A neural network architecture for noise prediction

Automated identification and assessment of environmental noise sources

Noise pollution is one of the major health risks in urban life. The approach to measurement and identification of noise sources needs to be improved and enhanced to reduce high costs. Long measurement times and the need for expensive equipment and trained personnel must be automated. Simplifying the identification of main noise sources and excluding residual and background noise allows more effective measures. By spatially filtering the acoustic scene and combining unsupervised learning with psychoacoustic features, this paper presents a prototype system capable of automated calculation of the contribution of individual noise sources to the total noise level. Pilot measurements were performed at three different locations in the city of Ljubljana, Slovenia. Equivalent sound pressure levels obtained with the device were compared to the results obtained by manually marking individual parts of each of the three measurements. The proposed approach correctly identified the main noise sources in the vicinity of the measurement points.

Vehicular traffic noise modelling of urban area-a contouring and artificial neural network based approach

Road traffic vehicular noise is one of the main sources of environmental pollution in urban areas of India. Also, steadily increasing urbanization, industrialization, infrastructures around city condition causes health risks among the urban populations. In this study, we have explored noise descriptors (L₁₀, L₉₀, L_dn, LNI, TNI, NC), contour plotting and find the suitability of artificial neural networks (ANN) for the prediction of traffic noise all around the Dhanbad township in 15 monitoring stations. In order to develop the prediction model, measuring noise levels of five different hours, speed of vehicles, and traffic volume in every monitoring point have been studied and analyzed. Traffic volume, percent of heavy vehicles, speed, traffic flow, road gradient, pavement, road side carriageway distance factors were taken as input parameter, whereas L_Aeq as output parameter for formation of neural network architecture. As traffic flow is heterogenous which mainly contains 59%, two wheelers and different vehicle specifications with varying speeds also affect driving and honking behavior which constantly changing noise characteristics. From radial noise diagrams shown that average noise levels of all the stations beyond permissible limit and the highest noise levels were found at the speed of 50-55 km/h in both peak and non-peak hours. Noise descriptors clearly indicate high annoyance level in the study area. Artificial neural network with 7-7-5 formation has been developed and found as optimum due to its sum of square and overall relative error 0.858 and .029 in training and 0.458 and 0.862 in testing phase respectively. Comparative analysis between observed and predicted noise level shows very less deviation up to ± 0.6 dB(A) and the R² linear values are more than 0.9 in all five noise hours indicating the accuracy of model. Also, it can be concluded that ANN approach is much superior in prediction of traffic noise level to any other statistical method.

Modelling road traffic Noise under heterogeneous traffic conditions using the graph-theoretic approach

A traffic noise system involves several subsystems like road traffic subsystem, human subsystem, environment subsystem, traffic network subsystem, and urban prosperity subsystem. The study's main aim was to develop road traffic noise models using a graph theory approach involving the parameters related to road traffic subsystem. The road traffic subsystem variables selected for the modeling purposes included vehicular speed, traffic volume, carriageway width, number of heavy vehicles, and number of honking events. The interaction of the selected variables considered in the form of permanent noise function is given in the matrix form. Eigenvalues and corresponding eigenvectors are calculated for removing any human judgmental error. The permanent noise function matrix was then updated using the eigenvectors, which was ultimately utilized for obtaining the permanent noise index. Data regarding the selected variables were collected for three months, and the noise parameters included in the study were equivalent noise level (Leq,1h), maximum noise level (L10,1h), and background noise level (L90,1h). A logarithmic transformation was applied to the permanent noise index and linear regression models were developed for Leq,1h , L90,1h , and L10,1h respectively. The models were validated using the data collected from the same locations for nine months. The models were found to provide satisfactory results, although the results were somewhat overestimated. The method can prove beneficial for estimating future noise levels, given the expected changes in values for the independent variables considered in the study.

Application of Machine Learning to Include Honking Effect in Vehicular Traffic Noise Prediction

A vehicular road traffic noise prediction methodology based on machine learning techniques has been presented. The road traffic parameters that have been considered are traffic volume, percentage of heavy vehicles, honking occurrences and the equivalent continuous sound pressure level. Leq A method to include the honking effect in the traffic noise prediction has been illustrated. The techniques that have been used for the prediction of traffic noise are decision trees, random forests, generalized linear models and artificial neural networks. The results obtained by using these methods have been compared on the basis of mean square error, correlation coefficient, coefficient of determination and accuracy. It has been observed that honking is an important parameter and contributes to the overall traffic noise, especially in congested Indian road traffic conditions. The effects of honking noise on the human health cannot be ignored and it should be included as a parameter in the future traffic noise prediction models.

Comparison of land use regression and random forests models on estimating noise levels in five Canadian cities

Chronic exposure to environment noise is associated with sleep disturbance and cardiovascular diseases. Assessment of population exposed to environmental noise is limited by a lack of routine noise sampling and is critical for controlling exposure and mitigating adverse health effects. Land use regression (LUR) model is newly applied in estimating environmental exposures to noise. Machine-learning approaches offer opportunities to improve the noise estimations from LUR model. In this study, we employed random forests (RF) model to estimate environmental noise levels in five Canadian cities and compared noise estimations between RF and LUR models. A total of 729 measurements and 33 built environment-related variables were used to estimate spatial variation in environmental noise at the global (multi-city) and local (individual city) scales. Leave one out cross-validation suggested that noise estimates derived from the RF global model explained a greater proportion of variation (R²: RF = 0.58, LUR = 0.47) with lower root mean squared errors (RF = 4.44 dB(A), LUR = 4.99 dB(A)). The cross-validation also indicated the RF models had better general performance than the LUR models at the city scale. By applying the global models to estimate noise levels at the postal code level, we found noise levels were higher in Montreal and Longueuil than in other major Canadian cities.

Vehicular traffic noise prediction using soft computing approach

A new approach for the development of vehicular traffic noise prediction models is presented. Four different soft computing methods, namely, Generalized Linear Model, Decision Trees, Random Forests and Neural Networks, have been used to develop models to predict the hourly equivalent continuous sound pressure level, Leq, at different locations in the Patiala city in India. The input variables include the traffic volume per hour, percentage of heavy vehicles and average speed of vehicles. The performance of the four models is compared on the basis of performance criteria of coefficient of determination, mean square error and accuracy. 10-fold cross validation is done to check the stability of the Random Forest model, which gave the best results. A t-test is performed to check the fit of the model with the field data.

Modelling and evaluation of light railway system's noise using neural predictors

BACKGROUND

Noise is defined as a sound or series of sounds that are considered to be invasive, irritating, objectionable and disruptive to the quality of daily life. Noise is one of the environmental pollutants, and in cities it is usually originated from road traffic, railway traffic, airports, industry etc. The tram is generally considered as environmentally friendly, namely non-polluting and silent. However complaints from residents living along the tramway lines prove that it may sometimes cause annoyance. In this study, a Global Pointing System (GPS) receiver for determining the sampling locations and a frequency based noise measurement system for collecting the noise data are used to analyse the noise level in the city centre. Both environmental (background) and tram noises are measured.

RESULTS

Three types of neural networks are used to predict the noises of the tram and environment. The results of three approaches indicate that the proposed neural network with Radial Basis Function (RBF) has superior performance to predict the noises of the tram and environment.

CONCLUSIONS

For making a decision about transportation planning, this network model can help urban planners for evaluating and/or isolating the tram noise in terms of human health.

A basic neural traffic noise prediction model for Tehran's roads

We present an artificial neural network model to predict hourly A-weighted equivalent sound pressure levels (L(Aeq,1h)) for roads in Tehran at distances less than 4 m from the nearside carriageway edge. Our model uses the UK Calculation of Road Traffic Noise (CORTN) approach. Data were obtained from 50 sampling locations near five roads in Tehran at nearside carriageway edge distances of less than 4 m. The data were randomly assigned to training, testing, and holdout subsets. Model training was carried out using the training and testing subsets and comprised 60% and 20% of the data, respectively. Model validation was performed using the remaining 20% of data as a holdout subset. We examine the overall model efficiency using non-parametric tests, such as the Wilcoxon matched-pairs signed-rank test for the training step and the Kolmogorov-Smirnov test for two independent samples for the validation step. Our results indicate that a neural network approach can be applied for traffic noise prediction in Tehran in a statistically sound manner. The Wilcoxon matched-pairs signed-ranks test detects no significant difference between the absolute testing set errors of the developed neural network and a calibrated version of the CORTN model.

A neural network based model for urban noise prediction

Noise is a global problem. In 1972 the World Health Organization (WHO) classified noise as a pollutant. Since then, most industrialized countries have enacted laws and local regulations to prevent and reduce acoustic environmental pollution. A further aim is to alert people to the dangers of this type of pollution. In this context, urban planners need to have tools that allow them to evaluate the degree of acoustic pollution. Scientists in many countries have modeled urban noise, using a wide range of approaches, but their results have not been as good as expected. This paper describes a model developed for the prediction of environmental urban noise using Soft Computing techniques, namely Artificial Neural Networks (ANN). The model is based on the analysis of variables regarded as influential by experts in the field and was applied to data collected on different types of streets. The results were compared to those obtained with other models. The study found that the ANN system was able to predict urban noise with greater accuracy, and thus, was an improvement over those models. The principal component analysis (PCA) was also used to try to simplify the model. Although there was a slight decline in the accuracy of the results, the values obtained were also quite acceptable.

Short-term model of the production of construction aggregates in Taiwan based on artificial neural networks

BACKGROUND

Taiwan's geography and limited stock of sandstone have caused sandstone resources to gradually decline to the point of exhaustion after long-term excavation. Moreover, the Taiwanese government has continuously increased the amount of land area near rivers that cannot be excavated to facilitate riverbed remediation and promote conservation of water resources. Accordingly, predicting and managing the annual production of construction aggregates in future construction projects, and dealing appropriately with some thorny problems, for instance, demand that excess supply, excessive excavation, unregulated excavation, and the consequent environmental damage, will significantly affect the efficient use of natural resources in a manner that accords with the national policy of Sustainable Development (SD).

METHODS

. This study establishes an empirical model for forecasting the annual production of future construction aggregates using Artificial Neural Networks (ANN), based on 15 relevant socio-economic indicators, such as indicator of annual consumption of cement. A sensitivity analysis is then performed on these indicators.

RESULTS AND DISCUSSION

This work applies ANN to estimate the annual production of construction aggregates; the estimates, the verification of the model and the sensitivity analysis are all acceptable. Furthermore, sensitivity analysis results indicate that the annual consumption of cement is the indicator that most strongly influences the production of construction aggregates, as well as whether construction waste can be recycled and steel structures can be used in buildings, helping to reduce the future production of construction aggregates in Taiwan.

CONCLUSIONS

The elaborate prediction methodology presented in this study avoids some of the weaknesses or limitations of conventional linear statistics, linear programming or system dynamics. Additionally, the results not only provide a short-term prediction of the production of construction aggregates in Taiwan, but also provide a viable and flexible means of verifying quality certification of the production data of construction aggregates in the future by incorporating those relevant socio-economic indicators.

RECOMMENDATIONS AND OUTLOOK

The continuity and quality of the database of relevant indicators used in this study should be closely scrutinized in order to ensure the SD means of exploiting resources.

Sound propagation in urban areas: a periodic disposition of buildings

A numerical simulation of background noise propagation is performed for a network of hexagonal buildings. The obtained results suggest that the prediction of background noise in urban spaces is possible by means of a modified diffusion equation using two parameters: the diffusion coefficient that expresses the spreading out of noise resulting from diffuse scattering and multiple reflections by buildings, and an attenuation term accounting for the wall absorption, atmospheric attenuation, and absorption by the open top. The dependence of the diffusion coefficient with geometrical shapes and the diffusive nature of the buildings are investigated in the case of a periodic disposition of hexagonal buildings.