Use of neurofuzzy networks to improve wastewater flow-rate forecasting
Introduction
During the recent years, a large number of control algorithms have been developed for controlling the biological processes of wastewater treatment facilities (WWTFs). Probably, the more significant one is the predictive model-based controller (PMBC). This algorithm uses computer simulation to predict the behaviour of a facility and later, with this information, modifies the inputs of the process to optimise the operational results and/or costs. In order to use properly a predictive model-based controller in a WWTF, a forecasting tool that let the computer to know future flow-rates and pollutant loads, is needed. The most frequently used forecasting methods are decomposition/recomposition of time series models (BLS, Census Methods, etc), auto regressive moving average (ARMA) models and auto regressive integrated moving average (ARIMA) models (Koutroumanidis et al., 2006). In addition, new models based on artificial intelligence are used as predictive models, especially neural networks (NNs). These techniques have been compared with the traditional ones showing significant improvements (White, 1988, Wong et al., 1992, Weigend et al., 1992, Tamada et al., 1993, Onkal-Engin et al., 2005, Raduly et al., 2007, Al-Alawi et al., 2008), which makes their application in the PMBC very interesting. In the same way that NN, fuzzy logic (another branch of the artificial intelligence) has important characteristics that make it suitable for different uses (Yong et al., 2006, Barreto-Neto and de Souza Filho, 2008). Moreover, neural networks and fuzzy logic share the common ability to deal with difficulties arising from uncertainty, imprecision, and noise in a natural environment. In this context, learning capabilities of NN, and as a consequence their potential use as universal approximators (Hornik et al., 1989, Poggio and Girosi, 1990), is the main factor in their selection as forecasting methods. About fuzzy systems, they are also used as approximators, since they can easily approximate any continuous function on a compact set (Kosko, 1992).
In order to get both, the benefits of neural networks and fuzzy logic systems, avoiding at the same time their respective problems, many authors proposed to combine them into an integrated system, such that the low level-learning and computation power of neural networks can be implemented into the fuzzy logic systems, and also, provide the high-level human-like thinking and reasoning of fuzzy logic systems into the neural networks. These systems are called fuzzy neural network (FNN) and they have been widely used in many different applications during the last decades (Enbutsu et al., 1993, Tanaka et al., 1995, Hiraga et al., 1995, Gobi and Pedrycz, 2006, Luo et al., 2007, Modi et al., 2008).
The main objective of this work has been to use FNN to model the influent flow-rates in different WWTFs. The input data used consisted of the influent flow-rate to the WWTFs during the previous day and the day of the week. The single output consists of the influent flow-rate forecast. Daily data during 3 months were used for training the network (parameter estimation) whereas a different set of flow-rate data (which contains the flow-rates of one month) was used to test the generalisation ability of the FNN at various stages of learning (model validation). The Performance Index (P.I.) as defined by Lin and Cunningham (1995) was used to evaluate the generalisation ability of the FNN, being the cross-validation criterion (Amari et al., 1997) used to determine the best moment to stop the learning process.
In order to increase the credibility and impact of the model presented here, the 10 iterative steps for good, disciplined development of models (Jakeman et al., 2006) were taken into account. Thus, the objective proposed for the model was not only to reproduce the observed flow-rates but also to be a first stage in the development of a procedure to obtain rules with a physical meaning from raw flow-rate data. These rules would be subsequently used in a PMBC in order to enhance the WWTF control. Because of their final users, personnel of WWTFs, one of the constrains of the model is that it should not be too complex. Taking this in mind, the topology of the network used corresponded to the three stages in the development of a fuzzy system: fuzzification, rules and defuzzification. This topology was selected because it allowed to consider the neural network not as a black box model but as a model that can obtain rules with a physical meaning from raw flow-rate data. The parameter estimation was obtained by applying two learning algorithms, a self-organised and a back-propagation one and the train a validation of the model was carried out using experimental observations in the WWTFs. In addition, it is important to bear in mind that the model proposed is not the final step of our research line but just a stage in a more ambitious program that aims to:
- •
Develop a procedure that allows to obtain meaningful rules (expert system) from flow-rate data without the direct cooperation of human thinking (primary fundamental objective). In a next step (future work) this will allow to simplify the design of expert systems to control of WWTP.
- •
Develop a procedure that can be implemented in a supervisory control scheme particularly inside a model predictive controller for optimizing the performance of WWTF (primary practical objective).
For this reason, and in order to avoid any problem due to software connectivity a home-made code was developed and used instead of using a commercial software package.
Section snippets
Formulation of the neurofuzzy wastewater flow-rate forecasting tool
The objective of this neural network is not only to model flow-rates (to be used in a PMBC) but to try to use the topology of the network to obtain rules with a physical meaning. Obviously, this work is a first step and the topology of the network has been specially designed to accomplish this objective. This allows to explain the topology of the FNN used and some of the assumptions taken into account on the model formulation (number of layers, neurons, etc.). In order to define the neural
Training and validation of the wastewater flow-rate forecasting FNN
One important point to be noticed is that the number of data for training and validating the FNN should be small because the FNN is developed for controlling processes and long-term changes should not be fed to the model because they can lead to wrong control actions. Taking this into account, a period of four months was considered good for training and validation of the model (3 months for training and 1 month for validation).
Conclusions
From this work the following conclusions can be drawn.
- •
Fuzzy neural networks are a suitable method for forecasting urban wastewater flow-rates.
- •
Using only two input variables and small number of neurones, average errors below 10% (expressed as PI) can be obtained in the foresight of urban wastewater flow-rates. Maximum errors are always under 22% (expressed as relative error).
- •
The artificial intelligence model cannot predict flow-rates under or over the values contained in the training data set,
References (25)
- et al.
Combining principal component regression and artificial neural networks for more accurate predictions of ground-level ozone
Environmental Modelling & Software
(2008) - et al.
Application of fuzzy logic to the evaluation of runoff in a tropical watershed
Environmental Modelling & Software
(2008) - et al.
The potential of fuzzy neural networks in the realization of approximate reasoning engines
Fuzzy Sets and Systems
(2006) - et al.
Multilayer feedforward networks as universal aproximators
Neural Networks
(1989) - et al.
Ten iterative steps in development and evaluation of environmental models
Environmental Modelling & Software
(2006) - et al.
Time-series modeling of fishery landings using ARIMA models and fuzzy expected intervals software
Environmental Modelling & Software
(2006) - et al.
A fuzzy neural network model for predicting clothing thermal comfort
Computers and Mathematics with Applications
(2007) - et al.
Fuzzy neural network based voltage stability evaluation of power systems with SVC
Applied Soft Computing Journal
(2008) - et al.
Determination of the relationship between sewage odour and BOD by neural networks
Environmental Modelling & Software
(2005) - et al.
Artificial neural networks for rapid WWTP performance evaluation: methodology and case study
Environmental Modelling & Software
(2007)
Intelligent control aeration and external carbon addition for improving nitrogen removal
Environmental Modeling & Software
Asymptotic statistical theory of overtraining and cross-validation
IEEE Transactions on Neural Networks
Cited by (49)
Dynamic nonlinear effects of urbanization on wastewater discharge based on inertial characteristics of wastewater discharge
2023, Science of the Total EnvironmentData to intelligence: The role of data-driven models in wastewater treatment
2023, Expert Systems with ApplicationsOptimal control of sewage treatment process using a dynamic multi-objective particle swarm optimization based on crowding distance
2023, Journal of Environmental Chemical EngineeringLeveraging water-wastewater data interdependencies to understand infrastructure systems’ behaviors during COVID-19 pandemic
2022, Journal of Cleaner ProductionCitation Excerpt :We plotted the relationships between wastewater flow and previous wastewater flows across multiple lag periods—e.g., 1-day lag of flow (i.e., flow in the previous day), 2-day lag of flow—to identify the lag with the highest correlation. For our wastewater-flow time series, 1-day lag turned to be the best lagged flow determinant, aligning with the literature (Fernandez et al., 2009; Zhang et al., 2019). Further, we plotted the wastewater flow with respect to the various factors to determine possible types of relationships.
Forecasting China's wastewater discharge using dynamic factors and mixed-frequency data
2019, Environmental PollutionCitation Excerpt :This technique has been widely applied to research on global climate changes because it has larger advantages in portraying nonlinear characteristics. Similarly, it has been used to forecast the amount of wastewater discharge (Fernandeza et al., 2009) including methane emissions (Du et al., 2018) and performs well. Indeed, the above Grey model and neural networks utilizing the past change trends of wastewater discharge do not take relative factors into consideration.