INTEC Acoustics
Effects of noise
The (possible) effects of environmental noise exposure on people are multiple: annoyance, sleep disturbance, stress, cardiovascular illness, mental illness, reduced cognitive capabilities (in children) and so on.
Quantifying these effects is largely complicated by the importance of inter-individual differences determining how people react to person-environment stress. But also the description of the exposure itself needs special care. First, different sound sources (like road traffic noise, railway noise or windturbines) might affect people in different ways. In addition, people are often exposed to many adverse environmental factors at the same time: air pollution, dust, noise and odour, emerging from several sources. Adding more detail to the description of exposure takes the most detailed form in soundscape research and design (please choose the "soundscapes" button on the right to find out more).
Research at INTEC focuses detailed noise assessment, modelling and simulation. Models serve a double purpose.
First, they improve fundamental understanding of the underlying mechanisms of exposure-effect relationships. Accurate computer simulations are impossible to make without thorough understanding of what is going on in real-life.
Second, they allow to predict future evolution, possibly under different scenarios.
Different types of models carry our interest:
Models that match an average trend to epidemiological data and thus allow to predict the effect on an "average person".
Models that aim at predicting the effect at the level of an individual or small group based on general field knowledge and epidemiological data. Such models have to take into account vagueness and imprecision.
Simulation based on knowledge on the auditory system, psycho-acoustics, neurology, psychology and so on, that tell us why and how the observed effect occurs (please click the "Perception and appraisal" topic to find out more about this).
Fuzzy Rule based models
Fuzzy Rule Based (FRB) models are often used in Artificial Intelligence systems to represent complex knowledge. Such models make field knowledge explicit in the form of a set of rules that can be read by humans easily - in contrast to e.g. neural networks. In the noise annoyance projects where we used FRB, epidemiologic databases were not used to extract rules but rather to validate rules that were formulated by field experts against the data. A Genetic Algorithm is used to tune the importance of the rules. We used FRB models to: model the cognitive processes involved in reporting overall annoyance based on combined exposure to sound from different sources; model the effect of person-related and contextual factors on annoyance by noise; to some extend for modelling coping with environmental noise and odour.
Fuzzy Integrals for aggregating Quality of Life indicators
Fuzzy integrals (Choquet and Sugeno type) can be used for aggregation. There applicability for aggregating "human style" has been investigated for several application related to effect of environmental stress prediction. One application of interest concerns the aggregation of various aspects of life and in particular the living environment to more overall indicators for the quality of life based on social surveying. Combined with models for average annoyance with noise and odor they can be used to predict the influence of these environmental stressors on overall quality of life in an area.
Integrated noise assessment
Within INTEC, modelling is combined with and applied to real-life environmental noise exposure; a more recent project involves annoyance due to wind turbine noise. Here, the relationship is investigated between the inhabitants' wind turbine noise annoyance, exposure indicators, operational characteristics and environmental variables. Combining data from regular on-line annoyance reports, long-term sound registration and analysis with electricity production and meteorological data not only provides insight in the exposure--effect relationship, but is also useful for the designing adequate operational restrictions.