CN100476644C - Utmost PID control method of single input single output system - Google Patents

Utmost PID control method of single input single output system Download PDF

Info

Publication number
CN100476644C
CN100476644C CN 200510112230 CN200510112230A CN100476644C CN 100476644 C CN100476644 C CN 100476644C CN 200510112230 CN200510112230 CN 200510112230 CN 200510112230 A CN200510112230 A CN 200510112230A CN 100476644 C CN100476644 C CN 100476644C
Authority
CN
China
Prior art keywords
controller
control
system
pid
signal
Prior art date
Application number
CN 200510112230
Other languages
Chinese (zh)
Other versions
CN1794118A (en
Inventor
张卫东
陈培颖
曹春生
尹汝泼
青 赵
Original Assignee
上海交通大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 上海交通大学 filed Critical 上海交通大学
Priority to CN 200510112230 priority Critical patent/CN100476644C/en
Publication of CN1794118A publication Critical patent/CN1794118A/en
Application granted granted Critical
Publication of CN100476644C publication Critical patent/CN100476644C/en

Links

Abstract

一种工业过程控制技术领域的单输入单输出系统的极限PID控制方法,步骤如下:1)当工控机的检测部分接到主机发出的采样命令后,对被控制对象进行采样滤波,由模拟量输入通道将采样信号送入检测变送装置,再经A/D转换后得到数字信号后对对象进行辨识;2)通过确定控制器可调参数λ的下限,保证对应的PID控制器的四个参数取值为正;3)判断经过A/D转换后的数字量输入信号极性,据此计算误差信号;4)计算极限PID控制器参数值;5)计算控制信号增量的值,由D/A转换后输出至执行器,使被控对象运行在给定的范围内。 Limit PID control method for an industrial process control art single-input single-output system, the following steps: 1) when the detecting portion to the IPC sample command sent by the host, the control target of the filtering is sampled by the analog the sampled input channel signals into transmission detecting means for object re-identification of the digital signal obtained after a / D conversion via; 2) λ tunable by determining the lower limit of the controller, to ensure that corresponding to the four PID controller the parameter n; 3) Analyzing the digital value through a / D conversion polarity of the input signal, calculating an error signal accordingly; 4) calculating the limit value of PID controller parameters; 5) calculates a control signal value increment by D / a converter outputs to the actuator, the controlled object is operating within a given range. 本发明得到的控制器逼近理论设计出的最优控制器性能,可达到用户满意的标称性能和鲁棒性能,实现更好的控制效果。 The controller of the present invention is obtained by approximation theory devised optimal controller performance, user satisfaction can be achieved nominal performance and robustness, better control effect.

Description

单输入单输出系统的极限PID控制方法 Limit PID control of single-input single-output system

技术领域 FIELD

本发明涉及的是一种工业过程控制技术领域的方法,特别是一种单输入单输出系统的极限PID控制方法。 The present invention relates to a method of industrial process control art, in particular a single-input single-output PID control limit system.

背景技术 Background technique

PID控制器整定方法的目的就是要获得最好的比例、积分、微分控制的时间常数项,来使实际过程的闭环响应尽可能好的与理想的闭环响应相一致。 PID controller tuning methods aim is to obtain the best ratio, the time constant of the integral term, a differential control, to make the closed loop response of the real process as good as possible consistent with the desired closed loop response. 目前,使用较多的PID控制器整定方法还是以传统的经验方法为主。 Currently, more use of PID controller tuning method is still dominated by traditional empirical methods. 然而传统的基于经验的整定方法都只利用了系统动态特性的部分信息,无法按指定性能和鲁棒性设计控制器。 However, based on the experience of conventional tuning method using only part of the information system dynamics, not specified performance and robustness of the controller design. 近年来发展起来的基于IMC控制结构的控制器设计方法,因为充分利用了系统的动态特性信息,力图设计出最优控制器而受到广泛的重视。 Developed in recent years IMC controller design method based control structure, since full use of the dynamic characteristics of the information system, trying to design the optimal controller received widespread attention. 为了在物理上易于实现,通常使用两种方法设计最优PID控制器。 To facilitate physically implemented, usually optimal PID controller design using two methods. 一种是采用一般的近似方法(如Pade近似或Taylor近似等)展开被控对象的纯滞后项,即将被控对象近似转化成有理形式,然后针对有理控制对象推导最优控制器,再等效为PID控制器的形式;一种是先推导控制器,然后用有理近似方法对控制器所包含的纯滞后项进行近似降阶。 One is a method using a general approximation (e.g., Taylor approximation Pade approximation or the like) of pure expanded lags of the controlled object, the controlled object is about to be converted into rational approximation form, and then derive the optimal controller for rational control object, then the equivalent in the form of PID controller; one is the first derivation of the controller, and then pure rational approximation lags controller comprises reduced-order approximation is performed. 前一种方法比较简单,后一种方法则具有较好的标称性能。 The former method is relatively simple, the latter method has better nominal performance. 但是由于两种方法都是针对模型或推导的最优控制器中的无理项(即纯滞后项)进行逼近,也存在不足。 However, because both methods to approximate the optimal controller for the derivation of the model or unreasonable terms (ie net lags), there are also inadequate. 因此在各产业的控制要求逐渐细化提高的情况下,需要更为精确的方法来获得满足要求的PID控制器。 Therefore, when the control requirements of various industries to improve gradually refined the need for a more accurate way to get to meet the requirements of the PID controller.

经对现有技术的文献检索发现,Yougho Lee等人在《AIChE Journal》(美国化学工程协会杂志)(1998年I月,第I期,总第44卷,第106—115页)上发表的“PID controller tuning for desired closed loop response for SI/SO systems” (单输入单数出系统理想闭环响应的PID控制器调节方法),该文提出一种基于Maclaurin展开式的PID控制器(以下简称Madaurin PID控制器)设计方法,该方法在理论上能达到更好的系统性能,其不足是文章中没有给出具体的控制器参数的整定方法。 Literature search of the prior art found, Yougho Lee et al., "AIChE Journal" (Journal of American Chemical Engineering Association) (In January 1998 I, phase I, the total Volume 44, pp. 106-115) published in the "PID controller tuning for desired closed loop response for SI / SO systems" (the number of single-input single-out system over the closed loop response of the PID controller regulating method), based on the proposed Maclaurin expanding PID controller (hereinafter referred to Madaurin PID controller) design method can theoretically achieve better system performance, which is less than the setting method of the article did not give a specific controller parameters. 发明专利“定量整定鲁棒性的智能比例积分微分控制方法”(申请号03115673.8)用Madaurin展开式对控制器的最优解的逼近,虽然控制器参数表达式相对复杂,但所得的控制器要比仅使用一阶Pade近似或Taylor近似所得到的H00PID控制器和/Z2PID控制器效果更好。 Patent invention "Quantitative tuning robust PID control method smart" (Application No. 03115673.8) by expanding Madaurin controller approximation of the optimal solution, although relatively complex expression controller parameters, but resulting to the controller a first order Taylor approximation Pade approximation or obtained H00PID controller and / Z2PID better than using only the controller. 尽管如此,Madaurin PID控制器与近似前的原始控制器还是存在一定的误差。 Nevertheless, Madaurin PID controller before the controller is still some approximation of the original error.

发明内容 SUMMARY

本发明的目的在于针对现有技术的不足,提出一种单输入单输出系统的极限PID控制方法,使其依据解析的方法设计最优控制器,再采用Madaurin展开序列和Pade近似理论对包含有纯滞后的控制器进行降阶,得到的控制器可以最大可能的逼近理论设计出的最优控制器性能。 Object of the present invention is for the deficiencies of the prior art, PID control method of a proposed limit SISO system, it parses the method according to the optimal controller design, and then using Madaurin deployment sequence with Pade approximation theory pure lag controller reduction, the controller can get the maximum possible approximation theory to design optimal controller performance. 通过定量调节控制器参数可以达到用户满意的标称性能和鲁棒性能,实现更好的控制效果。 Quantitative parameters may be adjusted by the controller to achieve customer satisfaction and the nominal performance robustness, better control effect.

本发明是通过以下技术方案实现的,本发明在现有的鲁棒控制器设计方法以及单参数PID整定方法的基础上,将提出的极限PID控制方法集成在工业控制系统的监控模块中实现控制作用。 The present invention is achieved by the following technical solutions, the present invention limits the PID control method based on the existing Robust controller design and tuning method of a single PID parameters on the proposed integrated monitoring module implemented in the control system of the industrial control effect. 采用单位反馈控制结构,在工控系统辨识出控对象模型的基础上,根据用户对系统响应的实际要求和辨识出的模型参数得到控制器调节参数的所取的初值,然后将这些数据送入监控模块,自动执行事先编制好的PID控制程序,计算出最佳PID控制器参数并得到控制信号,从而实现对系统鲁棒性的定量整定,使系统的标称性能和鲁棒性以最佳的方式达到折中。 Using feedback control unit structure, identified on the basis of the control object model in the industrial system, to obtain the initial value taken controller tuning parameters based on the actual requirements of the user and the system response identified model parameter, and then these data into a monitoring module automatically performs PID control program pre-programmed to calculate the optimum parameters of the PID controller signal and controlled to achieve quantitative robust tuning system, the system performance and robustness nominal best the way to achieve a compromise. 同时用户还可通过在线调节控制器参数来调节控制效果,获得要求的标称性能和鲁棒性。 While the user is online but also by adjusting the parameters of the controller to adjust the control performance and robustness to obtain nominal performance requirements. 具体步骤如下: Specific steps are as follows:

1)当工控机的检测部分接到主机发出的釆样命令后,对被控制对象进行釆样滤波,由模拟量输入通道将采样信号送入检测变送装置,再经A/D转换后得到数字信号后对对象进行辨识,对象辨识模块基于阶跃响应法辨识出一阶加纯滞后模型 1) when the detection section to preclude the kind of the IPC command sent by the host, to the control target Bian filtered sample by sampling the analog input signals into the detection channel transmission means, and then to give after A / D conversion identification of an object is a digital signal, the module-based object recognition step response method identified a model order plus dead

0(8)=〜的参数,稳态增益尺、时间常数Γ和纯滞后时间0,并将辨识出的模型 0 (8) = - parameters, the steady-state gain scale, and the time constant of the pure time delay Γ 0, and the identified model

T5 + 1 T5 + 1

参数送到主机的存储单元RAM中。 The parameter storage unit of the RAM to the host. 若模型为高阶对象则要先进行降阶为一阶加纯滞后的形式,再辨识出相应的模型参数。 If the high order model is a first order to a lower order plus dead form, and then identify the corresponding model parameters.

2)控制器参数的调节和整定:为保证本发明中极限PID控制器的可实现性,即保证对应的PID控制器的四个参数比例增益仏、积分时间常数7;、微分时间常数&以及7>取值为正,需要加一个约束条件,即控制器可调参数2有一个下限,下式给出了A最小值的解析解。 2) adjusting the parameters of the controller and setting: To ensure that the present invention may be implemented Limit PID controller, i.e., to ensure that the four parameters of the PID controller proportional gain corresponding to Fo, 7 ;, integral time constant and derivative time constant & 7> is a positive value, it is necessary to add a constraint, i.e., the controller has a tunable limit 2, the analytical solution is given by a minimum.

式中:0—控制对象纯滞后,r_控制对象时间常数,A —系统性能度,即系统可调参数。 Where: 0 Delay control object, the control object R_ time constant, A - of the performance of the system, i.e., tunable system.

整定控制器时可以先将Amin和辨识出的模型参数输送到极限PID控制器参数的计算单元,由第4)步给出的公式计算出PID控制器的四个参数,再送入控制单元实现闭环控制,观测响应曲线,通过在线的从小到大单调增加;I来整定控制器,直至获得要求的响应。 When tuning controller may be delivered first and Amin identified model parameter calculation unit to the limit of PID controller, is calculated by the PID controller four parameters equation 4 given step), and then into the closed-loop control unit control, observer response curve, increased by the monotonous ascending line; the I to tune the controller, until obtaining the required response. 需要指出的是,采用较大的;I值可以获得较小的超调量和较好的鲁棒性能,但同时会导致较长的上升时间;而采用较小的;I值可以增加系统的响应速度,但是牺牲了一定的鲁棒性,同时可能会导致系统存在较大的超调。 It should be noted that the use of large; I can get the value of a smaller overshoot and robust performance, but will also lead to a longer rise time; and using small; I can increase the value of the system response speed, but at the expense of certain robustness, and could cause a large overshoot of the system. 调节;I的步长可设定为0.010或更小。 Regulation; the I step size may be set to 0.010 or less. 对于纯滞后很小的控制对象还可以通过时间常数r来确定典型步长,譬如PSO.lr时可以考虑取典型步长为O.OOlr或O.Olr。 For pure small hysteresis control target can also be determined by the time constant of a typical step size r, for example, may be considered when PSO.lr step is typically taken or O.OOlr O.Olr.

如果控制系统所需达到的标称性能指标己指定,还可以按照以下经验公式得到A的初值或A调节的参考范围: If the control system required to achieve the specified nominal performance hexyl, may be obtained initial value or reference range A A regulating according to the following empirical formula:

式中:σ —超调,^一上升时间。 Where: σ - overshoot, the rise time of a ^. 此经验公式构建起了系统所需达到的标称性能指标和最佳控制器参数之间的直观联系,连同本发明中;Itnin计算方法一同装入监控模块中,并将二者计算结果中较大值送入主机存储单元RAM中。 This empirical formula build a visual link between the nominal system required to achieve optimum performance and controller parameters, together with the present invention; Itnin calculated with the load monitoring module, and the calculated results representing the two a large value into the main storage RAM. 这种调节参数的经验算法为工程技术的调试工作提供了快速有效的解决方案。 This adjustment empirical algorithm parameters provide a quick and effective solution for debugging engineering.

3)由监控模块执行事先编制好的极限PID控制程序:首先对控制系统输出采样滤波,经模拟量输入通道传输信号,并将信号接入检测变送装置,再经A/D转换后得到数字量输入信号与此时的系统输入信号相减。 3) by the monitoring module performs pre-programmed limit PID control routine: sampling the filtered first control system output, the input analog signal transmission channel, and access detection signal transmission means, and then after A / D conversion to obtain a digital At this time, the input signal and the system input signal subtracted. 判断该信号极性,并据此计算误差信号,以保证控制器比例增益取值为正。 The polarity determination signal, and calculate the error signal to ensure that the controller proportional gain value is positive.

具体规则是:如果是正作用,则误差信号等于系统输出扒》)减去系统输入r(«)后的差值;如果是负作用,则<«)等于r(«)减:μ(«)后的差值。 Specific rules are: If positive effect, the system outputs an error signal equal to Pa ') after subtracting a difference R & lt system input ( «); if negative effect, the <«) R & lt equal ( «) Save: [mu] («) after the difference.

4)读取存储单元RAM中的被控对象模型参数和根据解析公式和经验公式得出相应的控制器参数A参考取值,按下式计算极限PID控制器的参数值: 4) Read the plant model parameter storage unit and the RAM controller parameters derived in accordance with the respective analytical formulas and the empirical formula A a reference value, the limit value of the PID controller parameters calculated as follows:

其中:尺一控制对象增益,0—控制对象纯滞后, r一控制对象时间常数,义一系统性能度,&一控制器增益, Wherein: a foot control object gain, 0- pure time delay control object, a control object R & lt time constants, a system-defined performance metric, a & gain controller,

T7—控制器积分时间,—控制器微分时间,7;—滤波器时间常数虽然由本发明所推导出的PID控制器参数计算的公式相对于前面提到的PID控制器、H2 PID控制器和Madaurin PID控制器中四个参数的计算方法要复杂,但所给公式中各个变量都是己知的,因此同样具备操作简便直观的优点,而且极限PID控制器的计算方法相对于前三种方法,可以更加精确的达到理论推导出的控制器的最优性能。 T7- controller integral time, - the controller derivative time, 7; - the filter time constant although the equation derived by the present invention PID controller parameter calculation with respect to the aforementioned PID controllers, PID controllers and H2 of Madaurin four PID controller parameter calculation method complex, but the equation given in the respective variables are known, and therefore also have the advantage of intuitive and easy operation, and it is calculated limit of PID controller with respect to the first three methods, We can achieve more accurate theory derive optimal performance of the controller.

5)按照离散域PID控制算式计算控制信号增量Δ«(»)的值,与前一时刻的控制信号M(«-l)通过加法器进行加法运算就得到当前时刻的输出控制信号<„)。Au(n)计算公式如下: 5) discrete domains of the PID control according to a control signal equation calculating the value of the increment Δ «(»), and the control signal M and the previous time ( «- for adding l) is obtained by the adder outputs a control signal for the current time <" ) .Au (n) is calculated as follows:

并对μ(«)进行限幅,防止积分饱和,由D/A转换后输出至执行器,由执行器作用到被控对象,使被控对象运行在给定的范围内,同时显示现时的状态等参数,原始数据系列向前滚动一个单元。 And [mu] ( «) clipped to prevent windup, the output from the D / A converter to the actuator, by the action of the actuator to the controlled object, the controlled object is operating within a given range, while displaying the current status parameters, scroll forward a series of original data unit. 如此周而复始,整个工控系统就实现了有效控制。 Again and again, the entire industrial system to achieve effective control.

本发明通过编制成相应的极限PID控制监控模块,在现有的工控系统中直接实施。 The present invention controls the monitoring module via a corresponding PID limit compiled directly implemented in existing industrial control systems. 并明确给出了系统所需达到的标称性能指标和最佳控制器参数之间的经验公式用以得到控制器调节参数,从而构建起这两者之间的直观联系,连同本发明所釆用的PID控制算法一同装入监控模块中。 And clearly empirical formula between the nominal system required to achieve optimum performance and a controller for obtaining parameters of the controller tuning parameters, and thus construct visual link between the two, together with the present invention preclude together with the PID control algorithm loaded into the PCU.

在工业控制现场采用本发明提出的极限PID控制方法,最大的特点就是采用Maclaurin展开序列和Pade近似理论对包含有纯滞后的最优控制器以解析方式进行近似和降阶,得到的控制器基本上达到了这个阶次控制器所能达到的标称性能极限。 In the field controller of an industrial control PID control using the limit proposed according to the present invention, the biggest feature is the use of Maclaurin series to expand and Pade approximation theory optimal controller with pure time delay and reduced-order approximated analytically obtained substantially on reaching the order of the controller can be achieved nominal performance limits. 通过定量调节控制器参数可以用户满意的标称性能和鲁棒性。 The controller parameters can be adjusted by quantifying user satisfaction with the nominal performance and robustness. 同时这种方法也适用于其他高阶控制器,降阶或近似后的控制可以最大可能的逼近理论设计出的控制器性能,在实际应用中能达到更好的控制效果,且用户操作起来也更简便直观,从而显著地克服了传统方法的主要缺点。 At the same time this method is also applicable to other high-end controller, or reduced-order approximation may control the maximum possible approximation theory to design a controller performance in practical applications to achieve better control effect, and the user can operate more simple and intuitive, thereby significantly overcome the main disadvantage of the conventional method. 本发明给串的极限PID控制器的设计方法可广泛应用于能源、冶金、石化、轻工、医药、建材、纺织等行业中单输入单输出生产过程的控制和调节。 Design of the PID controller to limit the present invention, the string can be widely applied energy, metallurgy, petrochemicals, light industry, medicine, construction materials, textile and other industries in a single input single output control and regulation of the production process.

附图说明 BRIEF DESCRIPTION

图I是采用本发明方法的工控机系统的工作流程图。 Using a flowchart of FIG. I is a method IPC system according to the invention. 图2为本发明采用极限PID控制器的设计方法所用的闭环控制结构图。 FIG 2 using closed loop control configuration diagram of the design method limits the PID controller used in the present invention.

其中C为控制器,G为被控对象,r和;;分别为闭环系统的输入和输出,e为偏差信号,W为控制器输出,^为干扰信号。 Wherein the controller C, G controlled object, r and the input and output, respectively ;; closed loop system, e is the error signal, W is the controller output, ^ is the interference signal.

图3为本发明实施例中系统在标称情况下的给定值响应曲线 Response curve of the present invention FIG 3 setpoint embodiment the system in the case of the embodiment of nominal

其中实线为应用本发明的极限PID控制器控制系统的响应曲线,虚线为应用Zi2PID控制器控制系统的响应曲线。 Wherein the solid line is the response curve of a PID controller control system limits the application of the present invention, the broken line curve in response to the application Zi2PID system controller. 从图中可看出,在超调相同的情况下,应用本发明的控制方法可获得相对于后者较快的响应。 As it can be seen from the figure, in the case where the same overshoot, the control method of the present invention can be obtained with respect to the latter a faster response.

图4为本发明实施例中系统在标称情况下,在ί = 25秒时加入幅值为0.1的反向阶跃负载干扰信号后的响应曲线。 Example 4 In the nominal system, the amplitude response curve after addition of the reverse step load disturbance signal at 0.1 seconds ί = 25 the present invention. 从图中同样可看出本发明的控制方法响应较IX。 The same can be seen that the control method of the present invention is more responsive IX from FIG. 图5为本发明实施例中系统在模型不匹配情况下的响应曲线 Figure 5 embodiment the system response curve model mismatch in the embodiment of the present invention.

图中,虚线为模型参数存在误差时,控制器参数依然取;1 = 1.5时的响应曲线,实线为适当调节控制参数至;I = 2.6时的响应曲线。 Figure, a broken line when there is an error model parameters, controller parameters still take; 1 = 1.5 response curve, the solid line to the appropriate control parameters are adjusted; the I = 2.6 when the response curve. 从图中可看出当模型出现不匹配时,通过适当的调节控制器参数;I,依然可获得符合用户要求的响应曲线。 As it can be seen from the figure, when model mismatch occurs, by appropriately adjusting the control parameters; the I, can be obtained still meet the requirements of the user response curve.

具体实施方式 Detailed ways

以下结合附图和实施例对本发明的技术方案作进一步描述。 Hereinafter, the technical solution of the present invention will be further described in conjunction with the accompanying drawings and embodiments.

如图I所示,系统进入控制阶段后,用户根据实际工况通过辨识被控对象的模型参数,同时工控机读取监控模块中的程序,顺序执行控制过程:首先对采样信号进行检测变送和A/D转换等输入预处理得到数字量输入信号,判断该信号极性,并据此计算误差信号。 As shown in FIG. I, the system enters the control phase, controlled according to the actual conditions of the object by identifying the model parameters, while the IPC program reads in the monitoring module, a control procedure performed sequentially: first transmission sampled signals detected and A / D conversion to obtain a digital input pre-input signal, determines the polarity of the signal and calculate the error signal. 然后判断是否是初次调节被控对象,如果是,则先计算出控制器参数;I可取的最小值Anun和指定标称性能指标对应的参数值;两者取其大,如果否,则读取用户在上述I取值的基础上,按步长调节后;I值。 Then determines whether the first regulating controlled object, if so, to calculate the controller parameters; parameter value I Anun The minimum value and a specified nominal performance indicators corresponding to; whichever is greater, if not, read the based on the above the user on the I value, adjustable in steps; I value. 然后依据确定好的控制参数和模型参数计算极限PID控制器的参数,最终得到控制信号。 Then good basis for determining the control parameters and the model parameter calculation parameter limits PID controller, to obtain a final control signal. 该输出信号经过限幅和D/A转换等处理后去驱动被控系统的执行机构便实现了对被控对象的极限PID闭环控制,同时可对系统鲁棒性实现定量整定,达到最佳控制效果。 The output signal after the clipping and D / A conversion processing and the like to drive the actuators of the controlled system to achieve a closed-loop control will limit the PID controlled object, while on the robustness of the system to achieve quantitative tuning to achieve optimum control effect.

上述工控过程中若将辨识模块设成自动测试模式,那么就可形成自整定极限PID控制系统。 If the above-mentioned industrial process into an automatic recognition module test mode, then the limit can be formed self-tuning PID control system.

实施例:将本发明提出的控制方法用于造纸生产过程自动控制系统,其目的是生产具有恒定定量的纸张。 Example: The proposed control method of the present invention for paper production process control systems, which aim is to produce a constant amount of paper. 此处所谓的定量即是每平方米纸张的重量。 Here it is called quantitative i.e. weight per square meter of paper. 系统的控制对象是典型的长网纸机。 Control object system is a typical Fourdrinier paper machine. 在造纸生产过程中,从打浆工段送来的中浓纸浆与网下回收白水混合成为适合上网的低浓纸浆,输送到流浆箱中,可成为是该造纸机的输入。 In the paper production process, the beating section sent from the white water consistency pulp become mixed with the net recovery of low consistency suitable for the Internet, and supplied to the headbox, the input can be a paper machine. 再从流浆箱的堰板喷出口喷射到网上,纸浆在网上经滤水脱去绝大部分水分形成湿纸页后进入压榨部,湿纸页在压榨部经过压锟多次压榨又脱去剩余的大部分水分,然后进入干燥部,在一系列的充满高温蒸汽的烘缸中被加热,于是湿纸中的水分逐渐蒸发掉,最终得到成品纸,也就是造纸机的输出。 Reinjection headbox slice from the discharge port to the Internet, the pulp web is formed by wet paper off most of the drainage of water into the press part, the wet paper in the press section and after the press pressure repeatedly removed Kun most of the remaining water, and then enters the drying section, is heated in a series of full high-temperature steam in the dryer, the wet paper thus gradually evaporate the water, eventually to give the finished paper, which is the output of the paper machine. 工控系统的主机部分采用研华工控机,实现控制器功能。 Host part of the industrial control system uses Advantech IPC, implement the controller function. 执行器是调节纸浆流量的调节阀,采用ZBJV精密电磁阀。 The actuator is adjusted to a pulp flow regulating valve using a solenoid valve ZBJV precision. 传感器是测量纸张定量的WDT-3P定量仪。 Quantitative measuring sensor is a sheet WDT-3P quantitative instrument. 由RTU-88组态硬件实现A/D、D/A转换。 To achieve A / D, D / A converted by the RTU-88 hardware configuration. 设计要求为:保证系统响应无超调和尽可能小的上升时间。 Design requirements to: ensure that the system response without overshoot and rise time as small as possible.

以工控机中采用极限PID控制过程为例,控制结构如图2所示,介绍具体实施步骤: PID control process to limit IPC used as an example, the control structure shown in Figure 2, the specific embodiment described the steps of:

1.在系统进入整定前,先利用该造纸生产工控系统对控制对象一典型的长网纸机进行自动开环辨识,其辨识原理是开环阶跃响应面积法,结果得到定量控制 1. Before entering the system setting, the first use of the paper production system for industrial control object a typical Fourdrinier paper machine to automatically open loop identification, identification principle which is an open loop step response area method, quantitative control results

的造纸机动态模型为 The paper machine dynamic model

即一阶加纯滞后模型参数为:火= 5.15, That is the first order plus time delay model parameters are: fire = 5.15,

T = LS, 6» = 2.8。 T = LS, 6 »= 2.8. 该控制过程具有典型的纯滞后特性。 The control process is a typical pure hysteresis characteristic. 工控系统将辨识出的模型参数送到主机的存储单元RAM中。 The industrial control system identified model parameter storage unit to the host's RAM. 在此基础上,进入极限PID控制过程。 On this basis, the process enters the limit of PID control.

2.从主机的存储单元RAM中读取辨识出模型参数,依据公式计算2min为 2. Read identified from the host RAM memory unit model parameters, 2min is calculated according to the formula

1^=0.9436,读取用户设定标称性能指标,超调量σ和上升时间由于本例超调量定为零,所以依据经验公式取;I, =1.5左右即可,并与Imin相比较,取两者中较大者即人=1.5,送入存储单元RAM中,用以计算极限PID控制器的比例、积分和微分相的时间常数。 1 ^ = 0.9436, reading the user set the nominal performance metrics, overshoot and rise time σ overshoot because the present embodiment is set to zero, so take based on empirical formula; the I, to = 1.5, and phase and Imin comparing, whichever is greater man = 1.5 i.e., into the RAM storage means, for calculating the time constant of the proportional limit of the PID controller, the integral and differential phase. 3.同时,监控模块开始执行事先编制好的极限PID控制程序:首先对系统输出采样滤波,经模拟量输入通道传输信号,并将信号接入检测变送装置,再经A/D转换后得到数字量输入信号减去此时系统的输入信号。 3. Meanwhile, the monitoring module starts executing pre-programmed limit PID control procedures: First, the system output sampling filtering, the input analog signal transmission channel, and access detection signal transmission means, and then obtained after A / D conversion digital input signal minus the input signal of the system at this time. 判断该信号极性,并据此计算误差信号。 The polarity determination signal, and calculate the error signal. 如果是正作用,则误差信号e(n)等于系统输出><«)减去系统输入r(«)后的差值;如果是负作用,则e(«)等于〃(《)减><«)后的差值。 If the action is positive, then the error signal e (n) is equal to the system output> < «) after subtracting a difference R & lt system input («); if negative effect, then E ( «) equal 〃 (") Save> < « ) after the difference value.

4.然后读取存储单元RAM中的被控对象模型参数和调节参数A的参考值,依据公式: 4. Then read reference value of the controlled object model parameters and tuning parameters A memory cell in the RAM, according to the formula:

计算PID控制器参数&,7},,7;,对系统施加控制作用,然后按照步长0.01沒, & PID controller parameters calculated, 7} ,, ;, 7 applies the control action of the system, and not in steps of 0.01,

从小到大单调增加A来整定控制器,直至获得要求的响应。 A monotonic increase from small to large to tune the controller, until obtaining the required response. 对于该造纸生产工控系统,当被控对象模型精确时(即为标称对象时),取;1 = 1.5,系统可达到响应要求。 The industrial control systems for paper production, when the controlled object model accuracy (that is, when the nominal object), to take; 1 = 1.5, the system can be achieved in response to the requirements. 此时再次将A送到存储单元RAM中,依据公式计算得到极限PID控制器参数为:ATc =0.1290,7; =2.8558,J70 =0.7046,7> =0.1442 A case again to the memory location in RAM, the limit is calculated according to the formula of the PID controller parameters: ATc = 0.1290,7; = 2.8558, J70 = 0.7046,7> = 0.1442

5.系统采样周期取r = 0.b进一步按照离散域PID控制算式: The sampling period in accordance with a further take r = 0.b discrete domains of PID control equation:

计算系数值6ρ62,ί>3,α为: Calculated coefficient values ​​6ρ62, ί> 3, α is:

b, = 0.4268,¾ = -0.7970, b3 = 0.3721,α = 0.5904 b, = 0.4268, ¾ = -0.7970, b3 = 0.3721, α = 0.5904

则所得Δμ(«)为: The resulting Delta] [mu ( «) is:

Au(n) = 0.4268e(n) - 0.7970e(nI) + 0.372 le(n-2) + 0.5904Δ· -1)与前一时刻的控制信号^ -1)通过加法器进行加法运算就得到当前时刻的输出控制信号《(«)。 Au (n) = 0.4268e (n) - 0.7970e (nI) + 0.372 le (n-2) + 0.5904Δ · -1) -1) for adding the control signal of the previous time is obtained by the adder ^ output control signal at the current time "(«). 同时对w(»)进行限幅,防止积分饱和,由D/A转换后经模拟量输出通道送到调节纸衆流量的调节阀,再由调节阀作用到造纸机,便可使造纸机运行在给定的范围内,生产出定量的纸张,同时显示现时的状态等参数,原始数据系列向前滚动一个单元。 While the W ( ») clipped to prevent windup, is pumped by the D / A converts the analog output channel modulating flow valve all paper, then the action of the control valve to the papermaking machine, papermaking machine can run within a given range, the amount of paper produced, while displaying the current status and other parameters, the original data series scrolling forward one cell. 如此周而复始,整个造纸控制系统便实现了对纸张生产过程稳定可靠的闭环控制。 Again and again, throughout the papermaking process control system will achieve a stable and reliable closed-loop control for paper production. 得到的系统阶跃响应曲线如图3中实线所示,图3中还给出由TZ2PID控制器对同一对象进行控制的响应曲线(如图3虚线所示),当采用/Z2PID The step response to obtain the solid-line curve in FIG. 3, the figure 3 also shows the response curve for the control (shown in phantom in FIG. 3) by the same subject TZ2PID controller, when using / Z2PID

控制器时,取;I = 1.83时满足系统响应要求,此时PID控制器参数为 The controller, to take; I = 1.83 is satisfied in response to system requirements, then the PID controller parameters

对应的离散控制器为 Corresponding to a discrete controller

Δμ(«) = 0.1829e(«) - 0.344 \e{n-1) + 0.161 Se(n-2) + 0.8469Δ«(« -1) Δμ ( «) = 0.1829e («) - 0.344 \ e {n-1) + 0.161 Se (n-2) + 0.8469Δ «(« -1)

同时为测试系统的抗干扰性能,在ί = 25秒时加入幅值为0.1的反向阶跃负载干扰信号,采用上述两种控制方法所得到的千扰响应曲线也如图4。 At the same time as the anti-interference performance of the test system, when added ί = 25 seconds reverse step load disturbance magnitude signals of 0.1, using one thousand interference response curve obtained above two control methods is also shown in Figure 4. 从图3和图4中可看出,在超调量相同时,相对于F2PID控制器,极限PID控制系统的给定值响应和干扰响应都具有较快的上升时间。 As can be seen from FIG. 3 and FIG. 4, in the same overshoot, with respect F2PID controller, in response to a given value and the disturbance response limit of the PID control system has a faster rise time.

在实际生产过程中,通常会由于设备磨损等原因使工况发生变化,导致模型参数发生变化,系统响应也随之产生波动。 In the actual production process, generally due to wear and other reasons the device conditions change, resulting in changes in the model parameters, the system will also fluctuate in response. 假设模型参数中稳态增益尤、时间常数和纯滞后时间沒存在10%的摄动,即参数不确定范围为尤e [4.635 5.665], Suppose the model parameters, especially in a steady state gain, lag time constant and the presence of 10% of the time not perturbed, i.e., the range uncertainty parameters, especially e [4.635 5.665],

[2.52 3.08], re[1.62 1.98]。 [2.52 3.08], re [1.62 1.98]. 图5中虚线为当存在模型误差时依然采用控制参数为A = 1.5的响应曲线,从图中可看出响应超调远远大于系统所要求的范围。 FIG 5 is a broken line when the model error is still used as the control parameter A = 1.5 response curve, it can be seen from FIG overshooting far more widespread than required by the system. 所以需要调节;I。 It needs to be adjusted; I. 调节方法为:依据模型失配最差的情况,即增益K和纯滞后项0分别在其不确定范围的最高边界,而时间常数r在其不确定最低边界的原理,按照步长0.010调节控制器参数;I,直至获得最佳响应要求。 Adjustment method is: based on the worst case of model mismatch, that is pure gain K and lags 0, respectively, in its highest boundary of the range of uncertainty, and the time constant r in principle its lowest boundary of uncertainty, regulation and control in steps of 0.010 parameters; I, until an optimal response to the request. 对于本例,当;1=2.6时满足系统响应要求,则新的一组PID控制器参数为: For this example, when; 1 = 2.6 is satisfied in response to the system requirements, a new set of parameters for the PID controller:

所以离散域PID控制算式为: So discrete domains of PID control formula is:

与前一时刻的控制信号通过加法器进行加法运算就得到当前时刻的输出控制信号u(n))。 Control signal output by the adder adding the control signal of the previous time is obtained at the current time u (n)). 系统经过该控制信号作用即可地实现了有效控制,控制效果如图 The system through which the control signal is applied to achieve effective control, as shown in FIG control effect

5中实线所示。 5 shown in solid lines. 从图中可看出当模型出现不匹配时,通过适当的调节控制器参数;I,依然可获得符合用户要求的响应曲线。 As it can be seen from the figure, when model mismatch occurs, by appropriately adjusting the control parameters; the I, can be obtained still meet the requirements of the user response curve.

正因为本发明在设计控制器的过程中考虑的是被控对象可能产生的模型失配的最差情况,而这种最差情况对系统的鲁棒性影响最大,本发明所设计的控制器在不改变控制器结构的情况下,通过调节控制器参数能使系统在最差情况下达到标称性能和鲁棒性的最佳折中,那么当实际对象的模型失配程度不是最差情况时,该控制器的性能将比预期的还好。 Because the present invention is considered in the design of process control is the worst possible case where the controlled object model mismatch, and the maximum effect of this worst case the robustness of the system, the design of the present invention, the controller in the controller without changing the structure, by adjusting the parameters of the controller enables the system to achieve the best compromise nominal performance and robustness in the worst case, then when the model is not the worst case Small mismatches actual object , the performance of the controller better than expected. 因此保证了本发明获得的是最佳控制器。 The present invention therefore guarantees optimum control is obtained.

以上阐述的是本发明给出的一个实施例表现出的优良控制效果,显然本发明不只限于上述实施例,在不偏离本发明基本精神及不超出本发明实质内容所涉及范围的前提下对其可作种种变形加以实施。 Is set forth above exhibit an excellent control effect embodiment of the present invention, a given embodiment, the present invention is clearly not limited to the embodiments described above, the substance thereof on the premise of the present invention without departing from the basic scope and spirit of the invention without departing from It can be used for various modifications to be implemented.

而且,尽管上述实施例所说明的极限PID控制方法是在现有工控系统硬件的基础上以编制软件来进行的,目的是使说明简洁清楚,此外也可以根据同样的原理通过采用单片机等相关硬件在其他场合实现。 Further, although the limit of PID control described in the above-described embodiment is based on the existing industrial control system hardware to software to the preparation, object of simplicity and clarity of description, in addition also be based on the same principle by using hardware associated SCM achieved on other occasions. 采取本发明控制方法的工控系统可广泛应用于能源、冶金、石化、轻工、医药、建材、纺织等行业中各类企业的生产过程控制。 Take control method of the invention it can be widely used in industrial control systems energy, metallurgy, petrochemicals, light industry, medicine, building materials, textile and other industries in the production of various types of process control.

Claims (7)

1.一种单输入单输出系统的极限PID控制方法,其特征在于,具体步骤如下: 1)当工控机的检测部分接到主机发出的采样命令后,对被控制对象进行采样滤波,由模拟量输入通道将采样信号送入检测变送装置,再经A/D转换后得到数字信号后对对象进行辨识,对象辨识模块基于阶跃响应法辨识出一阶加纯滞后模型G(S) = ^l的参数,稳态增益尺、时间常数r和纯滞后时间0,并将辨识出的模型r^ + l参数送到主机的存储单元RAM中; 2)控制器参数的调节和整定:通过确定控制器可调参数/I的下限,保证对应的PID控制器的四个参数比例增益&、积分时间常数7;、微分时间常数Td以及7>取值为正,下式给出了/I最小值的解析解: Limit PID control method for a single input single output system, wherein the following steps: 1) when the detecting portion to the IPC sample command sent by the host, the control target of the filtering is sampled by an analog sampling the input channel signals into transmission detecting means for object re-identification of the digital signal obtained after a / D conversion by the object recognition module based on a step response method identified order plus dead model G (S) = ^ l parameter, the steady-state gain scale, the time constant r 0 and the pure time delay, and the identified model parameters r ^ + l to the host RAM memory cell; and 2) adjusting and setting the controller parameters: by the controller determines tunable / I lower limit to ensure that the four parameters corresponding to the PID controller & proportional gain, integral time constant of the derivative time constant Td ;, 7 and 7> value of n, given a / I the minimum analytical solution:
式中:0一控制对象纯滞后,r一控制对象时间常数,A —系统性能度,即系统可调参数; 3)由监控模块执行事先编制好的极限PID控制程序:首先对控制系统输出采样滤波,经模拟量输入通道传输信号,并将信号接入检测变送装置,再经A/D转换后得到数字量输入信号与此时的系统输入信号相减,判断该信号极性,并据此计算误差信号,以保证控制器比例增益取值为正; 4)读取存储单元RAM中的被控对象模型参数和得到的控制器参数/I参考取值,按下式计算极限PID控制结构和公式计算PID控制器系数值: Where: 0 a pure time delay control object, a control object R & lt time constant, A - of the performance of the system, i.e., the adjustable parameters of the system; 3) by the monitoring module performs pre-programmed limit PID control procedures: First, an output sample control system filtering, the input analog signal transmission channel, and access detection signal transmission means, and then after a / D conversion to obtain a digital input signal and the system input signal at this time is subtracted from the signal polarity determination, and accordingly this error signal is calculated to ensure that the controller proportional gain value is positive; 4) reads the plant model parameter storage unit and the RAM controller parameters obtained / I reference value, calculated as PID control structure limits and coefficient values ​​of the PID controller formula:
式中· Where ·
其中:尺一控制对象增益,0—控制对象纯滞后,r一控制对象时间常数,/I一系统性能度,^一控制器增益,7;—控制器积分时间,7^—控制器微分时间,TF—滤波器时间常数; 5)按照离散域PID控制算式计算控制信号增量Δ«(«)的值,与前一时刻的控制信号M(«-l)通过加法器进行加法运算就得到当前时刻的输出控制信号<«);并对Μ(«)进行限幅,防止积分饱和,由D/A转换后输出至执行器,由执行器作用到被控对象,使被控对象运行在给定的范围内,同时显示现时的状态参数,原始数据系列向前滚动一个单元,如此周而复始,实现整个工控系统的有效控制。 Wherein: a foot control object gain, 0- pure time delay control object, a control object R & lt time constant, / I performance of a system, the controller gain a ^, 7; - a controller integral time, 7 ^ - controller derivative time , the TF- filter time constant; 5) in accordance with the discrete domains of the PID control equation calculates a control signal value delta Δ «(«), the control signal M and the previous time ( «- for adding l) is obtained by the adder the current control signal output timing < «); and [mu] («) clipped to prevent windup, the output from the D / a converter to the actuator, by the action of the actuator to the controlled object, the controlled object runs in within a given range, while displaying the current state parameter, scroll forward a series of original data unit, again and again, to achieve effective control of the entire industrial system.
2、根据权利要求I所述的单输入单输出系统的极限PID控制方法,其特征是,整定控制器时,或者先将/Lmin和辨识出的模型参数输送到极限PID控制器参数的计算单元,由第4)步给出的公式计算出PID控制器的四个参数,再送入控制单元实现闭环控制,观测响应曲线,通过在线的从小到大单调增加I来整定控制器,直至获得要求的响应。 Limit PID control method, according to claim I of the single input single output system, characterized in that, when the tuning controller, or first / Lmin identified and supplied to the model parameter calculation unit limit of PID controller , calculated by the formula 4 given step) out of the four parameters of the PID controller, and then fed to the control unit closed-loop control, the observed response curve, from small to large online I monotonously increases to tune the controller, until the required response.
3、根据权利要求I或者2所述的单输入单输出系统的极限PID控制方法,其特征是,控制器可调参数I的步长小于0.0 W,对于纯滞后很小的控制对象通过时间常数r来确定典型步长。 3, I, or PID control limit SISO system according to claim 2, wherein the controller I tunable step length of less than 0.0 W, for the pure delay control target by a small time constant r is typically determined step size.
4、根据权利要求I所述的单输入单输出系统的极限PID控制方法,其特征是,如果控制系统所需达到的标称性能指标已指定,按照以下经验公式得到A的初值或/I调节的参考范围: Limit PID control method, according to claim I SISO system, wherein, if the control system required to achieve the nominal performance indicators have been designated, the initial value obtained in accordance with the following empirical formula A or / I adjusting the reference range:
式中:σ一超调量,t一上升时间。 Where: σ overshoot a, t a rise time.
5、根据权利要求4所述的单输入单输出系统的极限PID控制方法,其特征是,所述的经验公式,构建起了系统所需达到的标称性能指标和最佳控制器参数之间的直观联系,连同Ainin计算方法一同装入监控模块中,并将二者计算结果中较大值送入主机存储单元RAM中。 Limit PID control of single-input single-output system 5, according to claim 4, characterized in that, between the nominal system required to achieve optimum performance and controller parameters of the empirical formula, built up visual contact, together with Ainin calculation load monitoring module, and the calculation result in both a large value into the main storage RAM.
6、根据权利要求I所述的单输入单输出系统的极限PID控制方法,其特征是,所述的步骤3)中,判断该信号极性,并据此计算误差信号,具体规则是:如果是正作用,则误差信号<«)等于系统输出><»)减系统输入十O后的差值;如果是负作用,则e(n)等于r(«)减:K»)后的差值。 6. Limit PID control according to claim I SISO system, wherein said step 3), it is determined that the signal polarity, and calculate the error signal, the specific rule is: if difference value): If the negative effect, the e (n) is equal to r ( «K) minus»; is a positive effect, the signal < «) equal to the system output> <» the error) Save system input difference value ten O .
7、根据权利要求I所述的单输入单输出系统的极限PID控制方法,其特征是,所述步骤5)中,Δμ(«)计算公式如下: Au(n) = bte(n) + b2e(nI) + b3e(n -2) + aAu(n -1) 其中, 7 Limit PID control according to claim I of the single-input single-output system, and wherein said step 5), Delta] [mu ( «) is calculated as follows: Au (n) = bte (n) + b2e (nI) + b3e (n -2) + aAu (n -1) where,
.1 尺£.一控制器增益,7;—控制器积分时间,Td—控制器微分时间,7>—滤波器时间常数,Γ一系统采样周期,Δμ(«)—当前《时刻控制器输出信号增量,Διφ-I)—«-I时刻控制器输出信号增量,e(«)—«时刻跟踪误差,e{n-\)—n-\时刻跟踪误差,e(«-2)-«-2时刻跟踪误差。 .1 foot £ a gain controller, 7; - a controller integral time, controller derivative time Td - 7> - the time constant of the filter, a Gamma] sampling period, Delta] [mu ( «) - current" time controller output increment signal, Διφ-I) - «- I incremental timing controller output signal, e («) - «time tracking error, e {n - \) - n- \ time tracking error, e (« - 2) - «- second time a tracking error.
CN 200510112230 2005-12-29 2005-12-29 Utmost PID control method of single input single output system CN100476644C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 200510112230 CN100476644C (en) 2005-12-29 2005-12-29 Utmost PID control method of single input single output system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 200510112230 CN100476644C (en) 2005-12-29 2005-12-29 Utmost PID control method of single input single output system

Publications (2)

Publication Number Publication Date
CN1794118A CN1794118A (en) 2006-06-28
CN100476644C true CN100476644C (en) 2009-04-08

Family

ID=36805621

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 200510112230 CN100476644C (en) 2005-12-29 2005-12-29 Utmost PID control method of single input single output system

Country Status (1)

Country Link
CN (1) CN100476644C (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100428090C (en) 2006-10-19 2008-10-22 上海交通大学 Distribution type PI and PID controller quantitative setting method for industrial multi-input-output process
KR100866213B1 (en) 2007-02-09 2008-10-30 삼성전자주식회사 Proportional-integrate-derivative conrol apparatus and method
CN103294073B (en) * 2013-05-13 2016-04-27 广东电网公司电力科学研究院 First Order Time Delay ADRC control system and control method
CN103426344B (en) * 2013-05-23 2016-03-02 浙江工业大学 Automatic control bench
CN103558755B (en) * 2013-11-05 2017-02-08 四川理工学院 Fractional integral pid controller tuning and self-tuning method
CN103576711B (en) * 2013-11-14 2015-09-09 上海交通大学 Chemical reactor temperature was controlled based on the quantitative control of pid single
CN105599439A (en) * 2015-10-29 2016-05-25 华中科技大学 Online-parameter-identified self-adapting color register control system
CN106094509A (en) * 2016-06-22 2016-11-09 广东电网有限责任公司电力科学研究院 LCR filter control method and LCR filter control device

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5687077A (en) * 1991-07-31 1997-11-11 Universal Dynamics Limited Method and apparatus for adaptive control
CN1410853A (en) * 2002-11-21 2003-04-16 上海交通大学 Industrial process single parameter ration integrating and differential control method
CN1445629A (en) * 2003-03-06 2003-10-01 上海交通大学 Intelligence Proportional-integral-aifferential control method with quantitative positive definite robustness
CN1604092A (en) * 2004-11-04 2005-04-06 复旦大学 Single input single output RCS interconnection circuit degradation method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5687077A (en) * 1991-07-31 1997-11-11 Universal Dynamics Limited Method and apparatus for adaptive control
CN1410853A (en) * 2002-11-21 2003-04-16 上海交通大学 Industrial process single parameter ration integrating and differential control method
CN1445629A (en) * 2003-03-06 2003-10-01 上海交通大学 Intelligence Proportional-integral-aifferential control method with quantitative positive definite robustness
CN1604092A (en) * 2004-11-04 2005-04-06 复旦大学 Single input single output RCS interconnection circuit degradation method

Also Published As

Publication number Publication date
CN1794118A (en) 2006-06-28

Similar Documents

Publication Publication Date Title
CA2380413C (en) Statistical determination of estimates of process control loop parameters
Miao et al. Automatic detection of excessively oscillatory feedback control loops
EP2083153B1 (en) Method of control and real-time comparison of a turbine using an alternate control strategy
AU743549B2 (en) Method of predicting overshoot in a control system response
JP3881871B2 (en) Fuel control method for a gas turbine, and a control device provided for it
US8478433B2 (en) Fault detection systems and methods for self-optimizing heating, ventilation, and air conditioning controls
US4314878A (en) Method of operating a papermachine drying line
Åström et al. Design of decoupled PI controllers for two-by-two systems
CN1143971C (en) Diagnostic device and method for pressure regulator
JP5180520B2 (en) Method for controlling the load on the machine and the turbine engine control system
US5355305A (en) Pattern recognition adaptive controller
CN101482448B (en) Method for determining and controlling fatigue load of a wind turbine, and wind turbines therefor
US6330484B1 (en) Method and apparatus for fuzzy logic control with automatic tuning
US6937909B2 (en) Pattern recognition adaptive controller
Araki PID control
JPH07191710A (en) Controller
CN103370538B (en) A method and apparatus for using a variation characteristic curve of the equivalent system, i.e., the adaptive control of the pump control curve
Zhang et al. Adaptive predictive functional control of a class of nonlinear systems
Huang A pragmatic approach towards assessment of control loop performance
Hägglund Automatic on-line estimation of backlash in control loops
Liu et al. Optimal-tuning PID controller design in the frequency domain with application to a rotary hydraulic system
CN101346675A (en) Automated tuning of large-scale multivariable model predictive controllers for spatially-distributed processes
CN1599889B (en) Detection and discrimination of instabilities in process control loops
US5568377A (en) Fast automatic tuning of a feedback controller
HU225571B1 (en) Process controller

Legal Events

Date Code Title Description
C06 Publication
C10 Request of examination as to substance
C14 Granted
C17 Cessation of patent right