The IAEA DEEP desalination economic model: A critical review

Abstract

The IAEA DEEP software has been used worldwide for the economic evaluation of desalination plants (thermal or electrical) coupled with various energy sources (nuclear, fossil fueled or renewable). Throughout the years, the software was updated constantly. Such updates included the user interface and model structure but not the economic models. Previous continuous development was culminated in the development of the DEEP 3.2 version which has been recently released in 2009.

This paper presents a step forwards in the continuous effort to maintain high standards and reliability of DEEP. It also scrutinizes methods used, assumptions made, and constants or default values originally used. The validity of calculations as well as the identification of the most important parameters is presented. Sensitivity analysis is used to identify the most important parameters in the DEEP model.

Overall, the review proves that both the DEEP economic model and software implementation are solid for economic evaluation of dual purpose plants. Based on results presented and recommendations made, a new version of DEEP is expected to be released in 2010 which will address minor issues and improvements.

Introduction

Technological advances of the last decade have succeeded in making desalination spreading faster and becoming a reliable source for the supply of water, and consequently for sustainable development. Yet, minimizing the cost of seawater desalination is recognized as one of the most important technology challenges. With the rising energy costs and water demands, the energy consumed and subsequently the costs involved in any desalination plant may play an important role in any economic feasibility and optimization studies of desalination systems.

In the last decade, the total contracted desalination capacity has almost tripled (see Fig. 1). The desalination technology with the greatest share is 60% for RO, 30% for MSF and 10% for MED [1]. The average capacity per project has also dramatically increased (see Fig. 2). Consequently, the energy needs of each project have become significantly larger creating the necessity for larger and more reliable energy sources. Moreover, the increase in energy costs and the uncertainty in fossil fuel prices have multiplied the expenditures of constructing and operating a desalination plant [2].

The economics of desalination could be enhanced further through cogeneration i.e. the use of dual purpose plants (e.g. for electricity generation and water production). Sustainability, environmental considerations, and large-scale economic aspects have made nuclear energy a promising energy source candidate for desalination, based on previous experience with nuclear desalination (see Table 1) [1], [3]. Currently, there is a growing interest in the use of nuclear energy for various non-electrical applications such as desalination, hydrogen production, and process heat applications [4]. Among other drivers for this interest are cheaper energy, less uncertainty on energy costs, higher load factor of the desalination plant, better load factor of the nuclear unit, utilization of nuclear plant's free land, and reduction of the desalination carbon footprint [5], [6], [7].

The attractiveness of using nuclear energy for seawater desalination on large scale [8], [9] has led the International Atomic Energy Agency (IAEA) to develop and distribute freely the Desalination Economic Evaluation Program (DEEP). DEEP was originally derived from the desalination cost evaluation package developed in the eighties by General Atomics on behalf of the IAEA [10]. The old version, named “Co-generation and Desalination Economic Evaluation” Spreadsheet, (CDEE) which was used for feasibility studies related to nuclear desalination in the IAEA and other Member States. Subsequently, with its increasing popularity, a user-friendly version was issued by the Agency towards the end of 1998 under its current name of “Desalination Economic Evaluation Program” (DEEP).

The DEEP software is usually used for the following [11]:

• Calculation of the levelized cost of electricity and desalted water as a function of quantity, site specific parameters, energy source, and desalination technology.

• Side-by-side comparison of a large number of design alternatives on a consistent basis with common assumptions.

• Quick identification of the lowest cost options for providing specified quantities of desalted water and/or power at a given location.

Despite the fact that DEEP is not a design code, it has been used worldwide for the economic evaluation of desalination plants (thermal or electrical) coupled with various energy sources (nuclear, fossil fueled or renewable) for site specific project feasibility analysis [12], [13], [14], what-if analysis [15], [16] or even for conceptual research studies [17]. Throughout the years, the software was updated constantly. Such updates included the user interface and model structure but not the economic models. One of the most salient features of DEEP was the complete modularization of various cases. As the user group enlarged, new ideas as well as criticisms of the DEEP models appeared. Some of them were implemented gradually in different working versions (versions 2.0 [11], 2.1, 2.2, 2.3, 2.4, 2.6, 3.0 [18], 3.1). The previous continuous development culminated in the development of the DEEP 3.2 version which has been recently released in 2009. The DEEP main calculation sheet supports both nuclear and fossil power options. It considers heating and power plants as well as heat-only plants, distillation processes MSF and MED and membrane process reverse osmosis.

The scope of this work is to review the overall economic model and parameters used in DEEP, and evaluate the validity and reliability of DEEP through comparative results. The review scrutinizes methods used, assumptions made, and constants or default values originally used. As a part of the review process, concepts and methodologies of the economic sub-models used in DEEP are presented, and the model results are discussed comparatively. Moreover, the sensitivity of the models to its parameters is examined for various characteristic cases. As an important goal of the review is to verify that the model expressions have been encoded correctly into the computer software. For the sake of clarity, the detailed model equations are not presented here and are available in the DEEP computer manual [11]. This paper presents a step forwards in the continuous effort to maintain high standards and reliability of DEEP.