Heat consumption forecasting using partial least squares, artificial neural network and support vector regression techniques in district heating systems

Park, Tae Chang; Kim, Ui Sik; Kim, Lae-Hyun; Jo, Byung-Wan; Yeo, Yeong Koo

doi:10.1007/s11814-010-0220-9

Cited by 34 publications

(18 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Application of PLS in electricity-load prediction [131] PLS Calculation of failure rate of power equipments [132] PLS Natural gas load forecasting based on least-squares PLS support-vector machine [133] Heat consumption forecasting in district heating systems [134] PLS Classification…”

Section: Discussionmentioning

confidence: 99%

The diversity in the applications of partial least squares: an overview

Mehmood

Ahmed

2015

Journal of Chemometrics

View full text Add to dashboard Cite

Advances in technology make it happen to have massive amount of information in the form of multiple variables per object. The use of multivariate approaches for modeling the real-life phenomena is natural in such situation. There are numerous multivariate approaches in the literature, and its a challenge to stay updated on the possibilities. Partial least squares (PLSs) are one of the many modeling approaches for high-throughput data, and its use in different fields to address the variety of problems has been increased in recent years. We therefore present an overview of PLS's applications. The objective of this paper is to give a comprehensive overview on the advances in PLS algorithm together with its applications for regression, classification, variable selection, and survival analysis problems covering genomics, chemometrics, neuroinformatics, process control, computer vision, econometric, environmental studies, and so on. We have mainly presented different PLS approaches and their applications, so that the reader can easily get an understanding of possibility to use PLS for their own field. For further reading, literature references together with software availability are provided. Figure 2. Illustration of partial least squares (PLS) algorithm, and how to use it for regression, classification, variable selection, and survival analysis, is presented. For regression, trained regression coefficients together with test data provide the fitted response, while variable selection is actually to find the subset of X. In classification and survival analysis, usually, the influential PLS scores are respectively used with linear discriminant analysis (LDA) or with quadratic discriminant analysis (QDA) and proportional hazard regression.

show abstract

Section: Discussionmentioning

confidence: 99%

The diversity in the applications of partial least squares: an overview

Mehmood

Ahmed

2015

Journal of Chemometrics

View full text Add to dashboard Cite

show abstract

“…As discoursed in the previous section, exergy analysis is an influential tool in the design, optimization, and performance evaluation of energy systems. Exergy is the maximum theoretical useful work obtained if a system S is brought into thermodynamic equilibrium with the environment by means of processes in which the system S interacts only with this environment . Exergy is expressible in terms of four components: physical exergy, kinetic exergy, potential exergy, and chemical exergy.…”

Section: Theoretical Backgroundmentioning

confidence: 99%

“…The specific physical exergy X Ph can finally be defined as follows:

X = X^{P h} = h ((),,, T, P, z) - T_{0} S ((),,, T, P, z) - ([], h ((), T_{0}, P_{0}, z) - T_{0} S ((), T_{0}, P_{0}, z))

where h and S are the enthalpy and entropy, respectively, and T 0 and P 0 are the reference environmental temperature and pressure. Another form of exergy is associated with the heat transfer out of or into a control surface called thermal exergy:

X_{Q} = Q ((), 1 - \frac{T_{0}}{T})

where T is the uniform temperature at the control surface. Exergy of work is defined as the equivalent work of a given energy form.…”

Section: Theoretical Backgroundmentioning

confidence: 99%

“…Performing nonlinear, multidimensional interpolations between input and output parameters makes it possible to identify nonlinear relationships that exist between operational parameters and total exergy efficiency. In recent years, intelligent techniques such as ANNs have been increasingly applied to identify nonlinear and complex energy systems behavior …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Exergy analysis of NGL recovery plant using a hybrid ACO_R‐BP neural network modeling: a case study

Safarvand

Aliazdeh

Giri

et al. 2014

Asia-Pacific J Chem Eng

View full text Add to dashboard Cite

The major objective of the study is to make exergy analysis of natural gas liquid (NGL) process more understandable by coupling it with the use of an artificial neural network modeling. The presented method permits to provide an energy diagnosis of the process under a wide range of operating conditions. As a case study, Siri Island NGL Recovery in Iran is considered. The Aspen Plus process simulator linked with MATLAB Software was used to obtain thermodynamic properties of the process streams and to perform exergy balances. The results are validated with industrial data. The exergy destruction and exergetic efficiency for the main system components and for the entire system were calculated. Major sources of irreversibility in the process are identified, and the best conditions for process improvement are presented. The simulation results reveal that the exergetic losses of the separation towers, heating/cooling equipment, and compression/expansion section obtain the highest rank among the other components of the plant. The results show that the overall exergetic efficiency of the system is about 61%. After proposing new operational conditions, another exergetic analysis was made that caused a decrease of 6% in the exergetic losses of the entire system. Then, the recorded and calculated data are used as inputs for the neural network. The results show that ACO R is highly effective to optimize the performance of the neural networks to predict overall exergy efficiency. This method was compared with other current methods, and the results indicated that the integrated n Ant Colony Optimization-Back Propagation (ACO R -BP) provides the least error on the testing dataset.

show abstract

“…Such regression models are fitted to the real data according to a minimization criterion related to a norm of the differences (i.e., residuals) between estimates and observations [4]. The most commonly used regression approaches in the chemical processes are the artificial neural networks (ANN) [5][6][7] and the support vector regression (SVR) [8][9][10][11][12] because of their capability to accurately model the nonlinear and multi-dimensional processes [13,14]. However, the reliable predictions of the regression models highly depend on the adequate measurement data, and the predictions of such models are valid only inside the training data domain.…”

Section: Introductionmentioning

confidence: 99%

A comparative study of hybrid models combining various kinetic and regression models for p-xylene oxidation

Dong

Yan

2014

Korean J. Chem. Eng.

View full text Add to dashboard Cite

The hybrid modeling approach, which combines kinetic and regression module parts to an overall process model, is an attractive process modeling approach. However, the selection of various kinetic and regression models affects the hybrid model performance. As an illustrating example, we investigated the p-xylene (PX) oxidation process and summarized the published results of the PX oxidation kinetic model. The kinetic parameters of three kinetic models (i.e., three frequently used kinetic models of PX oxidation) were estimated and the fitting results were evaluated. Six hybrid models were then developed based on these three kinetic models and two regression models (artificial neural network and support vector regression). Afterwards, a comparative study of the six hybrid models was carried out based on various kinetic and regression models. The performances of these kinetic and regression models on the hybrid models were also evaluated. Finally, the best suitable hybrid model was obtained for the PX oxidation process.

show abstract

Heat consumption forecasting using partial least squares, artificial neural network and support vector regression techniques in district heating systems

Cited by 34 publications

References 3 publications

The diversity in the applications of partial least squares: an overview

The diversity in the applications of partial least squares: an overview

Exergy analysis of NGL recovery plant using a hybrid ACO_R‐BP neural network modeling: a case study

A comparative study of hybrid models combining various kinetic and regression models for p-xylene oxidation

Contact Info

Product

Resources

About

Heat consumption forecasting using partial least squares, artificial neural network and support vector regression techniques in district heating systems

Cited by 34 publications

References 3 publications

The diversity in the applications of partial least squares: an overview

The diversity in the applications of partial least squares: an overview

Exergy analysis of NGL recovery plant using a hybrid ACOR‐BP neural network modeling: a case study

A comparative study of hybrid models combining various kinetic and regression models for p-xylene oxidation

Contact Info

Product

Resources

About

Exergy analysis of NGL recovery plant using a hybrid ACO_R‐BP neural network modeling: a case study