Prediction of parameters characterizing the state of a pollution removal biologic process

Abstract

International audienceThis work is devoted to the prediction, based on neural networks, of physicochemical parameters impossible to measure on-line. These parameters-the Chemical Oxygen Demand (COD) and the ammonia NH 4-characterize the organic matter and nitrogen removal biological process carried out at the Saint Cyprien WWTP (France). Their knowledge make it possible to estimate the process quality and efficiency. First, the data are treated by K-Means clustering then by principal components analysis in order to optimize the Multi-Level Perceptron (MLP) learning phase. K-Means clustering makes it possible to highlight different operations within the Saint Cyprien treatment plant. The Principal Components Analysis (PCA) is used to eliminate redundancies and synthesizes the information expressed by a data set. With respect to the neural network used, these techniques facilitate the pollution removal process understanding and the identification of existing relations between the predictive variables and the variables to be predicted

Wavelet-based multi-resolution analysis and artificial neural networks for forecasting temperature and thermal power consumption

“…Because artificial neural networks also suffer from the difficulty of finding both an optimal topology and adequate training parameters [27], some different approaches were proposed. In 1957, Allen [28] tested a simple parametric model for minimum temperature prediction.…”

“…According to previous tests, more than one hidden layer proved to cause slower convergence during the learning phase because intermediate neurons not directly connected to output neurons learn very slowly. Based on the principle of generalization versus convergence, both number of hidden neurons and iterations completed during the training phase were optimized [68]. The multi-layer Perceptron neural network learns using an algorithm called backpropagation.…”

A comparative study: Prediction of constructed treatment wetland performance with k-nearest neighbors and neural networks

“…Artificial neural networks such as feed-forward neural networks were developed to predict the effluent concentrations including BOD, chemical oxygen demand (COD), and SS for wastewater treatment plants (Grieu et al, 2005;Hamed et al, 2004;Onkal-Engin et al, 2005), and to control water treatment processes automatically by modeling, for example, the alum dose (Maier et al, 2004). These studies have shown that ANN could be applied to establish a mathematical relationship between variables describing a process state and different measured quantities.…”

Artificial intelligence tools and inverse methods for estimating the thermal diffusivity of building materials

“…According to previous tests, more than one hidden layer proved to cause slower convergence during the learning phase because intermediate neurons not directly connected to output neurons learn very slowly. Based on the principle of generalization versus convergence, both number of hidden neurons and iterations completed during the training phase were optimized [19]. The multi-layer Perceptron neural network learns using an algorithm called backpropagation.…”

“…That is why the main objective of the present work is to test some tools belonging to the field of artificial intelligence (artificial neural networks and neuro-fuzzy systems) [14,15] with the aim of rebuilding the impulse response of a sample or for estimating directly its properties from its response to a random excitation. Artificial neural networks, a useful tool for modeling and controlling non-linear systems [16][17][18][19][20][21][22], are known as universal and parsimonious approximators. They present some interesting attributes, mostly their learning and generalization capabilities, to be used for rebuilding impulse responses of building materials (in this case, the thermal diffusivity of the concerned materials is thereafter determined by means of inverse methods; one could speak of "neuro-inverse" approach) or for directly estimating the above-mentioned thermophysical properties.…”