This paper gives the comparison between RETSCREEN and SAM(System Advisor Model), the two softwares used for the energy calculations for the resources.

1. Comparison between System Advisor Model (SAM) and RET Screen Software
Ankit Thiranh
2011CS10210 Indian Institute of Technology, Delhi cs1110210@cse.iitd.ernet.in
ABSTRACT
In this paper, a comparison between the two soft wares, System Advisor Model (SAM) and RET Screen Plus for designing and performance evaluation of renewable Energy Resources.
Keywords
Renewable Energy, Clean Energy Technology, RETScreen, SAM.
1. INTRODUCTION
In today world, renewable energy is one of the most important energy source to be discussed. With rapid consumption of non- renewable energy sources, renewable energy sources are very important to be focused upon. For installation of various renewable energy resources, there is certain performance prediction and cost of energy estimates that has to be done. In this paper, we discuss the two very famous performance calculation softwares, RETScreen4 International and System Advisor Model (SAM) which are broadly used for renewable energy calculations.
2. RET SCREEN 4
Ret Screen 4 is a software which creates energy efficiency models for residential, industrial and commercial buildings, and for industrial usage. It includes climate database which is expanded to 4,700 ground measurement stations. It also NASA satellite database of over 20 years which is useful for calculations when there is no ground station nearby and also this data is directly integrated into the software and there is no need to download from the NASA website. In the fourth version of the software, all the calculation including renewable energy, cogeneration and efficiency calculation can be done in one software file. It includes project database of over 100 case studies and templates to which user can add their own projects. After doing the analysis, it can be saved in a compact .ret file. The software is integrated over 35 languages that cover 2/3 of the world’s population.
2.1 Project Analysis
Various things have to be considered in all clean energy projects. Mainly, these are:
 Energy Resource at the project site.
 Performance of the equipment.
 Initial project costs.
 “Base case” Credits.
 Annual and periodic costs.
2.2 Objectives of RETSCREEN tool
RETSCREEN is not a design tool but it does feasibility and pre- feasibility. The three main objectives of the tool are:
1) Improve accuracy of estimates.
2) Speeding up the project analysis.
3) Reducing the cost of doing feasibility and pre-feasibility studies.
2.3 Financial Analysis
1) Base case Energy Cost: The cost of conventional source of energy. When it is high, clean energy technology gives more benefit.
2) Financing: This includes debt ratio, the term of the loan and interest rate which also have a big effect on profitability of the project.
3) Taxes on equipment and income: Sales taxes and income taxes should also be included. Energy efficient projects reduces operating costs, which raises profits and income taxes.
4) Environmental characteristics of the energy replaced: This should also be included. For example, coal produces more pollution than natural gas in any terms.
5) Environmental credits: Greenhouse gases Reduction credits and deployment incentives.
6) Decision maker’s definition of cost-effective: It can be payback period.
2.4 Five Step Standard Analysis
When RET SCREEN opens, the user specifies username and type of the project, language, currency, unit system and the climate data. The user can choose between two available options, a simplified single spreadsheet or a detailed approach. Then a five step RETSCREEN analysis begins:
1) Energy Model: It determines the energy benefits of proposed project compared to conventional alternative.
2) Cost Analysis: The incremental cost of the clean energy project are calculated.
3) Emission Analysis: It is an optional Green House Gas analysis which calculated the emission reduction associated with the project according to a standardized methodology developed in collaboration with United Nation Environment Program and the World Bank prototype Carbon fund.
4) Financial Analysis: It indicates whether the project is financially attractive.
5) Sensitivity and Risk Analysis: It calculates how changes in inputs effect the viability in the project through Monte-Carlo simulation.

2. Figure 1: Cumulative Cash flow Graph
The latest version also includes a product database of over 7,000 clean energy devices ranging from wind turbines to fuel cells.
2.5 EXAMPLE (RETSCREEN 4)
Figure 2: Start Page of RETScreen
Let us take a wind turbine project. This example gives a comparison between two scenarios. The proposed project is a 50MW wind turbine to be installed on Ottawa International Airport. A capacity factor of 30% is estimated based on other projects, i.e., the winds are able to produce 30% of the 50MW rated power on average. The installed turbine cost are estimated to be $2000/KW. In the first scenario, it is calculated whether it is financially good to sell this electricity at $0.10/ KWh. In the second scenario, the calculation is done on whether a $0.01/KWh green power production incentive would be preferable to a payment of $15/ton GHG emissions reduction.
RET SCREEN tells us that the turbine generates 1,31,400 MWh per year. Then we enter the 10,000 million installation cost of the project. Then we add the tarrif rate of $0.10/ Kwh, an inflation rate of 2% and a project lifetime of 25 years. It is assumed that 70% of the project will be financed by loan at an interest rate of 8% and a term of 15 years. After this, RET SCREEN provides a simplified assessment of turbine’s financial viability. A cumulative cash flow graph shows that the revenue will cover the loan paid off in 15 years after which yearly net income increases sharply. The pre-tax IRR on equity is 13.9%.
Out of the two products, which would increase our project interest more, a $0.01/KWh green power production incentive or carbon trade valuing emission reductions at the cost of $15/ton. The first possibility is examined by adding $10 to our base tarrif of $100/MWh. To see this $/ KWh, one can click on the option ‘Select alternative unit.’ The pre-tax IRR on equity rises from 13.9% to 17.5%.
The second possibility requires a greenhouse emission analysis which appear in simplified form when the checkbox is clicked. The user must apply a reasonable emissions factor for the base case, i.e., the tons of C02 equivalent emitted per MWh of conventional electricity generated. This will reflect the fuels that the wind turbine will displace. RETSCREEN lists average emission factor for grid electricity for each country. Between the power plant and the end user, there are transmission and distribution losses of say 5%. These increase the emissions factor from the consumer point of view. The wind turbine also suffers T&D losses, so for a fair comparison, RETSCREEN assumes that the base case quid makes up for these losses with the associated emissions attributed to the wind turbine. The wind turbine reduces emissions by 1000 tons of CO2 equivalent per year. RETSCREEN gives another analogy that it is equivalent to 4,628 light truck taken out of the road for a year or carbon absorbed annually by 2,324 hectares of forest. So, the incentive is removed and carbon trade is added and it is found that the IRR is now 15.0% less than the incentive.
Figure 3: After-tax IRR euquity distribution calculated using Risk Analysis (Method 2)
We can add more details by selecting method2 and then doing the calculations. Method 1 deals with capacity factor by itself. But it can be calculated by using method2 which uses annual wind data and turbine power curve. Method 3 deals with more complications such as monthly wind data and tarrifs according to the season.
A user can go back to the start page and then she can also use the method 2 for detailed analysis. This creates in depth cost, financial, emission and risk analysis pages where she can refine the estimates. Further, there are two more methods within cost analysis and also all other pages also have more than or equal to two methods using which a user can define her estimates clearly. RETSCREEN provides a very good cost and other analysis for various clean energy projects. Therefore, it has a large number of users which is shown in figure3.
3. SYSTEM ADVISOR MODEL (SAM)
The System Advisor Model (SAM) is a performance and financial model designed to facilitate decision making for people involved in

3. renewable energy industry. It does more or less the same work as RETSCREEN and makes predictions for performance and cost of energy estimates for various clean energy projects.
3.1 SAM Models and Databases
The cost and performance of clean energy projects are represented using various computer models developed at NREL, Sandia National Laboratories, the University of Wisconsin, and various other organizations. The SAM’s interface makes it quite easy for people with no computer experience in building a model of renewable energy project, and to estimate cost and performance evaluations based on these model results.
The weather file required by SAM to describe the renewable energy resource and weather conditions at a project location can be either downloaded by choosing a weather data file from a list or can be created by using the user’s data. SAM automatically download data and populate input variable values from various online databases such as DSIRE for U.S. incentives, NREL Solar prospector for solar resources data and ambient weather conditions, etc.
3.2 Results: Tables, Graphs and Reports
Various tables and graphs displayed by SAM display levelized cost of energy, first year annual production and various other important measurements. After simulation is run, SAM displays a set of default curves which can be customized further.
3.3 Performance Models SAM’s performance models make hour-by-hour calculations of a power system's electric output, generating a set of 8,760 hourly values that represent the system's electricity production over a single year. The latest version of SAM has performance models for the following technologies:
o Photovoltaic Systems o Parabolic trough concentrating solar power o Power tower concentrating solar power (molten salt and direct steam) o Linear Fresnel concentrating solar power o Dish-Stirling concentrating solar power o Conventional thermal o Solar water heating for residential or commercial buildings o Large and small wind power o Geothermal power and geothermal co-production o Biomass power
3.4 Financial Models
The financial model of SAM calculates financial metrics for various kinds of power projects based on a project’s cash flow over an analysis period which is specified by the user.
3.5 Incentives
SAM’s financial model can account for a wide range of incentive payments and tax credits: o Investment based incentives (IBI) o Capacity-based incentives (CBI) o Production-based incentives (PBI) o Investment tax credits (ITC) o Production tax credits (PTC) o Depreciation (MACRS, Straight-line, custom, bonus, etc.)
3.6 Analysis options
The analysis from SAM makes it possible for the user to conduct studies which involve multiple simulations. Some typical types of analysis are: o Parametric Analysis: Assign multiple values to input variables to create graphs and tables showing the value of output metrics for each value of the input variable. Useful for optimization and exploring relationships between input variables and results. o Sensitivity Analysis: Create tornado graphs by specifying a range of values for input variables as a percentage. o Statistical: Create histograms showing the sensitivity of output metrics to variations in input values. o P50/P90: For locations with weather data available for many years, calculate the probability that the system's total annual output will exceed a certain value.
3.7 Example (SAM)
Let us create a wind turbine project using the default values taken by SAM for various calculations and the weather file for choosing a wind resource be NREL AWS True Power Data: representative of southern WY –flat lands (Canada). The turbine cost is estimated to be $56076.00. The financial parameters taken are that we have a loan term of 25 years and a loan rate of 7.5% per year. We are assuming the analysis period to be 25 years. The electricity rate is taken as 0.12$/KWh. The various graphs obtained after starting simulation are given below.
After simulation, we have various options for plotting graphs, seeing the various kinds of data, plotting a customized data, cash flow, new features like time series, Daily - analysis monthly profile for various things like hour energy, wind direction, wind speed, etc.
We can also have the scatter of various options such as hourly energy, wind direction, wind speed, air temperature, pressure, etc.
Figure 4: Monthly Energy Calculation using SAM

4. Figure 5: Annual energy Production using SAM
Figure 6: After tax cash-flow using SAM
.
4. COMPARISON BETWEEN SAM AND RETSCREEN
Both the tools are very good and easy to use to do calculations and optimizing the cost. Both of have some unique features which prove the dominancy of one tool over another.
For the comparison of two softwares, the calculations are done on both softwares using the same data. The calculations are done for a wind turbine project. Both RETSCREEN and SAM uses different approaches to calculate some parameters like estimation of wind speed and direction for any given hour. SAM prediction is limited only within US, whereas RETSCREEN make predictions all over the globe. Therefore, RETSCREEN has more temporal and spatial locality than SAM.
4.1 RETSCREEN’s Calculation
4.1.1 Wind Data
The climate database of RETSCREEN is taken from two resources, from ground monitoring stations and from model predictions based on NASA´s global satellite data. If climate data is not available from a specific ground monitoring station, it is then obtained from NASA’s satellite data. It can also be manually entered.
4.1.2 Wind Speed input
It can be obtained from program’s climate database or can be manually entered. The wind speed input is monthly. Weibull distribution is used to convert the monthly average wind speed to a distribution of hourly values.
4.1.3 Turbine Selection
A large database of information from all turbine producers exists for RETScreen. The information that comes with each turbine is: Capacity, Hub height, Rotor diameter and Swept area which are perpendicular to the wind direction that the rotor will cover during one complete rotation. Most of the turbines also have power curve data which is necessary to find the production.
4.2 SAM’s Calculation
4.2.1 Wind Data
The weather data for SAM depends on what kind of energy resource is being analyzed. The wind resource data is taken from NREL’s Western Wind Dataset.
4.2.2 Wind Speed Input
The wind power model uses wind resource data from files in the ‘swrf’ format, which is a tab-delimited text format. Most database and spreadsheet programs are able to read or save in a delimited format, for example Excel. Since the model absorbs the wind speed hourly, there is no need for a probability density function.
4.2.3 Turbine Selection
The turbine selection available in SAM is a list of wind turbines with a variety of capacities. In addition to the turbine’s nameplate capacity, SAM includes information on rotor diameter, cut-in wind speed, hub height and the turbine power curve.
4.3 Model Comparison
The inputs which are kept same for both SAM and RETScreen are described below:
1. Wind Data location.
2. The type of turbine.
3. Wind shear exponent.
4.3.1 Choice of Wind Shear Exponent
The wind speed data available in the software is either measured at airports at a height of 10 meters or predicted from a meteorological interpolation model at elevations much greater than 10 meters. Therefore conversion of the wind speed to the turbine hub height (67 meters) is needed. A common method to describe the relationship of wind speed and height is the “power law”. 푣1 푣2= ( ℎ1 ℎ2) 훾
The variables v1 and v1 are wind speeds at heights h1 and h2, and exponent γ is termed the wind shear exponent. v1 indicates the known wind speed at known height (h1) and v2 indicates the unknown wind speed at desired height (h2).

5. 4.4 Results
4.4.1 Measured Wind Speed vs. Predicted Wind speed
The average wind speed per month averaged over a year measured at 67 m came out to be 9.1 m/s.
4.4.1.1 Predicted Wind Speed by RETScreen
67 m wind speed predicted by RETScreen vs. observed wind farm wind speed (percent difference) came out 21% at wind farm location and 35% at the airport.
Figure 7: Predicted Wind Speed by RETScreen
4.4.1.2 Predicted Wind Speed by SAM
For the wind farm location, three years of data were available and two years data for airfield. Wind speed predicted by SAM vs. measured data at the wind farm (% difference) came out to be 14% while for the airfield, it came out to be 26%.
Figure 8: Predicted Wind Speed by SAM
4.4.2 Measured electrical production vs. estimated production
In this research, data from the year 2011 is used. The capacity factor represent “the ratio of the net electricity generated, for the time considered, to the energy that could have been generated at continuous full-power operation during the same period”.
4.4.2.1 Predicted Production from RETScreen
Figure shows the difference in percentage on electrical production for RETScreen vs. measured data from the wind turbines. RETScreen prediction from measured data (%difference) came out to be 13%. The prediction by RETScreen at wind farm came out to be 35% while at airport came to be 51%.
Figure 9: Comparison on production, RETScreen vs. measured data (% difference)
4.4.2.2 Predicted Production from SAM
Figure shows the difference in percentage on estimated electrical production for SAM vs. measured data from the wind turbines. SAM prediction at wind farm came out to be 26% while at airfield, it came out to be 36%.
Figure 10: Comparison on electricity production, SAM vs. measured data
4.5 RETScreen vs. SAM
With regard to comparison on the two computer programs, numerous factors must be taken into consideration. The wind farm location wind speed is underestimated according to RETScreen

6. even though for a few months (August through January) it is closer to the measured wind speed. Predicted wind data for wind farm location by SAM, follows the measured data quite well even though it fluctuates over a few months (October through February). The reason why the computer programs differ could be that NASA data set has information from locations that are far apart while SAM has information from sites (locations) closer to each other. Information from SAM comes closer to measured data than from RETScreen so that measuring technique might be more accurate.
It appears that the airfield wind speed data from SAM have a tendency for higher wind during the spring/summer months while the airfield wind speed data from RETScreen have it through the fall/winter months. Again, it could be that the data from NASA is not taking the terrain (environment) into full consideration.
4.6 Final Results
Final results from both programs on wind speed and electrical production vs. measured data in percentages. Both programs assume no losses even though measured data includes losses within the farm. By taking losses into account estimated production would be reduced and result would be closer to measured data. Therefore by overestimating, programs deliver more reliable data. Results from SAM are the average from 3 years period (2004-2006) that was available in the climate database. Only SAM delivered overestimated results on electrical production compared to measured data. Table 16 shows the production results and if the programs would take losses into account SAM would deliver more accurate data and RETScreen less accurate.
Figure 11: Production Results
Figure 12: Wind Speed Results
5. CONCLUSION
According to results on electrical production SAM will be better than RETScreen given that SAM includes a much more detailed topography and a much finer scale wind field.
RETScreen is straightforward and convenient to use. It offers a variety of options such as losses and turbine parameters. RETScreen accesses wind data from ground stations and NASA, the ground stations are located near populated areas and are often on an airfield. Those locations do not necessarily provide the best wind data for a specific, nearby location as illustrated by the assessment provided in this thesis. The data resolution is fairly coarse for the NASA database and does not work well for complex terrain.
SAM is easy and convenient to use like RETScreen. The wind data base includes fairly dense resolution and offers data from three years (2004-2006) for comparison. The results from both programs for annual average wind speed are compared to measured wind farm data. RETScreen underestimates the wind speed by 21% (wind farm location and source data from NASA) compared to measured wind speed from the wind farm. SAM on the other hand overestimates the wind speed by about 14% (wind farm location) compared to measured wind speed from the wind farm. The outcomes associated with SAM appear more accurate and the western wind database appears to predict wind quite well, although different terrain might impact this finding. RETScreen underestimates electrical production by 35% on wind farm location compared to measured production from turbines while SAM overestimates it by 26%. SAM is more accurate in estimating electrical production for a potential wind farm. Based on this comparison SAM contains more accurate wind data to compare with measured data and could be used in the future to develop a wind farm. By adding information on array loss, airfoil loss and availability loss to the programs like addressed in the paper, they could deliver more accurate results on production. This would deliver more accurate results on production for SAM while less accurate for RETScreen.
By looking at the purpose of those computer programs it has to be considered that they should only predict the feasibility and potential profitability for a wind farm location. Each and every one then has to evaluate how accurate the data has to be to decide on further investigation/measures.
6. REFERENCES
[1] www.retscreen.net/ang/home.php
[2] sam.nrel.gov
[3] Analysis of the Renewable Energy Assessment Programs RETScreen and System Advisor Model (SAM) - Wind Energy Model Predictions Comparison with Measured Operational Data - Gudmundsson, Sigurdur Oli