Telling Tales: Models, Stories and Meanings

摘要: ion versus idealisation Applying both the abstraction (physical modelling) and (system identification) processes to same situation can often be very revealing. For example, ignoring friction, inertia compressibility using physical reasoning, leads an integrator model of actuator. Including viscous applying Newton's Laws, a form Figure 5; while allowing for fluid as well admits possibility oscillation. Yet numerical values parameters obtained by two routes given system vary considerably. System identification tends give 'best fit' order model, analysis provides useful information about sensitivity behaviour individual components. The approaches are not equivalent. In one practical example modelling robot arm, estimates important such natural frequency differed more than 50% approaches. [5] Modelling design significance process technological will considered in detail below. But set scene, consider following statements: When we cannot change, aim type handle easily. building new device or system, try it behave like suitable ideal model. inflation modelled (idealised) geometric progression, leading easy mathematics. A savings account, on other hand, is designed progression: mathematics defines system. Or, technical Ohm's Relationship brief call resistor. V/I resistor constant: feature conductors general. Similarly, inductors, capacitors, servomotors whole range artifacts carefully engineered way defined advance models with particular properties. Telling tales stories Let us now step back from problems process(es) pose general question: What position engineering! Our starting point observation that educators professional engineering bodies argue formal central essential part engineer's education (e.g. IMA, 1995; Croft et al, 2000). However, special status does seem rub off students who, after graduation, claim lot their mathematical expertise largely irrelevant jobs. seems missing educational practice any link between procedures taught practising engineers actually use. It used at all, but there clearly significant difference what mathematician calls 'doing mathematics' engineer mathematics'. We want suggest overly approach obscure those areas which supposed illuminate. Attention diverted away concentrated details Mathematics then becomes goal itself forget ignore quite fragile links were originally up its As result, easily confusion results relate primarily structure may interpreted (with care) terms Engineered systems complex structures own right, built people fulfil functions. From view, produce whose achieves function. Focusing too closely helpful where understand better: needs able think talk ways appropriate this goal. counts valid approach, therefore, likely differ approach. Engineers use variety they build. They mathematical, computer-based physical; exist equations, graphs, diagrams, description programs scale models. Models various purposes: insight, provide basis explanations performance, artefacts perform functions predict existing non-existing under conditions. neither itself, nor self-evident representation it. autonomous, stand-alone entities encapsulating all worth knowing component convenient package. Rather, points conversations among practitioners claimed represent. have mediated negotiated within community make sense. development, learn how models: tell them recognise sorts legitimate. Engineering why happens, future behaviour, as-yet-unbuilt devices. Explanations answers 'how',