Graeme Cole

A 5-L bioreactor was set up with an impeller, cooling and heating coils, probes for measuring pH, temperature and dissolved oxygen concentration. The reactor instrumentation was interfaced with a computer via data acquisition hardware. LabVIEW, a graphical programming language was used in all measurement and control programs. The reactor temperature control could be achieved by a conventional proportional-integral (PI) controller. In contrast controlling pH was difficult due to a considerable variation in the slope of the titration curve. A simple PI loop resulted in aggressive pH control actions, causing excessive base addition and likely demise of the microorganisms. In this work synthetic waste water was used for testing control schemes. The Generic Model Control (GMC) and gain-scheduling adaptive control strategies were investigated and compared with a PI control scheme. In these strategies, the rate change of pH was assumed to be a function of residence time, base addition, and pH, allowing the continuous calculation of base added against the pH measured. Some experimental and simulated results are presented for comparison.

The Instrumentation and Control Laboratory (IC-lab) was designed to provide separated modules of different types of heat exchangers, air- pressured vessels, water tanks, pumps, flow meters and control valves. These module s can be connected to provide various conventional pieces of process control equipment. Through the experiments designed for the IC-lab, students can obtain real hands-on experience. In the first two years of the IC course students learn to perform step-tests, develop linear and non-linear model, then design and implement P, PI and PID cont rollers in the real time systems. More sophisticated control schemes such as Cascade, Feedforward, Generic Model Control (GMC) or Fuzzy Logic Control can also be achieved in the lab. This paper demonstrated an example of generating a nonlinear system by connecting two pressured- air tanks and other modules. A step test is perform ed to determine an approximate model for the PI controller design. At the same time a mathematical model of the system is developed based on measurements and calibrations of the system. A GMC scheme is applied based on this model. Comparison of performances of the PI and GMC controllers to control the pressure in the vessel are finally presented and discussed.

The Pilot Plant at Murdoch University is equipped with a Honeywell Supervisory Control System, known as SCAN3000. The plant was designed to simulate three stages: grinding and digestion, clarification, and precipitation of the Bayer Process to manufacture alumina from bauxite. This paper presents some examples in design and implementation of a great number of feature-based feedback PI controllers and the implementation of sophisticated control laws such as Generic Model Control (GMC), Dynamic Matrix Control (DMC), Supervisory Control and Fuzzy Control. Although all features of the pilot plant are not currently available for teaching purposes, Process Control Engineering (PCE) students still have the opportunity to operate and implement many different control strategies to various unit operations such as working with a real process to obtain real hands-on experience. A different example is the upgrade and maintenance program based on Programmable Logic Controllers (PLC) which highlights the learning and developing of skills in using PLC and Supervisory Control and Data Acquisition systems (SCADA).

Control system architectures are expanding significantly in terms of functionality, complexity and scalability. Real-time programming methodologies have evolved in the last decade with increasing complexity of control system and scalability of equipment in the industrial automation domain. The paper proffers an analytical survey of tools and methodologies that can enhance the control action simulation and software design. The paper includes a diagnostic view for application of Real Time- Languages like Giotto and RT-Java for time critical tasks within the control system. Secondly, the paper uses the statistical approach to model the network induced delays to manifest their impact on control system operation. The aim of the paper is to analyze, model and validate the timing aspects within a control loop and physical plant operation.

Middleware allows the co-operating components of a distributed system to interact through a well-defined request/response architecture thus masking the issues related to heterogeneity, concurrency, and distribution of resources. In order to support real-time applications for high performance networks the model driven development needs to transit to a service oriented architectures. The main trend emerging from the middleware technology evolution is that objects are being replaced by components. Component based modelling is an efficient mechanism for building complex real-time applications. This paper expounds on component based middleware strategies like CCM and Enterprise Java Beans. The novelty of the paper is analysis of component based middleware for optimization of real-time applications, specifically related to business and distributed control.

The recent pioneering initiatives undertaken at the School of Engineering Science at Murdoch University associated with the way our students are taught in the last two years of their study will be introduced in this paper. The conversion of the third and fourth year of the study programme from the traditional model of teaching that incorporates lectures, tutorials and laboratory work into the engineering studio mode will be presented. The use of industry based projects and problems to devise a modern curriculum for the most advanced training of instrumentation and control engineers will be explained. The model is mapped across the attributes of Engineers Australia as well as the Bloom’s taxonomy of the cognitive domain which includes the six major levels of such domain, namely, knowledge, understanding, application, analysis, synthesis and evaluation. The paper will present these pioneering initiatives currently being used as a pilot program by Engineers Australia. By the end of the year such approach will be fully evaluated by Engineers Australia. It may become the basis of the education and training across many Engineering Schools in Australia, mainly due to the fact that it resembles as close and realistic as possible industry practice. It is undoubtedly useful as an advanced approach to graduate engineers with the attributes the industry requires. The various components of this approach, such as brainstorming, analysis, report making, progress reporting writing, presentations and project evaluation will be discussed. Examples will be provided so that one can appreciate the value of this approach. The final semester intership programme will also be briefly discussed to present the complete picture of the way Murdoch University’s Engineering students are encouraged to enter the profession with a higher level of experience when compared with many other programmes within the country.

A number of constraints have initiated a major curriculum revision of the Bachelor of Engineering degrees offered at Murdoch University (MU). The new curriculum will have mostly traditionally taught units in first and second year followed by design studios in third and fourth year, incorporating problem and project-based learning. Two examples from European Universities where problem and project based learning have been embedded in engineering curricula are reviewed. They highlight many positive aspects of student-focused teaching methods, but also show that there is inevitably some loss of technical competency for engineering graduates. The paper then investigates the transition from two traditionally taught units in Photovoltaics and Wind Energy Engineering at MU to an engineering design studio that is part of the Renewable Energy Engineering specialisation. First core content centred learning objectives are determined before a number of possible projects are evaluated. It is found that, in order to achieve technical as well as process skill competency, students should be exposed to a number of projects and a combination of traditional and problem/project-based learning methods should be applied. The paper also discusses the challenges of a successful implementation of the design studio model at Engineering MU, which will require resources, training and support for staff so that they are prepared for their new roles in a more student focused learning environment.

This paper presents the design and implementation of a biologically inspired four legged walking robot, with a certain level of comparable complexities and similarities to its biological counterpart. A four-phase walking strategy inspired from four legged animals has been proposed and implemented in the robot. Four parallel Subsumption Architectures and a simple Central Pattern Generator are used in the robot for physical implementation. Experimental results demonstrate that the robot employs the proposed walking strategy and can successfully carry out its walking behaviours under various experimental terrain conditions, such as flat ground, incline, decline and uneven ground.

The Renewable energy sector has been growing rapidly internationally for several years, with Australia being no exception. The rapid growth is leading to a demand of trained engineers in this field and a number of initiatives have taken place in Australia to address this need with new educational programs. In Western Australia, Murdoch University is offering an undergraduate engineering degree in Renewable Energy Engineering since 2001. This paper gives an overview on the new degree, its aims, and provides details on the specific renewable energy engineering units. Background information on the School of Engineering at Murdoch University and its facilities is also provided. First experience with the new degree shows that the program has attracted highly motivated students from a variety of backgrounds including students from Western Australia, Northern Territory and Canada. The program receives strong financial support from Western Power, the electricity utility in Western Australia, as well as the Alternative Energy Development Board (AEDB).