The forecasting of future value of water consumption in an urban area is highly complex and nonli... more The forecasting of future value of water consumption in an urban area is highly complex and nonlinear. It often exhibits a high degree of spatial and temporal variability. It is a crucial factor for long-term sustainable management and improvement of the operation of urban water allocation system. This chapter will study the application of two preprocessing phase space reconstruction (PSR) and wavelet decomposition transform (WDT) methods to investigate the behavior of time series to forecast short-term water demand value of Kelowna City (BC, Canada). The research proposes two pre-process technique to improve the accuracy of the models. Artificial neural networks (ANNs), gene expression programming (GEP) and multi linear regression (MLR) methods are the tools that considered for forecasting the demand values. Evaluation of the tools is based on two steps with and without applying the pre-processing methods. Moreover, auto correlation function (ACF) is used to calculate the lag time. Correlation dimension is used to study the chaotic behavior of the dataset. The models’ relative performance is compared using three different fitness indexes; coefficient of determination (CD), root mean square error (RMSE) and mean absolute error (MAE). The results showed how pre-processing combination of WDT and PSR improved the performance of the models in forecasting short-term demand values.
13th International Conference on Hydroinformatics, 2018
A comprehensive understanding of water demand and its availability is essential for decision-make... more A comprehensive understanding of water demand and its availability is essential for decision-makers to manage their resources and understand related risks effectively. Historical data play a crucial role in developing an integrated plan for management of water distribution system. The key is to provide high-resolution temporal-scale of demand data in urban areas. In the literature, many studies on water demand forecasting are available; most of them were focused on monthly-scales. Since monitoring of time series is a prolonged and costly procedure, the popularity of disaggregation methods is a most recent desirable trend. The objective of this research is to transfer low-resolution into high-resolution temporal scale using random cascade disaggregation and non-linear deterministic methods. This study defines a new technique to apply previously proposed random cascade method to disaggregate continuous data of the city of Peachland. The accuracy of the results is more than 90%. It represents a satisfactory application of the models. The proposed approach helps operators to have access to daily demand without acquiring high-resolution temporal scale values. Although the disaggregated values may not be precisely equal with observed values, it offers a practical solution for the low equipped WDS and leads to lesser number of drinking water-related problems.
1st International WDSA / CCWI 2018 Joint Conference, 2018
Water demand forecasting is a nonlinear and non-stationary process. Different variables affect th... more Water demand forecasting is a nonlinear and non-stationary process. Different variables affect the value of demand directly or indirectly. These variables have non-stationary behavior relates to the demand time series, which should be considered in forecasting of water demand. Thus, in the first stage, chaotic behavior of water demand data was investigated to determine the number of effective variables by the value of correlation exponent. Then, the nonlinear process of water demand was forecasted by nonlinear local approximation (NLA) method using reconstructed phase space of input variables. Next, to improve the accuracy of the method, selected embedding dimension of input variables was decomposed by wavelet transformers. The application of phase space reconstruction (PSR) and wavelet decomposition were compared with the results of the models without pre-processing to evaluate their application in improving the models' accuracy in forecasting water consumption of Kelowna City (BC, Canada). The results showed that NLA forecasted more accurate than the results in previous literature in the same case study.
Increasing draught seasons and lack of access to potable water reserves have been the major risks... more Increasing draught seasons and lack of access to potable water reserves have been the major risks threatening water authorities and governments over the recent years. Therefore, long term water forecasts are receiving much more attention nowadays. Unlike the conventional projection of historical water demand, researchers have tried to implement sophisticated mathematical models to predict demand of water. Gene expression programming (GEP), as a relatively new forecasting technique, remains to be explored in this endeavor. The main purpose of this research was to assess the performance of GEP models using wavelet decomposition with 2 transfer functions (db2 and haar) and 3 levels. Results of this study showed GEP models can be highly sensitive to wavelet decomposition if all combinations of proper lag times are used as inputs feeding these models.
High complexity of water distribution systems' (WDS) dynamics has convinced researchers to look f... more High complexity of water distribution systems' (WDS) dynamics has convinced researchers to look for more sophisticated statistical approaches in urban water demand forecasting. Given the huge threat to major water reserves and ongoing draughts, water authorities are concerned with long term analysis of water demand to deal with uncertain future of this dynamic infrastructure. Researchers have tried a wide range of modelling techniques to propose an accurate model. However, applications of machine learning techniques are yet to be explored in detail. This research proposes a support vector machine (SVM) model, using polynomial kernel function to forecast monthly water demand of City of Kelowna (CKD), Canada. The prime objective of this research is to assess the use of phase space reconstruction prior to design of models' input variables combinations. Results of this study proved optimum lag time of the input variables can significantly improve the performance of SVM models.
The forecasting of future value of water consumption in an urban area is highly complex and nonli... more The forecasting of future value of water consumption in an urban area is highly complex and nonlinear. It often exhibits a high degree of spatial and temporal variability. It is a crucial factor for long-term sustainable management and improvement of the operation of urban water allocation system. This chapter will study the application of two preprocessing phase space reconstruction (PSR) and wavelet decomposition transform (WDT) methods to investigate the behavior of time series to forecast short-term water demand value of Kelowna City (BC, Canada). The research proposes two pre-process technique to improve the accuracy of the models. Artificial neural networks (ANNs), gene expression programming (GEP) and multi linear regression (MLR) methods are the tools that considered for forecasting the demand values. Evaluation of the tools is based on two steps with and without applying the pre-processing methods. Moreover, auto correlation function (ACF) is used to calculate the lag time. Correlation dimension is used to study the chaotic behavior of the dataset. The models’ relative performance is compared using three different fitness indexes; coefficient of determination (CD), root mean square error (RMSE) and mean absolute error (MAE). The results showed how pre-processing combination of WDT and PSR improved the performance of the models in forecasting short-term demand values.
13th International Conference on Hydroinformatics, 2018
A comprehensive understanding of water demand and its availability is essential for decision-make... more A comprehensive understanding of water demand and its availability is essential for decision-makers to manage their resources and understand related risks effectively. Historical data play a crucial role in developing an integrated plan for management of water distribution system. The key is to provide high-resolution temporal-scale of demand data in urban areas. In the literature, many studies on water demand forecasting are available; most of them were focused on monthly-scales. Since monitoring of time series is a prolonged and costly procedure, the popularity of disaggregation methods is a most recent desirable trend. The objective of this research is to transfer low-resolution into high-resolution temporal scale using random cascade disaggregation and non-linear deterministic methods. This study defines a new technique to apply previously proposed random cascade method to disaggregate continuous data of the city of Peachland. The accuracy of the results is more than 90%. It represents a satisfactory application of the models. The proposed approach helps operators to have access to daily demand without acquiring high-resolution temporal scale values. Although the disaggregated values may not be precisely equal with observed values, it offers a practical solution for the low equipped WDS and leads to lesser number of drinking water-related problems.
1st International WDSA / CCWI 2018 Joint Conference, 2018
Water demand forecasting is a nonlinear and non-stationary process. Different variables affect th... more Water demand forecasting is a nonlinear and non-stationary process. Different variables affect the value of demand directly or indirectly. These variables have non-stationary behavior relates to the demand time series, which should be considered in forecasting of water demand. Thus, in the first stage, chaotic behavior of water demand data was investigated to determine the number of effective variables by the value of correlation exponent. Then, the nonlinear process of water demand was forecasted by nonlinear local approximation (NLA) method using reconstructed phase space of input variables. Next, to improve the accuracy of the method, selected embedding dimension of input variables was decomposed by wavelet transformers. The application of phase space reconstruction (PSR) and wavelet decomposition were compared with the results of the models without pre-processing to evaluate their application in improving the models' accuracy in forecasting water consumption of Kelowna City (BC, Canada). The results showed that NLA forecasted more accurate than the results in previous literature in the same case study.
Increasing draught seasons and lack of access to potable water reserves have been the major risks... more Increasing draught seasons and lack of access to potable water reserves have been the major risks threatening water authorities and governments over the recent years. Therefore, long term water forecasts are receiving much more attention nowadays. Unlike the conventional projection of historical water demand, researchers have tried to implement sophisticated mathematical models to predict demand of water. Gene expression programming (GEP), as a relatively new forecasting technique, remains to be explored in this endeavor. The main purpose of this research was to assess the performance of GEP models using wavelet decomposition with 2 transfer functions (db2 and haar) and 3 levels. Results of this study showed GEP models can be highly sensitive to wavelet decomposition if all combinations of proper lag times are used as inputs feeding these models.
High complexity of water distribution systems' (WDS) dynamics has convinced researchers to look f... more High complexity of water distribution systems' (WDS) dynamics has convinced researchers to look for more sophisticated statistical approaches in urban water demand forecasting. Given the huge threat to major water reserves and ongoing draughts, water authorities are concerned with long term analysis of water demand to deal with uncertain future of this dynamic infrastructure. Researchers have tried a wide range of modelling techniques to propose an accurate model. However, applications of machine learning techniques are yet to be explored in detail. This research proposes a support vector machine (SVM) model, using polynomial kernel function to forecast monthly water demand of City of Kelowna (CKD), Canada. The prime objective of this research is to assess the use of phase space reconstruction prior to design of models' input variables combinations. Results of this study proved optimum lag time of the input variables can significantly improve the performance of SVM models.
Uploads
and mean absolute error (MAE). The results showed how pre-processing combination of WDT and PSR improved the performance of the models in forecasting short-term demand values.
and mean absolute error (MAE). The results showed how pre-processing combination of WDT and PSR improved the performance of the models in forecasting short-term demand values.