Some personal insights on compressor anti-surge control (thanks to everyone for pointing out errors)
Thread Content
This post was last edited by yy1007 on 2010-10-23 at 22:57. Dear colleagues, I hope you can help explain the concept of decoupling. I also ask that you point out any issues present here and share the differences in anti-surge control methods used in various power plants. What follows is merely my personal summary; there are certainly many shortcomings, so I hope you will point them out. Mechanism of compressor anti-surge control: During the design phase, the limit flow rate is determined. In actual operation, standard flow rates are calculated based on actual data such as suction pressure, discharge pressure, suction temperature, discharge temperature, and flow rate, thereby determining the polytropic head and identifying the operating point. The distance between this operating point and both the anti-surge control line and the surge line is then assessed. Between these two lines, there is also a BOOST line; I consider it to be the control line in between. When the system crosses this control line and reaches the BOOST line, it determines that a dangerous condition has occurred, and the BOOST line promptly sends a signal to activate the anti-surge valve. That is, by staying away from the surge control line, when the operating point moves rapidly to the left at a certain speed or faster, the rate control mechanism will adjust the opening degree to prevent surge. Once surge occurs, the system determines the minimum opening degree of the SMP anti-surge valve; it is calculated that if this degree is exceeded, the unit will continue to experience surge. Therefore, this minimum control valve position is set. If the system reaches a stable state, manual reset is required before the anti-surge valve can be closed, and care must be taken to ensure no disturbances occur during this process. To prevent the anti-surge valve from continuously opening and closing in response to even minor changes in the output signal, thereby shortening its lifespan, the anti-surge system includes a function to freeze out such minor movements; when the output signal
Increasing the speed first when raising pressure, and reducing the pressure first when decreasing speed – this can be considered the basic rules for operating centrifugal compressors. However, I disagree with the poster on one point: \"It is not allowed to close the anti-surge valve on one’s own without direct instructions from senior management.\" The senior supervisor cannot always be in the compressor control room. If the focus is solely on keeping the compressor control points away from the surge line, steam consumption will inevitably increase. In my opinion, it should be up to the production team leader to decide, based on the operating conditions, to close the anti-surge valve as much as possible while still preventing surges. It is necessary to consider not only the safe operation of the compressor but also the production costs
Reply to 2# 115115ddd: Oh, I guess I didn’t explain myself clearly. What I mean is that when there’s a failure in the flow transmitter and the flow rate can’t be measured, it might seem as though the flow rate is normal, and one might think it’s okay to shut down the anti-surge valve forcibly. But it’s better not to do that; follow the instructions given. I work in an ethylene plant, so I’m speaking based on the operations of the ethylene units. From personal experience, in state-owned enterprises, it’s best for management to make decisions regarding such incidents. Moreover, based on past experiences, if there’s a problem with the anti-surge valve of those units, senior management will arrive within 10 minutes, without any doubt. “When the flow transmitter fails, the anti-surge valve will open to a very large extent; this situation must be resolved promptly. Out-of-range conditions and instrument failures are common. In such situations, it is not allowed to close the anti-surge valve on one’s own without direct instructions from senior management. ”
This post was last edited by julius6054 on 2012-2-16 at 10:13. Our control system uses the pressure ratio and flow ratio for each section (the measured flow value compared to the maximum allowable flow rate; this maximum value can be adjusted, and it corresponds to the range of the instrument used to measure the anti-surge flow rate in that section. Strictly speaking, the anti-surge flow rate is determined through experimental tests conducted by the manufacturer. However, in practical use, there are significant differences between the conditions indicated by the instruments and the actual operating conditions as well as the experimental conditions. It’s not clear whether the anti-surge control calculation module includes any adjustments to account for these differences. In practice, we have seen cases of manual modifications, which don’t seem quite appropriate. ), is there a minimum anti-surge opening value for the SMP? We have not encountered this issue. As for our control system, as long as it is still on the BOOST line and on the left side, the anti-surge valve will automatically open unless manual control is used; in semi-automatic or fully automatic mode, the anti-surge valve remains open and cannot be closed. Once the operating conditions stabilize, there is no need to reset it. In automatic control mode, the anti-surge valve closes on its own, while in semi-automatic and manual control modes it does not close and remains at the opening level set in manual mode. Personally, I think this control logic is quite good. The minimum anti-surge valve opening is accurate to 0.1%, allowing for smooth operation. As for operation at low load, it mainly depends on the output requirements of the compressor for the subsequent processes (including pressure and flow rate; pressure is particularly important. Theoretically, when operating at low load, the compressor is not at its optimal operating point, and due to the conditions in the subsequent processes, it is necessary to activate anti-surge mechanisms to ensure that the outlet pressure meets the requirements of the production process) ； ““If there is no direct instruction from a senior manager, it is not allowed to close the anti-surge valve on one’s own.” Well, that’s hard to say; it depends on the readings of the instruments and the skill level of the person in charge at that time. It’s not advisable to wait for instructions from a senior manager, but when it comes to matters related to responsibilities, well, it’s tough to say.
Reply to 4# julius6054: After determining this maximum flow rate during the initial tests, the manufacturer still conducts online tests to find the surge limit flow rate once the unit is installed, in order to artificially identify the surge point. The SMP control functions as a kind of safety mechanism; it sets a minimum opening degree for the anti-surge valve when the unit is unstable, thereby preventing periodic openings – surge elimination – closures – and subsequent surging. To prevent this additional element, it seems to be quite effective; the logic you mentioned is quite similar to what we use here. As for the last sentence, regarding accident handling in special situations, one really needs to be cautious; it’s fine if nothing happens, but if an accident does occur, everything will be ruined.
From a safety perspective, it is not allowed to manually reduce the anti-surge valve in case of a flow transmitter failure; the original poster’s view is reasonable. After all, it takes only a few hours to resolve such faults. However, an experienced shift supervisor should be able to make conservative manual adjustments by referring to previous operation records as well as flow meters from other compression sections (which are used to calculate the actual flow rate). I also have a few questions for the original poster: 1. When such a fault occurs, is it necessary to switch to manual mode on-site? 2. When driving in your area originally, was venting allowed solely to prevent stalling? Was the timing of venting determined by on-site operators or mandated by dispatchers?
Reply to 6# 115115ddd: I agree with your view of adopting a conservative approach. 1. When a fault occurs in the flow transmitter, the anti-surge system automatically switches to full manual mode. As an operator, one does not have the authority to switch the anti-surge system to manual mode; only the instrumentation engineer, who possesses a higher-level password, can perform this action. Such operations must be approved by management ； 2. I’m not sure what you mean by \"venting\" – do you mean reducing the back pressure of the system by venting to a flare? During operation, we generally use speed adjustment and anti-surge valves to prevent surge. Although pressure differences before and after have a certain impact on anti-surge protection, they cannot be used as means of adjustment; they serve only as a preventive measure, or their range of effectiveness is very limited.
Reply to 5# yy1007: During the initial testing, online tests were indeed carried out to determine the anti-surge flow rate. However, after a certain period of time, it was necessary to calibrate the measuring instruments on site before reinstalling them, and this process was quite complicated. It was not possible to manually identify the surge point; it was a very time-consuming task. The only options available were to adjust the pressure ratio between the inlet and outlet, as well as to adjust the range of the anti-surge flow meter used in the module for calculating the anti-surge flow rate. After all, the surge line and BOOST line need to be modified, and there are too many aspects involved; it requires personnel from the ITCC manufacturer to handle this, as it is difficult for the instrument technicians at the production facility alone to accomplish it. SMP control is necessary, but it can be improved. Since the operating conditions are stable, there is no need to perform a reset in order to make adjustments; at this point, semi-automatic control for anti-surge protection should be used to prevent the surge valve from opening and closing periodically. Under semi-automatic control, SMP still imposes a requirement for a minimum valve opening degree, and if the flow rate is not sufficient, the valve will continue to be forced open. (Of course, if you manually enter a higher value, such as 100%, it will achieve the anti-surge flow rate more quickly, but it will also cause significant stress on the moving parts at the compressor inlet.) ) Once it operates safely and stably outside the surge zone, the anti-surge valve does not close automatically even when far away from that zone; I think this is much more user-friendly. Generally, a handwheel is provided on the surge valve at the installation site, allowing it to be turned open directly; of course, this is only done as a last resort. Under normal circumstances, there are three control modes available on the control screen: fully automatic, semi-automatic, and manual. The venting process should be ordered by the scheduler, but it is possible to achieve zero venting gradually by adjusting the load of the entire process; this can be done within half an hour. Compared to protecting core equipment, this is still a worthwhile approach. PS: Our turbo compressors are controlled using ITCC, and a common problem is that in emergency situations, some anti-surge valves show as fully open on the control panel, while in reality the valves are fully closed, which is very dangerous. It’s because the valve is stuck, and in such cases the only option is to press the emergency shutdown button of the unit directly.
Reply to 8# julius6054: Initially, the vehicle was tuned by the compressor manufacturer together with the anti-surge software provider and the people in charge of the instrumentation; relying solely on the instrumentation provided by the manufacturing unit would result in a far inferior outcome. Here, there is basically no difference between fully automatic and semi-automatic systems; the semi-automatic version allows for manual output via the human-machine interface (the output must be greater than all other outputs to be valid), and such output can also be generated from the DCS. The anti-surge valve with a handwheel is indeed meant to be used only in extremely urgent situations; under normal circumstances, no one should touch it. I’m not quite sure what you mean by “let your mind wander”; the terms used vary from place to place.
I’m very grateful for everyone’s participation. Since my own points allowance is limited, I hope the moderators of this forum will give more points to those who have taken part. Thank you.
Reply to 9# yy1007: It’s definitely the pressure relief at the front end of the compressor; there’s no need for relief further down the line. Adjustment is certainly necessary. In an emergency, the anti-surge valves are usually opened quickly one after another, from the high-pressure side to the low-pressure side, followed by a reduction in speed. Of course, at this point, the backpressure of the upstream process, that is, the pressure at the compressor inlet, must also be adjusted right away. This is done by adjusting the pressure control valve at the compressor inlet in order to stabilize the pressure in the upstream process. After all, it’s impossible to reduce the load that quickly; it takes a considerable amount of time to do so. Otherwise, the only option would be to shut everything down at once, which would be too extreme.