- 上一篇：采暖通风与空调装置英文文献和中文翻译
- 下一篇：提高级进模性能英文文献中文翻译
Fig. 3. Hierarchy of KSs using object-oriented knowledge organization
The lower level classes including “Unfolding_KS”, “Nesting_ KS”, “Mapping_KS”, “Staging_KS” and “Strategy_KS” can be represented as the child class objects of the Abstract_KS. They can inherit the latter’s attributes and methods, and also add specific slots and methods pertinent to themselves.
3.3 Agenda-based control module
The specialist KSs respond opportunistically to the events, i.e., changes on the blackboard due to new data or KS action. An agenda-based control module is used to monitor the events and decides the actions to be taken next. The agenda keeps track of all the events on the blackboard, and serves as a repository of specialist knowledge source activation records (KSARs) that can be selected for execution, and calculates the priority of execution.
The control module uses heuristic control rules as the strategy KS to set the above agenda, e.g., by defining the dynamic priorities of activated KSARs at the particular point in different stamping process planning stages, and invoking execution of a KSAR with the highest priority.
3.4Validation of the prototype system
An important aspect of the quality assurance of a knowledge based system is the validation. In this case validation consists of the following:
1. Check that the perse KS rules have been correctly elicited. The rules are reviewed with the designer to ensure the design intent has been accurately captured.
2. Check the technical feasibility of solutions generated by the prototype system. This is a manual post-solution check. The solution is checked against the rules to ensure all rules have been met by the solution.
3. Check the practicality of the solution against a known, proven solution. In this case, a design situation with a known, proven solution is used as the reference. The system solutions compared to the known solution. Differences are noted and discussed with the designer to ensure the solution will work in practice. The system solution ought to be as good as or better than the existing solution.
Note the validation is based on a sample set. The assurance that the system will generate “good, feasible, practical solution” outside this set is based on the underlying ontology and methods of inference. The present doesn’t focus on this aspect, but it has been discussed elsewhere in our previous work [22].
4 An illustrative example
A typical stamped metal part modeled in Solid Edge CAD system (Fig. 4) is taken as an example to demonstrate the blackboard-based stamping process planning approach in the prototype. The system starts with the retrieval of required geometrical information from the part CAD model, and user input of other technical information (e.g., part weight, surface treatments, blank material, annual production, press type, press tonnage, bolster dimensions, bed open dimensions, shut height, etc.) to produce the first level of hierarchy on the blackboard, i.e., stamped part level.
Then, the CAD KS (CAD API functions) analyzes the geometrical and technological information of the part and press objects, and extracts stamping feature objects to form the second level of hierarchy on the blackboard, i.e., stamping feature level (Fig. 5). The user may also add or revise the feature information interactively by means of an interactive tool residing in the blackboard environment. In this user interface, the right-hand window shows the generated feature relation graph, which represents interactions among stamping features (including three hole features, one slot feature, four bend features, three emboss features, two extrusion hole features, and five flat features) through four types of feature relations, i.e., “is-in”, “is-on”, “adjacent to”, and “precision-associated”. The left-hand window shows detailed information about a selected feature object.