APPLICATION OF THE PROMETHEE AND VIKOR METHODS FOR SELECTING THE MOST SUITABLE CARBON DIOXIDE GEOLOGICAL STORAGE OPTION

: CO 2 storage in geological formations is one of the leading solutions for mitigation of greenhouse gas emissions. Types of geological formations that can be used for CO 2 storage, that are discussed in this paper are: depleted oil and gas reservoirs, saline aquifers and injection CO 2 in partially depleted oil reservoirs for enhanced oil recovery (EOR–CO 2 method). In order to select the most suitable geological storage of CO 2 , the ranking of these storage options was performed using two methods of multi criteria analysis, PROMETHEE and VIKOR. This paper presents an overview of considered criteria (storage capacity, total storage costs, risk assessment costs, storage time dynamics, risk of CO 2 leakage from geological formation and risk of CO 2 leakage through the well), description of applied multi criteria analysis methods, selection of optimal CO 2 storage option and results of their application.


INTRODUCTION
The concentrations of CO2 in the atmosphere that are the main cause of global warming and climate change are constantly increasing, resulting from the combustion of fossil fuels during the process of electricity generation, certain industrial processes and transport. It is considered that one of the leading solutions for mitigation of CO2 emissions is geological storage of CO2, which includes three phases: capturing anthropogenic CO2, transporting and injection into different types of geological formations such as: depleted oil and gas reservoirs, saline aquifers, unmined coal beds, injection CO2 in partially depleted oil reservoirs for enhanced oil recovery (EOR-CO2 method), as well as storage in salt caverns, basalt formations and oil or gas rich shale.
In order to select the most suitable geological storage of carbon dioxide, using two methods of multi criteria analysis, PROMETHEE and VIKOR, these storage options:

OVERVIEW OF ANALYZED CRITERIA
The criteria that are considered in this paper in order to choose the optimal CO2 storage option are geological, techno-economic parameters and risk factors.
All parameters are significant, but the most significant from the safety aspect, which is the most important characteristic of the geological storage, can be identified risk factors for migration and leakage of CO2 (risk of CO2 leakage from geological formation and risk of CO2 leakage through the well).

Storage capacity
A number of methods have been developed to estimate the storage capacity based on different characteristics depending on the type of storage. Capacity assessment requires a multidisciplinary approach to the analysis of available data. The accuracy of the storage capacity assessment depends on the availability of data on the potential storage.
The analyzed parameters for estimating the storage capacity in depleted oil and gas reservoirs and during EOR-CO2 process, include: original oil (or gas) in place, recoverable oil or gas reserves, porosity, reservoir rock volume, reservoir temperature and pressure, water saturation, potential water inflow, phase behavior of CO2, solubility in water and possible spill point. The most important parameters for saline aquifers are: aquifer characteristics, water salinity, CO2 solubility in water, as well as the presence of cap rocks continuity (Bachu,2002;Tomić, 2018).
The estimated storage capacity in depleted oil and gas reservoirs varies between 675 и 900 Gt CO2 (Vishal, 2016). Storage capacity in saline aquifers is estimated at 1.000-10.000 Gt CO2 (Cook, 2012), while the storage capacity during EOR-CO2 process is the least and is 140 Gt (IEAGHG, 2009).

Total storage costs
Total storage costs include the costs related to the infrastructure for CO2 capture and transportation, the costs of injection wells, as well as costs of field facilities whose value varies depending on storage option (Tomić, 2018).
Based on Hendriks (2002) if we neglect the capture and transportation costs, the most probably estimated storage costs, for "onshore" saline formations in Europe, located at Application of the PROMETHEE and VIKOR … 45 depth of 2 km, are 2,7 euro/t CO2 stored. Offshore storage is more expensive than onshore storage. For offshore saline aquifers at the same depths, the most probable value is 7,3 euro/t CO2 stored. Storage costs for onshore depleted oil and gas fields are 1,6 euro/tCO2 stored, and for offshore formations the most probable value is 5,7 euro/tCO2 stored.
Primary purpose of EOR-CO2 projects are additional oil production, and therefore storage of CO2 during the EOR-CO2 process requires lower costs compared to storing in other formations. Storage costs for EOR-CO2 at depths up to 2 km are reduced to minimum. The costs of storage depending on the actual oil price, as well as the depth of the storage reservoir (Hendriks, 2002).

Risk assessment costs
Risk assessment costs are significantly lower in researched storage, where a large number of reservoir and fluid data is available, which is the case for the storage of CO2 in depleted hydrocarbon reservoirs and during the EOR-CO2 process. Insufficient research of saline aquifers increases the risks of storage, and therefore the costs of risk assessment are very high. In order to reduce risks, large investments are needed in researching, collecting and analyzing the necessary data (IPCC, 2005).

Storage time dynamics
Storage time dynamics relates to the probability that a particular geological formation will become a geological storage in the near future. Aquifers are found in almost all sediments that are widespread around the world, and possess the largest storage capacities. It is believed that these geological formations need to be further explored, and exploit their potential.
After completing primary oil recovery, resorts to secondary oil production (waterflooding, ie. injection of water or gas) and tertiary (enhanced) oil recovery (EOR), therefore the reservoirs remain longer in production, so it cannot be said that a large number of depleted reservoirs will become a geological storage in the near future (IPCC, 2005).

Risk of CO2 leakage from geological formation
The existence of impermeable barriers, i.e. cap rocks is a very important parameter for ensuring a long term CO2 storage (Mortezaei, 2018). For this reason, cap rocks should have the following characteristics: to be laterally homogeneous, with low permeability and porosity which means that there is a high value of capillary entry pressure and to be ductile and resistant to high injection pressure (Rutqvist, 2012).
From the beginning of the CO2 injection process, available monitoring methods are used at certain time intervals to monitor the CO2 movement. If CO2 leakage is detected, 46 Tomić L., Karović Maričić V., Danilović D., Leković B., Crnogorac M. appropriate remediation techniques are used depending on the type of leakage, in order to prevent further leakage to the surface (IPCC, 2005). Insufficient research of saline aquifers i.e. a small number of reservoir data increases the risks of storage, as opposed to storage in depleted oil and gas reservoirs and during EOR-CO2 process. Figure 1 shows the possible leakage pathways for injected CO2 into saline aquifers: • CO2 leaks through the cap rock due to injection pressure increase above the capillary pressure, • Presence of faults enables CO2 leakage into a freshwater aquifers, • Leakage through natural fractures, • Injected CO2 goes up due to the presence of fault, • CO2 leakage through poorly plugged abandoned well, • Due to natural water flow into the stored formation, saline water dissolves at the water/ interface and transports it out of closure, • CO2 migration through the spill point of the storage geological formation.

Risk of CO2 leakage through the well
The greatest risk of CO2 leakage is the possibility of migration through active or abandoned wells because they represent a direct connection between the land surface and deep subsurface (IPCC, 2005). It is important to ensure well integrity in order to keep CO2 safe inside the formation.
Modern well technology, and wells completion with corrosion-resistant materials, significantly reduces the risks of CO2 migration. Since there is no existing infrastructure in saline aquifers, the use of new equipment reduces risks to a minimum.
Possible leakage pathways through an active and an abandoned well can be ( Figure 2): through deterioration (corrosion) of the casing, between the cement and the outside of the casing, in cement fractures, between the cement and the formation. In an active well CO2 can also leakage through deterioration (corrosion) of the tubing and around the packer, and in an abandoned well through the cement plug and between the cement and the inside of the casing (Mortezaei, 2018).

MULTI CRITERIA DECISION MAKING (MCDM) METHODS
This section presents the theoretical basis of PROMETHEE and VIKOR methods.

PROMETHEE
The PROMETHEE (The Preference Ranking Organization METHod for Enrichment of Evaluations) is an outranking method (Papathanasious, 2018). There are 4 variants of the PROMETHEE method: PROMETHEE I for partial ranking of the alternatives, PROMETHEE II for complete ranking of the alternatives (Medić, 2017), PROMETHEE III for ranking based on intervals, and PROMETHEE IV for continuous case (Brans, 2005).
Below is a brief description of the PROMETHEE II method.
The PROMETHEE procedure is based on mutual comparison of every pair of alternatives on each criterion (Medić, 2017).
In order to rank the alternatives, PROMETHEE method introduces the preference function P(a,b), for alternative a and b. Alternatives a and b are evaluated according to The preference function is defined as follows (Opricović, 1998): (2) Brans and Mareschal (2005) proposed six types of generalized criteria:

Criterion with linear preference and indifference area
The multicriteria preference index of alternative a over alternative b is calculated in the following way (Brans, 2005;Papathanasious, 2018): Where wi is the weight of i-th criterion.
Outranking flows (Medić, 2017): The positive outranking flow of alternative aj expresses how an alternative aj is outranking all the others.
Based on positive and negative outranking flows, the net flow is defined as follows: Alternative aj is better than alternative ak, if j > k is ranked higher on the list.

VIKOR
A multicriteria compromise ranking method VIKOR (the acronym: VIšekriterijumsko KOmpromisno Rangiranje, in Serbian) was first introduced by S. Opricović and was developed for multicriteria optimization of complex systems (Papathanasious, 2018;Opricović, 2004). This method has to provide compromise solutions in terms of noncommensurable and conflicting criteria (Papathanasious, 2018;Opricović, 2007). Compromise solutions is the closest to the ideal one, where the ideal solution is determined based on the best values of criteria. (Opricović, 2007;Papathanasious, 2018;Marković, 2013) The compromise ranking procedure has the following steps (Marković, 2013;Medić, 2017;Opricović, 2007;Papathanasious, 2018): Step 1: Determine the best xi * and the worst values xiof all criteria functions: Step 2 where wi is weight of criteria and expresses the preference of a decision-maker.
Step 4: Alternatives are ranked based on the values of Rj, Sj and Qj. The first on the ranking list is the alternative whose values Rj, Sj and Qj are the least, and it represents the best alternative. Alternative aj is better than alternative ak, if Qj<Qk and is ranked higher on the list.
VIKOR is very useful method, specifically in situation when decision-maker is not able or doesn't know to express preference for criteria at the beginning of system design (Opricović, 2004).

COMPARATIVE ANALYSIS OF PROMETHEE AND VIKOR METHODS ON AN EXAMPLE
Based on PROMETHEE and VIKOR models, selection of the most suitable formation for storage CO2 was performed. Several alternatives have been evaluated, such as: storage CO2 in depleted oil and gas reservoirs, saline aquifers, and partially depleted oil reservoirs for enhanced oil recovery (EOR-CO2 method). The selection of the optimal geological formation is based on the following criteria: storage capacity, total storage 52 Tomić L., Karović Maričić V., Danilović D., Leković B., Crnogorac M.
costs, risk assessment costs, storage time dynamics, risk of CO2 leakage from geological formation and risk of CO2 leakage through the well. Preference presents the importance of each criteria expresses by the decision-maker, in this case the most important criteria are risk of CO2 leakage from geological formation and through the well.
The initial and the quantified decision-making matrix are shown in Table 1 and Table 2.

PROMETHEE
The values of the preference function, the preference index and complete ranking of alternatives are shown in Table 3, Table 4 and Table 5.     Table 6, Table 7 and Table 8.

DISCUSSION AND CONCLUSION
In this paper a multi criteria analysis of selection the most suitable CO2 geological storage formation was carried out by means of a comparative analysis of PROMETHEE and VIKOR methods. The selection of the most suitable option for storage of CO2 based on the given criteria was made.
The complete ranking of geological formations for the storage of carbon dioxide using the PROMETHEE method show that storage in the depleted oil and gas reservoirs is in 56 Tomić L., Karović Maričić V., Danilović D., Leković B., Crnogorac M.
the first place. In the second place there is EOR-CO2 method and finally, storage in saline aquifers.
By applying VIKOR method, ranking of the alternatives is as follow: EOR-CO2 method, storage in the depleted oil and gas reservoirs and storage in saline aquifers. The only difference is in reverse order of EOR-CO2 and depleted oil and gas reservoir compared to results of PROMETHEE method.
Ranking of the alternative may vary depending on the approach and method used, which means that the decision-maker have to decide which method is the most appropriate for solving the problem.