文献翻译

外文文献原稿和译文

原 稿

INTRODUCTION

Nowadays, advanced control systems are playing a major role in plant operations because they allow for effective plant management. The prime advantage of a full plant automation stands in the improvement of plant profitability and productivity that lead to short term payoffs of the investment required for the implementation of advanced automation systems.

Typically, advanced control systems rely heavily on real-time process modeling, and this puts strong demands on developing effective process models that, as a prime requirement, have to exhibit real-time responses. Because in many instances detailed process modeling is not viable, efforts have been devoted towards the development of approximate dynamic models. The models suitable for real-time applications result from a trade-off between model complexity and degree of representability of the actual process dynamics.

Approximate process models can be classified in structured and unstructured models. While the former are based on first principles, and thus require good understanding of the process physics, the latter are based on some sort of black-box modeling and, in principle, would not require any a priori knowledge of the process. Neural network modeling represents an effective framework to develop unstructured models when relying on an incomplete knowledge of the process under examination. Because of the simplicity of neural models, they exhibit great potentials in all those model-based control applications that require real-time solutions of dynamic process models. The better understanding acquired on neural network modeling has driven its exploitation in many chemical engineering applications: data reconciliation and rectification, process identification and optimization, software sensor development (inferential measurements), state estimation, fault analysis, multivariable control (for a list of references cf. the recent review by Stephanopoulos and Han). The common

1

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

belief that effective and efficient neural can be developed without any a priori understanding of the investigated process is not fully correct. This misconception may lead to neural applications exhibiting poor performance, and this may have been the reason for preventing the full exploitation of neural modeling potentials in real-world applications. Indeed, the selection of the proper inputs and outputs to be fed to the network represents a critical step when developing a neural network model, and an adequate understanding of the process under examination is demanding.

Recently Baratti et al. have demonstrated the potentials of neural modeling to develop effective inferential control strategies of industrial multicomponent distillation columns. Figure 1 schematically shows a tray distillation column with the typical controlled (temperatures on specified trays, the so called pilot temperatures) and manipulated (reflux, L0, and boilup, V, flowrates) variables. Figure 1 refers to a case where the pilot temperatures are the ones of the 16th and 3rd trays.

The prime aim of this work is to investigate and fully exploit neural modeling in advanced model-based control applications of distillation columns. The basic idea is to construct a neural tool-box that, on the basis of detected disturbances, is capable of

2

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

acting properly on the manipulated variable so as to compensate for the upsetting loads as soon as they are detected. This action relies on a process modeling that must provide the description of the effect of disturbances and manipulated variables on the process performance. The direct identification of the action on the manipulated variable so as to meet the specifications requires a process inverse model. In this work, we discuss the use of neural modeling to develop an inverse model of a multicomponent distillation column. The neural model provides the controller action law. The direct action on the manipulated variable allows the control engineer to overcome the problems associated with the optimal tuning of the parameters of conventional feedback controllers. Furthermore, the action of the neural controller resembles that of a feedforward strategy, and thus it may be expected to exhibit a good control performance also in the case of a crude description of the process inverse model.

For the case of a single composition control problem of a distillation column, neural network modeling is used to develop a model-based neural control strategy to compensate for upsets in the operating pressure, feed flowrate, and feed composition so as to hold constant the content of the key component in the distillate stream. The neural controller is developed for a five-component distillation column (propane, butane, n-pentane, i-pentane and hexane) that is representative of actual stabilizer units, so commonly encountered in the refinery industries. The performance of the neural controller is compared with that of a conventional temperature control loop and of a neural inferential controller.

CASE STUDY

The neural controller is developed for a multicomponent distillation column for which the set of calibration data are produced making use of a dynamic simulator. A 18-trays (ideal stages), five component (propane, hexane, butane, n- and i-pentane) distillation unit is selected as prototype of a gasoline stabilizer tower. The column is also equipped with a reboiler and a total condenser. The feed (saturated liquid) is fed at the 11th stage from the bottom and is constituted of a mixture of propane (l0%), butane (5%), n-pentane (15%), i-pentane (30%) and hexane (40%) (the composition is given on a mole basis). The control objective is to maintain as low as possible the

3

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

content of i-pentane (i-C5) is the distillate and that of propane (C3) is the residue when the main upsets are due to variations or fluctuations in the feed composition, operating pressure and feed flowrate. The variation in the feed composition is the result of mixing the primary stream with a secondary one (slop) whose composition (on mole basis) is: C3, 2%; C4, 2%; C6, 45%; n-C5, 20%; i-C5 31%.Typically, the secondary stream accounts for about 5-15% of the total feed flowrate entering the distillation unit.

The dynamics of distillation columns can be simulated by various model approaches, involving different levels of complexity. Here, we are mainly concerned with addressing the issue of process inversion via neural modeling, thus we have adopted a simple description of the distillation unit that relies on the following assumptions: each tray is an adiabatic ideal equilibrium stage, all the involved species have the same latent heat of vaporization, the vapor and liquid are ideal mixtures, the vapor and liquid holdups are negligible. From these assumptions it follows that the vapor and liquid flowrates are constant along the column but different in the stripping and enriching sections. Moreover, the reboiler and the condenser dynamics have been described as first order processes. Under the prescribed assumptions, the mass and energy balance equations are decoupled, and the process model is simply constituted of the transient liquid mass balances. Within the framework of this description the tray temperatures are evaluated by computing the bubble points corresponding to the composition of the liquids on the trays. The equilibrium vapor pressures are calculated through the Antoine equation. The adopted dynamic model of the distillation column is rather simplified but, nevertheless, is capable of capturing the most significant dynamical features of the unit that are mainly related to the dynamics of the liquid compositions on the tray. Indeed, these are known to exhibit the largest characteristic times.

The design specifications are: i-C5 content in the distillate less than 0.5% (on mole basis) and C3 content in the residue less than 0.1%. The parameters corresponding to the reference case are summarized in Table 1.

MODEL-BASED NEURAL CONTROLLER

The relative gain array analysis has been used to select the most proper control

4

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

strategy, and has indicated the E (reflux ratio)/V (vapor boilup) strategy as the most appropriate for the prescribed objectives. In particular, the distillate prescription should be controlled with V while E should be preferred to regulate the residue specification.

The control of the residue specification exhibits rather serious problems due to the high volatility of propane that makes the residue C3 content almost insensitive to E. This results in high gains that lead to serious pitfalls when trying to implement the neural controller that is based on process inversion. Overall, the dual composition control problem exhibits rather ill-conditioned characteristics so as to prevent the neural controller from achieving a reasonable performance. Indeed, for the reference case, the condition number of the steadystate gain matrix is >> 1.

In this work, we use a multilayer feed-forward neural network trained through the backpropagation algorithm. The adopted network has three layers: the first layer (input) contains n1 nodes corresponding to the actual net inputs, the second layer (hidden) contains n2 nodes, and the third layer (output) contains n3 nodes that correspond to the number of the monitored state variables. The configuration of such a network is simply indicated as (n1–n2–n3 ). The feed and hidden layers are also augmented with a bias unit that represents the threshold value. The input value of the bias node is held constant and equal to 1. Two pilot temperatures, located at the 3rd and 16th tray from the bottom, are also used as inputs to the neural network. It is worth pointing out that we have selected as network inputs those that closely resemble the field measurements available for actual distillation units. Further details on the development of the multilayer neural network, on the selection, and on the pre-processing of the inputs are given elsewhere.

For the dual composition problem, three control systems based on the E／V strategy have been developed: a conventional temperature control loop, a neural inferential system, and a neural controller based on process inversion. The inferential control strategy relies on a neural observer that provides compositions as set points to a conventional PI feedback controller. The actions of the inferential and neural controllers are schematically shown in Fig. 2.

5

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

In the neural controller the actuator actions (on V and E) are modeled as first order filters, with a time constant typical of the industrial control valves. The actuator model is given by the following first order filter equation:

τdy/dt+y=x

where τ represents the time constant of the actuator device (control valve) and is given by Δtacq ／α；Δtacq represents the acquisition interval; y is the filtered signal (actuation); x is the manipulated signal, that is, the reflux ratio (E) or the boil-up rate (V) as predicted by the process inverse model. The parameter α governs the response speed of the actuator, in other words, it determines the rate at which the control action is able to compensate for the upsetting loads to the unit. The process model inversion is simply carried out by inverting the trained neural network exchanging the manipulated variables (E and V) with the controlled ones (the C3 content in the residue and the i-C5 content in the distillate). As anticipated earlier, the dual composition control problem exhibited an ill-character that has prevented us from efficiently performing the process inversion, thus, making unsuccessful any attempt to develop a model-based neural controller. The problem has been overcome by reconsidering the control objectives and discovering that the ill-condition character of the distillation unit is mainly related to the weak sensitivity of the residue composition (C, content) on the reflux ratio, E. This suggested releasing the control specification on the residue by only maintaining the one on the distillate, thus reducing the control

6

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

problem from dual composition to single composition. For the single composition control problem, the analysis of the steady-state gain matrix has led to selection of the reflux flowrate as a manipulated variable to control the distillate specification. The neural network used for the single composition problem has been furtherly simplified to a (5-l-l) configuration that has proved to be very efficient for process inversion, even though it was less accurate than the (7-2-2) network in representing the distillation unit.

DISCUSSION OF THE RESULTS

For the reference case, the relevant parameters used for simulating the distillation unit are summarized in Table 1. All the results discussed for the neural controller refer to the single composition control problem where the i-C5 content in the distillate (controlled variable) is regulated by acting on the reflux flowrate (manipulated variable).

The neural networks used to develop the inferential control strategy and the neural controller were trained using sets of calibration data spanning 50 h of operation. The calibration data were generated by making use of the dynamic simulator previously described. The simulations have been designed so as to capture the principal response features of the distillation units to variations in feed flowrate, feed composition, distillate and residue specifications, reflux and boil-up flowrates, and operating pressure.

At first, the performance of the neural controller has been evaluated with respect to the value of the parameter α, to test the response sensitivity of the neural controller to changes in the time constant of the actuator device. Increasing α, the response speed of the neural controller increases, at least up to a critical valueα0, above which the controller response is not significantly affected anymore. In reality, for values of α very close to 1 the neural controller exhibits an unstable behavior. For the case of a change of set point in the i-C5, content in the distillate, the response of the neural controller is shown in Fig. 3 for two values of a, corresponding to valve time constants of 50s (a) and 125s (b). As expected, increasing the value of α, the response rate of the neural controller increases, and thus it takes a shorter time to compensate for the set point change. Above the value of α= 0.1 no detectable changes in the

7

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

controller response occurred. By further increasing α the neural controller exhibited an unstable behavior above the value of 0.9. This is the reason why in all the following simulations the value of α= 0.1 has been adopted. The results of Fig. 3 also show the insensitivity of the residue composition (C3, mole fraction) that does not exhibit any significant change when the operating conditions are changed to accommodate for the new set point in the i-C5 content in the distillate stream. This explains why the neural controller is capable of meeting both the specifications on distillate and residue, even though it directly acts only on the manipulated variable that regulates the distillate composition.

A second series of tests have been carried out to compare the performance of the neural controller with that of the inferential strategy. The results, shown in Fig. 4, refer to a set point change in the distillate composition. The neural observer is based on a (7-2-2) neural network configuration, while the neural controller relies on a (5- 1-1) network topology. The responses of the two control systems (inferential and neural controller) are compared in terms of the capability in set point tracking. The neural controller exhibits a much faster response than the inferential strategy. The first is able to accommodate to the new operating conditions within about 8min, while the

8

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

second requires about 20min. To verify that the control action of the neural controller is indeed achievable, the required load on the manipulated variable is shown in Fig. 5 in terms of the percentage load variation with respect to the reference case. The results show that the neural controller is able to achieve a good setpoint tracking performance by requiring at most a 3% change in the manipulated variable.

9

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

Finally, the performance of the three control structures (temperature loop, inferential and neural controllers) are compared in Fig. 6 in terms of the disturbance rejection characteristics. The data of Fig. 6, spanning over an operating window of about 1000 h, refer to unit upsets due to fluctuations in feed flowrate (±1-2%), feed composition (±5-15%) and operating pressure (±3-4%). The results show that the control system based on the temperature loop (Fig. 6(a)) fails to meet the desired distillate specification (i-C5 composition), even though the temperature set point, not shown here, is met. The nonunique relationship between temperature and composition, for multicomponent systems, is the prime reason for the observed faulty performance of the temperature control strategy. On the other hand, the C3 specification on the residue stream is well satisfied, and this is due to its insensitivity to the operating conditions rather than to an actual efficacy of the temperature loop policy. The tall peaks exhibited by the i-C5 response correspond to perturbations in the feed composition that represent the most critical type of upsets to be faced by the temperature loop control.

The response of the inferential system is illustrated in Fig. 6(b), which clearly shows the improvement achieved over the temperature loop performance. The

10

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

inferential controller also performs well in rejecting the critical disturbances in the feed composition, the response peaks are attenuated by a factor of about 4, with respect to those detected with the temperature loop strategy. The neural controller response is illustrated in Fig. 6(c). The results show that this control strategy by far outperforms the previous two policies since it is capable of completely compensating for the upsetting loadings. In practice, with the neural controller a complete disturbance rejection is achieved. This is mainly due to the intrinsic characteristics of the neural controller that acts on the manipulated variable as soon as the upsetting conditions are detected, and this makes the correction action much faster.

CONCLUSIONS

For a multicomponent distillation column representative of an actual stabilizer unit, three control policies are discussed: a conventional temperature loop, an inferential strategy and a neural controller. The results show that control strategies based on process modeling (the inferential and neural controllers) perform better than the conventional temperature control loop.

The neural controller is based on process inversion that performed by simply inverting the neural network model of the distillation unit. The results indicate that for ill-conditioned processes, such as the one investigate here, the implementation of nonlinear control strategies based on process inversion requires a careful revision of the control objectives as well as of the input set to be fed to the neural network.

The neural controller by far outperforms the temperature loop and the inferential control strategies. In reality, the neural controller is the only one capable of achieving an almost complete disturbance rejection, even for the most critical upsetting loadings represented by fluctuations in the feed composition. . All this suggests that for multicomponent distillation units the conventional temperature loop strategies should be abandoned in favor of more advanced control policies.

The results clearly demonstrate the potentials of neural modeling for developing advanced nonlinear control strategies based on process inversion. It is remarkable that even for ill-conditioned processes, such as the one investigated here, neural modeling provides an efficient framework to perform process inversion.

11

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

译 文

概论

如今，先进的控制系统在工厂运营中发挥着重要的作用，因为它们可以使工厂进行有效的管理。一个全自动化工厂站的主要优点在盈利能力和生产力的提高，得益与短期的先进的自动化系统的实施。

典型的，先进的控制系统非常依赖于实时的流程建模，这将要求制定有效的过程模型，作为一个主要的条件，须显示有实时响应。因为在许多情况下详细的流程建模是不可行的，应该向近似的动态模型的发展。该模型适用于实时应用源自于的实际过程动态模型的复杂性和程度的表示性之间的权衡。

大致过程模型可分为结构化和非结构化模型。虽然前者是基于第一原则，因此需要很好的了解的过程中物理，后者则是基于某种黑盒模型，并在原则上是不需要任何先验经验的过程。神经式网络模型是一种有效的框架来开发非结构化模型依赖于一个不完整的知识的过程检查时，由于简单的神经模型，他们表现出巨大的潜力，在所有这些基于模型的控制应用程序需要动态的实时解决方案流程模型中。更好地理解神经网络模型需要开发在许多化学工程应用：数据的和解与整顿，识别和优化过程中，软件传感器发展（推理性测量），状态估计，故障分析，多变量控制（引用列表比照最近的审查由普洛斯和汉）。共同的想法，有效和高效的神经式的开发的调查过程中没有任何先验的认识是不完全正确的。这种误解可能会导致神经的应用效果变差，可能是这个原因，阻碍了在现实中的应用神经建模潜力的充分开发。实际上，选择适当的输入和输出被反馈到的网络是发展的神经网络模型的一个关键步骤，并根据检查的过程中有足够的了解要求。 最近巴拉荻等，已经证明了的神经建模来开发多组分推理控制策略的精馏塔的潜力。图1示意性地示出了塔板与典型的控制（温度在指定的塔板，所谓的导频的温度）的蒸馏塔和操纵（回流，L0和沸腾，V，流率）的变量。图1中是指这样一种情况，其中的试验温度的第16块和第3块塔板的温度。

12

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

图1 传统精馏塔的温度和操纵变量控制回路原理图

这项工作的主要目的是研究和充分利用先进的基于模型的控制应用程序的蒸馏塔的神经建模。基本的想法是构造一个神经工具，检测到的干扰的基础上，是能够形迹上正确操纵变量，以便补偿载荷和检测。此操作依赖于一个过程建模，必须提供的说明对过程性能的干扰和操纵变量的效果。直接识别的操纵变量的行动，以满足规格需要一个过程的逆模型。在这项工作中，我们讨论了使用神经建模开发的逆模型的多组分蒸馏塔。神经网络模型提供了控制器的操作法。操纵变量的直接作用于允许控制工程师，以克服这些问题，与传统的反馈控制器的参数的最佳调谐。此外，神经式网络控制器的动作的类似于一个前馈策略，因此，可以预期的粗描述的过程的逆模型的情况下也表现出良好的控制性能。 对于蒸馏塔的单一组分控制问题，用于开发基于模型的神经网络控制策略，来补偿操作压力，进料流量，进料组成，以保持恒定的馏出物流中的关键组分。神经网络控制器开发五组分的蒸馏塔（丙烷，丁烷，正戊烷，异戊烷和己烷），代表实际的稳定剂单元，因此通常在炼油工业中遇到。与常规的温度控制回路和一个神经推理控制器的神经网络控制器的性能进行分析。

案例分析

开发一种多组分的蒸馏塔，该校准数据集产生一个动态模拟器利用神经式网络控制器。一个18块塔板（理想状态），选择（丙烷，己烷，丁烷，n-及i-戊烷）5种组分蒸馏单元的汽油稳定塔的原型。精馏塔还配备了总的再沸器和冷凝器。饱和液体在第11块塔板的阶段从底部送入，丙烷的混合物（10％），丁烷（5％），（15％）n-戊烷，异戊烷（30％）和构成己烷（40％）（以摩尔组成为基础）。操作目标是保持尽可能低的i-戊烷第（i-C5）的内容是馏出物和丙烷（C3）的残基

13

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

时，主要是由于进料组合物中的变化或波动，操作压力和进料流量。加料混合物中的变化的结果的二次（斜率）的组成（按摩尔计）是：C3，2％；C4，2％； C6，45％；C5，20％，i-C5，31％。代表性地，约5-15％的总进料流量，进入蒸馏装置

的

精

馏

段

。

蒸馏塔的动态，可以模拟各种模型的方法，涉及到不同层次的复杂性。这里，我们主要是通过神经建模解决问题的过程转化有关的问题，因此，我们采用了一个简单的描述蒸馏装置依赖于以下假设：每个塔板是一个的绝热理想的平衡阶段，所涉及的品种有相同的汽化潜热，蒸汽和液体是理想的混合物，汽相和液相的含率可忽略不计的。从这些假设如下沿列的蒸汽和液体的流量是恒定的，但在剥离和丰富部分不同。此外，再沸器和冷凝器动力学已被描述为一阶过程。规定的假设下，去耦的质量和能量平衡方程，和过程模型是简单构成，瞬态的液体物料平衡。在此描述的框架内的塔板温度进行评估，通过计算相应的塔板上的液体组合物的气泡点。平衡蒸气压力计算通过所采用的动态模型的蒸馏塔是相当简化的Antoine平衡。但是，尽管如此，能够捕获的最显着的动力学特征的单元的主要是有关的液体混合物的动态塔板上。事实上，这些是已知的，表现出广泛时间的

特

征

。

设计的详细规格是：i-C5含量小于0.5％（按摩尔计）的馏出液和C3的残基的含量小于0.1％。总结于表1中的参数对应的参考情况。

基于模型的神经控制器

相对增加排列分析来选择最合适的控制方法，并表示E（回流比）/ V（蒸气蒸出）的方法，最适合的预定目标。特别是，馏出物处方应控制与V，而E应优选调

节

残

余

物

。

残留物的控制表明出相当严重的问题，是由于丙烷的高波动性，使残留C3的含量几乎不对回流敏感。这将导致高收效，导致严重的缺陷时，反转实施过程的神经网络控制器。总体而言，双组合物控制问题，而表现出病态特性实现合理的性能，以便防止神经控制器控制。事实上，参考的情况下，稳态增加排列的条件

>>

1

。

在这项工作中，我们使用了一个多层前馈神经网络通过反向推算法。所采用的网络有三层：第一层（输入）包含n1对应于实际的净输入节点，所述第二层（隐藏）包含n2的节点，并且所述第三层（输出）包含n3相对应的节点的数目监控状态变量。这样的网络的结构的是简单地表示为（n1-n2-n3）。也增加与表示所述阈

14

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

值的偏置单元的进料和隐藏层。偏压节点的输入值被保持恒定等于1。两项试验的温度下，位于从底部的第3和第16块塔板，也被用来作为输入的神经式网络。值得指出的是，我们选择了网络输入那些类似实地测量，可用于实际的蒸馏装置。多层神经式网络的发展，进一步细节上的选择，并在预先处理的输入给定的其他地

方

。

对于双组分的问题，已经开发了三个基于E/V控制系统的方法：一个常规的温度控制回路，一个神经推理系统和一个基于神经网络控制器的程序反转。推理控制方法依赖于一个神经观察装置提供的组合作为一个传统的PI反馈控制器的设定值。推理过程和神经式控制器见图2。

不平衡因素 推理式控制器

神经式控制器

图2 推理控制器与神经式控制器示意图

15

设定点 PI控制器 操作变量 过程 控制变量 检测 神经控制系统 干扰变量 不平衡因素 过程 操作变量 检测 神经控制器 设定点 北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

在神经式网络控制器的过程（在V和E）被建模为第一阶过滤器，与典型的工业控制阀的时间常数。以下的第一阶滤波器方程由下式给出的致动器模型：

τdy/dt+y=x

其中，τ表示的制动器的移动设备（控制阀）的时间常数，由下式给出Δtacq/α；Δtacq表示采集间隔； y是滤波后的信号（驱动）；x是操纵的信号，即，回流比（E）或蒸发速率（V），如所预测的过程的逆模型。参数α控制着制动器的响应速度，换句话说，它确定在其中，控制操作是能够补偿的单元的负载率。由反相受的神经式网络交换的操纵变量（E和V）控制的（C3含量在馏出物中的残余物和第i-C5含量）的方法，简单地进行模型反演。正如所预期的那样，的双组分控制问题表现出不良特性，使我们无法有效地进行过程中反转，因此，不能成功开发一种基于模型的神经网络控制器。此问题已得到解决，重新考虑控制目标和发现，不良条件下的蒸馏装置主要涉及的残留物组成的回流比（C，内容）的弱感光，建议释放控制规范上的残余物仅通过保持一个上的馏出物，从而减少从双组合物以单一组合物的控制问题。对于单一的组合物控制问题，导致的稳态增益矩阵的分析来选择作为一个操作变量来规范控制的馏出液中的回流流量。该神经网络用于单一组合物的问题已进一步简化至（5-1-1），已被证明是非常有效的过程反转结构，即使它相比（7-2-2）在代表网络蒸馏装置。

讨论的结果

有关的参考的情况下，用于模拟蒸馏单元的相关参数总结于表1中。神经网络控制器的所有讨论的结果是指单一组分混合物的回流流量（受操纵变量）作用于受控制问题的，其中的i-C5中的馏出液（受控变量）的含量。 神经网络用于开发的推理过程和神经网络控制器使用套校准的数据，50小时的操作进行了设置。生成的校准数据，通过使用先前描述的动态模拟器。仿真已设计，以便捕获在进料流量的变化，进料组合物，馏出液和残余物的规格，回流的蒸馏装置的主要响应功能和恒沸流速，和工作压力。 起初，神经网络控制器的性能评估相对于参数α值，测试的神经网络控制器向致动器装置的时间常数的变化的响应灵敏度。增加α，神经网络控制器的增加的响应速度，至少在一个关键参数α，上述控制器响应还没有显着地影响。在现实中，α接近1的值的神经网络控制器具有不稳定性。在i-C5中，如果在一个给定值的变化，在馏出液中的内容的情况下，神经网络控制器的响应示于图3。 两个值a，对应于阀的时间常数的50（a）和125S（b）。正如预期的那样，增加α的

16

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

值，增加神经控制器的响应速率，因此，它所需的时间较短，为给定值变化以补偿。以上的值α= 0.1控制器响应中没有可检测到的变化发生。通过进一步增加α值在0.9以上的表现出不稳定的现象。这是为什么在以下所有模拟值α= 0.1被采纳的原因。图3还示出的残留的组份（C3，摩尔分数）不表现出任何显着的变化的操作条件时，在i-C5的馏出物流中的内容为新的设定值被改变，以适应不敏感。这就解释了为什么神经式网络控制器是能够同时满足的规格对馏出物和残留物，它又能直接作用操纵变量，调节馏出物组合物。

摩尔分数

时间，min

图3 神经控制器单一关键组分变化响应值

第二系列试验已进行了的神经网络控制器与推理策略的性能进行比较。结果，示于图4，参考到一个给定值变化的馏出物中组合物。观察者是根据在（7-2-2）神经网络的配置，而神经控制器依赖于（5-1-2）的网络拓扑结构。两个控制系统（推理和神经网络控制器）的响应进行比较的设定值跟踪能力。相比推理方法神经网络控制器具有更快的响应速度。首先是能够容纳到新的操作条件下，在约8分钟内，而第二个需要约20min。为了验证神经网络控制器的控制动作的确实是可以实现的，所要求的负载上的操纵变量示于图5。结果表明，神经网络控制器能够实现一个良好的设定值跟踪性能要求最多有3％的可调节变量变化。

17

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

摩尔 分 数 摩 尔 分数 时间，min

图4 推理控制方法响应（a）和神经网络控制器（b）在馏出物的关键成分的变化设定值；表示法如图3

操作变量负荷 ％

时间，min

图5 负载操做变量百分比（回流流量）的神经控制器，数据参考图4（b）

最后，三种控制结构（温度循环，推理与神经控制器）的性能比较图6的干

18

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

扰抑制特性的数据的图6，跨越约1000小时的操作窗口中，提及的单元搅得由于在进料流量（为±1-2％），进料组合物（±5-15％）和工作压力（±3-4％的波动）。结果表明，该控制系统的基础上的温度循环（图6（a））不足以满足所需的馏出物规范（i-C5的组合物），即使温度设定点，此处未示出，满足。非唯一的温度和成分之间的关系，对于多组分系统，是所观察到的故障的温度控制策略的性能的首要原因。另一方面，残余物流的C3规范以及满意，这是由于其不敏感的工作条件下，而不是实际疗效的温度回路。高大的由第i-C5的响应表现出的峰对应于扰动的进料混合物中，代表所面临的温度循环控制的最关键的类型。

图6 时间，min

推理系统的反应示于图6（b），清楚地看出的温度循环性能取得了改善。推理控制器也表现良好在远离的临界的混合物中的干扰，响应峰是由大约4倍衰减，相对于检测的温度循环策略的那些。神经控制器响应示于图6（c）。结果表明，该控制方法远远优于先前两项的，因为它是能够完全补偿负荷。在实践中，与神经控制器实现一个完整的扰动抑制。这主要是由于神经网络控制器，作用于操纵变量尽快条件被检测到的固有特性，并且这使得校正的动作快得多。

摩尔分数％ 结论

对于多组分蒸馏塔代表实际的稳定器，三个控制方法进行了讨论：传统的温度循环，推理策略和神经式网络控制器。结果表明，控制策略，流程建模（推理和神经控制器）的基础上相比传统温度控制回路有更好的表现。

神经式网络控制器的基础上进行简单的倒置蒸馏装置神经网络模型的过程中反转，结果表明，错误情况下的过程，如一个调查这里，非线性控制策略的实

19

北京化工大学北方学院毕业设计（论文）——外文文献原稿和译文

施过程中反转的基础上，需要仔细修订的控制目标，以及将被输送到神经式网络的设备。

远远优于神经网络控制器的温度循环和推理控制策略。在现实中，神经网络控制器是唯一一个能够实现一个几乎完整的抗干扰性，即使是最关键的进料成分的波动所代表的负荷。所有这一切都表明，多组分蒸馏装置的策略应该放弃传统的温度循环，有利于更先进的控制方法。

结果清楚地表明神经式建模开发先进的非线性控制策略的过程反转的潜力。值得注意的是，即使对于错误情况下的过程，如这里研究的一个，神经建模提供了一种高效的框架来执行过程的反演。

20