Introduction

Automated design (CAD) is been long and successfully applied to engineering design in a wide variety of fields, ranging from concepts spacecraft to the landscape design. Now a procent of the computer usage in engineering has been highly increased and with the growth of computer power with its relative cheapness it became normall that even a small business can afford to automate the design, and, accordingly, there is increasing specialization of the CAD up to tailor it with the requirements of a particular firm. Appointment of intellectual superstructure of the existing problem-oriented CAD is really that existing knowledge modules from modern CAD easily create CAD system for a specific task. The formal language to describe the model here is very important thing because for the crossing, association, additions and other manipulations with the knowledge there was a need for a clear mechanism for their description.

Relevance of the topic

Relevance of the topic Creating intellectual superstructure above problem-oriented CAD is not new, it is, in fact, emerged almost simultaneously with the CAD. But over the past decade it hasn't lost its relevance, because the need for specialized and localized to specific tasks of the business name since that time has only grown. A satisfactory solution to the task at the moment does not exist. This situation relates to the complexity of addressing plural theoretic operations on the text of the formal language of models. The proposed approach uses semiotic model [1], a modified model of SAU [2] and based on grammatical approach may be solution for most of these problems.

Planned research and development

Plans to automate the creation of models in the language describing patterns associated with the selected class of problem-solving CAD (in this work it will be VHDL) with the semiotic model for knowledge and a modified model of SAU as the contextual model of the subject area. The main components of the modified theory of SAU are :

• Target Space Systems Headquarters (TSSH), which represents a number of well-known prototypes;
• Space of the System's Faces(SSF), which includes a number of possible technical specifications for prototypes of TSSH.

There are styles SSF and GCHQ :

• words in the language of the specification, by then automatically builds and grammar used shadow products.
• grammar and clear productions, then asked TSSH and SSF will no longer change the system;

We need both options, the first for an expert who deals in the subject area, but doesn't understand the grammars, second for expert in knowledge. SSF should be linked to the principle : for any of TT SSF must have at least one prototype of TSSH (in the general case they might be plural), but they generally can be as dependent (if one of them based on the maintenance of another), and independent. The proposed development would have had TSSH based on a SSF. The second important component of the system is to use semiotic model, which includes :

• textual representation models
• separate module in the knowledge base in the form of breed context-free grammar equivalent AND-OR-tree (that is, we choose the form of withdrawal for our future CAD) and the products it.

The semiotic model uses these types of products :

• clear-sharing existence alternatives in OR-nods in AND-OR tree, introduces the user;
• shadow-set of rules applied by default over lists numbers of prototypes, which are related to each element of grammar;

Shadow spaces will be used for the automatic generation of text-based grammar program clearly will ask for expert knowledge.
In the proposed system experts will have such ways of developing CAD :

• Assigning grammar and products for TSSH and SSF; This path requires expert knowledge not only of the subject, but also in grammars;
• Assigning set of keywords for TSSH and the SSF, and the relationship between them, that the system should automatically build a grammar;

marketability TSSH-SF pair (for each element TT was at least one prototype in TSSH) to provide:

• variant with automatic grammar for the construction of a set of words that should provide the mechanism for building, will also be checked by a full search of the elements of both sets;
• In the version with the introduction of user grammar and marketability of products "on the conscience of the" expert system will only be obtained verification model and issue a warning that there are "empty" elements TT.

Scheme how user deals with a system would be look like this. User calls to SSF of the system using instrumental envelope, SSF by the marketability appealed to the TSSH, the decision of the TSSH, in turn, goes to a problem-oriented CAD for modeling and documentation, after which the user has ready-made solution (or solutions).

Conclusions

The existing system for the design of CAD have a great variety of disabilities. Planned research and development will make its contribution in their elimination. The research is planned to be finished in December of 2007.

References

1. Pospelov D.A. Situational management: theory and practices - Ì.: Nauka, 1986.
2. Solodovnikov V.V., Tumarkin V.I., Theory of complicity and projecting of management systems - Ì. Nauka. 1990. - 186p.
3. Grigoriev A.V. Semiotical model of the knowlage base in CAD. Scientific articles of the Donetsk National Technical University #10. : - Donetsk, DonNTU, 1999. - 30-37p. UDK 381.3
4. Grigoriev A.V. Levels of the expert's qualification and the ways of developing intellectual CAD. In the same compilation