Comparative study of different wavelet-based neural network models to predict sewage sludge quantity in wastewater treatment plant

Zeinolabedini, Maryam; Najafzadeh, Mohammad

doi:10.1007/s10661-019-7196-7

Cited by 56 publications

(20 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(7) can be rewritten in a simplified form as :

indicates the approximation sub-signal of

and level

and

is the detailed sub-signal at levels

. The dyadic wavelet transform

that can be represented as [29] :

where

.…”

Section: Related Workmentioning

confidence: 99%

Modelling and forecasting of COVID-19 spread using wavelet-coupled random vector functional link networks

Hazarika

Gupta

2020

Applied Soft Computing

105

View full text Add to dashboard Cite

Researchers around the world are applying various prediction models for COVID-19 to make informed decisions and impose appropriate control measures. Because of a high degree of uncertainty and lack of necessary data, the traditional models showed low accuracy over the long term forecast. Although the literature contains several attempts to address this issue, there is a need to improve the essential prediction capability of existing models. Therefore, this study focuses on modelling and forecasting of COVID-19 spread in the top 5 worst-hit countries as per the reports on 10th July 2020. They are Brazil, India, Peru, Russia and the USA. For this purpose, the popular and powerful random vector functional link (RVFL) network is hybridized with 1-D discrete wavelet transform and a wavelet-coupled RVFL (WCRVFL) network is proposed. The prediction performance of the proposed model is compared with the state-of-the-art support vector regression (SVR) model and the conventional RVFL model. A 60 day ahead daily forecasting is also shown for the proposed model. Experimental results indicate the potential of the WCRVFL model for COVID-19 spread forecasting.

show abstract

“…(7) can be rewritten in a simplified form as :

indicates the approximation sub-signal of

and level

and

is the detailed sub-signal at levels

. The dyadic wavelet transform

that can be represented as [29] :

where

.…”

Section: Related Workmentioning

confidence: 99%

Modelling and forecasting of COVID-19 spread using wavelet-coupled random vector functional link networks

Hazarika

Gupta

2020

Applied Soft Computing

105

View full text Add to dashboard Cite

show abstract

“…Analysis conducted on synthetic time series shows how a wavelet with a reasonable support and good time-frequency localization property is able to capture the underlying trend and short time variability in the time series signal. In [16], a total of 32 wavelet functions from 6 wavelet families are applied as part of a methodology to predict sewage sludge parameters from a given sewage sludge data set. With respect to determining the decomposition level (L), the authors indicate that the number of data points (N) in the time series signal determines the maximum decomposition level based on the following equation: L = int(Log(N)).…”

Section: A Applications Of Wavelet Transform In Time Series Data Dommentioning

confidence: 99%

“…The weather data is used with the approximation signal that represent the smoothed version of the load series. In [16], the authors use multiresolution analysis (MRA) to extract load characteristics and use for precise forecasting and provide tools for short-term load forecasting. The forecasting analysis is based on predicting the load's future behavior by independently forecasting the subseries of the load that is generated using wavelets.…”

Section: A Applications Of Wavelet Transform In Time Series Data Dommentioning

confidence: 99%

“…For each DWT function, three decomposition levels (level 1, 2 and 3) are applied. Equations (11)- (16) show the details and approximation when applied to the measured signal Pmeas, where Pmeas=[P0,P1, where N=sample size = 35,040 (see Algorithm 1 for more info) and i is the index for ith mother wavelet function among the 34 DWT functions. This will generate a taxonomy of 102 unique load variability profiles (34 unique load variability signals at level 1, 34 unique load variability signals at level 2, and 34 unique load variability signals at level 3) for each building, for each day of the year.…”

Section: Measured Load Profile Decomposition and Feature Extractionmentioning

confidence: 99%

See 1 more Smart Citation

Framework for Extracting and Characterizing Load Profile Variability Based on a Comparative Study of Different Wavelet Functions

et al. 2020

View full text Add to dashboard Cite

The penetration of distributed energy resources (DERs) on the electric power system is changing traditional power flow and analysis studies. DERs may cause the systems' protection and control equipment to operate outside their intended parameters, due to DERs' variability and dispatchability. As this penetration grows, hosting capacity studies as well as protection and control impact mitigation become critical components to advance this penetration. In order to conduct such studies accurately, the electric power system's distribution components should be modeled correctly, and will require realistic time series loads at varying temporal and spatial conditions. The load component consists of the built environment and its load profiles. However, large-scale building load profiles are scarce, expensive, and hard to obtain. This paper proposes a framework to fill this gap by developing detailed and scalable synthesized building load profile data sets. Specifically, a framework to extract load variability characteristics from a subset of buildings' empirical load profiles is presented. Thirty-four discrete wavelet transform functions with three levels of decomposition are used to extract a taxonomy of load variability profiles. The profiles are then applied to modeled building load profiles, developed using the energy simulation program EnergyPlus ® , to generate synthetic load profiles. The synthesized load profiles are variations of realistic representations of measured load profiles, containing load variabilities observed in actual buildings served by the electric power system. The paper focuses on the framework development with emphasis on variability extraction and application to develop 750 synthesized load profiles at a 15-minute time resolution. INDEX TERMS Load profile, building energy modeling, electric power system, distributed energy resources, wavelet functions, discrete wavelet transform.

show abstract

“…The capacity of the wavelet transformation method is superior to that of other preprocessing methods due to its capability to abstract non-trivial and significant time series information [2]. Extracting explicit information from historical data time series can resolve the non-linearity and non-stationarity [26].…”

Section: Research Motivation and Enthusiasmmentioning

confidence: 99%

The Capacity of the Hybridizing Wavelet Transformation Approach With Data-Driven Models for Modeling Monthly-Scale Streamflow

et al. 2020

View full text Add to dashboard Cite

Hybrid models that combine wavelet transformation (WT) as a pre-processing tool with datadriven models (DDMs) as modeling approaches have been widely investigated for forecasting streamflow. The WT approach has been applied to original time series for decomposing processes prior to the application of DDM modeling. This procedure has been applied to eliminate redundant patterns or information that lead to a dramatic increase in the model performance. In this study, three experiments were implemented, including stand-alone data-driven modeling, hind cast decomposing using WT divided and entered into the extreme learning machine (ELM), and the extreme gradient boosting (XGB) model to forecast streamflow data. The WT method was applied in two forms: discrete and continuous (DWT and CWT). In this paper, a new hybrid model is proposed based on an integrative prediction model where XGB is used as an input selection tool for the importance attributes of the prediction matrix that are then supplied to the ELM model as a predictive model. The monthly streamflow, upstream flow, rainfall, temperature, and potential evapotranspiration of a basin named in 1805 and located in the south east of Turkey, are used for development of the model. The modeling results show that applying the WT method improved the performance in the hindcast experiment based on the CWT form with minimum root mean square error (RMSE = 4.910 m 3 /s). On the contrary, WT deteriorated the performance of the forecasting and the stand-alone models exhibited a better performance. WT increased the performance of the hindcast experiment due to the inclusion of future information caused by convolution of the time series. However, the forecast experiment experienced deterioration due to the border effect at the end of the time series. Hence, WT was found not to be a useful pre-processing technique in forecasting the streamflow.

show abstract

Comparative study of different wavelet-based neural network models to predict sewage sludge quantity in wastewater treatment plant

Cited by 56 publications

References 58 publications

Modelling and forecasting of COVID-19 spread using wavelet-coupled random vector functional link networks

Modelling and forecasting of COVID-19 spread using wavelet-coupled random vector functional link networks

Framework for Extracting and Characterizing Load Profile Variability Based on a Comparative Study of Different Wavelet Functions

The Capacity of the Hybridizing Wavelet Transformation Approach With Data-Driven Models for Modeling Monthly-Scale Streamflow

Contact Info

Product

Resources

About