Development of a feasibility prediction tool for solar power plant installation analyses

doi:10.1016/j.apenergy.2011.04.047

Applied Energy

Volume 88, Issue 11, November 2011, Pages 4078-4086

https://doi.org/10.1016/j.apenergy.2011.04.047 Get rights and content

Abstract

The solar energy becomes a challenging area among other renewable sources since the solar energy sources have the advantages of not causing pollution, having low maintenance cost, and not producing noise due to the absence of the moving parts. Although these advantages, the installation cost of a solar power plant is considerably high. However, feasibility analyses have a great role before installation in order to determine the most appropriate power plant site. Despite there are many methods used in feasibility analysis, this paper is focused on a new intelligent method based on an agglomerative hierarchical clustering approach. The solar irradiation and insolation parameters of Central Anatolian Region of Turkey are evaluated utilizing the intelligent feasibility analysis tool developed in this study. The clustering operation in the tool is performed by using the nearest neighbor algorithm. At the stage of determining the optimum hierarchical clustering results, Euclidean, Manhattan and Minkowski distance metrics are adapted to the tool. The achieved clustering results based on Minkowski distance metric provide the most feasible inferences to knowledge domain expert according to other distance metrics.

Highlights

► An agglomerative hierarchical clustering tool is designed for renewable energy sources in this study. ► In the model, nearest neighbor approach is used as clustering algorithm and Euclidean, Manhattan, and Minkowski distance metrics as distance equations. ► The developed tool assists knowledge domain expert in terms of analysing extensive datasets. ► The developed tool clusters the given sample data efficiently and successfully using each distance metrics. ► The clustering results are compared according to success rates.

Introduction

The increased energy demands and limited quotes of fossil fuels direct governments and individuals to renewable energy sources (RES) in electricity generation issues. The entire world embraces the “Clean Energy” and “Green Energy” mottos in order to support the RES usage. Furthermore generating the household or industrial electricity using RES, the latest trend in spreading clean energy to daily life is electrical vehicles (EV). Due to solar energy is the most abundant energy source, the usage of solar energy conversion systems are rapidly increasing among other renewable energy sources [1], [2], [3], [4]. Although the annual photovoltaic (PV) power generation is currently around 37 TW h in the World, it is estimated to be reached to 1247 TW h by 2030 and to 4572 TW h by 2050. The cumulative installed PV capacity that is around 27 GW in 2010 is expected to rise 872 GW by 2030 and 3155 GW by 2050, while this value was around 3.145 GW in 2003. The primary economic goal about PVs is to reduce constructed system prices and electricity generation costs by more than two-thirds by 2030. Installed plant prices are expected to drop by 70% from current USD 6000 per kW down to between USD 1200 and USD 1800 per kW by 2030, with a major price reduction already achieved by 2020. These prospects about solar energy are also related to PV panels and control techniques utilized to increase the efficiency of a panel while decreasing the installation costs dramatically [1], [5], [6]. The technological structures of solar energy conversion systems are also studied intensively besides short and mid-term planning studies [7], [8]. The efficiency rate of commercial PV modules is considered about 14–16% [9]. There are several control techniques are performed to increase the efficiency both in grid-connected or island mode solar power plants. The most widely used control techniques are maximum power point tracking (MPPT) and intelligent control methods such as fuzzy logic or artificial neural network (ANN) [10], [11], [12], [13].

Solar irradiation value and daily average insolation times are main parameters determining the efficiency of a PV module. Due to this situation, these parameters of the field where the solar power plant is intended to be installed should be analysed in detail. The feasibility analyses of the field should be acquired during a year at least. Since the hourly and daily measurements of the required parameters will cause to an overloaded data repository, it is not possible to infer for an expert by investigating himself/herself. The data stored in the defined repository is processed by a decision support system to assist knowledge domain expert in order to infer. This operation involves data mining applications such as evolutionary computation, adaptive fuzzy inference system, auto regression methods, and neural networks (NN) [14], [15], [16], [17]. In addition to these methods, hierarchical clustering (HC) methods that are based on unsupervised learning provide data-views at different levels of abstraction, and making them ideal for people to visualize. Although there are numerous studies are performed with HC method [18], [19], [20], [21], [22], [23], there is not any study performed in solar power plant analysis.

In this study, an intelligent feasibility analysis tool developed using the agglomerative HC method. In the model, nearest neighbor approach is used as clustering algorithm and Euclidean, Manhattan, and Minkowski distance metrics as distance equations. In the case study of the developed model, the sample dataset containing the solar irradiation and insolation parameters of Central Anatolian Region of Turkey are evaluated in the tool. The tool successfully clusters the dataset related to solar power plant installation. Furthermore, the tool assists to knowledge domain expert during decision-making stage of feasibility analysis. The obtained results are presented to expert numerically and visually. It is shown that the developed tool clusters the given sample data efficiently and successfully using each distance metrics. The clustering results obtained with Minkowski distance metric presented the most accurate results according to Euclidean and Manhattan used by the developed tool.

Section snippets

The substructure of prediction tool developed

Clustering is a process that is aimed to group a set of given objects according to similarity or dissimilarity in classes or clusters. Clustering process has a wide application area containing image processing, data mining, machine learning, and bioinformatics beside computer applications such as document or search engine listings. Due to clustering the large data repositories is required to provide meaningful data results to user, a variety of algorithm and method developing studies are

The solar feasibility prediction tool

The preparatory studies and case studies of developed tool are analysed in this section.

Conclusion

The portion of renewable energy sources in the electricity generation is rapidly increasing due to their natural abundance. In contrast to their increased utilization, installation and maintenance cost analyses should be carried out to increase the potential efficiency of a renewable energy plant. International authorities emphasize that the utilization of solar energy will be reached to a grand value with an increment rate that is not seen ever before up to 2050. An agglomerative hierarchical

References (44)

D.A. Chwieduk
Recommendation on modelling of solar energy incident on a building envelope
Renew Energy
(2009)
J. Margeta et al.
Feasibility of the green energy production by hybrid solar + hydro power system in Europe and similar climate areas
Renew Sust Energy Rev
(2010)
X.Q. Zhai et al.
Solar integrated energy system for a green building
Energy Build
(2007)
A. Zahedi
Solar photovoltaic (PV) energy; latest developments in the building integrated and hybrid PV systems
Renew Energy
(2006)
C.H. Lung et al.
Using hierarchical agglomerative clustering in wireless sensor networks: an energy-efficient and flexible approach
Ad Hoc Networks
(2010)
G. Martelet et al.
Classifying airborne radiometry data with agglomerative hierarchical clustering: a tool for geological mapping in context of rainforest (French Guiana)
Int J Appl Earth Obs Geoinf
(2006)
R. Gil-Garcia et al.
Dynamic hierarchical algorithms for document clustering
Pattern Recognit Lett
(2010)
M.H. Chehreghani et al.
Improving density-based methods for hierarchical clustering of web pages
Data Knowl Eng
(2008)
G. Mecca et al.
A new algorithm for clustering search results
Data Knowl Eng
(2007)
J. Goldberger et al.
A hierarchical clustering algorithm based on the Hungarian method
Pattern Recognit Lett
(2008)

A. Kusiak et al.

Models for monitoring wind farm power

Renew Energy

(2009)

S. Mingoti et al.

Comparing SOM neural network with Fuzzy c-means, K-means and traditional hierarchical clustering algorithms

Eur J Oper Res

(2006)

Y. Dong et al.

A hierarchical clustering algorithm based on fuzzy graph connectedness

Fuzzy Set Syst

(2006)

J. Wu et al.

Towards understanding hierarchical clustering: a data distribution perspective

Neurocomputing

(2009)

X. Geng et al.

Neural spike sorting using mathematical morphology, multiwavelets transform and hierarchical clustering

Neurocomputing

(2010)

J.A.S. Almeida et al.

Improving hierarchical cluster analysis: a new method with outlier detection and automatic clustering

Chemom Intell Lab Syst

(2007)

C.C. Hsu et al.

Hierarchical clustering of mixed data based on distance hierarchy

Inf Sci

(2007)

Tanaka N. Technology roadmap-solar photovoltaic energy. International energy agency report. Paris/France;...

J.W. Kimball et al.

A system design approach for unattended solar energy harvesting supply

IEEE Trans Power Electron

(2009)

T.J. Liang et al.

Single-stage photovoltaic energy conversion system

Electr Power Appl IEE Proc

(2001)

T.L. Kottas et al.

New maximum power point tracker for pv arrays using fuzzy controller in close cooperation with fuzzy cognitive networks

IEEE Trans Energy Convers

(2006)

E. Roman et al.

Intelligent PV module for grid-connected PV systems

IEEE Trans Ind Electron

(2006)

Cited by (12)

A fuzzy regression with support vector machine approach to the estimation of horizontal global solar radiation
2017, Energy
Citation Excerpt :
They applied stepwise model selection procedures over a set of terrestrial and climatic variables and obtained a model including latitude, longitude, and altitude as the most accurate one. Kabalci [23] used estimates of solar radiation to conduct a feasibility analysis of a solar power plant over an agglomerative hierarchical clustering approach based on hierarchical clustering results with Euclidean, Manhattan, and Minkowski distances. As the result, it was shown that the hierarchical clustering analysis using the Minkowski distance gives the most accurate clustering results.
Accurate estimation of the amount of horizontal global solar radiation for a particular field is an important input for decision processes in solar radiation investments. In this article, we focus on the estimation of yearly mean daily horizontal global solar radiation by using an approach that utilizes fuzzy regression functions with support vector machine (FRF-SVM). This approach is not seriously affected by outlier observations and does not suffer from the over-fitting problem. To demonstrate the utility of the FRF-SVM approach in the estimation of horizontal global solar radiation, we conduct an empirical study over a dataset collected in Turkey and applied the FRF-SVM approach with several kernel functions. Then, we compare the estimation accuracy of the FRF-SVM approach to an adaptive neuro-fuzzy system and a coplot supported-genetic programming approach. We observe that the FRF-SVM approach with a Gaussian kernel function is not affected by both outliers and over-fitting problem and gives the most accurate estimates of horizontal global solar radiation among the applied approaches. Consequently, the use of hybrid fuzzy functions and support vector machine approaches is found beneficial in long-term forecasting of horizontal global solar radiation over a region with complex climatic and terrestrial characteristics.
Design and implementation of a solar plant and irrigation system with remote monitoring and remote control infrastructures
2016, Solar Energy
Citation Excerpt :
In one hand, the integration of different RES in a distributed generation system (DG) provides great advance in terms of source management (Spagnuolo et al., 2010; Qiang et al., 2014; Jain and Agarwal, 2008). On the other hand, the communication infrastructures are also required to manage sources remotely when they are geographically spanned or instant monitoring is required (Yu et al., 2011; Kabalci, 2011; Kabalci et al., 2012). The stand-alone renewable energy plants require battery and energy storage systems support to supply loads continuously as well as conventional power grids.
One of the most prominent applications of solar plants is the solar pumping systems that are required in rural areas while it is almost impossible to connect to the utility grid. A solar plant and irrigation system with several control features are designed and implemented in this study. The technical aspects of the study include sizing of a solar plant, feasibility analyses, system planning and implementation, remote monitoring, and remote control issues. The sizing and feasibility analyses are performed by using highly reliable software. The remote monitoring system is implemented with several sensing and communication boards designed by authors, and server-client software. The remote control operation is performed owing to implemented 8-channel relaying device that is reached over GSM networks. Each part of the entire system is implemented and is optimized to provide an observable and controllable solar irrigation system in a county yard. The presented study is aimed to express the potential capabilities of a solar plant in irrigation, and integration capabilities to numerous communication systems such as 3G, GSM, and Wi-Fi to implement remote monitoring and remote control infrastructures.
Comparison of integrated clustering methods for accurate and stable prediction of building energy consumption data
2015, Applied Energy
Citation Excerpt :
Clustering methods have been used widely throughout the energy consumption literature. A number of articles in this journal have used clustering to extract similar groups out of overall population data: examples include searching for groups composed of similar energy consumers, load or generation profiles, building or site feasibility [4–9]. Since the overall literature that uses clustering for energy analysis is quite large, this review will focus on building energy consumption as a particular example to illustrate how these clustering methods, which are often thought of as unsupervised learning, are often used in predictive analyses.
Clustering methods are often used to model energy consumption for two reasons. First, clustering is often used to process data and to improve the predictive accuracy of subsequent energy models. Second, stable clusters that are reproducible with respect to non-essential changes can be used to group, target, and interpret observed subjects. However, it is well known that clustering methods are highly sensitive to the choice of algorithms and variables. This can lead to misleading assessments of predictive accuracy and mis-interpretation of clusters in policymaking.
This paper therefore introduces two methods to the modeling of energy consumption in buildings: clusterwise regression, also known as latent class regression, which integrates clustering and regression simultaneously; and cluster validation methods to measure stability. Using a large dataset of multifamily buildings in New York City, clusterwise regression is compared to common two-stage algorithms that use K-means and model-based clustering with linear regression. Predictive accuracy is evaluated using 20-fold cross validation, and the stability of the perturbed clusters is measured using the Jaccard coefficient. These results show that there seems to be an inherent tradeoff between prediction accuracy and cluster stability. This paper concludes by discussing which clustering methods may be appropriate for different analytical purposes.
On the determination of coherent solar microclimates for utility planning and operations
2014, Solar Energy
Citation Excerpt :
Schmalwieser and Schauberger (2001) proposed an objective method for selecting sites for the Austrian UV monitoring network based on de-correlation distances. Kabalci (2011) developed an intelligent analysis tool based on an agglomerative hierarchical clustering approach to provide a mapping of solar radiation parameters of the Central Anatolian Region of Turkey. In addition, methods for producing physical mappings of regions that exhibit similar climatological attributes have also been proposed in the literature.
This work presents a cluster analysis for the determination of coherent zones of Global Horizontal Irradiance (GHI) for a utility scale territory in California, which is serviced by San Diego Gas & Electric. Knowledge of these coherent zones, or clusters, would allow utilities and power plants to realize cost savings through regional planning and operation activities such as the mitigation of solar power variability through the intelligent placement of solar farms and the optimal placement of radiometric stations. In order to determine such clusters, two years of gridded satellite data were used to describe the evolution of GHI over a portion of Southern California. Step changes of the average daily clear-sky index at each location are used to characterize the fluctuation of GHI. The k-means clustering algorithm is applied in conjunction with a stable initialization method to diminish its dependency to random initial conditions. Two validity indices are then used to define the quality of the cluster partitions as well as the appropriate number of clusters. The clustering algorithm determined an optimal number of 14 coherent spatial clusters of similar GHI variability as the most appropriate segmentation of the service territory map. In addition, 14 cluster centers are selected whose radiometric observations may serve as a proxy for the rest of the cluster. A correlation analysis, within and between the proposed clusters, based both on single-point ground-based and satellite-derived measurements evaluates positively the coherence of the conducted clustering. This method could easily be applied to any other utility scale region and is not dependent on GHI data which shows promise for the application of such clustering methods to load data and/or other renewable resources such as wind.
Sustainable feasibility of solar photovoltaic powered street lighting systems
2014, International Journal of Electrical Power and Energy Systems
In the last few years, due to soaring fuel prices and gas emissions, renewable energy technologies have been suggested as the power source for infrastructures. The interest in solar photovoltaic (PV) assisted street lighting systems stems from the fact that they are sustainable and environmentally friendly compared to conventional energy powered systems. The present paper investigates and compares the economic feasibility of two types of systems: islanded and grid-connected system, for the street lighting systems in Hunan Province, China. Based on two options of solar panel materials, a simulation model of the system is developed for economic, technical and environmental feasibility. The comprehensively sustainability feasibility of these systems is conducted taking into account the cost, energy generation, CO₂ emissions and renewable fraction. Radar plot is employed to integrate all the sustainability indicators into a general indicator, which presents system’s sustainability as a real number in the interval [0, 2]. Results show that for street lighting systems of all the cities, single crystal panel has a larger number of annual electricity generation, less emissions and higher environmental performance, but is more expansive than polycrystalline. It is also found that when the feed-in tariff higher than 1.27 CNY/kW h, the cost of energy (COE) of the solar powered lighting systems is less than a pure grid powered system. This will incite the use of solar PV in infrastructures. Through comparing the scores of sustainability, it is found that the Loudi system has the highest feasibility while the Yongzhou system has the lowest in the province.
Framework for the implementation of a new renewable energy system in an educational facility
2013, Applied Energy
New Renewable Energy (NRE) has attracted considerable attention as a future energy source that might be a significant factor in establishing a sustainable energy supply system. This issue is becoming increasingly important due to the instability of oil prices and the issues surrounding climate change. The purpose of this research is to develop a framework for the implementation of building-scale NRE systems (solar photovoltaic, solar thermal, geothermal, and wind energy systems). For this research, ‘C’ High School facility in Seoul was selected as a case study, from where project characteristics and energy consumption data in 2010 were collected. This research (i) collected facility usage information and carried out market research on NRE systems; (ii) conducted a feasibility study of NRE systems; (iii) designed a detailed NRE system through energy simulation by using the software program ‘RETScreen’; and (iv) selected the optimal NRE system through a life cycle cost analysis and life cycle CO₂ analysis.
This research can be useful for educational facility managers or policymakers to determine whether or not an NRE system need to be introduced in a preliminary feasibility study, and to find out which NRE system is the optimal one for a given facility in the detailed design process. Also, it would benefit contractors in a competitive bidding process, where as one of the CO₂ emission reduction methods, the results of this research could be a part of an educational facility improvement program.

View all citing articles on Scopus

View full text

Development of a feasibility prediction tool for solar power plant installation analyses

Abstract

Highlights

Introduction

Section snippets

The substructure of prediction tool developed

The solar feasibility prediction tool

Conclusion

Renew Energy

Renew Sust Energy Rev

Energy Build

Renew Energy

Ad Hoc Networks

Int J Appl Earth Obs Geoinf

Pattern Recognit Lett

Data Knowl Eng

Data Knowl Eng

Pattern Recognit Lett

Renew Energy

Eur J Oper Res

Fuzzy Set Syst

Neurocomputing

Neurocomputing

Chemom Intell Lab Syst

Inf Sci

A system design approach for unattended solar energy harvesting supply

IEEE Trans Power Electron

Single-stage photovoltaic energy conversion system

Electr Power Appl IEE Proc

New maximum power point tracker for pv arrays using fuzzy controller in close cooperation with fuzzy cognitive networks

IEEE Trans Energy Convers

Intelligent PV module for grid-connected PV systems

IEEE Trans Ind Electron