Optimization of Large-Scale Solar Hot Water System Using Non-Traditional Optimization Technique

The depleting fossil fuel concerns, pollution emissions, and global warming force each and everyone towards using clean and renewable energy sources. Hot water production is one of such applications where solar energy can be used most effec-tively. In this project work, design, analysis, and optimization of large-scale Solar Hot Water System (SHWS) are attempted for a college hostel (having 700 students), which requires hot water for bathing purposes. The preliminary SHWS is designed based on the existing design methods. The system is modeled mathematically and a simulation program has been developed to predict thermal performance and pressure drop aspect. The preliminary calculations for determining performance characteristics and simulation methodology were carried out using MATLAB. The system performance is predicted over the life span of 10 years. Initially, the effect of collector area, storage tank volume on the performance and life-cycle cost of the system are analyzed using robust design methodology. Then the parameter is increased to four and optimization was carried out. Using the simulation, the performance of the system is analyzed for different settings. Using S/N ratio, more control parameter settings are identified so that the system is robust against the variation of noise factors. The optimum system configuration is achieved with less number of design iterations. The comparisons have been made between the optimum under four parameters using Taguchi's robust design methodology.


Introduction
Alternative energy has become important and relevant in to-day's world due to the problems associated with the use of fossil fuels [1,2]. Non-Conventional Energy Sources, which comprise both renewable and nonrenewable sources, should play an increasing role in the coming periods in view of fast depleting fossil-fuel reserves and growing concerns for environmental protection. Perennial energy shortages and resulting inflation have adversely affected the balance of payment position in energy scare economies of the developing countries [3,4].
How to cost-effectively design a high-performance solar energy conversion system has long been a challenge. Solarwater heater (SWH), as a typical solar energy conversion system, has complicated heat transfer and storage properties that are not easy to be measured and predicted by conventional ways. In general, an SWH system uses solar collectors and concentrators to gather, store, and use solar radiation to heat air or water in domestic, commercial, or industrial plants [3]. For the design of high-performance SWH, the knowledge about correlations between the external settings and coefficients of thermal performance (CTP) is required. However, some of the correlations are hard to know for the following reasons: (i) measurements are time-consuming [5]; (ii) control experiments are usually difficult to perform; and (iii) there is no current physical model that can precisely connect the relationships between external settings and intrinsic properties for SWH. Currently, there are some state-of-the-art methods for the estimation of energy system properties [6,7,4] and for the optimization of performances [8][9][10][11][12]. However, most of them are not suitable for the solar energy system. These problems, together with the economic concerns, significantly hinder the rational design of high-performance SWH. The solar radiant flux reaches the atmosphere [11,12] is at its greatest when the earth is closest to the sun [2,6]. The areas lying between the latitudes 30 0 N and 30 0 S, which have at least 2000 hours of bright sunshine per year. The daylight hours vary from month to month. The peak insola-. 02 . tion received on the earth's surface is approximately 1 kW/ m2. e mean value of the ux reaching the outside earth's atmosphere is called the solar constant, which is estimated as 1.353 kW/m2 [6,10].

Large-Scale Solar Water System
ermosyphon Systems [4] use a separate storage tank located above the collector. e liquid which gets heated in the collector rises naturally to the tank where it is kept until needed. A bene t of the systems is that they require no moving parts.
e ( Figure 1) shows a natural circulating type solar hot water system. It consists of a tilted collector, with transparent cover plates, a separate, highly insulated water storage tank, and well-insulated pipes connecting the two. e bottom of the storage tank is located at least 30cm higher than the top of the collector. ermosiphon Systems use the natural tendency of heated water to rise and cooler water to fall to perform the heat-trapping task. e cold water pumps the water which was heated through the collector outlet and into the top of the tank, thus heating the water in the tank. e cold-water line from the accommodation location ows directly to the tank. Solar heated water ows from the tank to load whenever water is used. circulation is used. A schematic diagram of a typical closed loop system is shown in the ( Figure 2). Water from a storage tank is pumped through a collector array, which performs heating and exit of the collectors and a suitable location inside the storage tank. As the solar radiation and atmospheric air temperature vary over a day and over the year, each component needs to be mathematically modeled and assembled in an appropriate manner to simulate the total system.

Design Methodology Utilized in LS-SWHS Modeling
pressure drop behavior of the total system is investigated using a simulation approach. It has become the most familiar method for investigating the dynamic solar water heater operational behavior. MATLAB is the simulation package used throughout in order to simulate the total performance of the solar water heater system. in college hostel having 600 students.

(a) Estimation of hot water requirement
No. of a student in the hostel (Ns) = 700 About 10% of students may not take bath using hot water. Hence the amount of quantity has to be reduced. However, there will losses from the system, which are about 10%, thus both mutually cancel out.

(b) Based on the detail presented in Sukhatme (1996)
Yearly average solar radiations available on horizontal surface (Ig-day) = 5 kW hr = 5 * 3600 = 18000 kJ/day About 60% of the above quality is available on the collector sur- Collector pump will be operated if the total temperature rise of water across the collector circuit is 5 C. Hot water is drawn from the tank during 6.00am to 8.00am (2 hours) for each day. parameters of the pipe, the friction factor and also the environmental conditions on the pipe. . 04 . Calculating the resistance of pipe (Rh) Rh = (2 * f * lp * V)/(dp ) 3

Modeling of Collector
A solar collector is a special kind of heat exchanger that makes use of the high energy infra-red radiations of the solar spectrum which is a negligible factor compared to solar energy radiations. collector in (Figure 4).
. 05 . e radiations absorbed by at-plate solar collectors lie in the range from infra-red to visible radiations. e radiation heat transfer is thus used in the calculation of the absorbed solar radiation and the heat gain in the solar collector. While the equations for collector performance are reduced to relatively simple forms in many practical cases of design calculations, they are presented below. Each collector is modeled using a lumped model. It requires solar radiation on the tilt surface and atmosphere temperature as inputs. . 06 . amount is in excess of the requirement of the application and discharges energy when the collected amount is inadequate. then all the levels that optimize at least one object are selected.

2.
factor level that optimizes this objective is selected regard-3. then the designer's discretion is used to determine the ob- computer-aided-design etc. Taguchi method is useful for 'tuning' a given process for 'best' results. Taguchi proposed a standard 8-step procedure for applying his method for optimizing any that parameter design using noises that are deliberately created experiments can be performed under various levels of noise i.e. with positive induction of noise to the design, we can obtain a chi's parameter design is the deliberate creation of noise for the sis-of-Variance (ANOVA) approach, was carried out to improve the performance of the system. control parameters in such a way that any variation in the noise of each collector.
matic conditions for winter and summer, ambient temperature and 24 hours solar radiation data. It has three level of the experiment the control parameters and two levels of noise factor (winthe (Table.1 Steps in taguchi methodology evaluating and implementing improvements in required per-sired characteristics and simultaneously reducing the number of defects by studying the key variables controlling the process and method can be applied to any process in engineering fabrication,

Orthogonal array (oa) selection and conducting experiments
number of an experiment that must be performed to reach a near optimum parameter set. For large scale solar hot water system, the suitable orthogonal array is L9 (for control factors of 4 and experimentation pattern and results of experimentation is shown in (

Result and Discussion
Determination of the optimum levels of control parameters sign method employs a signal -to -noise ratio to include the used in the analysis of the result is given in (Table 3). Since our objective is to minimize the cost we follow the smallest is best Since log is a monotone function, maximizing S/N is equivalent to minimizing the quality characteristic. For example for experi-(n = 2) is computed as follows S/N = -10 log (1/2{(3.9865e 6 ) 2 + (7.28426e 6 ) 2 }) = -135.36 ...… (6.1)    Table 4 for a parameter at a given level at every time it spective S/N ratio is -135.14 which is shown in the response table under B at level 2. Analysis of variance is a computational technique that quantitatively estimates the relative contribution of each parameter variation of each parameter variation makes to can be obtained by the formula: . 09 .

Data analysis using the S/N and ANOVA
Parameter setting (c): 100, 24000, 0.06, and (0.675, 6.6) Using the average S/N ratios from the response table 4, the graphs are plotted for the four control parameters as shown in (Figure 8).  . 010 .

Conclusions
ulation program has been developed using MATLAB to preperformance of the system is predicted over the lifespan of tion is optimized based on the total life-cycle cost using robust ly designed for producing hot water for a hostel having about summer and winter were considered based on the location and the energy required for producing hot water was calculated for both the conventional electrical heating system and solar hot water system.
Initially, four control parameters were taken and the optimiza-rameters were achieved with the objective to minimize the total life-cycle cost of the system and with a minimum pay-back be practical if applied for real-time systems using solar energy concepts thus saving the depleting conventional energy sourc-sulted in a cost saving of Rs. 14,10,600 for the total span of ten years considered, which is about (32%) and a slight reduction in standard deviation.
In this paper, we have summarized our recent studies on the predictive performance of machine learning on an energy system and proposed a framework of SWH design using a MAT-LAB and Taguchi optimization technique. A combined computational and experimental case study on LS-SWH shows that mized performance without knowing the complicated knowledge of the physical relationship between the SWH settings and blank of the HTS applications on optimizing energy systems and provide new insight into the design of the high-performance energy system. contributed to this research work.