# Investigating the potential of residential district energy.

INTRODUCTION

The International District Energy Association, in its website [1], states "District energy systems produce steam, hot water or chilled water at a central plant. The steam, hot water or chilled water is then piped underground to individual buildings for space heating, domestic hot water heating and air conditioning." This type of system has been implemented in the U.S.A. in different applications. Based on 2009 data, the 837 systems reported in the U.S. are distributed as Colleges and Universities 400, Community Utilities 119, Healthcare Installations 251, Military/Gov Installations 41, Airports 10, Industrial 13, other 3 [2]. The community utilities refer to downtown utilities.

District energy systems embrace efficient and renewable energy technologies, including combined heat and power, solar hot water, biomass, and thermal storage, among others. These technologies make possible to achieve benefits such as improved energy efficiency, fuel flexibility, and reliability. However, this study is limited to the consideration of the use of an efficient electric chiller with variable speed and a boiler using natural gas to provide chilled water and hot water, respectively.

As the authors' first approach to investigate the potential of district energy in residential buildings, a simplified system and analysis was developed. For the economic analysis, the Internal Rate of Return (IRR) is used to investigate the economic feasibility of the system comprised of a central utility plant and piping distribution system. The plant consists of an electrical chiller and a boiler to supply chilled water and hot water, with their ancillary equipment.

APPROACH

A reference home is used to estimate energy consumption and equipment capital cost. Although it is clear that identical homes will not have the same energy consumption due to different schedules of occupancy and internal loads, as well as preference on thermostat setpoints among other variables, in this study the hourly loads and energy consumption from the reference home is used independently of the number of homes. However, a factor of simultaneity of 0.8 was used in the simulations for the district energy system.

Reference Home

The systems capacities and energy consumption are obtained using a reference hypothetical home (RHH). The software BEopt [3] version 2.0.0 was used to develop a model to obtaining the electricity and natural gas consumption, as well as the cooling and heating loads. The simulation of the district energy system was performed using code developed for this research, and it is described in Section "District Energy Model."

This section presents parameters describing the RHH. Any parameters needed by the software that are not described in this section match those of the Building America Benchmark (B10 Benchmark) which defines a new construction (2010) building, based on the 2009 IECC code. More information on the B10 Benchmark can be found on Ref. [4]. Some specifications used for the development of the model and perform simulations are:

* Slab dimensions: 15 m (50 ft) by 15 m (50 ft); with 3 bedrooms and 2 bathrooms (front faces North)

* Garage area: 6.1 m (20 ft) by 6.1 m (20 ft) (located on the front left corner)

* Wall Height: 3 m (10 ft)

* Attic: unfinished attic over the house and garage

* Roof Type and Pitch: Hip and 10:12

* Location is Tyler, TX (weather file: USA_TX_Tyler-Pounds.Field.722448)

* Air conditioning COP 3 and furnace efficiency 80%

The cost of the HVAC system of the RHH is estimated to be $8,000 (indoor unit $2,500, outdoor unit $2,500, and ductwork $3,000). On the other hand, the homes connected to the district energy system differ only from the RHH on the substitution of the indoor and outdoor units by an air handler unit. The ducts of the air distribution system are identical.

Neighborhood layout

Preliminary analysis indicates that the piping system is a major factor affecting the economics of the district energy system. In this study it was assumed that the homes in a new neighborhood will have the layout shown in Figure 1. For the analysis, the number of homes increases from the central utility plant (CUP) through the main line along the street. As it can be noticed, sections of pipes with 15 m (50 ft) length are added for each two homes.

Due to the large different diameters that can be found for analysis of large number of homes, to compute the capital cost and pumping energy consumption, an average normalized diameter was used for the main pipe line along the street.

District Energy System

The district energy system is composed of an electric chiller, cooling tower, boiler, and pumps located on the building of the CUP, and the distribution system (piping system) supplying chilled and hot water to the air handler units (AHU). Figure 2 illustrates a schematic of the systems.

Cooling system

Chilled water is produced by an electrical centrifugal chiller, which includes a variable speed drive (VSD). The nominal efficiency of the chiller is estimated to be 0.45 kW (0.60 hp) per ton of refrigeration (TR). The temperature difference for the chilled water is set to be 8.9[degrees]C (~16[degrees]F). The flow rate is proportional to the number of houses. Heat transfer losses along the pipes are neglected in this study since the pipe is buried underground. The chiller uses a cooling tower with variable speed fan to keep the chiller cooling water at a constant inlet temperature.

Heating system

Hot water is produced using a boiler. The boiler fuel efficiency is estimated to be 85%. The design temperature difference for the hot water is set to be 14.0[degrees]C (~60[degrees]F). The flow rate is proportional to the number of houses.

Air Handler Unit

Each home has its own 4 pipes AHU connected to the chilled water loop, hot water loop, and the air distribution. Chilled water needed to satisfy the cooling demand is controlled by a 2 way- valve with temperature setpoint of 12.8[degrees]C (55[degrees]F) (6.5[degrees]C/11.7[degrees]F above the constant chilled water inlet temperature). Hot water needed to satisfy the heating demand is controlled by a 2 way-valve with temperature setpoint of 82.2[degrees]C (180[degrees]F) (13.9[degrees]C/25.0[degrees]F above the constant hot water inlet temperature).

Hydronic system

The hydronic system consists of the pumps located in the CUP and the piping system. Pumps are assumed to be variable speed to manage the variable flow rates. Pump efficiency is assumed to be constant at 0.756 (0.84 pump efficiency and 0.9 motor efficiency). The pipes are steel with an estimated roughness of 152 pm (0.006 in). For residential application water velocities between 3 and 5 m/s (10 ft/s and 17 ft/s) are allowable for acceptable noise levels, but to minimize erosion in systems operating more than 6000 hr/yr a velocity of 3 m/s (10 ft/s) is recommended. Therefore, a design velocity of 3 m/s (10 ft/s) was set for piping system computations.

DISTRICT ENERGY MODEL

Cooling system

The chiller hourly cooling demand ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is computed based on the home's hourly cooling demand ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) and the number of homes (N) connected to the system

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

where (t) corresponds to the hour of analysis based on the 8,760 hours of the year.

The chiller nominal capacity (C[H.sub.cap]) is defined as the maximum cooling demand based on the number of homes

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)

The chiller hourly part-load efficiency ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is computed based on the chiller nominal efficiency ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) and the hourly part-load-factor ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

The hourly part-load-factor is estimated based on the hourly partial-load-ratio ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) and the efficiency curve shown in Figure 3, which is adapted from Ref. [5] for a condenser- water temperature of 18.3[degrees]C (65[degrees]F). Since the chiller capacity is defined based on the maximum cooling demand, which may occur only during a few hours during the year, it is expected that the chiller will operate most of the time at a part-load ratio lower than 0.8.

The hourly partial-load-ratio is computed as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)

The hourly chiller electricity consumption ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is computed as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)

The hourly cooling tower electricity consumption ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]), due to condensing pumps and fans, is assumed to be a fraction ([f.sub.ct]) of chiller electricity consumption, 10% in this study,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (6)

Heating system

The hourly thermal energy needed from the boiler ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is proportional to heating demand ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) and affected by the boiler fuel efficiency ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII])

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (7)

Hydronic system

The hourly volumetric flow rate ([??]) at each house is computed based on the cooling or heating demand and the temperature chance across the cooling ([DELTA][T.sub.C]) or heating ([DELTA][T.sub.h]) coils as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (8)

The volumetric flow rate is increased for the street lines based on the number of homes been fed by the street line, until the section going into the plant for which the flow rate becomes N[??].

The distribution pump power ([P.sub.p]) is computed as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (9)

where [gamma] is the water specific weight (assumed constant at 1000 kgf/[m.sup.3] (62.4 lb/[ft.sup.3])), and H ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is the total head loss for the cooling or heating loops.

The total head loss is computed as the sum of all head losses on the loop

H = [H.sub.f] + [H.sub.fit] + [H.sub.coil] + [H.sub.eq] (10)

where [H.sub.f] is the head loss due to friction, [H.sub.fit] is the head loss due to fittings (30% of [H.sub.f]), [H.sub.coil] is the head loss at the AHU cooling or heating coil (4.6 m(15 ft)), and [H.sub.eq] is the head loss at the equipment used to produce the chilled water (chiller) or hot water (boiler). For this study, [H.sub.eq] is assumed 15.2 m (50 ft) at the chiller and boiler.

The hourly energy consumption from the pumps is computed accounting for the pump efficiency

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (11)

Grid electric power

The annual electricity purchased by the CUP to operate all equipment is estimated as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (12)

Fuel consumption

In this study, natural gas (NG) is used as fuel with a heating value (HV) of 10.27 kWh/[m.sup.3] (1027 Btu/[ft.sup.3]). Therefore, the annual fuel consumption of NG is estimated as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (13)

ECONOMIC ANALYSIS

This section presents parameters and criteria used to perform the economic analysis based on the Internal Rate of Return (IRR). When no information is given on the source of the information presented, it means that the parameters and criteria are based on authors' experience.

The IRR is found by varying the discount rate used to compute the NPV of the difference of positive and negative cash flow between the standard (decentralized) system and the district energy system. When the discount rate makes the NPV of the project equal to zero for the period of analysis, the IRR has been found. If the IRR is equal or greater than the Minimum Rate of Return (MRR) the project is attractive.

For this study, the period of analysis is 30 years. On the other hand, using the ASHRAE recommended service life of systems components, a replacement of the indoor and outdoor units for the standard system is done at year 15 and 30, and a replacement of CUP equipment and home air handler unit is done at year 20 for the district energy system. All replacement of equipment is done by increasing the capital cost with inflation. To compute the NPV, the inflation and fuels (electricity and natural gas) escalation rates were set equal to 3%.

Home System

A preliminary analysis allowed to consider that the total installed cost of the standard HVAC system in a single home can be compared with the cost of the installed system (AHU and pipes) needed by the district energy system at each home. Therefore, their costs are equivalent and consequently not necessary to be included to estimate the net present value (NPV). For the economic analysis this means that the initial capital cost per home of the district energy system is associated only to the central utility plant and piping system at the home's door.

Central Utility Plant

Since the size of the equipment will vary with the number of homes served by the CUP, equations of cost as a function of capacity were used. The equations were obtained from curve fits of data from a specialized manual of costs [6].

The cost of pipes varies with diameter and is given per linear meter of 4 steel pipes, 2 for chilled water and 2 for hot water. The costs include trench to bury the pipes, as well as the cost of fittings and welding the pipes. The costs are estimated based on the assumption that the neighborhood is a new development. On the other hand, the cost of the pipes of the main line along the street is estimated using the average normalized diameter as mentioned previously.

The cost of the building of the CUP is estimated based on capacity of chiller (includes the cooling tower) and boiler as

$ = [[(0.044[[m.sup.2]/kW]Cap).sub.chiller] + [(0.038[[m.sup.2]/kW]Cap).sub.boiler]] x 1,615 [$/[m.sup.2]] (18a-SI)

$ = [[(1.65[[ft.sup.2]/ton]Cap).sub.chiller] + [(0.12 [[ft.sup.2]/MBH]).sub.boiler]] x 150 [$/[ft.sup.2]] (18b-IP)

Rates

* The electric rate for the reference home is assumed at $0.1117/kWh [7], which corresponds to the average residential retail price for Texas during September 2012.

* The natural gas rate for the reference home is assumed at $436/1000[m.sup.3] ($12.36/1000[ft.sup.3]) [8], which corresponds to 2012 annual average for residential consumers in Texas [10].

* The electric rate for the CUP is assumed at $0.0817/kWh [7], which corresponds to the average commercial retail price for Texas during September 2012.

* The natural gas rate for the CUP is assumed at $233/1000[m.sup.3] ($6.59/1000[ft.sup.3]) [9], which corresponds to 2012 annual average for commercial consumers in Texas.

RESULTS AND DISCUSSION

From Figures 4 and 5, it can be noticed that the number of homes in the abscissa starts at 200 homes, which is the estimated number homes that will give enough cooling demand to justify an available chiller with the characteristics considered in this study. Preliminary results showed that the cost of the 4 pipes of the main line to which the homes are connected involves a large cost that makes difficult the district energy system to compete with the standard (decentralized) HVAC system from an economic point of view. To illustrate this, Figure 4 shows the capital cost associated to the CUP, the piping system, and the district energy system as a function of the number of homes. As expected from the equations of costs for the CUP, capital cost per home decreases as the number of homes increases with a trend towards an asymptotic value. The capital cost of the piping system is more than double of the CUP cost and increases with a ladder steps trend. This trend is defined by the average normalized diameter used to estimate the cost of the pipes of the main line along the street. Since the water design velocity is 3 m/s (10 ft/s) for the main line pipes along the street, the normalized diameter of the pipes and consequently the piping cost will be constant until a new diameter is needed to avoid exceeding the design water velocity. Since the variation of the CUP cost as a function of homes is negligible when compared to the piping system, the district energy system follows the trend of the piping system cost.

Figure 5 shows the IRR for the district energy project as a function of the number of homes. As the number of homes increases, the IRR decreases, making the project less economically attractive for the homeowner. The shape of the curve is explained by the average normalized approach used to dimension the pipes for the main line along the street. Large variations of IRR occur when a change on diameter is needed to avoid exceeding the maximum design velocity. Simulations were done in increments of 100 homes, and Figure 5 shows that for more than 300 homes the district energy system may be considered economically unfavorable because the IRR start to be lower than the inflation rate (3%) which could be the minimum rate of return (MRR)expected by a homeowner.

As mentioned and illustrated by the results, the capital cost of the piping system is the key factor defining the economic feasibility of a residential district energy system. Therefore, options to reduce the cost should be considered at the design stage. The following are some options that could be considered:

* Installation of pipes when other services are installed to distribute the cost of burying the pipes.

* Implementation of the district energy system in multifamily buildings instead of single homes.

* Use of a two pipes distribution system instead of a four pipes system. This option may reduce the cost of piping by half, but decreases the thermal comfort.

* The use of PVC or other materials instead of steel.

* Optimization of the neighborhood layout and piping system.

CONCLUSIONS

This paper presented an approach to investigate the economic feasibility of residential district energy. A hypothetical home and neighborhood layout were used to generate results from simulations of the model developed. The results show that the system may not be economically attractive for a number of homes greater than 300, and that the high capital cost of the piping system used to distribute the chilled and hot water to the air handler units located in each home is the major constrain to make district energy systems economically feasible. Although not considered in this study, the feasibility of residential energy system can be improved by considering socioeconomic benefits such as emission reduction. In this paper renewable energy and thermal storage energy systems were not considered. However, these systems should be included when possible to justify the district energy systems because they further increase the efficiency and/or environmental benefits.

REFERENCES

[1] International District Energy Association, http://www.districtenergy.org/

[2] U.S. District Energy Systems, International District Energy Association, http://www.districtenergy.org/u-s-districtenergy-systems-map

[3] Building Energy Optimization (BEopt) software, National Renewable Energy Laboratory Available at: http://beopt.nrel.gov/

[4] Robert Hendron and Cheryn Engebrecht (2010). "Building America House Simulation Protocols." National Renewable Energy Laboratory, Building Technology Program. Available at: http://www.nrel.gov/docs/fy11osti/49246.pdf

[5] Kevin M. Kuretich (May 7, 2010). "Large-Campus District Cooling." Online article found on HPAC Engineering (http://hpac.com/air-conditioning/large-campus-district-cooling-0510). Consulted on December 2013.

[6] Mechanical Cost Data 2009, RSMeans, 32nd Edition.

[7] Average Retail Price of Electricity to Ultimate Customers by End-Use Sector, Energy Information Agency, Department of Energy. Available at: http://www.eia.gov/electricity/monthly/epm_table_grapher.cfm?t=epmt_5_6_a

[8] Texas Price of Natural gas Delivered to Residential Consumers, Energy Information Agency, Department of Energy. Available at: http://www.eia.gov/dnav/ng/hist/n3010tx3m.htm

[9] Texas Price of Natural gas Delivered to Residential Consumers, Energy Information Agency, Department of Energy. Available at: http://www.eia.gov/dnav/ng/hist/n3020tx3m.htm

Nelson Fumo

ASHRAE Member

Vicente Bortone

ASHRAE Member

Juan Zambrano

ASHRAE Member

Aleyani Zambrano

N. Fumo is an associate professor in the Department of Mechanical Engineering, The University of Texas at Tyler, Tyler, Texas, USA. V. Bortone is a Project Development Consultant, Johnson Controls Inc., Lenexa, Kansas, USA. J. Zambrano and A. Zambrano are associate professors, Universidad Nacional Experimental del Tachira, San Cristobal, Tachira, Venezuela.

The International District Energy Association, in its website [1], states "District energy systems produce steam, hot water or chilled water at a central plant. The steam, hot water or chilled water is then piped underground to individual buildings for space heating, domestic hot water heating and air conditioning." This type of system has been implemented in the U.S.A. in different applications. Based on 2009 data, the 837 systems reported in the U.S. are distributed as Colleges and Universities 400, Community Utilities 119, Healthcare Installations 251, Military/Gov Installations 41, Airports 10, Industrial 13, other 3 [2]. The community utilities refer to downtown utilities.

District energy systems embrace efficient and renewable energy technologies, including combined heat and power, solar hot water, biomass, and thermal storage, among others. These technologies make possible to achieve benefits such as improved energy efficiency, fuel flexibility, and reliability. However, this study is limited to the consideration of the use of an efficient electric chiller with variable speed and a boiler using natural gas to provide chilled water and hot water, respectively.

As the authors' first approach to investigate the potential of district energy in residential buildings, a simplified system and analysis was developed. For the economic analysis, the Internal Rate of Return (IRR) is used to investigate the economic feasibility of the system comprised of a central utility plant and piping distribution system. The plant consists of an electrical chiller and a boiler to supply chilled water and hot water, with their ancillary equipment.

APPROACH

A reference home is used to estimate energy consumption and equipment capital cost. Although it is clear that identical homes will not have the same energy consumption due to different schedules of occupancy and internal loads, as well as preference on thermostat setpoints among other variables, in this study the hourly loads and energy consumption from the reference home is used independently of the number of homes. However, a factor of simultaneity of 0.8 was used in the simulations for the district energy system.

Reference Home

The systems capacities and energy consumption are obtained using a reference hypothetical home (RHH). The software BEopt [3] version 2.0.0 was used to develop a model to obtaining the electricity and natural gas consumption, as well as the cooling and heating loads. The simulation of the district energy system was performed using code developed for this research, and it is described in Section "District Energy Model."

This section presents parameters describing the RHH. Any parameters needed by the software that are not described in this section match those of the Building America Benchmark (B10 Benchmark) which defines a new construction (2010) building, based on the 2009 IECC code. More information on the B10 Benchmark can be found on Ref. [4]. Some specifications used for the development of the model and perform simulations are:

* Slab dimensions: 15 m (50 ft) by 15 m (50 ft); with 3 bedrooms and 2 bathrooms (front faces North)

* Garage area: 6.1 m (20 ft) by 6.1 m (20 ft) (located on the front left corner)

* Wall Height: 3 m (10 ft)

* Attic: unfinished attic over the house and garage

* Roof Type and Pitch: Hip and 10:12

* Location is Tyler, TX (weather file: USA_TX_Tyler-Pounds.Field.722448)

* Air conditioning COP 3 and furnace efficiency 80%

The cost of the HVAC system of the RHH is estimated to be $8,000 (indoor unit $2,500, outdoor unit $2,500, and ductwork $3,000). On the other hand, the homes connected to the district energy system differ only from the RHH on the substitution of the indoor and outdoor units by an air handler unit. The ducts of the air distribution system are identical.

Neighborhood layout

Preliminary analysis indicates that the piping system is a major factor affecting the economics of the district energy system. In this study it was assumed that the homes in a new neighborhood will have the layout shown in Figure 1. For the analysis, the number of homes increases from the central utility plant (CUP) through the main line along the street. As it can be noticed, sections of pipes with 15 m (50 ft) length are added for each two homes.

Due to the large different diameters that can be found for analysis of large number of homes, to compute the capital cost and pumping energy consumption, an average normalized diameter was used for the main pipe line along the street.

District Energy System

The district energy system is composed of an electric chiller, cooling tower, boiler, and pumps located on the building of the CUP, and the distribution system (piping system) supplying chilled and hot water to the air handler units (AHU). Figure 2 illustrates a schematic of the systems.

Cooling system

Chilled water is produced by an electrical centrifugal chiller, which includes a variable speed drive (VSD). The nominal efficiency of the chiller is estimated to be 0.45 kW (0.60 hp) per ton of refrigeration (TR). The temperature difference for the chilled water is set to be 8.9[degrees]C (~16[degrees]F). The flow rate is proportional to the number of houses. Heat transfer losses along the pipes are neglected in this study since the pipe is buried underground. The chiller uses a cooling tower with variable speed fan to keep the chiller cooling water at a constant inlet temperature.

Heating system

Hot water is produced using a boiler. The boiler fuel efficiency is estimated to be 85%. The design temperature difference for the hot water is set to be 14.0[degrees]C (~60[degrees]F). The flow rate is proportional to the number of houses.

Air Handler Unit

Each home has its own 4 pipes AHU connected to the chilled water loop, hot water loop, and the air distribution. Chilled water needed to satisfy the cooling demand is controlled by a 2 way- valve with temperature setpoint of 12.8[degrees]C (55[degrees]F) (6.5[degrees]C/11.7[degrees]F above the constant chilled water inlet temperature). Hot water needed to satisfy the heating demand is controlled by a 2 way-valve with temperature setpoint of 82.2[degrees]C (180[degrees]F) (13.9[degrees]C/25.0[degrees]F above the constant hot water inlet temperature).

Hydronic system

The hydronic system consists of the pumps located in the CUP and the piping system. Pumps are assumed to be variable speed to manage the variable flow rates. Pump efficiency is assumed to be constant at 0.756 (0.84 pump efficiency and 0.9 motor efficiency). The pipes are steel with an estimated roughness of 152 pm (0.006 in). For residential application water velocities between 3 and 5 m/s (10 ft/s and 17 ft/s) are allowable for acceptable noise levels, but to minimize erosion in systems operating more than 6000 hr/yr a velocity of 3 m/s (10 ft/s) is recommended. Therefore, a design velocity of 3 m/s (10 ft/s) was set for piping system computations.

DISTRICT ENERGY MODEL

Cooling system

The chiller hourly cooling demand ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is computed based on the home's hourly cooling demand ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) and the number of homes (N) connected to the system

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

where (t) corresponds to the hour of analysis based on the 8,760 hours of the year.

The chiller nominal capacity (C[H.sub.cap]) is defined as the maximum cooling demand based on the number of homes

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)

The chiller hourly part-load efficiency ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is computed based on the chiller nominal efficiency ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) and the hourly part-load-factor ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

The hourly part-load-factor is estimated based on the hourly partial-load-ratio ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) and the efficiency curve shown in Figure 3, which is adapted from Ref. [5] for a condenser- water temperature of 18.3[degrees]C (65[degrees]F). Since the chiller capacity is defined based on the maximum cooling demand, which may occur only during a few hours during the year, it is expected that the chiller will operate most of the time at a part-load ratio lower than 0.8.

The hourly partial-load-ratio is computed as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)

The hourly chiller electricity consumption ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is computed as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)

The hourly cooling tower electricity consumption ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]), due to condensing pumps and fans, is assumed to be a fraction ([f.sub.ct]) of chiller electricity consumption, 10% in this study,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (6)

Heating system

The hourly thermal energy needed from the boiler ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is proportional to heating demand ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) and affected by the boiler fuel efficiency ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII])

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (7)

Hydronic system

The hourly volumetric flow rate ([??]) at each house is computed based on the cooling or heating demand and the temperature chance across the cooling ([DELTA][T.sub.C]) or heating ([DELTA][T.sub.h]) coils as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (8)

The volumetric flow rate is increased for the street lines based on the number of homes been fed by the street line, until the section going into the plant for which the flow rate becomes N[??].

The distribution pump power ([P.sub.p]) is computed as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (9)

where [gamma] is the water specific weight (assumed constant at 1000 kgf/[m.sup.3] (62.4 lb/[ft.sup.3])), and H ([MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]) is the total head loss for the cooling or heating loops.

The total head loss is computed as the sum of all head losses on the loop

H = [H.sub.f] + [H.sub.fit] + [H.sub.coil] + [H.sub.eq] (10)

where [H.sub.f] is the head loss due to friction, [H.sub.fit] is the head loss due to fittings (30% of [H.sub.f]), [H.sub.coil] is the head loss at the AHU cooling or heating coil (4.6 m(15 ft)), and [H.sub.eq] is the head loss at the equipment used to produce the chilled water (chiller) or hot water (boiler). For this study, [H.sub.eq] is assumed 15.2 m (50 ft) at the chiller and boiler.

The hourly energy consumption from the pumps is computed accounting for the pump efficiency

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (11)

Grid electric power

The annual electricity purchased by the CUP to operate all equipment is estimated as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (12)

Fuel consumption

In this study, natural gas (NG) is used as fuel with a heating value (HV) of 10.27 kWh/[m.sup.3] (1027 Btu/[ft.sup.3]). Therefore, the annual fuel consumption of NG is estimated as

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (13)

ECONOMIC ANALYSIS

This section presents parameters and criteria used to perform the economic analysis based on the Internal Rate of Return (IRR). When no information is given on the source of the information presented, it means that the parameters and criteria are based on authors' experience.

The IRR is found by varying the discount rate used to compute the NPV of the difference of positive and negative cash flow between the standard (decentralized) system and the district energy system. When the discount rate makes the NPV of the project equal to zero for the period of analysis, the IRR has been found. If the IRR is equal or greater than the Minimum Rate of Return (MRR) the project is attractive.

For this study, the period of analysis is 30 years. On the other hand, using the ASHRAE recommended service life of systems components, a replacement of the indoor and outdoor units for the standard system is done at year 15 and 30, and a replacement of CUP equipment and home air handler unit is done at year 20 for the district energy system. All replacement of equipment is done by increasing the capital cost with inflation. To compute the NPV, the inflation and fuels (electricity and natural gas) escalation rates were set equal to 3%.

Home System

A preliminary analysis allowed to consider that the total installed cost of the standard HVAC system in a single home can be compared with the cost of the installed system (AHU and pipes) needed by the district energy system at each home. Therefore, their costs are equivalent and consequently not necessary to be included to estimate the net present value (NPV). For the economic analysis this means that the initial capital cost per home of the district energy system is associated only to the central utility plant and piping system at the home's door.

Central Utility Plant

Since the size of the equipment will vary with the number of homes served by the CUP, equations of cost as a function of capacity were used. The equations were obtained from curve fits of data from a specialized manual of costs [6].

* Chiller (capacity, Cap, in kW of refrigeration): $ = Cap[767[(Cap).sup.-0.19]] (14a-SI) (capacity, Cap, in tons of refrigeration): $ = Cap[2124[(Cap).sup.-0.19]] (14b-IP) * Cooling tower (capacity, Cap, in kW of refrigeration): $ = Cap[1140[(Cap).sup.-0.358]] (15a-SI) (capacity, Cap, in ton of refrigeration): $ = Cap[2556[(Cap).sup.-0.358]] (15b-IP) * Boiler (capacity, Cap, in kW): $ = Cap[1457[(Cap).sup.-0.373]] (16a-SI) (capacity, Cap, in 1000 Btu per hour): $ = Cap[675[(Cap).sup.-0.373]] (17b-SI) * Pumps (capacity, Cap, in kW): $ = Cap[10455[(Cap).sup.-0.561]] (17a-IP) (capacity, Cap, in hp): $ = Cap[9193[(Cap).sup.-0.561]] (17a-IP)

The cost of pipes varies with diameter and is given per linear meter of 4 steel pipes, 2 for chilled water and 2 for hot water. The costs include trench to bury the pipes, as well as the cost of fittings and welding the pipes. The costs are estimated based on the assumption that the neighborhood is a new development. On the other hand, the cost of the pipes of the main line along the street is estimated using the average normalized diameter as mentioned previously.

The cost of the building of the CUP is estimated based on capacity of chiller (includes the cooling tower) and boiler as

$ = [[(0.044[[m.sup.2]/kW]Cap).sub.chiller] + [(0.038[[m.sup.2]/kW]Cap).sub.boiler]] x 1,615 [$/[m.sup.2]] (18a-SI)

$ = [[(1.65[[ft.sup.2]/ton]Cap).sub.chiller] + [(0.12 [[ft.sup.2]/MBH]).sub.boiler]] x 150 [$/[ft.sup.2]] (18b-IP)

Rates

* The electric rate for the reference home is assumed at $0.1117/kWh [7], which corresponds to the average residential retail price for Texas during September 2012.

* The natural gas rate for the reference home is assumed at $436/1000[m.sup.3] ($12.36/1000[ft.sup.3]) [8], which corresponds to 2012 annual average for residential consumers in Texas [10].

* The electric rate for the CUP is assumed at $0.0817/kWh [7], which corresponds to the average commercial retail price for Texas during September 2012.

* The natural gas rate for the CUP is assumed at $233/1000[m.sup.3] ($6.59/1000[ft.sup.3]) [9], which corresponds to 2012 annual average for commercial consumers in Texas.

RESULTS AND DISCUSSION

From Figures 4 and 5, it can be noticed that the number of homes in the abscissa starts at 200 homes, which is the estimated number homes that will give enough cooling demand to justify an available chiller with the characteristics considered in this study. Preliminary results showed that the cost of the 4 pipes of the main line to which the homes are connected involves a large cost that makes difficult the district energy system to compete with the standard (decentralized) HVAC system from an economic point of view. To illustrate this, Figure 4 shows the capital cost associated to the CUP, the piping system, and the district energy system as a function of the number of homes. As expected from the equations of costs for the CUP, capital cost per home decreases as the number of homes increases with a trend towards an asymptotic value. The capital cost of the piping system is more than double of the CUP cost and increases with a ladder steps trend. This trend is defined by the average normalized diameter used to estimate the cost of the pipes of the main line along the street. Since the water design velocity is 3 m/s (10 ft/s) for the main line pipes along the street, the normalized diameter of the pipes and consequently the piping cost will be constant until a new diameter is needed to avoid exceeding the design water velocity. Since the variation of the CUP cost as a function of homes is negligible when compared to the piping system, the district energy system follows the trend of the piping system cost.

Figure 5 shows the IRR for the district energy project as a function of the number of homes. As the number of homes increases, the IRR decreases, making the project less economically attractive for the homeowner. The shape of the curve is explained by the average normalized approach used to dimension the pipes for the main line along the street. Large variations of IRR occur when a change on diameter is needed to avoid exceeding the maximum design velocity. Simulations were done in increments of 100 homes, and Figure 5 shows that for more than 300 homes the district energy system may be considered economically unfavorable because the IRR start to be lower than the inflation rate (3%) which could be the minimum rate of return (MRR)expected by a homeowner.

As mentioned and illustrated by the results, the capital cost of the piping system is the key factor defining the economic feasibility of a residential district energy system. Therefore, options to reduce the cost should be considered at the design stage. The following are some options that could be considered:

* Installation of pipes when other services are installed to distribute the cost of burying the pipes.

* Implementation of the district energy system in multifamily buildings instead of single homes.

* Use of a two pipes distribution system instead of a four pipes system. This option may reduce the cost of piping by half, but decreases the thermal comfort.

* The use of PVC or other materials instead of steel.

* Optimization of the neighborhood layout and piping system.

CONCLUSIONS

This paper presented an approach to investigate the economic feasibility of residential district energy. A hypothetical home and neighborhood layout were used to generate results from simulations of the model developed. The results show that the system may not be economically attractive for a number of homes greater than 300, and that the high capital cost of the piping system used to distribute the chilled and hot water to the air handler units located in each home is the major constrain to make district energy systems economically feasible. Although not considered in this study, the feasibility of residential energy system can be improved by considering socioeconomic benefits such as emission reduction. In this paper renewable energy and thermal storage energy systems were not considered. However, these systems should be included when possible to justify the district energy systems because they further increase the efficiency and/or environmental benefits.

NOMENCLATURE Cap = capacity El = electricity consumption F = fuel consumption H = head loss N = demand (number of homes) P = power Q = demand, thermal capacity T = temperature V = volumetric flow rate Subscript b = boiler c = cooling ch = chiller h = heating n = nominal p = pump

REFERENCES

[1] International District Energy Association, http://www.districtenergy.org/

[2] U.S. District Energy Systems, International District Energy Association, http://www.districtenergy.org/u-s-districtenergy-systems-map

[3] Building Energy Optimization (BEopt) software, National Renewable Energy Laboratory Available at: http://beopt.nrel.gov/

[4] Robert Hendron and Cheryn Engebrecht (2010). "Building America House Simulation Protocols." National Renewable Energy Laboratory, Building Technology Program. Available at: http://www.nrel.gov/docs/fy11osti/49246.pdf

[5] Kevin M. Kuretich (May 7, 2010). "Large-Campus District Cooling." Online article found on HPAC Engineering (http://hpac.com/air-conditioning/large-campus-district-cooling-0510). Consulted on December 2013.

[6] Mechanical Cost Data 2009, RSMeans, 32nd Edition.

[7] Average Retail Price of Electricity to Ultimate Customers by End-Use Sector, Energy Information Agency, Department of Energy. Available at: http://www.eia.gov/electricity/monthly/epm_table_grapher.cfm?t=epmt_5_6_a

[8] Texas Price of Natural gas Delivered to Residential Consumers, Energy Information Agency, Department of Energy. Available at: http://www.eia.gov/dnav/ng/hist/n3010tx3m.htm

[9] Texas Price of Natural gas Delivered to Residential Consumers, Energy Information Agency, Department of Energy. Available at: http://www.eia.gov/dnav/ng/hist/n3020tx3m.htm

Nelson Fumo

ASHRAE Member

Vicente Bortone

ASHRAE Member

Juan Zambrano

ASHRAE Member

Aleyani Zambrano

N. Fumo is an associate professor in the Department of Mechanical Engineering, The University of Texas at Tyler, Tyler, Texas, USA. V. Bortone is a Project Development Consultant, Johnson Controls Inc., Lenexa, Kansas, USA. J. Zambrano and A. Zambrano are associate professors, Universidad Nacional Experimental del Tachira, San Cristobal, Tachira, Venezuela.

Table 1. Piping costs Diameter (inches) Cost ($/m) Cost ($/ft) 2 686 209 4 1568 478 6 2453 748 8 3339 1018 10 4225 1288 12 5110 1558

Printer friendly Cite/link Email Feedback | |

Author: | Fumo, Nelson; Bortone, Vicente; Zambrano, Juan; Zambrano, Aleyani |
---|---|

Publication: | ASHRAE Transactions |

Article Type: | Report |

Geographic Code: | 1USA |

Date: | Jul 1, 2014 |

Words: | 3614 |

Previous Article: | Improving EER with off-peak radiant cooling. |

Next Article: | Polymeric hollow fiber heat exchangers liquid-to-gas application. |

Topics: |