自动化 自动控制 电气自动化毕业设计外文翻译.doc

毕 业 设 计外 文 文 献 翻 译学 号： 姓 名： 所在院系： 专业班级： 指导教师： 原文标题： Power-plant Control and Instrumentation - The Control of Boilers and HRSG Systems 2009年 4月 日译文： 给水控制和仪器仪表6.1给水控制原则控制给水控制系统的宗旨似乎很简单： 它是提供足够的水产生于锅炉匹配的挥发率。 但正像普通锅炉的情况，完成这个任务是一件很复杂的事。 有困难甚而在做控制系统取决于的基本的鼓筒水位测量。控制系统的设计是研究发生在锅炉系统之内和由实际的许多互作用，其中一些的互作用是在锅炉的装载范围的各种各样的点。控制系统设计师的任务是开发一个方案,提供足够的控制下的最大可行的操作情况,这样做,既安全合算。 要做到这些必须了解给水和蒸汽系统的详细的机制和充分地运作要求。在几乎所有的最小、最简单的锅炉,每个相关因素必须考虑,这是不够的,依靠简单的反应似乎是这三个参数相关的给水： 蒸流程、给水流程和水鼓的级别。6.2第一，二和三元素控制 一级水鼓提供了一个直接含水的锅炉。如果流进104发电厂控制和仪器仪表系统的水大于质量流量蒸汽，则水鼓中的水位将上升。相反，如果蒸汽输出大于饲料流入，水位也会下降。 如第2章，鼓的目的不仅是分开的水蒸汽，而且还提供一个储存库，使短期之间的不平衡饲料供水和蒸汽生产要处理无风险的植物。随着水位的鼓上升，风险增加的水结转到蒸汽电路。结果这种结转可以是灾难性的：冷水管道热冲击会导致极端的和局部应力金属，反过来说，如果水位下降，有可能锅炉损坏，部分原因损失的基本冷却炉水墙。 因此,给水控制系统的目标是要保持鼓中的水位大约在船的中点。鉴于这一目标，最简单的解决办法似乎是衡量一级的水鼓和调整给水，以保持这一价值的理想-给水更多的进入鼓中，如果水位上升，则进鼓的水位下降。不幸的是，水位受瞬态变化压力的影响，鼓和责任感在不同的水平不一定是相关的，其中饲料的流动必须调整。换句话说，假设它是不够的，只是因为水位提高给水流量必须降低，反之亦然。 这个奇怪的情况是由于膨胀和缩水影响的 。沸水包括大量的流体含有许多蒸汽气泡，随着沸腾速率上升泡沫生成数量也不断增加。水和泡沫的混合物类似于泡沫，它的体积取决于它所占领双方的水量和蒸汽气泡的数额。如果系统内的压力降低，饱和温度也降低，因此沸腾率增加（因为温度较高的混合物,现在是直接关系到饱和度温度比原来的压力变化） 。随着沸腾率的增加，水的密度减小，但由于大量的水蒸气和水并没有改变下跌的密度必须辅之以数量增加的混合物。 通过这种原理使鼓中水位上升的现象称为膨胀 。上升的水位是误导性的：它不是一个真正的指示性的增加，这就需要的减少供水量维持现状。事实上，如果下降的压力，是由于蒸汽的需求突然增加，水的供应需要增加，以配合增加蒸汽流量。 收缩是相反的膨胀：它发生在压力上升过程中。该机制是完全一样的，对于膨胀，但在相反方向。收缩导致水位在鼓下降时，蒸汽流量减少，并再次提供水的锅炉必须与实际需要，而不是可能误导性的说明所提供的鼓级别发射机。如果一个缓慢变化的蒸汽流量时，一切都很好，因为系统内的压力可以得到控制。这是当快速蒸汽流量变化的情况，即出现问题，因为，由于膨胀或收缩，对汽包水位提供了相反的迹象表明，对水的需求。后突然增加，蒸汽的需求，从而导致压力下降（因此，汽包水位上升） ，一个简单的一级控制器将作出反应，减少了流通给水。同样，突然下降，蒸汽流量，这将是伴随着上升的压力和随之而来的下降，汽包水位，将导致一级控制器增加水流。这两项行动，当然，在不正确的意义。膨胀的影响和收缩，除了正在确定的变化率的压力，还取决于政府的相对规模的鼓和压力，其运作。如果该卷的鼓大，涉及的数量，整个系统的影响将小于其他。如果系统压力低的影响将大于一个锅炉运行在更高的压力，因为影响的某一特定压力变化对密度的水将有更大的中低压锅炉将比如果同一压力的变化是发生在锅炉运行在更高的压力。面对这种情况，设计的控制系统已作出反应，实施了各种解决方案。其中最简单的方法是两个元素的系统，因为它是基于使用两个过程测量代替单一鼓级别测量上面使用。6.2.1二元给水控制记住的基本要求的给水控制系统，是维持一个恒定的水量锅炉，很明显，一个途径解决这一问题将是保持水流到系统中，价值匹配流动的蒸汽了。其中一个版本的这个系统是如图6.1 。在这里，流控制的一个容易确认的设备，一个阀门。我们将看看阀门更深入以后，但目前承担的版本中所使用图表率保持水流通过阀门时这个数字是成正比的需求信号控制器（即如果需求信号不同线性从0到100 ，流速也呈线性变化介于0和100 ） 。这种阀是说有一个线性特点和在系统中显示这是一起雇用发射机产生的信号比例蒸汽流量。一起使用，这两个器件参数保持在步骤。如果发射机产生的信号等于蒸汽流量在所有负载，如果流经阀是符合这个信号在每一个点的流动范围，控制器增益的团结将确保在整个动态范围该系统，水流将永远是平等的，以流动的蒸汽。 当然，规模因素的发射器和阀门，必须考虑到。如果一系列的流量变送器不同于阀的流量控制范围内，控制器增益将需要作相应调整，并在实际系统，这是永远需要的。为了提供足够的业务利润的信心，各种控制阀的设计总是大于流动范围锅炉。例如，在锅炉生产20公斤/ s的蒸汽阀门可能是中小企业提供二十二公斤/ s的水时，它是完全开放。在这个例子中，以线性阀的特点，开放约91 将需要通过一个流动的20公斤/秒。在这种情况下，如果蒸汽流量变送器生产的产量， 100 在20公斤/ s的流量，控制器增益必须是这样的，一个测量值的100 产生最大输出功率为91 。这是一个比例带110 （即增益20/22 ） ，如果这个增益分配给控制器饲料流将符合蒸汽流量的整个范围内的锅炉负荷（假设阀特点是直线，即输出的流量变送器是4毫安在零流量，而且零水流发生在阀信号为0 ） 。这个系统的问题是，它不仅符合蒸汽和饲料流率。如果在一开始，汽包水位低于所期望的价值，这就是它会留下来，因为如果一切设置正确的馈入锅炉将始终符合蒸汽流出，并没有任何机制介绍了小过多的饲料的蒸汽，或略有赤字，这是需要正确的鼓级别的错误。重要的是要考虑的现实的，会发生什么事情了，如果不被正确设置。在这种情况下，如果有一个小设置错误控制器增益，或如果饲料阀通行证或多或少水比它应该在考虑开放，或者在蒸汽流量变送器是尚未标定，将在汽包水位整合上涨或下跌的速度在确定规模的错误，没有将正确的这一不良的状况。换句话说，该系统无法正确的汽包水位如果此参数偏离理想的价值，可能是因为最初的错误，或者因为小错误在蒸汽流量测量或非线性阀之间的需求和实际的流量通过。在上述的例子，准确获得需要的是0.909 09 .因此，如果控制器增益被设定为0.91如上文建议，饲料，水流量将略有大于蒸汽流量和汽包水位将逐渐增加。为了对付这些影响，有必要增加一个反馈的内容，组成另一个控制器将采取行动，纠正任何不匹配。 108电厂控制和仪器仪表之间的实际和理想的鼓水平。图6.2显示一个不同的这样一个两个元素系统。 在这样一个系统，因为鼓数量和蒸汽和饲料流动形成一个综合系统，汽包水位将任何steam-flow/feed-flow不匹配，这是不必要的雇用额外的集成功能的控制器。因此，鼓级别控制器应的比例只有品种。正确的增益此控制器可确定从知识的膨胀和收缩的影响范围内的锅炉。如果这些是不知道他们可以决定的考验。一个适当的考验将是改变蒸汽流量，尽快通过，也就是说， 10 的最高蒸发量的锅炉，同时保持饲料流步骤的蒸汽流量。 （这可以通过连接供稿流信号进入系统，而暂时禁用鼓级别控制器） 。看看上的信息锅炉膨胀特性可用来帮助控制器调整，让我们来看看两个要素的制度的范围蒸汽流量变送器是介于上述（ 0-20公斤/ s ）和饲料阀是中小企业提供再次二二公斤/秒时，它是100 开放。假设鼓级别发射机是介于生产4毫安时水位已下降低于设定的250毫米，这是20毫安时的水平是250毫米以上的设定（即射程为500毫米） 。最后，假设测试如上所述确定，膨胀导致的突然变化， 10 的蒸汽流量提高了汽包水位由80毫米。如果鼓级别控制器是完全抵消影响的膨胀，它必须产生一个输出，取消了一步改变蒸汽流量，这是10 。该控制器的输出，因此必须改变10 时，输入错误的变化16 ，这意味着必须获得0.625 （ 10 + 16 ） 。当蒸汽流量和压力的变化外和解和水位已经恢复到设定的水平控制器输出将再次成为零。阀门打开，然后恢复到跟踪缓慢变化的蒸汽流量，如前所述。这种分析取决于影响正在不断膨胀的锅炉负载范围内，这可能是也可能不是真实的，但它提供了一个切实可行的方法，调整这种类型的系统，并能产生较好表现了广泛的条件。从理论上讲，更好的结果可以得到进行测试，以确定肿胀的影响在不同的负载范围内，并采用一种非线性函数的水平控制器，以弥补分歧的范围。但是，这是相当复杂的基本上是一个简单的系统，并在任何情况下，表现很可能受到限制的其他严重缺陷的制度，我们应审查在下一节中，其中讨论了更全面的系统，被称为三元素控制。原文： Feed-water control and instrumentation6.1 The principles of feed-water controlThe objective of a feed-water control system may seem simple: it is to supply enough water to the boiler to match the evaporation rate. But as is so often the case with boilers, this turns out to be a surprisingly complex mission to accomplish. There are difficulties even in making the basic drum-level measurement on which the control system depends. The design of the control system is then further complicated by the many interactions that occur within the boiler system and by the fact that the effects of some of these interactions are greater or smaller at various points in the boilers load range.The control-system designers task is to develop a scheme that provides adequate control under the widest practicable range of operational conditions,and to do so in a manner that is both safe and cost-effective. To do this it is necessary to understand the detailed mechanisms of the feed-water and steam systems and to be fully aware of the operational requirements.In all but the smallest and simplest boilers, each of the interrelated factors has to be taken into account, and it is insufficient to rely on simple responses to the three parameters which seem to be relevant to the supply of feed water: steam flow, feed-water flow and the level of water in the drum.6.2 One, two and three-element control The level of water in the drum provides an immediate indication of the water contained by the boiler. If the mass flow of water into the system is 104 Power-plant control and instrumentation greater than the mass flow of steam out of it, the level of water in the drum will rise. Conversely, if the steam output is greater than the feed inflow, the level will fall. As stated in Chapter 2, the purpose of the drum is not only to separate the steam from the water but also to provide a storage reservoir that allows short-term imbalances between feed-water supply and steam production to be handled without risk to the plant. As the level of water in the drum rises, the risk increases of water being carried over into the steam circuits. The results of such carry-over can be catastrophic: cool water impinging on hot pipework will cause extreme and localised stresses in the metal and, conversely, if the level of water falls there is a possibility of the boiler being damaged, partly because of the loss of essential cooling of the furnace water-walls. Therefore, the target of the feed-water control system is to keep the level of water in the drum at approximately the midpoint of the vessel.Given this objective, it would appear that the simplest solution would appear to be to measure the level of water in the drum and to adjust the delivery of water to keep this at the desired value-feeding more water into the drum if the level is falling, and less if the level is rising.Unfortunately, the level of water is affected by transient changes of the pressure within the drum and the sense in which the level varies is not necessarily related to the sense in which the feed flow must be adjusted. In other words, it is not sufficient to assume that simply because the level is increasing the feed-water flow must be decreased, and vice versa. This strange situation is due to effects known as swell and shrinkage. Boiling water comprises a turbulent mass of fluid containing many steam bubbles, and as the boiling rate increases the quantity of bubbles that is generated also increases. The mixture of water and bubbles resembles foam, and the volume it occupies is dictated both by the quantity of water and by the amount of the steam bubbles within it. If the pressure within the system is decreased, the saturation temperature is also lowered and the boiling rate therefore increases(because the temperature of the mixture is now higher in relation to the saturation temperature than it was before the pressure change occurred). As the boiling rate increases, the density of the water decreases, but since the mass of steam and water has not changed the decrease in density must be accompanied by an increase in the volume of the mixture. By this mechanism the level of water in the drum appears to rise, a phenomenon referred to as swell. The rise of level is misleading: it is not indicative of a real increase in the mass of water in the system, which would require the supply of water to be cut back to maintain the status quo. In fact, if the drop in pressure is the result of the steam demand suddenly increasing, the water supply will need to be increased to match the increased steam flow.Shrinkage is the opposite of swell: it occurs when the pressure rises. The mechanism is exactly the same as that for swell, but in the reverse direction. Shrinkage causes the level of water in the drum to fall when the steam flow decreases, and once again the delivery of water to the boiler must be related to the actual need rather than to the possibly misleading indication provided by the drum-level transmitter.If a slow change of steam flow occurs, all is well because the pressure within the system can be controlled. It is when rapid steam-flow changes happen that problems occur since, due to swell or shrinkage, the drum level indication provides a contrary indication of the water demand.Following a sudden increase in steam demand, which causes the pressure to drop (and therefore the drum level to rise), a simple level controller would respond by reducing the flow of feed water. Equally, a sudden decrease in steam flow, which would be accompanied by a rise in pressure and an attendant fall in the drum level, would cause a level controller to increase the flow of water. Both actions are, of course, in the incorrect sense.The effects of swell and shrinkage, in addition to being determined by the rate of change of pressure, also depend on the relative size of the drum and the pressure at which it operates. If the volume of the drum is large in relation to the volume of the whole system the effect will be smaller than otherwise. If the system pressure is low the effect will be larger than with a boiler operating at a higher pressure, since the effect of a given pressure change on the density of the water will be greater in the low-pressure boiler than it would if the same pressure change were to occur in a boiler operating at a higher pressure.Faced with this situation, designers of control systems have responded by implementing a variety of solutions. The simplest of these is a two-element system, since it is based on the use of two process measurements in place of the single drum-level measurement used above.6.2.1 Two-element feed-water controlRemembering that the basic requirement of a feed-water control system is to maintain a constant quantity of water in the boiler, it is apparent that one way of addressing the problem would be to maintain the flow of water into the system at a value which matches the flow of steam out of it. One version of this system is shown in Figure 6.1. Here, the flow is controlled by an easily recognised device, a valve. We shall look at valves in more depth later, but for the moment assume that the version used in the diagram maintains the rate of water flowing through the valve at a figure which is directly proportional to the demand signal from the controller (i.e. if the demand signal varies linearly from 0 to 100%, the flow rate also changes linearly between 0 and 100%). Such a valve is said to have a linear characteristic and in the system shown this is employed in conjunction with a transmitter that produces a signal proportional to steam flow. Used together, these two devices keep the parameters in step. If the transmitter produces a signal which is equal to the steam flow at all loads and if the flow through the valve is matched with this signal at every point in the flow range, a controller gain of unity will ensure that, throughout the dynamic range of the system, the flow of water will always be equal to the flow of steam.Naturally, scaling factors of the transmitter and the valve must be taken into account. If the range of the flow transmitter is different from the valves flow-control range, the controller gain will need to be adjusted accordingly, and in practical systems this is always necessary.In order to provide an adequate operational margin of confidence, the range of the control valve is always designed to be greater than the flow range of the boiler. For example, in a boiler producing 20 kg/s of steam, the valve may be sized to deliver 22 kg/s of water when it is fully open. In this example, with a linear valve characteristic, an opening of approximately 91% will be needed to pass a flow of 20 kg/s.In this case, if the steam-flow transmitter produces an output of 100% at 20 kg/s flow, the controller gain must be such that a measured value of 100% produces an output of 91%. This is a proportional band of 110 (i.e. a gain of 20/22) and if this gain is assigned to the controller the feed flow will match the steam flow over the entire range of boiler load (assuming that the valve characteristic is linear, that the flow transmitter output is 4 mA at zero flow, and that zero flow of water occurs with a valve signal of 0%).The problem with this system is that it only matches the steam- and feed-flow rates. If, at the outset, the drum level is below the desired value, that is where it will stay, because if everything is set up correctly the feed into the boiler will always match the steam flowing out of it, and there is no mechanism for introducing the small surfeit of feed over steam, or the slight deficit, that is needed to correct the drum-level error.It is important to consider the practical reality of what would happen if things were not to be set up correctly. In this situation, if there is a small setting error in the controller gain, or if the feed valve passes more or less water than it should at the given opening, or if the steam flow transmitter is slightly out of Calibration, the drum level will integrate up or down at a rate determined by the scale of the error, and nothing will correct for this undesirable state of affairs.In other words, the system cannot correct the drum level if this parameter deviates from the desired value either because of an initial error or because of small errors in the steam-flow measurement or nonlinearities between the valve demand and the actual flow through it. In the example given above, the exact gain required is 0.909 09 . Therefore, if the controller gain were to be set to 0.91 as suggested above, the feed-water flow would be slightly greater than the steam flow, and the drum level will gradually increase.To counter these effects it is necessary to add a feedback element, consisting of another controller which will act to correct for any mismatch.108 Power-plant control and instrumentation between the actual and desired drum levels. Figure 6.2 shows one variety of such a two-element system.In such a system, because the drum volume and the steam and feed flows form an integrating system, with the drum level integrating any steam-flow/feed-flow mismatch, it is unnecessary to employ an additional integration function in the controller. Therefore the drum-level controller should be of the proportional-only variety.The correct gain for this controller can be determined from a knowledge of the swell and shrinkage effects within the boiler. If these are not known they can be determined by test. A suitable test would be to change the steam flow as rapidly as possible by, say, 10% of the maximum evaporation rate of the boiler, while keeping the feed flow in step with the steam flow. (This can be achieved by hooking a feed-flow signal into the system while temporarily disabling the drum-level controller).To see how the information on the boilers swell characteristic can be used to help with controller tuning, let us examine a two-element system where the range of the steam-flow transmitter is ranged as above (0-20 kg/s), and the feed valve is again sized to deliver 22 kg/s when it