Potential and Impacts of Cogeneration in Tropical Climate Countries: Ecuador as a Case Study

: High dependency on fossil fuels, low energy e ﬃ ciency, poor diversiﬁcation of energy sources, and a low rate of access to electricity are challenges that need to be solved in many developing countries to make their energy systems more sustainable. Cogeneration has been identiﬁed as a key strategy for increasing energy generation capacity, reducing greenhouse gas (GHG) emissions, and improving energy e ﬃ ciency in industry, one of the most energy-demanding sectors worldwide. However, more studies are necessary to deﬁne approaches for implementing cogeneration, particularly in countries with tropical climates (such as Ecuador). In Ecuador, the National Plan of Energy E ﬃ ciency includes cogeneration as one of the four routes for making energy use more sustainable in the industrial sector. The objective of this paper is two-fold: (1) to identify the potential of cogeneration in the Ecuadorian industry, and (2) to show the positive impacts of cogeneration on power generation capacity, GHG emissions reduction, energy e ﬃ ciency, and the economy of the country. The study uses methodologies from works in speciﬁc types of industrial processes and puts them together to evaluate the potential and analyze the impacts of cogeneration at national level. The potential of cogeneration in Ecuador is ~600 MW el , which is 12% of Ecuador’s electricity generation capacity. This potential could save ~18.6 × 10 6 L / month of oil-derived fuels, avoiding up to 576,800 tCO 2 / year, and creating around 2600 direct jobs. Cogeneration could increase energy e ﬃ ciency in the Ecuadorian industry by up to 40%.


Introduction and Literature Review
Energy is key for people's well-being and for a countries' development. Still, current global energy use and production heavily relies on fossil derived fuels and electricity produced using this type of fuel. For instance, in 2018, 85% of the worldwide fuel consumption had its origin in fossil fuels. The total petroleum, coal, and natural gas consumption reached 4714 MTOE/year (Million tons of oil equivalent per year), 3744 MTOE/year, and 3328 MTOE/year, respectively [1]. One of the negative consequences  [14,15].
Cogeneration has been recognized as a key element for the diversification of the electricity generation matrix (to help balancing the seasonal hydropower generation), for the reduction of the costs of subsidies to energy in the Ecuadorian industry (by making a better use of fuels for heat production), for the increase in energy efficiency, and for reducing GHG emissions [8]. However, further work is required to determine how much the potential of cogeneration in the Ecuadorian industry is and to define strategies for implementing cogeneration in this sector. Year-round tropical climate, subsidies of the state to fossil fuels and electricity, and insufficient energy policies to promote investments in the energy sector are factors that have hindered the penetration of cogeneration in the country. Because of the relatively constant year-round temperature conditions, indoor heating is not required, even in the Andean highlands (where temperature normally varies between 7 and 23 °C). Thus, cogeneration has been adopted only marginally in the industrial sector. Our field work (see Section 2.1.2 for details) and [8,9] have identified that Ecuador's current installed cogeneration capacity is 172 MWel, which represents only 2% of the total (nominal) electricity generation capacity (i.e., 7361 MWel) [7]. Lignocellulosic biomass is the main fuel employed for cogeneration due to the utilization of bagasse in the sugarcane industry (Table 1). Although there are abundant lignocellulosic biomass resources in the country (e.g., oil palm, rice, banana, and wood residues), the use of these energy sources for cogeneration in the country is very low [7]. For example, in Ecuador, there are currently 35 companies that process oil palm fruit and 4 companies that produce oil from oil palm kernel, of which only 2 currently use cogeneration. Because of the positive impacts of biomass for cogeneration [16], the use of this fuel deserves more attention in the country. In addition to the existing installed cogeneration capacity in the country, there is a thermal power plant (Termogas Machala, 132 MWel of installed capacity) [15] that is currently being retrofitted for operating as a combined cycle (CC) plant by adding heat recovery steam generators (HRSG) and steam turbines. This plant runs with natural gas-NG (obtained from the Gulf of Guayaquil) and gas turbines.  Cogeneration has been recognized as a key element for the diversification of the electricity generation matrix (to help balancing the seasonal hydropower generation), for the reduction of the costs of subsidies to energy in the Ecuadorian industry (by making a better use of fuels for heat production), for the increase in energy efficiency, and for reducing GHG emissions [8]. However, further work is required to determine how much the potential of cogeneration in the Ecuadorian industry is and to define strategies for implementing cogeneration in this sector. Year-round tropical climate, subsidies of the state to fossil fuels and electricity, and insufficient energy policies to promote investments in the energy sector are factors that have hindered the penetration of cogeneration in the country. Because of the relatively constant year-round temperature conditions, indoor heating is not required, even in the Andean highlands (where temperature normally varies between 7 and 23 • C). Thus, cogeneration has been adopted only marginally in the industrial sector. Our field work (see Section 2.1.2 for details) and [8,9] have identified that Ecuador's current installed cogeneration capacity is 172 MW el , which represents only 2% of the total (nominal) electricity generation capacity (i.e., 7361 MW el ) [7]. Lignocellulosic biomass is the main fuel employed for cogeneration due to the utilization of bagasse in the sugarcane industry (Table 1). Although there are abundant lignocellulosic biomass resources in the country (e.g., oil palm, rice, banana, and wood residues), the use of these energy sources for cogeneration in the country is very low [7]. For example, in Ecuador, there are currently 35 companies that process oil palm fruit and 4 companies that produce oil from oil palm kernel, of which only 2 currently use cogeneration. Because of the positive impacts of biomass for cogeneration [16], the use of this fuel deserves more attention in the country. In addition to the existing installed cogeneration capacity in the country, there is a thermal power plant (Termogas Machala, 132 MW el of installed capacity) [15] that is currently being retrofitted for operating as a combined cycle (CC) plant by adding heat recovery steam generators (HRSG) and steam turbines. This plant runs with natural gas-NG (obtained from the Gulf of Guayaquil) and gas turbines.
Despite the positive reputation and the extended use of cogeneration worldwide (especially in temperate climate countries), there are not enough studies showing the potential of cogeneration of whole industrial sectors or how cogeneration, in the conditions of tropical climate countries, could contribute to meet energy requirements, help to increase energy efficiency, reduce national GHG emissions, and, thus, contribute to sustainable development. For some tropical climate countries, there exists some studies focused on cogeneration in specific industrial sectors, such as the sugarcane industry [17][18][19][20][21][22][23][24][25], the oil palm industry [26][27][28], and the wood processing industry [16,[29][30][31][32][33]. The methodologies and learnings from Energies 2020, 13, 5254 4 of 26 those works can be used to conduct a wider analysis on the impacts of cogeneration in a whole country or geographic region, although more research overall is necessary. Thus, the objective of this paper is two-fold: first, to compute the potential of cogeneration in the Ecuadorian industry, and, second, to show the positive impacts of cogeneration on power generation capacity, GHG emission reduction, industrial energy efficiency, and the economy of the country. The presence of subsidies from the state to both electricity and fuels in Ecuador, the seasonality of rains to run hydropower plants, and its year-round tropical weather are particular challenges considered in the study.

Materials and Methods
Our literature review suggests that there are not standardized methods for computing the potential of cogeneration/trigeneration in a specific geographical region or country, which is understandable since each country and its industrial sector have specific conditions that need to be taken into account. There are different aspects that need to be analyzed to determine the most suitable methodology to compute cogeneration potential at a country level (e.g., weather, types of energy sources available, altitude above the sea level, energy policies and incentives). In tropical climate countries such as Ecuador, the weather is an important factor that determines specific types of cogeneration schemes because, as previously mentioned, there is no need for indoor heating (an important energy requirement in tempered climate countries), but air conditioning is required instead [34][35][36]. Consequently, cogeneration projects are more suitable in the industrial sector and in other places where hot and cold fluids are used (e.g., hospitals, hotels, airports, shopping malls). These are the target places for cogeneration projects in tropical climate countries.
Another factor to consider for computing the potential of cogeneration is the pattern of energy consumption in the industrial sector, which in Ecuador is relatively constant throughout the year, reflecting a common feature of energy consumption in the industry of tropical countries. For Ecuador, and to illustrate this important point, Figure 2 shows two examples of energy consumption curves (both electricity and fuel) corresponding to two large Ecuadorian industrial companies (herein referred to as companies M and N) devoted to the production of tires (M) and pulp and paper (N). This energy consumption pattern of the industrial sector in Ecuador suggests that cogeneration plants in tropical climate countries could operate at approximately constant capacity year-round, which makes the sizing process of the cogeneration plants easier. The methodology adopted herein considers these elements.

Methodology
The potential of cogeneration in the whole industrial sector of a country can be obtained if the potential of cogeneration of each industrial plant in which cogeneration can be adopted is determined. The methods for sizing cogeneration plants for specific types of industries are based on their annual energy requirements (normally, heat for the industrial process and/or plant operation, since producing surplus heat will otherwise be wasted). Furthermore, producing electricity is not a priority in the industrial plants in the country due to its relatively low cost (i.e., due to subsidies). Table 2 presents a list of works devoted to determining the cogeneration capacity in specific types of industrial plants. These works served as the basis to compute the potential cogeneration capacity in industrial plants in Ecuador. In addition, a report on the potential of cogeneration in Spain [37] and a report by the Office of Environment and Heritage New South Wales [38] were used. Moreover, for sizing cogeneration plants, it is necessary to define the cogeneration schemes suitable to specific types of industries and the respective fuels available. In this study, such schemes are shown in Appendix A, while the main equations used are provided in Appendix B. Then, the potential of each industrial plant was added to obtain the potential of cogeneration by cluster of industries and the whole country's potential. The methodology adopted consisted of five stages (summarized in Figure 3) that are detailed in the following subsections.

Methodology
The potential of cogeneration in the whole industrial sector of a country can be obtained if the potential of cogeneration of each industrial plant in which cogeneration can be adopted is determined. The methods for sizing cogeneration plants for specific types of industries are based on their annual energy requirements (normally, heat for the industrial process and/or plant operation, since producing surplus heat will otherwise be wasted). Furthermore, producing electricity is not a priority in the industrial plants in the country due to its relatively low cost (i.e., due to subsidies). Table 2 presents a list of works devoted to determining the cogeneration capacity in specific types of industrial plants. These works served as the basis to compute the potential cogeneration capacity in industrial plants in Ecuador. In addition, a report on the potential of cogeneration in Spain [37] and a report by the Office of Environment and Heritage New South Wales [38] were used. Moreover, for sizing cogeneration plants, it is necessary to define the cogeneration schemes suitable to specific types of industries and the respective fuels available. In this study, such schemes are shown in Appendix 1, while the main equations used are provided in Appendix 2. Then, the potential of each industrial plant was added to obtain the potential of cogeneration by cluster of industries and the whole country's potential. The methodology adopted consisted of five stages (summarized in Figure 3) that are detailed in the following subsections.

Data Collection and Energy Consumption Baseline
The tasks described in Sections 2.1.1.1 and 2.1.1.2 aimed to determine which industrial plants could adopt cogeneration (or trigeneration) in the country. For this, information on electricity and fuel consumption was used to define a baseline that allows selecting prospective industrial companies. This information was obtained from two official sources, the Agency of Regulation and Control of Electricity-ARCONEL (in Spanish Agencia de Regulación y Control de Electricidad) and the Agency of Regulation and Control of Hydrocarbon Fuels-ARCH (in Spanish Agencia de Regulación y Control de Hidrocarburos), which are the institutions in charge of regulating and controlling the distribution and use of electricity and fossil-derived fuels, respectively. The data used corresponded to 2015 and were the information available at the time that this study was conducted (2017 and 2018).

Electricity Consumption Baseline
The initial list on electricity consumption from the ARCONEL contained clients/consumers reporting electricity consumption above 20,000 kWh/month. This electricity consumption baseline was established after analyzing the energy demand of a small food processing company with installed capacity of approximately 30 kWel, working 24 h/day the year-round (i.e., with electricity consumption of ~20,000 kWh/month). The company is located in the city of Cuenca, and herein it is referred to as Company A. The number of companies/consumers in the initial list was ~41,800. Next, the resulting list was analyzed and filtered again to remove companies and/or institutions (both public and private) in which, although their electricity consumption was >20,000 kWh/month, no fuels are required for their operation, except diesel for transportation and LPG (Liquid Petroleum Gas) for cooking at a small scale. This is the case of: (1) Elementary schools, high schools, colleges/universities, government buildings and offices at a national or municipal level where, as previously mentioned, due to climate conditions in Ecuador, there is no necessity of cogeneration intending, for example, indoor heating (which is common in temperate places) or water heating.

Data Collection and Energy Consumption Baseline
The tasks described in Sections 2.1.1.1 and 2.1.1.2 aimed to determine which industrial plants could adopt cogeneration (or trigeneration) in the country. For this, information on electricity and fuel consumption was used to define a baseline that allows selecting prospective industrial companies. This information was obtained from two official sources, the Agency of Regulation and Control of Electricity-ARCONEL (in Spanish Agencia de Regulación y Control de Electricidad) and the Agency of Regulation and Control of Hydrocarbon Fuels-ARCH (in Spanish Agencia de Regulación y Control de Hidrocarburos), which are the institutions in charge of regulating and controlling the distribution and use of electricity and fossil-derived fuels, respectively. The data used corresponded to 2015 and were the information available at the time that this study was conducted (2017 and 2018).

Electricity Consumption Baseline
The initial list on electricity consumption from the ARCONEL contained clients/consumers reporting electricity consumption above 20,000 kWh/month. This electricity consumption baseline was established after analyzing the energy demand of a small food processing company with installed capacity of approximately 30 kW el , working 24 h/day the year-round (i.e., with electricity consumption of~20,000 kWh/month). The company is located in the city of Cuenca, and herein it is referred to as Company A. The number of companies/consumers in the initial list was~41,800. Next, the resulting list was analyzed and filtered again to remove companies and/or institutions (both public and private) in which, although their electricity consumption was >20,000 kWh/month, no fuels are required for their operation, except diesel for transportation and LPG (Liquid Petroleum Gas) for cooking at a small scale. This is the case of: (1) Elementary schools, high schools, colleges/universities, government buildings and offices at a national or municipal level where, as previously mentioned, due to climate conditions in Ecuador, there is no necessity of cogeneration intending, for example, indoor heating (which is common in temperate places) or water heating. (2) Construction and civil engineering companies (e.g., roads construction companies) that report high electricity consumption (for example for reducing the particle size of rocks).
It was also observed that the possibilities of cogeneration in a few companies that process polymers/plastics (e.g., High Density Polyethylene-HDPE, Polypropylene-PP, Polyvinyl Chloride-PVC) for producing plastic toys, plastic bags and/or plastic furniture for both domestic and industrial use (with electricity consumption > 20,000 kWh/month) should be verified in situ. Thus, these companies were kept in the list. The amount of companies after this filtering process was approximately 2000.

Fuel Consumption Baseline
The fuel consumption baseline started by analyzing the possibilities of cogeneration in the representative Company A (Section 2.1.1.1), which uses heat (produced by burning diesel) for its manufacturing process. The fuel consumption of this company served as the basis to start filtering Energies 2020, 13, 5254 7 of 26 the data provided by the ARCH. The company uses a typical small boiler (186 kW th ) that produces saturated steam at 140-150 • C, working~6 h/day, 5 days/week, and employing up to 7570 L/month (i.e., 90,840 L/year) of diesel. A preliminary computation (following works of [43,44,47] and energy balances) showed that, if the company was interested in adopting cogeneration, the size of the cogeneration plant would be close to 300 kW el . This cogeneration unit could operate, for instance, on a diesel or a gas engine (depending on the fuel available) and use the waste heat for producing the steam for the process (in a HRSG). However, according to a study conducted in the industrial sector in Mexico (with weather conditions somehow similar to those in Ecuador), the projects on cogeneration that offer better prospective, from an economic viewpoint, are those larger than 500 kW el [72]. Therefore, the minimum capacity of the cogeneration plants in the Ecuadorian industry, in all cases and at this level of the study, should be 500 kW el , which corresponds to a cogeneration plant that demands 90,800 L/year of diesel (or any diesel equivalent fuel) Consequently, the fuel consumption data filtering process started by considering a baseline of diesel or fuel oil consumption of 90,800 L/year (76.19 t/year).
The information on fossil fuel consumption provided by the ARCH included data on type of fuel, amount, company's name, location and information on the main products of the company. This information was used to identify the location of each industrial plant. The types of fuels consumed in the country are as follows: fuel oil, diesel fuel (for both industry and transportation), gasoline (both regular and premium), liquefied petroleum gas (LPG), and NG in a smaller amount (all fuels were converted to diesel equivalent fuel). The initial list included~500,000 companies and institutions. An initial filtering process removed from the list companies that a) reported LPG consumption, since in the country LPG is not used for industrial processes, except some hotels, hospitals, and shopping malls that have centralized LPG supply in relatively small amounts, and b) companies that sell diesel and gasoline for transportation (i.e., gas stations). The resulting list was filtered again by removing institutions that reported large amounts of diesel consumption for transportation only (e.g., municipal governments; ministries from the Ecuadorian government; and civil engineering companies that use diesel for transport/operation of heavy machinery for the construction of roads, bridges, and large buildings in the country). After a quantitative analysis, similar to that conducted for company A, it was found that the cogeneration capacity in companies consuming <151,400 L/year of fuel-oil or diesel will be <500 kW el . Thus, the final fuel consumption baseline for selecting the companies where cogeneration could potentially be adopted was 151,400 L/year of diesel and/or fuel oil (both with approximately similar high heating value-HHV). Therefore, the list was reduced to~1000 companies.

Final List of Industrial Companies That Could Adopt Cogeneration
The resulting lists (after filtering the ARCONEL and the ARCH data) were put together to prepare a final list of industrial companies (including hotels and hospitals) at a national level. Although the majority of the companies from the filtered ARCH list were also present in the filtered ARCONEL list, some companies were present in one list only since they reported high electricity consumption but low fuel consumption (e.g., plastics processing and ice making companies) and vice versa (e.g., fishing companies). After a case by case analysis, the final list was comprised of 555 companies (See Figure 4). All the 555 companies from the list, except 2 (from the oil palm industry, which are located in the Amazonian region), are located in the coast (~57%) and in the Andean highlands (~43%) regions. Among this list, there were sixteen companies working on shrimp growing/processing and eight ice making plants. These companies reported both high electricity and diesel consumption, but the chances of cogeneration were apparently negligible, since it was identified that the fuels were used for water pumping using internal combustion (diesel) engines in places where no electricity grids were available for shrimp pools operation and/or for land transport (using trucks). Thus, we decided to keep these companies in the final list to confirm the possibilities of cogeneration after visiting some of those plants.
growing/processing and eight ice making plants. These companies reported both high electricity and diesel consumption, but the chances of cogeneration were apparently negligible, since it was identified that the fuels were used for water pumping using internal combustion (diesel) engines in places where no electricity grids were available for shrimp pools operation and/or for land transport (using trucks). Thus, we decided to keep these companies in the final list to confirm the possibilities of cogeneration after visiting some of those plants.

Classification of Companies by Clusters and Validation of Data
The 555 companies in the final list were classified by clusters, which helped to organize visits to confirm the energy consumption data and to identify and record the corresponding industrial processes, including the identification of hot/cold fluids and their characteristics. The companies were grouped into twelve categories or clusters of industries, following the International Standard Industrial Classification of All Economic Activities (ISIC) [73,74]. Airports, shopping malls, and oil refineries were included in the cluster "others". Table 3 shows the list of clusters and the number of companies in each cluster. The information provided by the ARCONEL and the ARCH was validated by visiting 162 companies (~30% of the total), as detailed in Table 3. The selection of the companies to visit considered the amount of companies per cluster, the sizes, location, and the types of manufacturing processes to guarantee that all types of industries were visited. Interview survey formats (asking about energy consumption, types and amounts of fuels, industrial process, types and conditions of industrial fluids, if cogeneration has been adopted in the plant and the corresponding conditions, and other aspects to determine cogeneration potential) were used to collect the information provided by the industrial companies.

Classification of Companies by Clusters and Validation of Data
The 555 companies in the final list were classified by clusters, which helped to organize visits to confirm the energy consumption data and to identify and record the corresponding industrial processes, including the identification of hot/cold fluids and their characteristics. The companies were grouped into twelve categories or clusters of industries, following the International Standard Industrial Classification of All Economic Activities (ISIC) [73,74]. Airports, shopping malls, and oil refineries were included in the cluster "others". Table 3 shows the list of clusters and the number of companies in each cluster. The information provided by the ARCONEL and the ARCH was validated by visiting 162 companies (~30% of the total), as detailed in Table 3. The selection of the companies to visit considered the amount of companies per cluster, the sizes, location, and the types of manufacturing processes to guarantee that all types of industries were visited. Interview survey formats (asking about energy consumption, types and amounts of fuels, industrial process, types and conditions of industrial fluids, if cogeneration has been adopted in the plant and the corresponding conditions, and other aspects to determine cogeneration potential) were used to collect the information provided by the industrial companies. * Three airports were included in the study: Guayaquil, Quito, and Cuenca. The rest of airports in the country operate only sporadically and are not candidates for cogeneration. ** The three main oil refineries in the country [6] were included. *** Air conditioning.

Selection of Cogeneration Technologies
The following considerations were made for selecting the cogeneration technology that fits into the industrial plants' requirements: (1) The proposed cogeneration/trigeneration system must fit into the current plant's requirements of heat (e.g., steam and/or hot water necessities) or cold fluids (including A/C) to guarantee cogeneration plants with high capacity factors. Therefore, the plant requirement of thermal energy with heating and/or cooling effect defined the cogeneration/trigeneration capacity of the plant. (2) The prime mover selected will allow one to cover the electricity requirements totally or partially.
In the case of deficit of electricity, and as long as the thermal energy production is met, it is preferred to import electricity from the national grid. If the cogeneration system produces electricity surplus, then it can be sold to the national grid. No sell or purchase of hot/cold fluids (i.e., transport of these fluids from or to the plant) were considered. (3) The type of fuel (e.g., biomass, biogas, NG, diesel, heavy oil) proposed for cogeneration should be readily available in the place the cogeneration plant will be located. Therefore, fuel availability is a key component for deciding on the technology proposed. (4) The yearly average thermal energy requirements (not the peak requirements) were used for sizing the cogeneration/trigeneration plant. (5) No indoor heating and/or district heating are required. This is expected due to geographical location [75]. (6) The selection of the prime movers considered the limitations imposed by geographical conditions, specifically altitude. For the case study, industrial plants in the Ecuadorian Andes highlands are located at approximately 2500 m above the sea level (m.a.s.l.); thus, in these places, it is preferred to use diesel engines, gas engines, or boiler and steam turbines instead of gas turbines to guarantee adequate levels of efficiency of the cogeneration plant [76][77][78]. (7) The selection of the prime movers also considered possible partial loads requirements (i.e., the ability to vary thermal and electrical output depending on hourly requirements, or the necessity for frequent stopping and starting). Consequently, diesel and/or gas engines are preferable for cogeneration instead of gas turbines or steam turbines coupled with boilers in companies that do not operate 24/7. Diesel and gas engines, additionally, are able to run with renewable fuels (biodiesel and biogas, respectively), which are expected to be available in the country in the future [79] (See Section 3.2). (8) Trigeneration can be projected only in industrial plants where air conditioning and/or process cooling fluids (above the water freezing temperature) for the industrial process are required. In this case, both air conditioning and/or cold fluids will be produced by using residual heat from the prime mover. The trigeneration system will mostly work on LiBr (lithium bromide) absorption equipment for air conditioning in the Coastal region and, in some cases, hotels, hospitals, and airports in the Andean highlands. Ammonia (NH 3 ) absorption systems are proposed only when fluids with low temperatures are required for the industrial process (e.g., for pasteurization in the beverages, food, and dairy industries). Freezing is not part of the proposed trigeneration systems.

Computation of the Potential of Cogeneration of Ecuador
The potential of cogeneration of Ecuador was determined in two steps. First, the sum of the potential of cogeneration of all industries by each cluster was conducted. Then, the potential of each cluster was added to obtain the potential at a national level. Regarding cogeneration sizing at the industry level, the computations were first conducted for the industrial plants that were visited (see Section 2.1.2), and computations were carried out for the rest of the plants, using the information on the fuels and electricity consumption, as well as its location, working conditions, and size in a case by case basis. The main steps for computing the potential of cogeneration of a specific company were as follows (see Appendix B for equations used):

1.
Identify the location of the industrial plant and the availability of electricity grids to ensure interconnection to import/export electricity when electricity deficit/surplus exists.

2.
Collect/verify data on electrical and thermal loads and types of fuels used. This information was compared with the data from the ARCONEL and the ARCH (Section 2.1.1).

3.
Gather data on the company's process: types of products, heat requirements (e.g., steam or hot gases) and other fluids used (e.g., cold fluids, air conditioning, hot water). 4.
Identify types of fuels that are or could be available in the company (or plant) location place.

5.
Select the appropriate cogeneration prime mover and the corresponding fuel. 6.
Compute the cogeneration plant capacity, based on the necessities of thermal energy. Table 4 presents equipment parameters used for the computations. 7.
Standardize the size of the equipment suggested for a specific company by using catalogues from companies that provide equipment for cogeneration/trigeneration (e.g., boilers, diesel engines, gas engines, steam turbines, HRSGs, and absorption chillers). 8.
Compute the amount of fuel that the cogeneration/trigeneration plant will require (Appendix B). 9.
Compute the amount of electricity that will be produced by the prime movers in the operating conditions of the cogeneration plant and how much of this electricity will be available for exporting to the national grid (if surplus electricity is available). Table 4. Parameters corresponding to the equipment used in the computations.

Equipment and Type Efficiency Comments
Diesel engine Up to 40% electric efficiency [78]) Expected heat recovery: up to 86% from the total heat released by the engine (i.e., heat from exhaust gases and heat from jacket coolant), depending on the size of the engine.
Gas engine (working with biogas) Up to 45% electric efficiency [78] Expected heat recovery: up to 88% from the total heat released by the engine (i.e., heat from exhaust gases and heat from jacket coolant), depending on the size of the engine. Steam turbine (back pressure)~5 5% [78] Heat recovery steam generator (HRSG) 82% [80] Absorption chillers (single effect in all cases) * The computation of the environmental impacts of cogeneration considered two types of impacts: (a) the GHG emissions resulting from the fuel burned in each cogeneration plant, and (b) the avoided GHG emissions resulting from the possible replacement of large thermal power plants in the country (that use fossil-derived fuels for electricity production) by cogeneration plants in the industry. It is expected that the availability of cogeneration plants could remove the necessity of installing a thermal power plant (that uses oil-derived fuels to run) with capacity equal to that corresponding to the total cogeneration potential. Both results were added to obtain the net GHG emissions.

(a) Emissions in cogeneration plants
The fuels required for cogeneration depend on the prime mover selected. Cogeneration in Ecuador will use diesel, biogas, and lignocellulosic biomass, which are the fuels available currently in the country (See Section 3.2). The GHG emissions were estimated for each type of fuel. The computations followed the concept of conservation of carbon, from the fuel combusted into CO 2 , according to the guidelines from the International Energy Agency [81]. For biogas, GHG emissions also considered the release of methane to the environment that can be avoided by using effluents in palm oil mills to produce biogas via anaerobic digestion [28].

(b) Emissions avoided by replacing thermal power plants
This computation consisted of determining how much fossil-derived fuels could save the country due to the substitution of existing or expected thermal power plants for electricity production (which could be a necessity to offset hydropower generation capacity in the country, especially during the dry season of the year) by cogeneration in industrial plants. To make easier the computations, it was assumed that the efficiency of large thermal power plants is~35% [80] (although the efficiency of some existing thermal power plants in Ecuador is lower). The expected efficiency of the cogeneration plants taken as a reference was calculated in five representative companies (including a hospital and a hotel, where trigeneration is possible). Results showed efficiencies >70% in all cases. Thus, the difference in efficiency in a scenario without cogeneration and a scenario with cogeneration was conservatively taken as 30%.

Economic Impacts
Economic analysis was carried out to understand the convenience of cogeneration in the country from an economic point of view. The analysis consisted of (a) estimating the costs avoided if cogeneration is used instead of large thermal power plants that operate on fossil fuels, and (b) computing the cost of generating electricity in cogeneration plants if the whole potential of cogeneration calculated is installed. Table 5 summarizes the parameters employed for conducting the economic analysis. Some of these parameters are in agreement with the work of [42]. The prices of fuels and electricity are similar in all regions of the country.  [84].

Operation and maintenance costs
Value varies from 2% of the investment during the first years of the projects to up to 7% after year 10. Values are in the range of those reported by [85], although a little higher after year 5 due to the necessity of importing parts. Expected capacity factor 95% to 60%, depending on the type of industry (see Table 6). discount rate (includes financial cost and financial risk) 12% (rate currently used for electricity projects in Ecuador).

Reinvestment
25% of the initial investment will be required on year 10. Projects lifetime 15 years.

Plant location and land requirements No land will be bought for cogeneration plants since the plant will be installed at existing companies' facilities. Substation and transmission facilities
Cost is included in the cost of prime movers. Insurance 0.5% of the investment per year Cost of diesel and natural gas USD 2.12/gallon (USD~0.57 US/L) and USD 0.45/kg, respectively (without subsidies) [86].
Cost of biomass *** USD 20/t, which is in the range of or above the costs of residues from the agroindustry (e.g., oil palm residues) in the Ecuadorian coast region (resulting from a field study).

Workforce salaries
Each cogeneration plant will require one employee per MW el installed per every 8 h of operation, with salaries of USD 1250/month (in the conditions of Ecuador), plus one supervisor and one person in charge of maintenance.
* Includes project management and design engineering as well as construction and start-up. This is a referential cost due to discrepancy of values in the literature. The authors of [78] show higher values, but [87] and [85] report values in the range of USD 1000/kW. However, the cost of a gas engine (1 MW) operating at a landfill in Cuenca was USD 450/kW. The value considered in this work could be adequate due to economy of scale when contracting and installing several cogeneration plants. ** TR refers to ton of refrigeration (equivalent to 3.52 kW). *** Electricity to be sold to the national electricity grid after operation of the plant and service loads are met. *** To operate cogeneration plants based on Rankine cycle. (1) Using only gas engines running with biogas. (2) Using biomass from the same plant. (3) Depending on the size of the company. (4) Further study is required to analyze the possibility of using biomass.

Social Impacts
According to [88] (p. 43), social impacts are the 'consequences of social relations (interactions) weaved in the context of an activity (production, consumption or disposal) and/or engendered by it and/or by preventive or reinforcing actions taken by stakeholders (ex. enforcing safety measures in a facility)'. A social life cycle analysis (SLCA) should consider the potential social impacts on local communities, workers, and consumers [89]. However, the literature shows that the social implications of projects related, for instance, with the use of lignocellulosic natural resources for energy [90] or wood-based products [91] are hard to estimate due to the difficulty of correlating cause-effect chains with regards to production activities and their potential social effects. Therefore, the computation of the social impacts of adopting cogeneration in a whole country is even more difficult. For this reason, in this work, the social impacts of cogeneration are focused on a preliminary estimation of such impacts on the creation of new jobs in the places where cogeneration plants could be installed. Such jobs are required, generally, for operating the cogeneration plants. Each plant will require at least five people: three for operation, one for maintenance, and one for management/supervision.

Current Electricity Demand and Fuel Consumption in the Industrial Sector of Ecuador
The electricity demand (from de National Interconnected System-SNI) and the fuel consumption in the 555 companies are 409,199 MWh/month and 61.73 × 10 6 L/month (51,773 t/month) of diesel equivalent, respectively. Figure 5 shows the electricity demand and fuel consumption by each type of cluster of companies (See Table 3). It is seen that the electricity consumption (Figure 5a) is higher in the clusters of food and construction materials industries, with 19% and 17% of the total, respectively. The fuel consumption, as seen in Figure 5b, is higher, again, in the cluster of companies of construction materials and in the cluster of food industries, with 17% and 16% of the total, respectively. The large amount of companies in the food industry cluster and the presence of energy intensive industries in the construction materials cluster (e.g., cement and ceramic tiles) explain these results.
is higher in the clusters of food and construction materials industries, with 19% and 17% of the total, respectively. The fuel consumption, as seen in Figure 5b, is higher, again, in the cluster of companies of construction materials and in the cluster of food industries, with 17% and 16% of the total, respectively. The large amount of companies in the food industry cluster and the presence of energy intensive industries in the construction materials cluster (e.g., cement and ceramic tiles) explain these results.  Table 6 presents the technologies suggested for cogeneration schemes in each type of industry in Ecuador. The table also shows the geographic location of each cluster of industries. Internal combustion engines (diesel and gas engines) are the most prominent prime movers suggested due to their advantages, as discussed in Section 2.1.3. In addition, these engines offer the possibility of working with biodiesel and biogas, in substitution of diesel and NG, respectively, which is of interest in Ecuador. Currently the country produces only ~30 t/year of biodiesel from Jatropha Curcas to operate diesel engines in thermal power plants the Galapagos Islands [92]. The program to produce  Table 6 presents the technologies suggested for cogeneration schemes in each type of industry in Ecuador. The table also shows the geographic location of each cluster of industries. Internal combustion engines (diesel and gas engines) are the most prominent prime movers suggested due to their advantages, as discussed in Section 2.1.3. In addition, these engines offer the possibility of working with biodiesel and biogas, in substitution of diesel and NG, respectively, which is of interest in Ecuador. Currently the country produces only~30 t/year of biodiesel from Jatropha curcas to operate diesel engines in thermal power plants the Galapagos Islands [92]. The program to produce biodiesel from this plant is in its infancy, but it is expected that the biodiesel production capacity will increase in coming years. The use of gas engines deserves further study since it is expected that the agroindustrial sector in Ecuador will start producing biogas using their residues via anaerobic digestion. However, this topic is out of the scope of this paper.

Potential of Cogeneration/Trigeneration
The estimated potential of cogeneration in Ecuador is 598 MW el , which, as mentioned in Section 2.1, consists of the potential of cogeneration of industries with expected installed cogeneration capacity above 0.5 MW el . The value excludes the existing cogeneration capacity shown in Table 1. This potential is~7% of the current electricity generation installed capacity in Ecuador and could produce up to 17% of the total electricity consumed in 2017 in the country. This last value is, interestingly, in the range of percentages of the cogeneration share (respect to the total electricity produced) in countries such as Germany (17%), Brazil (18%), Spain (12%), or the United States (12%) [58,[93][94][95]. Even though in the case of Ecuador this amount refers to potential cogeneration (i.e., not installed cogeneration capacity), such value is important because of the possibility of using cogeneration during the driest season of the year, when hydropower generation is negatively affected by weather conditions (See Section 1). For this reason, cogeneration has been seen in the country as an important strategy for electricity production in the near future, and new laws and regulations are under study to promote cogeneration/trigeneration. Figure 6 summarizes the potential of cogeneration in Ecuador by type of prime mover selected. Diesel engines are the predominant prime movers suggested for cogeneration (Section 3.2). These engines can run with biodiesel (mixed with diesel) when available. Figure 7 presents the potential of cogeneration by cluster, showing that the textile, food, and agroindustry industries are the clusters with higher potential. Moreover, the potential of trigeneration in the country is 212 MW el . Approximately 17% of the 555 companies identified in Section 2.1 could adopt trigeneration, especially in the food and beverages industries, as well as in hotels and hospitals (Figure 8).
Diesel engines are the predominant prime movers suggested for cogeneration (Section 3.2). These engines can run with biodiesel (mixed with diesel) when available. Figure 7 presents the potential of cogeneration by cluster, showing that the textile, food, and agroindustry industries are the clusters with higher potential. Moreover, the potential of trigeneration in the country is 212 MWel. Approximately 17% of the 555 companies identified in Section 2.1 could adopt trigeneration, especially in the food and beverages industries, as well as in hotels and hospitals (Figure 8).

Impacts of Cogeneration in Ecuador
3.4.1. Fuel Consumption, Improvement of Energy Efficiency, and GHG Emissions Reduction The adoption of cogeneration in Ecuador will require different types of fuels. Due to the lack of NG in the country (the preferred fuel for cogeneration in most countries from tempered regions), in

Fuel Consumption, Improvement of Energy Efficiency, and GHG Emissions Reduction
The adoption of cogeneration in Ecuador will require different types of fuels. Due to the lack of NG in the country (the preferred fuel for cogeneration in most countries from tempered regions), in the conditions of this study and considering current fuel availability in Ecuador (See Section 3.2), diesel has been selected. Diesel could comprise approximately 81% of the fuel requirements for cogeneration (if the whole potential of 598 MW el is installed), while biogas and biomass could, together, cover approximately 17% (as shown in Table 7). Biomass fuel is constituted by solid residues generated by the agroindustry (e.g., oil palm and rice), which are abundant biomass resources in the coast region. Although the potential of biomass for cogeneration can be higher than this value, its use deserves more analysis due to the difficulty of hauling and burning this fuel in industrial plants located in urban areas far away from biomass sources. The potential use of NG for cogeneration is very low (~2%). Because of NG is an important fuel for cogeneration in most countries (due to availability, competitive prices, and cleanliness during burning), Ecuador urgently needs to look for NG as an alternative (at least partially) to diesel. For this purpose, two options are being analyzed in the country: (a) importing NG from neighbor countries such as Peru, which, in addition to its high potential production [45], could also import it from Bolivia, as part of the so-called Latin America Energy Integration [96][97][98], and (b) exploring the Gulf of Guayaquil for more NG, since there is no certainty about the NG reserves in this part of the country. Table 7. Types and quantities of fuels required for cogeneration in Ecuador and potential contribution to greenhouse gas (GHG) generation/reduction.  Table 7 also shows the electricity that could be produced by type of fuel (column four) and the corresponding potential contribution to GHG emissions (Table 7, column five). The negative sign in the Table indicates avoided GHG emissions, which results from (1) burning biomass and biogas instead of oil-derived fuels to produce electricity (in cogeneration plants), and (2) the avoided methane formation from liquid effluents from the oil palm industry. Currently, although the majority of the 35 oil palm companies in the country (See Section 1) are aware about the necessity of using liquid effluents for biogas production, these effluents are discharged to pools for stabilization prior to final disposal due to the lack of incentives/regulations from the State to use them for energy.

Type of Fuel
The adoption of cogeneration could promote a reduction 18.55 million L/month (15,556 t/month) of diesel (and/or heavy fuel oil) and avoid up to 576,800 tCO 2 /year. This value results from considering that the country would need to install and operate a 600 MW el power plant (or several plants with equivalent total capacity) to offset the reduction of hydropower during the dry season and that, instead of installing such thermal power plant, cogeneration in the industry will be adopted. The positive impact of cogeneration in the industrial sector's energy efficiency of the country is proportional to the amount of fuels saved. Thus, in the conditions of this study, the increase in energy efficiency, if the whole cogeneration potential was installed, could reach between 35% and 40%. The net GHG emissions (i.e., total 1 in Table 7 minus 576,800) could be −296,007 tCO 2 /year (total 2), showing that installing cogeneration/trigeneration in the industry can be an important strategy to avoid GHG emissions in Ecuador. Figure 9 shows that the clusters in which fuel savings could be higher are the food industry, the beverage industry, and the agroindustry. Further study is necessary for analyzing the environmental positive impacts of changing diesel and natural gas by biodiesel and biogas, respectively. However, Table 7 shows that potential GHG emissions are reduced even using diesel and NG, as a consequence of higher efficiency on burning these fuels in cogeneration plants.

Economic Analysis
The economic analysis showed that an important consequence for Ecuador is that, if cogeneration is installed instead of a large thermal power plant to offset the future lack of hydroelectricity, the country could save up to USD 125 million per year by avoiding the use of oilderived fuels for electricity generation. The cost of the electricity produced in cogeneration plants will depend on the type of cogeneration scheme and the type of fuel used, as seen in Table 8. The cost for electricity produced in cogeneration plants (considering the cost of fuels shown in Table 5, but excluding NG), will vary from USD 0.09/kWh to USD0.17/kWh for electricity produced in the oil palm industry (using lignocellulosic biomass) and in hospitals (using diesel), respectively. Table 8 also shows that some types of cogeneration plants, even using diesel, could produce electricity at costs lower than USD 0.17/kWh. For instance, the hotels industry and the textile industry could produce electricity at USD 0.12/kWh and USD 0.13/kWh (using diesel as fuel), respectively. Although these values are higher than the cost of generating electricity in hydropower plants in Ecuador (up to 0.08 USD /kWh), cogeneration in these conditions is still of interest for Ecuador due to the necessity of diversification of electricity generation and the opportunity of having installed capacity for electricity generation during the dry season of the year. Because of insufficient electricity generation (especially before 2016), Ecuador has often required to import electricity from both Colombia and Peru at prices up to USD 0.28/kWh or to produce electricity using thermal power plants at even higher costs (up to USD 0.50/kWh in old thermal power plants).
An analysis of sensitivity was carried out to understand the effect of using NG (when available in the future) instead of diesel for cogeneration in the country. Results showed that NG could promote a substantial reduction of the costs of electricity production in cogeneration plants. For instance, the dairy industry could produce electricity at around USD 0.06/kWh, hotels at USD 0.08/kWh, and hospitals at USD 0.05/kWh (Table 8). These results reinforce the notion that the country must look for options for buying NG overseas, especially in neighboring countries (see Section 3.4.1). The production and use of biofuels for cogeneration requires further analysis.

Economic Analysis
The economic analysis showed that an important consequence for Ecuador is that, if cogeneration is installed instead of a large thermal power plant to offset the future lack of hydroelectricity, the country could save up to USD 125 million per year by avoiding the use of oil-derived fuels for electricity generation. The cost of the electricity produced in cogeneration plants will depend on the type of cogeneration scheme and the type of fuel used, as seen in Table 8. The cost for electricity produced in cogeneration plants (considering the cost of fuels shown in Table 5, but excluding NG), will vary from USD 0.09/kWh to USD0.17/kWh for electricity produced in the oil palm industry (using lignocellulosic biomass) and in hospitals (using diesel), respectively. Table 8 also shows that some types of cogeneration plants, even using diesel, could produce electricity at costs lower than USD 0.17/kWh. For instance, the hotels industry and the textile industry could produce electricity at USD 0.12/kWh and USD 0.13/kWh (using diesel as fuel), respectively. Although these values are higher than the cost of generating electricity in hydropower plants in Ecuador (up to 0.08 USD/kWh), cogeneration in these conditions is still of interest for Ecuador due to the necessity of diversification of electricity generation and the opportunity of having installed capacity for electricity generation during the dry season of the year. Because of insufficient electricity generation (especially before 2016), Ecuador has often required to import electricity from both Colombia and Peru at prices up to USD 0.28/kWh or to produce electricity using thermal power plants at even higher costs (up to USD 0.50/kWh in old thermal power plants).
An analysis of sensitivity was carried out to understand the effect of using NG (when available in the future) instead of diesel for cogeneration in the country. Results showed that NG could promote a substantial reduction of the costs of electricity production in cogeneration plants. For instance, the dairy industry could produce electricity at around USD 0.06/kWh, hotels at USD 0.08/kWh, and hospitals at USD 0.05/kWh (Table 8). These results reinforce the notion that the country must look for options for buying NG overseas, especially in neighboring countries (see Section 3.4.1). The production and use of biofuels for cogeneration requires further analysis. Table 8. Examples of costs of electricity generated in some types of clusters of industries in the conditions of the study (including the potential use of NG).

Social Impacts of Cogeneration
The adoption of cogeneration/trigeneration in Ecuador could promote more than 2600 new jobs. As mentioned in Section 2.1.5.3, these direct jobs are required for operating, managing, and maintaining the cogeneration plants. There is evidence showing positive impacts of energy efficiency measures on GDP, employment, economic structure, and welfare [99]. In addition, there is an important element that was not included in the economic analysis: the benefit to the state of avoiding the release of CO 2 by installing cogeneration plants, which is related to the "social cost of carbon" or marginal damage caused by an additional ton of carbon dioxide emissions [2][3][4]100]. Therefore, these and other benefits that are not considered at this level of the study (e.g., the impact on rural areas where some cogeneration will be installed, the benefits on health due to better air quality or the creation of indirect jobs) deserve further study.

Conclusions
In tropical climate countries, the potential of cogeneration (and as such, its calculation) of the industrial sector is dependent on particular climate conditions, consumption behavior, cogeneration schemes, and fuel availability. Tropical countries such as Ecuador do not necessitate indoor heating (an important energy requirement in tempered climate countries), although air conditioning is prominently used. Thus, large cogeneration projects are more suitable in the industrial sector and in places where hot and cold fluids are used (e.g., hospitals, hotels, airports, and shopping malls). This study has shown that the adoption of cogeneration at a large scale promotes environmental, economic, and social benefits to countries by reducing GHG emissions, promoting fuel savings and energy efficiency, and by creating new jobs, respectively. In the case of Ecuador, the potential of cogeneration in the industrial sector (including hospitals, hotels, shopping malls and two airports) is approximately 600 MW el , which is around 7% of the total electricity generation installed capacity in the country. If this cogeneration potential is implemented, the energy efficiency in the Ecuadorian industry could be increased by 35-40%. This potential could save up to 18.6 × 106 L/month of oil-derived fuels, avoiding up to 576,800 tCO 2 /year, and creating more than 2600 direct jobs. Lack of NG for cogeneration is seen as a problem that needs to be addressed in the future to reduce the cost of electricity generation in cogeneration plants. The use of diesel and gas engines (the main types of prime movers in the conditions of the industry in Ecuador) presents opportunities to easily move from fossil-derived fuels to renewable fuels, i.e., to use biodiesel and biogas in substitution of diesel and NG, respectively. This topic deserves further analysis, especially in identifying options for producing biofuels. Further studies should also address the logistics of integration of cogeneration with other electricity generation sources such as hydropower, or the logistics of biomass for cogeneration, to mention two aspects. Distributed generation through cogeneration offers opportunities to diversify local (small scale) electricity generation to optimize the use of the national grid and offset one of the Figure A1. Schematic of cogeneration system based on Rankine cycle for companies that can use biomass as fuel (e.g., sugarcane, pulp and paper, oil palm industries) and back pressure steam turbines. Adapted from [18,19,23,28,32,49,51] Figure A1. Schematic of cogeneration system based on Rankine cycle for companies that can use biomass as fuel (e.g., sugarcane, pulp and paper, oil palm industries) and back pressure steam turbines. Adapted from [18,19,23,28,32,49,51].    Figure A3. Schematic of a proposed trigeneration system using gas engines or diesel engines for the beverage industry, dairy industry, and food industry.  Figure A4. Schematic of a proposed trigeneration system using gas engines or diesel engines for service industries (e.g., hotels, hospitals). Adapted from [46]. Figure A5. Schematic of a proposed bottom cogeneration system used in cement industry. HRSGheat recovery steam generator (adapted from [55,101] Figure A4. Schematic of a proposed trigeneration system using gas engines or diesel engines for service industries (e.g., hotels, hospitals). Adapted from [46].
Energies 2020, 13, x FOR PEER REVIEW 23 of 30 Figure A4. Schematic of a proposed trigeneration system using gas engines or diesel engines for service industries (e.g., hotels, hospitals). Adapted from [46]. Figure A5. Schematic of a proposed bottom cogeneration system used in cement industry. HRSGheat recovery steam generator (adapted from [55,101] Figure A5. Schematic of a proposed bottom cogeneration system used in cement industry. HRSG-heat recovery steam generator (adapted from [55,101]).