes too long, it’s better to shut down the system – safety comes first.
That’s not objective; our combined systems use ABB, and there has never been any situation where the entire system came down
Please provide the following information; otherwise everyone will keep guessing, which is of no help in actually solving the problem. Also, looking at replies is just a waste of time – I feel sorry for that! 1. DCS model, brand. Architecture setup: Clearly describe the hardware aspects, including the power supply arrangement (presence of UPS or not), the composition of controllers (with or without redundancy), network settings, gateway configuration, and the status of monitors, etc. 2. The approximate data capacity of your factory’s control system – in other words, how many energy-saving functions, AO/AI/DO/DI/PID points, etc., it has. Are there any complex processing units (which consume CPU resources)? 3. What is the CPU usage rate of the host under normal conditions? 4. Provide, as much as possible, the solutions proposed by subsequent DCS manufacturers (for such critical equipment, technical cooperation agreements are definitely concluded with professional manufacturers). ***TO ALL: Recommendation 1: One must never rely solely on the DCS system for security purposes. During design: It is essential to consider the safe status of various valves and equipment in the event of a DCS failure. For example, in the case of a exothermic reaction vessel, when monitoring is not possible, we should keep the cooling valves fully open and the heating valves fully closed. Different valves can be selected to meet the requirements. This point must be considered when designing the factory. If it was missed, it’s also a good idea to review it before an accident occurs. When in use: Be sure to prepare contingency plans for accidents: plans for equipment failures, emergency startup procedures, plans for quickly clearing the production line, and so on… Otherwise, in the event of an accident, panic will set in. Routine maintenance: Necessary daily inspections are a prerequisite for ensuring safety. For automated equipment, the use of various electronic components and computing units involves a lifespan factor; this aspect can be determined quickly by consulting specialized manufacturers. Of course, daily maintenance should never be entirely left to the contract manufacturer; one’s own employees must understand it. 60~70%, the supervisor should understand 40%. 2. The use of DCS saves a great deal of labor and improves production efficiency. But never rely solely on the displayed data; it is essential to have appropriate display devices on site. As the saying goes: tradition is also classic. For everyone’s reference! When a DCS failure occurs, the instruments on site can help you make a judgment. 3. The design of the DCS-controlled process, the engineering work, the control personnel, and the production staff form a cohesive unit; it must not be entrusted to a single party to carry out independently. Remember!
In the event of a sudden shutdown: 1. The operators on site must remain at their posts, carefully monitor the instruments on site, and promptly report any data information to the control room. The control room will then issue operational instructions, which will be used to make adjustments on site; it is essential not to panic. 2. If there is a fault with the system’s hardware, such as communication cards or controllers, contact an instrumentation engineer immediately to handle the issue. 3. If the system cannot be restored within a short time despite attempts to fix it, proceed with cutting off power to the system. Our facility once experienced a screen blackout that lasted 2.5 hours; operations continued despite this, but it was not possible to restore normal operation for some time due to issues handled by the instrumentation engineers, so cutting operations had to be carried out. Everyone was drenched in sweat. It was finally determined that there was a problem with the DCs hardware, so the manufacturer was contacted to handle it.
Reply to Post 1#: 1. In the event of a DCS shutdown, the control valve positions of various control systems outside the unit should be at their normal operating settings. First, it is necessary to contact the instrumentation technicians to shut down the computer, restart it, and resume production. 2. If the above solution doesn’t work, proceed with an emergency shutdown