Traditional Ramps with No Imbalance 未失衡的传统斜坡

通常情况下,在系统内使用激增容量“滤波”过程扰动是有益的。DMCplus实现了这一大多数管理控制器并没有实现的目标。

对于任一斜坡变量,都有一个上限,一个下限,一个设定值,以及一个称为斜坡率的整定参数。上限和下限对应于系统经历一重大干扰后所能达到的最大和最小激增限制。

斜坡设定值和斜坡率(可调参数)的作用是在期待下一主要干扰时使控制器将斜坡变量逐渐移动到介于上下限之间的位置。(斜坡设定值必须设定在上下限之间,否则DMCplus界面将恢复到最近的限制内)。

DMCplus斜坡设定值(见DMCplus动作发生器)是基于斜坡变量当前值,操作员输入的斜坡变量设定点,以及斜坡率并根据下列简单公式计算的。

如果当前值等于操作限或在操作限之间,则:

实际的内部斜坡设定值=当前值+斜坡率*(操作员输入斜坡设定–当前值)

如果当前值高于上操作限,则:

实际的内部斜坡设定值=上操作限+斜坡率* (操作员输入斜坡设定点-上操作限)

如果当前值低于下操作限,则:

实际的内部斜坡设定值=下操作限+斜坡率* (操作员输入斜坡设定点–下操作限)

斜坡率值必须设在0和1之间。若斜坡率设为1.0,则内部斜坡设定值始终等于操作员值。若斜坡率设为0.0,则斜坡设定值在上下限之间漂移。因此,斜坡率可看作是一种强迫函数驱动所述斜坡变量值回到操作员输入的斜坡设定点。该功能允许DMCplus控制器保持在合理的水平范围内,但又不过度积极地执行它的设定值。

对于未失衡的传统斜坡当RAMPRT=0时,稳态优化找到一个平衡斜坡(消除稳态失衡)的稳态解决方案。DMCplus动作计算中斜坡的设定值由斜坡当前值所确定。若当前值在操作限之间,设定值就是当前值(参见图21)。

如果当前值超出了操作限,设定值呈现在违反限制的值(参见图22)。对于这种类型的斜坡,在限制之间并没有优选值;值被允许在限制间漂移。如果两个操作限制被夹为相同值,斜坡被控制为在设定值的固定值。倘若稳态优化无法平衡斜坡,DMCplus控制将被中止。


Figure 21: Traditional Ramp with No Imbalance, value between limits, RAMPRT=0

                                              图21:未失衡传统斜坡,值位于限制之间,RAMPRT=0


Figure 22: Traditional Ramp with No Imbalance, value outside limits, RAMPRT=0

                                              图22:未失衡传统斜坡,值位于限制之外,RAMPRT=0

传统的斜坡还支持平均液位控制,通过给RAMPRT参数一大于0且小于等于1的值。稳态优化将找到一个稳态解平衡坡道。

RAMPRT= 0与0<RAMPRT<=1的关键区别是后者在限制间存在优选值。RAMPSP的值是由用户指定的,并且必须是操作限之间。如果斜坡当前值处于操作限之间,RAMPRT指定了当前值与用于DMCplus动作计算设定值的RAMPSP距离分数。

例如,如果当前值是60.0,RAMPSP=40.0且RAMPRT= 0.25,则设定值等于60.0+0.25 *(40.0-60.0)=55.0(参见图23)。RAMPRT设置地越小,控制器尝试去驱动至RAMPSP的努力就越小。RAMPRT= 0表示RAMPSP完全被忽略,RAMPRT=1相当于将两个操作限夹成一个值RAMPSP。

如果当前值违反了操作限制,DMCplus动作计算的设定值被设定为朝向RAMPSP违反限制的距离分数(由RAMPRT指定)。

例如,如果上操作限制=80.0,而当前值大于80.0,RAMPSP=40.0,RAMPRT= 0.25,则DMCplus动作计算的设定点为80.0+0.25 *(40.0-80.0)=70.0(参见图24)。再次声明,如果稳态优化无法平衡斜坡,DMCplus控制器将被中止。


Figure 23: Traditional Ramp with No Imbalance, value between limits, RAMPRT=0.25

                                            图23:未失衡传统斜坡,值位于限制之间,RAMPRT=0.25


Figure 24: Traditional Ramp with No Imbalance, value outside limits, RAMPRT=0.25

                                             图24:未失衡传统斜坡,值超出限制,RAMPRT=0.25


附原文:

Normally, it is advantageous to make use of the surge capacity within a system for "smoothing"process disturbances. DMCplus achieves this goal while most regulatory controllers do not.

For each ramp variable, there is a high limit, a low limit, a setpoint, and a tuning parameter called the ramp rate. The high and low limits correspond to the maximum and minimum surge limits that can be reached after a major disturbance.

The ramp setpoint and ramp rate (tunable parameter) enable the controller to move the ramp variable gradually to a position between the high and low limits, in anticipation of the next major disturbance. (The ramp setpoint must be set between the high and low limits failing which the DMCplus interface will reset to the nearest limit).

The DMCplus value of the ramp setpoint (seen by the DMCplus move generator) is calculated based on the current value of the ramp variable, the operator-entered ramp variable setpoint, and the ramp rate according to the following simple formula.

If current value is at or between operating limits, then:

Actual internal ramp setpoint =current value + ramp rate * (operator entered ramp setpoint - current value)

If current value is above upper operating limit, then:

Actual internal ramp setpoint= upper operating limit + ramp rate *( operator entered ramp setpoint - upper operating limit)

If current value is below lower operating limit, then:

Actual internal ramp setpoint= lower operating limit + ramp rate * (operator entered ramp setpoint - lower operating limit )

The value of the ramp rate must be set between 0 and 1. For a ramp rate of 1.0, the internal ramp setpointis always the operator value.For a ramp rate of0.0, the ramp setpoint drifts between the lower and upper limit. Thus, the ramp rate can be seen as a kind of forcing function that drives the ramp variable value back towards the operator-entered ramp setpoint. This feature allows the DMCplus controller to keep level within a reasonable range without being overly aggressive about enforcing its setpoint.

For the Traditional Ramp with No Imbalance and RAMPRT=0, the steady-state optimization finds a steady-state solution that balances the ramp (eliminates steady-state imbalance). The setpoint for the ramp in the DMCplus move calculation is determined from the current value of the ramp. If the current value is between the operating limits, the setpoint is the current value (see Figure 21).

If the current value is outside the operating limits, the setpoint takes on the value of the violated limit(see Figure 22). For this type of ramp, there is no preferred value between the limits; the value is allowed to drift between limits. If the two operating limits are pinched to the same value, the ramp is controlled to a fixed value setpoint. In the event that the steady-state optimization is unable to balance the ramp, then DMCplus control is aborted.

The traditional ramp also supports averaging level control, specified by giving RAMPRT a value greater than zero(0) but not greater than one(1). The steady-state optimization finds a steady-state solution that balances the ramp.

The key difference between RAMPRT=0 and 0<RAMPRT<=1 is that a preferred value exists between the limits.This value, RAMPSP,is user-specified, and must be between the operating limits. If the current value of the ramp is between the operating limits then RAMPRT specifies the fraction of the distance from the current value back toward RAMPSP to be used as the setpoint for the DMCplus move calculation.

For example, if the current value is 60.0, RAMPSP=40.0 and RAMPRT=0.25, then the setpoint is60.0+0.25*(40.0-60.0)=55.0 (see Figure 23) The smaller RAMPRT is set, the less effort the controller makes in trying to get to RAMPSP. RAMPRT=0 means that RAMPSP is ignored, and RAMPRT=1 is equivalent to pinching both operating limits to the single value RAMPSP.

If the current value is violating an operating limit, the setpoint for the DMCplus move calculation is set at the fraction (specified by RAMPRT) of the distance back from the violated limit toward RAMPSP.

For example, if Upper Oper.Limit=80.0, the current value is greater than 80.0, RAMPSP=40.0, and RAMPRT=0.25, the setpoint for the DMCplus move calculation is 80.0+0.25*(40.0-80.0)=70.0 (see Figure 24). Again, if the steady-state optimization is unable to balance the ramp, DMCplus control is aborted.

                                                                         2015.10.3

最后编辑于
©著作权归作者所有,转载或内容合作请联系作者
  • 序言:七十年代末,一起剥皮案震惊了整个滨河市,随后出现的几起案子,更是在滨河造成了极大的恐慌,老刑警刘岩,带你破解...
    沈念sama阅读 204,732评论 6 478
  • 序言:滨河连续发生了三起死亡事件,死亡现场离奇诡异,居然都是意外死亡,警方通过查阅死者的电脑和手机,发现死者居然都...
    沈念sama阅读 87,496评论 2 381
  • 文/潘晓璐 我一进店门,熙熙楼的掌柜王于贵愁眉苦脸地迎上来,“玉大人,你说我怎么就摊上这事。” “怎么了?”我有些...
    开封第一讲书人阅读 151,264评论 0 338
  • 文/不坏的土叔 我叫张陵,是天一观的道长。 经常有香客问我,道长,这世上最难降的妖魔是什么? 我笑而不...
    开封第一讲书人阅读 54,807评论 1 277
  • 正文 为了忘掉前任,我火速办了婚礼,结果婚礼上,老公的妹妹穿的比我还像新娘。我一直安慰自己,他们只是感情好,可当我...
    茶点故事阅读 63,806评论 5 368
  • 文/花漫 我一把揭开白布。 她就那样静静地躺着,像睡着了一般。 火红的嫁衣衬着肌肤如雪。 梳的纹丝不乱的头发上,一...
    开封第一讲书人阅读 48,675评论 1 281
  • 那天,我揣着相机与录音,去河边找鬼。 笑死,一个胖子当着我的面吹牛,可吹牛的内容都是我干的。 我是一名探鬼主播,决...
    沈念sama阅读 38,029评论 3 399
  • 文/苍兰香墨 我猛地睁开眼,长吁一口气:“原来是场噩梦啊……” “哼!你这毒妇竟也来了?” 一声冷哼从身侧响起,我...
    开封第一讲书人阅读 36,683评论 0 258
  • 序言:老挝万荣一对情侣失踪,失踪者是张志新(化名)和其女友刘颖,没想到半个月后,有当地人在树林里发现了一具尸体,经...
    沈念sama阅读 41,704评论 1 299
  • 正文 独居荒郊野岭守林人离奇死亡,尸身上长有42处带血的脓包…… 初始之章·张勋 以下内容为张勋视角 年9月15日...
    茶点故事阅读 35,666评论 2 321
  • 正文 我和宋清朗相恋三年,在试婚纱的时候发现自己被绿了。 大学时的朋友给我发了我未婚夫和他白月光在一起吃饭的照片。...
    茶点故事阅读 37,773评论 1 332
  • 序言:一个原本活蹦乱跳的男人离奇死亡,死状恐怖,灵堂内的尸体忽然破棺而出,到底是诈尸还是另有隐情,我是刑警宁泽,带...
    沈念sama阅读 33,413评论 4 321
  • 正文 年R本政府宣布,位于F岛的核电站,受9级特大地震影响,放射性物质发生泄漏。R本人自食恶果不足惜,却给世界环境...
    茶点故事阅读 39,016评论 3 307
  • 文/蒙蒙 一、第九天 我趴在偏房一处隐蔽的房顶上张望。 院中可真热闹,春花似锦、人声如沸。这庄子的主人今日做“春日...
    开封第一讲书人阅读 29,978评论 0 19
  • 文/苍兰香墨 我抬头看了看天上的太阳。三九已至,却和暖如春,着一层夹袄步出监牢的瞬间,已是汗流浃背。 一阵脚步声响...
    开封第一讲书人阅读 31,204评论 1 260
  • 我被黑心中介骗来泰国打工, 没想到刚下飞机就差点儿被人妖公主榨干…… 1. 我叫王不留,地道东北人。 一个月前我还...
    沈念sama阅读 45,083评论 2 350
  • 正文 我出身青楼,却偏偏与公主长得像,于是被迫代替她去往敌国和亲。 传闻我的和亲对象是个残疾皇子,可洞房花烛夜当晚...
    茶点故事阅读 42,503评论 2 343

推荐阅读更多精彩内容