Artificial Intelligence in Engineering最新文献

A new fuzzy controller for stabilization control of inverted pendulum systems is presented based on the Single Input Rule Modules (SIRMs) dynamically connected fuzzy inference model. The fuzzy controller has four input items, each with a SIRM and a dynamic importance degree. The SIRMs and the dynamic importance degrees are designed such that pendulum angular control has priority over cart position control. It is made clear that the fuzzy controller performs the pendulum angular control and the cart position control in parallel, and switching between the two controls is realized by automatically tuning the dynamic importance degrees according to control situations. The simulation results show that the proposed fuzzy controller has a high generalization ability to stabilize completely a wide range of the inverted pendulum systems within 9.0 s for an initial angle up to 30.0°.

Gesture-based applications widely range from direct manipulation interfaces to speaking aids for the deaf. The crucial point in recognizing gestures is that it requires great computational power to deal with spatio-temporal patterns. In this paper, a syntactic approach is proposed to provide a simple recognition algorithm. In order to verify the proposed method, we apply it to recognize 3D arm movements involved in the Taiwanese Sign Language. We extract prime patterns from the input patterns. The classification is then accomplished by deciding which one of possible arm movements can produce the sequence of primary patterns. Experiments were conducted to confirm the effectiveness of the method.

The usefulness of the elastic net (EN) approach for the general sequencing problem is investigated. Initial parameter setting and subsequent adjustments need to be intelligently undertaken for various applications. Based on the experiments on the typical traveling salesman problem, helpful guidelines were suggested. Computational results indicate that EN may be accepted as a candidate approach to a class of sequencing problems where multiple attributes are considered.

The paper investigates applicability of different artificial intelligence (AI) techniques in the design of a speed controller for electric drives. A speed-sensorless drive system is considered. A controller structure consisting of a load torque observer, a speed estimator and a speed predictor is developed. Next, different AI based approaches to speed controller design are investigated. The speed controllers based on (1) feed-forward neural network, (2) neuro-fuzzy network, and (3) self-organising Takagi–Sugeno (TS) rule based model are designed. A comparative analysis of the drive behaviour with these three types of AI based speed controllers is performed. In addition, a comparison is made with respect to the drive performance obtained with a conventional optimised PI controller. A detailed simulation study of a number of transients indicates that the best performance, in terms of accuracy and computational complexity, is offered by the self-organising Takagi–Sugeno controller. The controllers are developed and tested for a plant comprising a variable-speed separately excited DC motor.

This paper presents a neural network approach with successful implementation for the robot task-sequencing problem. The problem addresses the sequencing of tasks comprising loading and unloading of parts into and from the machines by a material-handling robot. The performance criterion is to minimize a weighted objective of the total robot travel time for a set of tasks and the tardiness of the tasks being sequenced. A three-phased parallel implementation of the neural network algorithm on Thinking Machine's CM-5 parallel computer is also presented which resulted in a dramatic increase in the speed of finding solutions. To evaluate the performance of the neural network approach, a branch-and-bound method and a heuristic procedure have been developed for the problem. The neural network method is shown to give good results and is especially useful for solving large problems on a parallel-computing platform.

Product design is a complex process, requiring many design factors and knowledge areas to be considered simultaneously. An expert system with a blackboard architecture is appropriate for representing this variety of factors and areas. However, oftentimes it is unclear how to transform the knowledge into the component knowledge bases of a blackboard expert system. The paper describes a non-directed graph decomposition procedure for transforming the knowledge into knowledge bases as part of an approach for developing a product design blackboard expert system. A wood head golf club design example is provided.

To reduce the manufacturing time of a product, one effective way to develop a machining process plan for a new part is to retrieve a relevant case of process planning similar to a new desired part and then adapt the retrieved case to meet the new requirements. This paper proposes a mechanism for retrieval of process planning cases. The core of the retrieval mechanism contains: (1) a feature-based representation of a part and cutting processes; (2) indexing of a part; (3) a feature hierarchical structure based on cutting processes; and (4) a similarity metric used to measure the similarity between a new desired part and any old part in the case base. The application domain here is for axisymmetric part machining. A prototype based on the retrieval mechanism is implemented on a Sun workstation using the ACIS 3D-Toolkit from Spatial Technology Inc.

This paper presents a case study that explores the advantages that can be derived from the use of a design support system during the design of wastewater treatment plants (WWTP). With this objective in mind a simplified but plausible WWTP design case study has been generated with KBDS, a computer-based support system that maintains a historical record of the design process.

The study shows how, by employing such a historical record, it is possible to: (1) rank different design proposals responding to a design problem; (2) study the influence of changing the weight of the arguments used in the selection of the most adequate proposal; (3) take advantage of keywords to assist the designer in the search of specific items within the historical records; (4) evaluate automatically the compliance of alternative design proposals with respect to the design objectives; (5) verify the validity of previous decisions after the modification of the current constraints or specifications; (6) re-use the design records when upgrading an existing WWTP or when designing similar facilities; (7) generate documentation of the decision making process; and (8) associate a variety of documents as annotations to any component in the design history.

The paper also shows one possible future role of design support systems as they outgrow their current reactive role as repositories of historical information and start to proactively support the generation of new knowledge during the design process.

Artificial intelligence can play an important role in the reduction of manufacturing costs and the enhancement of production efficiency and product quality. In order to assist designers in the early stages of a product development this paper develops an intelligent methodology for integration of design and assembly planning processes, including product design, assembly evaluation and redesign, assembly process planning, design of assembly system and assembly simulation, subjected to both econo-technical and ergonomic evaluations. A new unified class of object-oriented knowledge based Petri nets called OOKPNs, incorporating knowledge based expert systems and fuzzy logic into ordinary place–transition Petri nets, is defined and used for the representation and modeling of the distributed design processes. A prototype intelligent integrated design and assembly planning system (IIDAP) is implemented through distributed blackboard structure with concurrent integration of multiple cooperative knowledge sources and software. It consists essentially of the networked agents and the meta-system, each of which is a knowledge Petri net system with the capabilities of problem solving, learning and conflict resolution, and can be obtained through the inheritance, polymorphism and dynamic binding of instances of OOKPNs. In IIDAP system, both C/C++ language and COOL (CLIPS object-oriented language) are used to incorporate a Petri net tool, a geometric modeling and design tool, a planner and simulator and an evaluation tool. By use of this system, product design and assembly planning can be carried out simultaneously and intelligently in an entirely computer-aided concurrent design and assembly planning system. The design of manufacturable, cost-effective, usable products can therefore be achieved rapidly and flexibly. The developed methodology and system have been successfully applied to assembly design and planning of a micro switch.