WO2000020938A1

WO2000020938A1 - Method for designing strategies for controlling industrial plants

Info

Publication number: WO2000020938A1
Application number: PCT/EP1999/007791
Authority: WO
Inventors: Guido Ceiner; Giuseppe Scafati; Massimo Danieli; Luca Ferrarini; Claudio Maffezzoni
Original assignee: Abb Research Ltd.
Priority date: 1998-10-06
Filing date: 1999-10-05
Publication date: 2000-04-13
Also published as: ITMI982150A1; IT1302614B1

Abstract

A method for designing industrial plant control and REGULATION strategies, comprising the steps of: decomposing a process to be controlled into a plurality of modules which contain a process partition and corresponding control functions; defining a hierarchical tree structure in which each node is constituted by one of the plurality of modules, the hierarchical relationship between the modules being defined by arcs which mutually connect the nodes; defining connecting terminals adapted for the exchange of control and/or process parameters and variables among the modules.

Description

METHOD FOR DESIGNING STRATEGIES FOR CONTROLLING

INDUSTRIAL PLANTS DESCRIPTION

The present invention relates to a method for designing strategies for the control and regulation of an industrial plant. More particularly, the invention relates to a method for designing plant control strategies, which is oriented to the industrial process performed in the plant.

It is known that in a cycle for developing the engineering of plant control systems, currently available tools allow to define plant control strategies as sets of blocks of control functions. The selection of these blocks is generally difficult due to their complexity and therefore a good result is very often dependent on the experience of the engineers assigned to studying the control and regulation functions. The documents, used to design the control strategies of the plant, are produced on the basic engineering phase and they contain purely functional information. It is known that, after the basic engineering phase, the workflow for designing said control strategies comprises a task where documents, able of describing the operation of the process being controlled, are prepared. Then, using these documents, the task of defining in detail the hardware and the software for implementing the control, monitoring and protection functions of the plant, is carried out.

It is well known that the integration between these two tasks is often very difficult. For example the first task can lead to an only partially correct definition of the plant control strategies. In fact, very often, it is carried on without having a complete and correct vision of the technological and practical problems that will be found for the actual implementation of the control system. This often leads a loss of efficiency in the cycle for developing the engineering of the control systems. Moreover, conventional design processes do not provide a method for designing control strategies, which is highly orientated to the characteristics of the specific industrial process. This fact has severe repercussions on the actual possibility to reuse solutions, already prepared and tested in the past. In order to evaluate whether a given already-defined control strategy is suitable for the requirements of a new plant it is in fact often necessary to read hundreds of logic and regulation diagrams, printed on a corresponding number of sheets of paper. This operation is obviously neither rapid nor simple to perform. Moreover, reading and interpreting a plant logic diagram often does not adequately guarantee that the prepared and tested solutions actually match the requirements of the specific plant being studied.

Therefore, in short, the designer is never absolutely certain that a previously used control strategy can adapt with the same assurances of good operation to the plant for which he is about to design the control system. The aim of the present invention is therefore to provide a method for designing plant control strategies which is highly orientated to the industrial process to be controlled.

Within the scope of this aim, an object of the present invention is to provide a method for designing plant control strategies, which allows deriving, from the structuring of the process, the architecture of the intended control functions.

Another object of the present invention is to provide a method for designing plant control strategies, which allows defining control functions easily reusable for subsequent control systems of other plants. Another object of the present invention is to provide a method for designing plant control strategies, which allows considerably increasing the efficiency of the cycle for developing the engineering of control systems. Another object of the present invention is to provide a method for designing plant control strategies, which is highly reliable, relatively easy to provide and at competitive costs.

Thus the present invention provides a method for designing strategies for controlling an industrial plant, characterized in that it comprises the steps that consist in: - decomposing the process to be controlled into a plurality of modules which contain a process partition and corresponding control functions; - defining a hierarchical structure in which each node is constituted by one of said plurality of modules, the hierarchical relationship between said modules being defined by arcs which mutually connect said nodes; - defining connecting terminals suitable for the exchange of control and/or process parameters and variables among said modules.

Further characteristics and advantages of the invention will become apparent from the description of a preferred but not exclusive embodiment of the method according to the invention, illustrated only by way of non-limitative example in the accompanying drawings, wherein:

Figure 1 is a general block diagram of the design method according to the present invention; and

Figure 2 is a block diagram of an example of one of the modules into which the industrial process to be controlled is decomposed by using the method according to the invention; and

Figure 3 is a view of another module obtained from the decomposition of the process by using the method according to the invention; and Figures 4, 5, and 6 are views of further modules produced by decomposing the process to be controlled; and Figure 7 is a view of a tree structure, which illustrates the connection of the various modules obtained by decomposing the process to be controlled. With reference to the above figures, the method according to the present invention comprises a step, designated by the reference numeral 1 , which leads to the creation of a description of the process by means of a plurality of modules 2. The plurality of modules 2 constitutes, as a whole, the process to be controlled. The decomposition of the process into the plurality of modules 2 is then used (reference numeral 3) to physically implement the software and the hardware for providing said modules. The implemented modules are designated by reference numeral 4.

The method according to the present invention provides the step of defining a hierarchical structure in which each node is constituted by one of said plurality of modules. Then the step of defining connecting terminals, suitable for the exchange of control and/or process parameters and variables among said modules, is foreseen.

The modeling of the process and of the corresponding control functions, provided by the method according to the present invention, is obviously dependent on the definition of module used to decompose the continuous process to be controlled into a hierarchical structure.

With reference now in detail to Figures 2 to 7, some preferred but not limiting embodiments of the method according to the present invention are presented. The hierarchical structure used can be, for example, the tree type shown in Figure 7. An attribute indicating the state of the various modules of the structure, can advantageously be used.

For example, the module which acts as root of the tree can be designated by the "hybrid aggregate" attribute and is designated by the reference numeral 15 in Figures 7 and 3. Accordingly, the state of the other modules can be defined with the attributes of "aggregate" or "hybrid". The module, whose state is aggregate, can be composed only of modules representing control functions (control modules), only of modules representing physical devices (physical modules), or of both. On the contrary, a module whose attribute is hybrid (designated by the reference numeral 5 in Figure 2) contains both a description of the process and a description of the control functions.

In detail, the module 5 can be provided with a topological description of the process. This can be composed by lists of the various physical elements that provide the plant and by an internal description of equations. These equations can be executed by an engine for simulating the process being controlled in order to simulate the behaviors of the functional elements of the plant. The block 6 in Figure 2 generally designates the topological description and the equations are generally.

The block 6 can also comprise a functional description of the control strategies that are pertinent to the process portion confined within said hybrid module 5. Therefore, advantageously, the hybrid module 5, and particularly the block 6, comprises a model (made of topology plus equations plus algorithms) as well as variables and parameters.

The variables can have a value, which is determined by the available simulator and have one or more of the following attributes: type, unit of measure, domain, name and comment.

The parameters are similar to the variables, except that they have a numeric value instead of a name. Only the operator can modify their value. Preferably, there can be physical and geometric parameters, and said parameters can be used to allow having different modules which have the same equations and variables but differ in the values of the parameters. The parameter definition can contain one or more the following attributes: type, unit of measure, domain, value and comment.

The model can preferably be assigned by means of the following attributes: - type: it contains information regarding the representation formalism which is actually used for the description of the model; and/or - description: it contains the relationships between the variables and the parameters. The typical descriptions of a model are equations (for physical process models), block diagrams or SFCs (Sequential Function Charts) (for control models), a possible combination of equations and block diagrams (for hybrid models in the strictest sense).

The hybrid module 5 also comprises a state variable 7, which indicates the state of a model. Said state variable can assume the following values:

- aggregate: the model comprises both a pure process part and/or a pure control part and/or - elementary control, the model comprises exclusively control functions; and/or

- elementary process: the model comprises only process information; and/or

- elementary hybrid: the model is different from an aggregate.

The method according to the present invention provides each module with terminals for exchanging parameters and/or variables that can be both physical terminals 8 and control terminals 9.

The terminals are a collection of variables, which can be exported externally to the module. The physical terminals 8 correspond to a non-causal physical connection by means of which power exchanges, due to the interaction of the physical quantities that are involved in the process, occur. The control terminals 9 instead correspond to a causal connection by means of which information exchanges occur. The control terminals are of the input and output types.

The hybrid module 5 also can preferably comprise an aggregation structure 10, which describes the composition of the aggregate module in terms of other modules. The aggregation structure 10 comprises advantageously:

- a list of component modules;

- connections between the terminals 8 and 9 of these modules.

Finally, each module can comprise an alphanumeric string 11 in which comments by the operator can be entered.

With the above definitions, an instrument, such as for example a sensor and an actuator, can be modeled by means of a module, typically with a physical terminal 8 and a control terminal (an input for an actuator and an output for a sensor). The model represents the physical behavior of the sensor or actuator. A module whose state is aggregate can be composed, as mentioned, only of control modules or only of physical modules or of both. There are no parameters in the aggregate module definition. This means that a precise composition/decomposition of the plant and of the control system is required. Where necessary, the parameters are made available in the elementary modules. Figure 3 shows, by way of example, the hybrid aggregate module 15, which comprises a controlled subsystem SI. The subsystem SI, represented in figure 4, comprises the controlled valves CV1 and CV2 respectively, a pipe PI while the reference numeral 6 indicates a block similar to the block 6 of Figures 2 and 3. The reference numeral 16 designates a coordinating submodule, subsequently illustrated in the tree structure of Figure 7 as the parent node of function nodes Fl described hereinafter. Figure 5 instead shows an exemplifying module related to one of the controlled valves, in particular CN1, which internally comprises a control portion 20, a module representing a valve 21 and a block 6 which is similar to the above-described ones.

Figure 6 instead shows an exemplifying control module, which corresponds to the control portion 20 and internally comprises control functions F1...F3 and a block 6, which is similar to the preceding blocks 6. In a further preferred embodiment, the method according to the invention can also allow the operator to perform editing, by means of external programs. The various aspects of the modules, identified in the hierarchical structure (topological description of the process, description in terms of process equations executable by a simulation engine, functional description of the control strategies), can therefore be inserted. It can be furthermore possible to manage all the kinds of connection provided by the modules by means of their physical terminals 8 and/or control terminals 9. The management of the libraries containing the prepared and already-tested solutions is particularly advantageous. It allows the operator to interface with the library modules by means of their topological process description. Accordingly, it is avoided the need of referring to control diagrams printed on paper, without having the certainty that said diagrams correspond to a control solution which can be implemented in the new process, which the engineer must deal with. In practice it has been observed that the method according to the invention fully achieves the intended aim and objects, since it allows to design plant control strategies in a highly process-orientated manner. An approach, which allows obtaining the architecture of the control functions from the structuring of the industrial process, is substantially implemented. The method thus conceived is susceptible of numerous modifications and variations, all of which are within the scope of the inventive concept. All the details may furthermore be replaced with other technically equivalent elements.

Claims

1. A method for designing strategies for controlling an industrial plant, characterized in that it comprises the steps of:

- decomposing the process to be controlled into a plurality of modules that contain a process partition and corresponding control functions;

- defining a hierarchical structure in which each node is constituted by one of said plurality of modules, the hierarchical relationship between said modules being defined by arcs which mutually connect said nodes;

- defining connecting terminals adapted for the exchange of control and/or process parameters and variables among said modules.

2. The method according to claim 1, characterized in that one or more of said modules comprise a topological description of the process portion with which they are associated.

3. The method according to claim 2, characterized in that one or more of said modules that comprise a topological description, comprise also description of the process portion in terms of equations which can be executed by a simulation engine.

4. The method according to one or more of the previous claims, characterized in that one or more of said modules comprise the functional description of the control strategies that are pertinent to the associated process portion.

5. The method according to one or more of the previous claims, characterized in that said exchanged variables are defined by one or more of the following attributes: type, unit of measure, domain, name and comment.

6. The method according to one or more of the previous claims, characterized in that said exchanged parameters are defined by one or more of the following attributes: type, unit of measure, domain, value and comment.

7. The method according to one or more of the previous claims, characterized in that each module comprises a state variable which defines the condition of said module.

8. The method according to one or more of the previous claims, characterized in that each module comprises an aggregation structure, which is suitable to describe the composition of a module in terms of other modules.