Next Article in Journal
Evaluation Model of Specific Indoor Environment Overall Comfort Based on Effective-Function Method
Previous Article in Journal
Using Modularity to Reduce Complexity of Industrialized Building Systems for Mass Customization
 
 
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Multi-Objective Analysis of a CHP Plant Integrated Microgrid in Pakistan

Department of Electrical Engineering, Bahria University, 44000 Islamabad, Pakistan
*
Author to whom correspondence should be addressed.
Energies 2017, 10(10), 1625; https://doi.org/10.3390/en10101625
Received: 15 September 2017 / Revised: 12 October 2017 / Accepted: 13 October 2017 / Published: 17 October 2017
(This article belongs to the Section F: Electrical Engineering)

Abstract

:
In developing countries like Pakistan, the capacity shortage (CS) of electricity is a critical problem. The frequent natural gas (NG) outages compel consumers to use electricity to fulfill the thermal loads, which ends up as an increase in electrical load. In this scenario, the authors have proposed the concept of a combined heat & power (CHP) plant to be a better option for supplying both electrical and thermal loads simultaneously. A CHP plant-based microgrid comprising a PV array, diesel generators and batteries (operating in grid-connected as well as islanded modes) has been simulated using the HOMER Pro software. Different configurations of distributed generators (DGs) with/without batteries have been evaluated considering multiple objectives. The multiple objectives include the minimization of the total net present cost (TNPC), cost of generated energy (COE) and the annual greenhouse gas (GHG) emissions, as well as the maximization of annual waste heat recovery (WHR) of thermal units and annual grid sales (GS). These objectives are subject to the constraints of power balance, battery operation within state of charge (SOC) limits, generator operation within capacity limits and zero capacity shortage. The simulations have been performed on six cities including Islamabad, Lahore, Karachi, Peshawar, Quetta and Gilgit. The simulation results have been analyzed to find the most optimal city for the CHP plant integrated microgrid.

1. Introduction

During the winter season in Pakistan, natural gas (NG) is used as a primary source to supply the thermal load in the form of heating [1]. However, due to frequent outages of NG during the winter season, consumers tend to switch to electricity to fulfil their heating needs [1]. As there is an average capacity shortage (CS) of around 27% in electricity, the supply of thermal load by electricity puts an extra burden on the grid [2]. In such a situation, a microgrid along with a combined heat and power (CHP) plant represents a sensible solution to utilize the wasted heat and improving the efficiency to 75–88% [3,4,5,6,7,8]. A conventional thermal generation system has an efficiency of 25–35%, whereas the rest of energy is wasted in the form of unhealthy pollutant emissions [3,4,5,7]. A CHP plant recovers the wasted heat by using a waste heat recovery (WHR) unit, and therefore helps in controlling the greenhouse gas (GHG) emissions [5,9,10]. A microgrid with a CHP plant ensures the supply of electrical and thermal loads at the same time. Microgrids can range from small units for a single home to larger units for an entire community [4,5,6,7,11]. Moreover, a microgrid with a CHP plant can operate both in grid-connected as well as islanded modes [3,7,8]. Currently, CHP plants have broadly grabbed the attention in various countries and pilot projects are being undertaken in Europe, the U.S. and Japan [8,12].
So far, various researchers have investigated the operation of a CHP plant from certain perspectives. Guo et al. [3] have conducted research on an isolated hybrid CHP system consisting of PV/wind/gas turbine generator with vanadium redox flow battery (VRB) for the Qingshan Hu Campus of Hangzhou Dianzi University in China. With the implementation of a CHP system, the efficiency of the gas turbine was significantly improved from 29.5% to 82%. Ebara-Ballard et al. [4] have reported the steady state electrical efficiency of a combined fuel cell/CHP plant operating on NG to be 31%. For this unit, a net increase in fuel consumption and CO2 emissions was expected. However, 52% of the fuel energy was recollected in the form of heat, which has improved the energy efficiency up to 83%. Boljevic et al. [5,11] have analyzed the impact of CHP plant on thermal and electrical energy supply systems for small and medium sized enterprises. The authors have shown that the mentioned system has improved the overall efficiency of the system to around 77.6% and reduced the emissions to 57.8%. Ivanova et al. [6] have increased the efficiency of a CHP plant integrated with renewable energy sources up to 88%, by proposing a flexible operation algorithm. Bjelic et al. [12] have developed a microgrid with a CHP plant in HOMER to assess the lowest total net present costs (TNPC) under the variation of CO2 reduction constraint. Ren et al. [13] have evaluated the economic as well as environmental effects of distributed energy resources (DER) on the power system by using a multi-objective linear programming (MILP) technique. An eco-campus in Japan was selected for case study while considering PV, fuel cell and gas turbine for the satisfaction of both electrical and thermal loads. Hossain et al. [14] have improved the efficiency of a diesel generator by utilizing the waste heat of a 4-stroke 4-cylinder water cooled direct injection Hino W04D internal combustion engine (usually known as diesel engine coupled with 50 kVA generator-set considering ammonia and HFC-134a), and finally compared their results with water. Hopulele et al. [15] have worked in the field of combined cool, heat and power (CCHP) plant using genetic algorithm and used HOMER as an optimization tool. The system has fulfilled 90% of electrical load and 75% of thermal load. Surdu et al. [16] have developed an optimization tool which focuses on CHP employment in a competitive energy market context. The authors have minimized the total operating cost by solving the long-term unit commitment involving a CHP plant. Colson et al. [7] have evaluated the benefits of a hybrid solid oxide fuel cell (SOFC) with CHP plant for energy sustainability and emissions control. The hybrid system fulfils the electricity as well as hot water needs for a residential community of 500 homes, more sustainably with less environmental emissions as compared to conventional power plants. Chernyaev et al. [17] have developed a load distribution optimization tool for a CHP plant. The tool optimizes the fuel consumption by using a CHP power plant. Dvorak et al. [18] have scheduled the operation of a CHP plant by using the decomposition methods based on the heat demand, fuel cost and electricity pricing. Sekgoele et al. [8] have carried out the assessment of land filled gas-based CHP plants in South Africa, both technically and economically. The authors have assumed that the stand-alone CHP plant will supply both heat and power to remote communities, while the grid-connected CHP plant will work only during the peak load periods. Chandan et al. [19] have modelled and optimized a CCHP plant to fulfil the cooling, heating and power needs of the University of California, Irvine (UCI) by using cogeneration and thermal storage capabilities. The authors have minimized the operating cost of the plant by forecasting the electrical and thermal loads. Ruieneanu et al. [9] have conducted the parallel operation of a CHP plant with wind farms and have reduced the CO2 emissions, and therefore have reduced the operating cost of the system. Dai et al. [20] have proposed a new dispatch model for a CHP plant considering the heat transfer process. Boljevic [11,21] has developed a planning algorithm for optimal sizing of CHP plant connected to an urban distributed network with least costs under long term network planning policy. Pierre et al. [10] have technically and economically accessed a flexible CHP plant with carbon capturing and storage. Their work resulted in gaining higher profits by reducing CO2 emissions. Scholz et al. [22] have evaluated a system consisting of a CHP plant and a conventional gas-fired boiler with a power to heat unit. The authors have evaluated the benefits of the flexibility of power to heat unit to gain the economic incentives during low electricity price hours. Buoro et al. in [23] have evaluated a distributed energy supply system consisting of a CHP plant with PV and thermal storage by using mixed integer linear programming (MILP). Pareto fronts have been applied to the results to find the most optimized solution. Somma et al. in [24] have investigated a sustainable hybrid CHP-PV system to supply the both electrical and thermal loads. The results have shown that about 21–36% of the total annual costs were minimized with the optimized solution. Somma et al. [25] have developed a multi-objective optimization problem to reduce the energy costs and CO2 emissions. The authors have considered various thermal energy storage systems to fulfill a time-varying load profile. The results have indicated that a reduction of 27% in costs and 26% in CO2 emissions was achieved. Zhang et al. [26] have proposed a CHP plant integrated microgrid with energy storage to satisfy the electricity and heat demand. In order to minimize the computational burden, the authors have used a stochastic non-convex optimization which results in minimum operating cost. Ping et al. [27] have proposed a CHP plant dispatch model while considering thermal performance of pipe line and building’s inertia. The model is executed by decoupling the electricity and heat supply and in this way the wind penetration is increased. The overall result is saving of operational costs of system. Zidan et al. [28] have proposed a multi-objective optimization problem to minimize the overall costs and CO2 emissions simultaneously. A genetic algorithm (GA) was applied to find the optimal generation-mix among the CHP plants (with various properties), renewable sources and energy storages. Hussain et al. [29] have proposed a CCHP plant for building microgrids (BMGs) in grid-connected mode. A mixed integer linear programming (MILP) based optimization model to minimize the day to day operational cost has been developed. The cost has been reduced by the energy exchange with the external grid and heat exchange with the prosumer. Alarcon et al. [30] have performed a detailed review of the distributed energy resources integration in distribution networks. Various optimization techniques have been discussed to gain benefits like low operational costs, minimum CO2 emissions, reduction in network losses, enhancement in power quality etc. Moreover, the challenges of grid-integration like voltage regulation, frequency stability, adequacy, system reliability etc. have been investigated. Somma et al. [31] have developed a Pareto frontier based stochastic optimization technique for the daily scheduling of distributed energy resources to minimize the operating costs and CO2 emissions. A sensitivity analysis has been carried out to investigate the impact of high renewable penetration on the economic and environmental aspects. Maroufmashat et al. [32] have proposed a multi-objective optimization based on augmented epsilon constraint technique to minimize the operating costs and GHG emissions.
Most of the abovementioned research has been conducted to improve the efficiency of systems with a CHP plant with waste heat recovery (WHR). This WHR results in minimizing the operating costs. However, from microgrid perspective, there are other parameters like total net present costs (TNPC), cost of generated energy (COE), greenhouse gas (GHG) emissions, WHR and grid sales (GS), which also effect the efficiency of a CHP plant integrated microgrid. Therefore, in this paper the authors have analyzed a CHP plant integrated microgrid considering multiple objectives. The multiple objectives include the minimization of TNPC, COE, annual GHG emissions, and the maximization of the annual WHR and annual GS. Moreover, the CHP plant integrated microgrid has never been evaluated for Pakistan. This multi-objective optimization problem has been simulated for six cities of Pakistan including Islamabad, Lahore, Karachi, Peshawar, Quetta and Gilgit. Different configurations of distributed generators (DGs) like PV/diesel generators with/without batteries have been simulated in both grid-connected as well as isolated modes, to find the optimal configuration. The final optimal solution concludes the most optimum city. These cities are considered for evaluation because these are the most populous cities and provincial capitals [33].
The rest of the paper is organized as follows: Section 2 presents an existing urban community load profile (electrical and thermal), solar energy resource (average monthly solar irradiance) for above mentioned six cities, and temperature resource for same cities. Section 3 shows the microgrid modeling in the form of different DG configurations with/without batteries. Section 4 shows the analysis of results considering multiple objectives. Section 5 concludes the paper.

2. Data Collection

2.1. Load Profile

An urban community identical load profile (electrical and thermal) has been assumed for all the above-mentioned cities. In case of a non-thermal system (without WHR), the scaled annual electrical energy utilization is 13,331.07 kWh/d, peak electric load is 1427.06 kW and an average load is 555.46 kW, as shown in Figure 1. In case of a thermal system (with WHR), the scaled annual electrical energy utilization is 10,911.02 kWh/d, peak electric load is 1073.99 kW and the average load is 454.63 kW, as shown in Figure 2. An equivalent scaled annual thermal energy utilization is 2419.8 kWh/d, peak thermal load is 407.45 kW and an average load is 100.83 kW, as shown in Figure 3.

2.2. Solar Energy Resource

The 22-years (from July 1983 to June 2005) average monthly solar irradiance profiles of the Islamabad, Lahore, Karachi, Peshawar, Quetta and Gilgit located at 33°43.8’ N, 73°5.6’ E; 31°33.3’ N, 74°21.4’ E; 24°51.7’ N, 67°0.6’ E, 34°0.9’ N, 71°34.8’ E, 30°11.0’ N, 66°59.9’ E and 35°55.2’ N, 74°18.5’ E respectively, are taken from NASA (National Aeronautics and Space Administration) database [34]. According to the NASA database, as shown in Figure 4, the average monthly solar irradiance in Karachi is greater than other cities during the winter months (January to April, October to December), whereas Peshawar has its peak during the peak summer months of June and July. Quetta has high irradiance in the months of August and September, Lahore has a high irradiance in the month of May.

2.3. Temperature Resource

The same NASA database is utilized for the 22-year (July 1983–June 2005) average monthly temperature of earth’ surface for the above-mentioned cities, as shown in Figure 5. Karachi has greater average temperature for eight months (January–April, September-December), while Lahore has its peak for four months (May–August). On the other hand, Gilgit has the lowest peak during the entire year.

3. Microgrid Modelling

A microgrid introduces the concept of operating the generating sources close to the loads. The generating sources could be thermal or renewable sources, supported by energy storage. This model enhances the efficiency, reliability and cost-effectiveness of the system, which cannot be achieved with a single generating source [25]. In this paper, the microgrid model composes of DGs including PV/diesel generators with/without batteries in both grid-connected as well as isolated modes, with and without considering the effect of WHR. In the context of a grid-connected system, the grid connection is utilized to sale the excess electricity of the microgrid in case of capacity shortage.
In total seventeen different configurations have been analyzed. The conventional diesel generators only system, as shown in Figure 6, has been considered as a first configuration. Among the remaining sixteen configurations, eight are analyzed in isolated mode, as shown in Figure 7a–h and rest of the eight are analyzed in grid-connected mode, as shown in Figure 8a–h. In addition, among the remaining sixteen configurations, eight configurations have been simulated while considering the WHR effect, as shown in Figure 7c,d,g,h and Figure 8c,d,g,h. The PV is the only renewable energy resources used in study (because of the 2.9 million MW solar potential in Pakistan [35]), while batteries are used as the only storage device.
Table 1 and Table 2 [36], highlight different costs and technical details of the DGs with/without batteries. These parameters are the input data to HOMER Pro software. The GHG emissions penalty and CS penalty has been set at $20/ton and $20/kWh respectively [36], whereas Pakistan’s fuel price has been set at 0.75 $/L [37]. Table 3, highlights the sizes of the DGs with/without batteries taken under consideration.

3.1. Multi-Objective Analysis Using HOMER Pro

HOMER is an optimization tool developed by the U.S. National Renewable Energy Laboratory (NREL). HOMER Pro is the latest version. HOMER models the physical behavior and the life-cycle cost of a power system [38,39]. In addition, HOMER allows the user to analyze different configurations of generating and storage units based on their technical and economic benefits. The user provides the resource data like average daily solar irradiance, load profile to be served, generating/storage units to be considered and their costs as input to the HOMER. The software then performs an hourly power balance calculation for each configuration for a year. After simulating all the possible configurations, the infeasible configurations are discarded and the feasible solutions are ranked according to the lowest total net present cost (TNPC).
HOMER uses an optimizer which is based on a derivative free optimization. The optimization algorithm uses a modified grid search algorithm. The user specifies different options (in the form of inputs) related to the generating/storage units in a searchable grid, while the algorithm searches for the optimal solution [40]. The HOMER initially takes input data in a table form, then it performs the simulations on the given data to find out all the possible configurations [40]. These configurations are then analyzed to shortlist the optimal configuration while considering the objectives. Figure 9 shows the flowchart for HOMER Pro.

3.2. Objective and Constraints

The optimal configuration for the CHP plant integrated microgrid is based on multiple objectives. The multiple objectives include the minimization of the, total net present cost (TNPC) of microgrid, cost of generated energy (COE) and the annual greenhouse gas (GHG) emissions, and the maximization of the, annual waste heat recovery (WHR) of thermal units and annual grid sales (GS). These multiple objectives are subject to the constraints of power balance with 25% of operating reserve, battery operation within state of charge (SOC) limits, generation capacity limit and zero capacity shortage.
The TNPC of generating/storage unit is the present value of all the costs that it acquires during its lifespan minus the present value of all the revenues that it earns over its lifetime. Revenues include salvage value and grid sales [38]. The COE is calculated based on the total annualized cost and the total load supplied including the grid sales (GS). The GHG emissions include the pollutants like carbon dioxide (CO2), carbon monoxide (CO), unburned hydrocarbons (HC), Sulphur dioxide (SO2) and nitric oxides (NOx). All emissions are calculated by multiplying the fuel quantity with the emission coefficients [38]. In case of a CHP plant, the generator’s heat is recovered to supply the thermal load. Normally the generator’s fuel curve is used to estimate the electricity production for a given fuel. It is assumed that the remaining fuel energy will be converted to heat. Waste heat recovery (WHR) is the energy that can be recovered to supply thermal load [38]. Excess of energy that can be sold to the grid is accounted as grid sales (GS). This energy is the difference between total annual load and total annual generation [38].
The power balance with 25% of operating reserve is an equality constraint which not only supplies the load demand, but also ensures 25% additional reserves. Battery operation within SOC limits is an inequality constraint which is maintained to achieve a prolonged battery life. Generator operation within capacity limits is also an inequality constraint to maintain fuel efficiency. Zero capacity shortage is a constraint to ensure the uninterrupted supply of load.

4. Results and Analysis

Table 4 and Table 5 indicate the optimal configurations of DGs with/without batteries for each mentioned city individually. The configurations include sizes of diesel generators (kW), PV arrays (kW), batteries (kWh) and converters (kW). Some configurations involve single generator (1Gen) and single PV (1PV), however some configurations involve double generator (2Gen) and double (PV). Table 6 and Table 7, show the values of the multiple objectives based on the optimal configurations for each city. Finally, a comparative analysis has been performed to identify best solution in terms of defined objectives. Table 8 shows the most optimal value of each individual objective from different optimal configurations.
Table 8 indicates the optimal solutions considering the objectives. Table 8 further shows that all the objectives are not pertaining to a single city. Gilgit has the most optimum values for TNPC and COE. However, its other three objectives are not optimum. Lahore has the most optimum value for annual GHG emissions. Similarly, Quetta has the most optimum values for annual WHR and annual grid sales.

4.1. Graphical Representation

4.1.1. Total Net Present Cost

Figure 10 shows the comparative analysis of TNPC for different DG configurations for the six cities. Gilgit has the lowest TNPC among all DG configurations. In grid-connected mode, double diesel generator (2Gen) and double PV (2PV) with WHR system costs about 5.79 million$, which is the lowest TNPC.

4.1.2. Cost of Generated Energy

Figure 11 shows the comparative analysis of COE for different DG configurations for the six cities. Gilgit has the lowest COE among all DG configurations. In grid-connected mode, double diesel generator (2Gen) and double PV (2PV) with WHR system costs about 0.049 $/kWh.

4.1.3. Greenhouse Gas Emissions

Figure 12 shows the annual GHG emissions for different DG configurations for the six cities. The DG configurations without WHR have lower GHG emissions. However, these configurations are excluded due to zero WHR. Lahore has the lowest annual GHG emissions among all DG configurations. In isolated mode, single generator (1Gen), single PV (1PV) and battery with WHR has the lowest GHG emissions of about 1000.214 tons/year.

4.1.4. Waste Hear Recovery

Figure 13 shows annual WHR for different DG configurations for the six cities. Configurations with WHR are only considered. Quetta has the highest annual WHR among all DG configurations. In grid-connected mode, double diesel generator (2Gen), double PV (2PV) and battery has the highest annual WHR of about 2,040,282 kWh/year.

4.1.5. Grid Sales

Figure 14 shows the annual grid sales for different DG configurations for the six cities. Quetta has the highest annual grid sales among all DG configurations. In grid-connected mode, double diesel generator (2Gen) and double PV (2PV) and battery with WHR has the highest annual grid sales of about 8,322,268 kWh/year.

5. Conclusions

In this research authors have demonstrated the use of CHP plant when the consumers of NG start using electricity due to outage of natural gas. This supply of thermal load by the electricity was putting an extra burden on the electricity grid. The authors have proposed a solution in the form of a CHP plant integrated microgrid to supply both electrical and thermal loads simultaneously. In this aspect HOMER Pro software has been used to simulate a CHP plant integrated microgrid. Different configurations of the DGs with/without batteries were evaluated considering multiple objectives, including the minimization of TNPC, COE and annual GHG emissions as well as the maximization of annual WHR and annual GS. These multiple objectives were subject to the constraints of power balance, battery operation within state of charge limits, generator operation within capacity limits and zero capacity shortage. The multi-objective analysis shows that a single city does not meet all the objectives in a single configuration. However, Gilgit and Quetta are two cities which satisfy more than one objective in a single configuration.
Gilgit has the lowest TNPC, in both grid-connected and isolated modes. It is because the temperature profile of Gilgit is very contented for solar PV power generation. As the operating cost of renewable energy is very low, and this in return makes the TNPC low. The value of TNPC is lower in configurations where double generators (2Gen) and double PV (2PV) systems of different ratings are used as compared to configurations with single generators (1Gen) and single PV (1PV) systems. This is because a single generator may operate inefficiently during low loads. A single generator (1Gen) with WHR has lower TNPC than a single generator (1Gen) with no WHR. It is because, a single generator (1Gen) with WHR supplies both the electrical as well as with WHR loads. Same is true for double generators (2Gen) with WHR. Among all the cities Gilgit has the lowest TNPC, followed by Peshawar, Islamabad, Quetta, Lahore and Karachi respectively.
Gilgit also has the lowest COE. It is because the temperature profile of Gilgit is very contented for solar PV power generation. As the operating cost of renewable energy is very low, and this in return makes the COE low. Authors concluded that the value of COE is lower in grid-connected configurations as compared to isolated configurations. This is due to the fact that the excess electricity could be sold to the grid. Furthermore, the value of COE is lower in double generator (2Gen) configurations as compared to single generator (1Gen) configurations, as during light loads, a single generator (1Gen) may operate inefficiently.
Lahore has the lowest annual GHG emissions. This is because the DG configuration in isolated mode (1Gen) with single generator has the lowest annual GHG emissions. It is concluded that GHG emissions are higher in grid-connected configurations as compared to isolated configurations. The configurations without battery have low GHG emissions as compared the configurations with battery. This is because battery charging from the generators results in higher GHG emissions. The configurations without WHR have lower GHG emissions as compared to the configurations with WHR. It is concluded that a configuration with double generator (2Gen) have more GHG emissions than configurations with single generator (1Gen).
Quetta has the highest annual WHR. A double generator (2Gen), double PV (2PV) and battery with WHR, recover the maximum heat in grid-connected mode.
Quetta also has the highest annual grid sales. A double generator (2Gen), double PV (2PV) and battery with WHR, sales maximum electricity to the grid on yearly basis.
The decision for selection of final most optimal city for the CHP integrated microgrid is left to the state authorities and the planning commission. In future work the authors expect a more detailed analysis on the effect of WHR on TNPC by varying the heat recovery ratio. Moreover, a more detailed analysis is expected by changing the generator fuel to biomass or natural gas.

Author Contributions

All authors have contributed to this work. Asad Waqar proposed the research idea and contributed to writing the paper. Shahbaz Tanveer, Fareeha Anwar performed the simulations. Jehanzeb Ahmad, Muhammad Aamir and Muneeb Yaqoob contributed to the analysis of results. All authors have approved this paper.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Diversifying Pakistan’s Energy Needs. Available online: http://naturalgasglobal.com/2016/06/20/diversifying-pakistans-energy-needs-a-case-of-natural-gas-lng (accessed on 20 June 2016).
  2. Dunya News. Available online: http://dunyanews.tv/en/Pakistan/285583-Electricity-shortfall-reaches-5500-MW-in-Pakistan (accessed on 18 December 2016).
  3. Tengfei, G.; Qiuxuan, W. Optimal design and feasibility analysis of a stand-alone hybrid CHP system-based on PV/Wind/gas turbine generator/VRB for a university in Hangzhou city. In Proceedings of the 2016 Chinese Control and Decision Conference (CCDC), Yinchuan, China, 28–30 May 2016; Volume 28, pp. 6187–6191. [Google Scholar]
  4. Adrain, B.D. Integration of Combined Heat and Power Generators into Small Buildings. Master’s Thesis, University of Waterloo, Waterloo, ON, Canada, 2006. [Google Scholar]
  5. Sreto, B.; Barry, N. Impact of combined heat and power plant on thermal and electrical energy supply for small and medium size enterprises. In Proceedings of the 42nd International Universities Power Engineering Conference, Brighton, UK, 4–6 September 2007; Volume 42, pp. 17–20. [Google Scholar]
  6. Plina, I.; Antans, S.; Olegs, L.; Maris, B. Combined heat and power plants towards efficient and flexible operation. In Proceedings of the IEEE 16th International Conference on Environment and Electrical Engineering (EEEIC), Florence, Italy, 7–10 June 2016; Volume 16. [Google Scholar]
  7. Colson, C.M.; Nehrir, M.H. Evaluating the Benefits of a hybrid solid oxide fuel cell combined heat and power plant for energy sustainability and emissions avoidance. IEEE Trans. Energy Convers. 2011, 26, 140–148. [Google Scholar] [CrossRef]
  8. Alexey, C.; Vladimir, M.; Alexander, A. Combined heat and power plants load distribution optimization. In Proceedings of the 45th International Universities Power Engineering Conference UPEC2010, Wales, UK, 31 August–3 September 2010; Volume 45. [Google Scholar]
  9. Michal, D.; Petr, H. Decomposition methods for operations scheduling problem in a combined heat and power plant. In Proceedings of the 10th International Conference on Environment and Electrical Engineering, Rome, Italy, 8–11 May 2011; Volume 10, pp. 1–4. [Google Scholar]
  10. Sekgogele, K.; Chowdhury, S.; Chowdhury, S.P. Technical and economic assessment of landfill gas-based CHP plants in South Africa. In Proceedings of the 46th International Universities’ Power Engineering Conference (UPEC), Soest, Germany, 5–8 September 2011; Volume 46, pp. 1–5. [Google Scholar]
  11. Vikas, C.; Anh-Tuan, D.; Baoduo, J.; Faryar, J.; Jack, B.; Ioannis, A.; Amit, C.; Andrew, A. Modeling and optimization of a combined cooling, heating and power plant system. In Proceedings of the 2012 American Control Conference (ACC), Montreal, QC, Canada, 27–29 June 2012. [Google Scholar]
  12. Ilija, B.B.; Rade, M.C. Optimal distributed generation planning at a local level—A review of Serbian renewable energy development. Renew. Sustain. Energy Rev. 2014, 39, 79–86. [Google Scholar]
  13. Liviu, R.; Mihai, P.M. Parallel operation of combined heat and power plants and wind farms. In Proceedings of the 2012 International Conference on Applied and Theoretical Electricity (ICATE), Craiova, Romania, 25–27 October 2012; pp. 1–4. [Google Scholar]
  14. Yuanhang, D.; Lei, C.; Yong, M.; Qun, C.; Kang, H.; Junhong, H.; Yiwei, Z.; Fei, X. Dispatch model of combined heat and power plant considering heat transfer process. IEEE Tarns. Sustain. Energy 2017, 8, 1225–1236. [Google Scholar]
  15. Ren, H.; Weisheng, Z.; Ken’ichi, N.; Weijun, G.; Qiong, W. Multi-onjective optimization for the operation of distributed energy systems considering economic and environmental aspects. Appl. Energy 2010, 87, 3642–3651. [Google Scholar] [CrossRef]
  16. Sreto, B. Planning algorithm for optimal combined heat & power generation plant connection in urban distribution network (UDN). In Proceedings of the 11th International Conference on the European Energy Market (EEM14), Krakow, Poland, 28–30 May 2014; Volume 11, pp. 1–7. [Google Scholar]
  17. Shekh, N.H.; Saiful, B. Waste heat recovery from exhaust of a diesel generator set using organic fluids. In Proceedings of the 10th International Conference on Mechanical Engineering, Dhaka, Bangladesh, 20–22 December 2013; Volume 10, pp. 439–444. [Google Scholar]
  18. Sreto, B. Combined heat & power plant connection into urban distribution network according to least cost technically acceptable principle. In Proceedings of the 50th International Universities Power Engineering Conference, Stoke on Trent, UK, 1–4 September 2015; Volume 50, pp. 1–6. [Google Scholar]
  19. Saint-Pierre, A.; Pierluigi, M. Techno-economic assessment of flexible combined heat and power plant with carbon capture and storage. In Proceedings of the 2014 Power Systems Computation Conference, Wroclaw, Poland, 18–22 August 2014; pp. 1–7. [Google Scholar]
  20. Constantin, S.; Nouredine, H.; Christophe, K.; Raphael, C. Combined heat and power plant optimization tool in a competitive energy market context. In Proceedings of the CIRED 2005—18th International Conference and Exhibition on Electricity Distribution, Turin, Italy, 6–9 June 2005; Volume 18. [Google Scholar]
  21. Eugen, H.; Mihai, G. Optimization of a combined cool, heat and power plant based on genetic algorithms and specialized software. In Proceedings of the 2014 International Conference and Exposition on Electrical and Power Engineering (EPE), Iasi, Romania, 16–18 October 2014; Volume 8, pp. 1030–1033. [Google Scholar]
  22. Daniel, S.; Felix, M. Increasing flexibility of combined heat and power plants with power-to-heat. In Proceedings of the 12th International Conference on the European Energy Market (EEM), Lisbon, Portugal, 19–22 May 2015; Volume 12, pp. 1–5. [Google Scholar]
  23. Buoro, D.; Casisi, M.; Nardi, D.A.; Pinamonti, P.; Reini, M. Multicriteria optimization of a distributed energy supply system for an industrial area. Energy 2013, 58, 128–137. [Google Scholar] [CrossRef]
  24. Somma, D.M.; Yan, B.; Bianco, N.; Graditi, G.; Luh, P.B.; Mongibello, L.; Naso, V. Multi-objective design optimization of distributed energy systems through cost and exergy assessments. Appl. Energy 2017, 204, 1299–1316. [Google Scholar] [CrossRef]
  25. Marialaura, D.S.; Bing, Y.; Nicola, B.; Peter, B.L.; Giorgio, G.; Luigi, M.; Vincenzo, N. Multi-objective operation optimization of a distributed energy system for a large-scale utility customer. Appl. Therm. Energy 2016, 101, 752–761. [Google Scholar]
  26. Zhang, G.; Cao, Y.; Cao, Y.; Li, D.; Wang, L. Optimal Energy Management for Microgrids with Combined Heat and Power (CHP) Generation Energy Storages and Renewable Energy Sources. Energies 2017, 10, 1288. [Google Scholar] [CrossRef]
  27. Li, P.; Wang, H.; Lv, Q.; Li, W. Combined Heat and Power Dispatch Considering Heat Storage of Both Buildings and Pipelines in District Heating System for Wind Power Integration. Energies 2017, 10, 893. [Google Scholar] [CrossRef]
  28. Zidan, A.; Gabbar, H. DG Mix and Energy Storage Units for Optimal Planning of Self-Sufficient Micro Energy Grids. Energies 2016, 9, 616. [Google Scholar] [CrossRef]
  29. Hussain, A.; Bui, V.; Kim, H.; Im, Y.; Lee, J. Optimal Energy Management of Combined Cooling Heat and Power in Different Demand Type Buildings Considering Seasonal Demand Variations. Energies 2017, 10, 789. [Google Scholar] [CrossRef]
  30. Alarcon-Rodriguez, A.; Ault, G.; Galloway, G. Multi-objective planning of distributed energy resources: A review of the state-of-the-art. Renew. Sustain. Energy Rev. 2010, 14, 1353–1566. [Google Scholar] [CrossRef]
  31. Di Somma, M.; Graditi, G.; Heydarian-Forushani, E.; Shafie-khah, M.; Siano, P. Stochastic optimal scheduling of distributed energy resources with renewables considering economic and environmental aspects. Renew. Energy 2018, 116, 272–287. [Google Scholar] [CrossRef]
  32. Maroufmashat, A.; Sattari, S.; Roshandel, R.; Fowler, M.; Elkamel, A. Multi-objective optimization for design and operation of distributed energy systems through the multi-energy hub network approach. Ind. Eng. Chem. Res. 2016, 55, 8950–8966. [Google Scholar] [CrossRef]
  33. Pakistan Bureau of Statistics. Available online: http://www.pbscensus.gov.pk/content/area-population-aadministrative-units (accessed on 23 August 2017).
  34. NASA Surface Meteorology and Solar Energy. Available online: https://eosweb.larc.nasa.gov/cgi-bin/sse/[email protected] (accessed on 2 January 2017).
  35. Tribune. Available online: https://www.tribune.com.pk/story/1357453/going-green-pakistan-1-9mw-solar-energy-potential/ (accessed on 17 March 2017).
  36. Fahd, D.; Hai, L.; Lijun, Z.; Salwa, A. An Environmentally-Friendly Tourist Village in Egypt Based on a Hybrid Renewable Energy System––Part One: What Is the Optimum City? Energies 2015, 8, 6926–6944. [Google Scholar]
  37. PAkBIZ. Available online: http://www.pakbiz.com/finance/petroleumprices/ (accessed on 3 January 2017).
  38. Pavan, Y.V.; Ravikumar, B. Renewable energy based microgrid system sizing and energy management for green buildings. J. Mod. Power Syst. Clean Energy 2015, 3, 1–13. [Google Scholar] [CrossRef][Green Version]
  39. Tom, L.; Paul, G.; Peter, L. Micropower system modeling with HOMER. Integr. Altern. Sour. Energy 2006, 379–418. [Google Scholar] [CrossRef]
  40. Homer Microgrid NEWS and Insight. Available online: http://microgridnews.com/homer-optimizer-a-faster-path-to-finding-least-cost-microgrid-options/ (accessed on 23 August 2017).
Figure 1. Electrical load in a without thermal system.
Figure 1. Electrical load in a without thermal system.
Energies 10 01625 g001
Figure 2. Electrical load in a CHP plant integrated microgrid.
Figure 2. Electrical load in a CHP plant integrated microgrid.
Energies 10 01625 g002
Figure 3. Thermal load in a CHP plant integrated microgrid.
Figure 3. Thermal load in a CHP plant integrated microgrid.
Energies 10 01625 g003
Figure 4. Average monthly solar irradiance data of Islamabad, Lahore, Karachi, Peshawar, Quetta and Gilgit.
Figure 4. Average monthly solar irradiance data of Islamabad, Lahore, Karachi, Peshawar, Quetta and Gilgit.
Energies 10 01625 g004
Figure 5. Average monthly earth’s surface temperature of Islamabad, Lahore, Karachi, Peshawar, Quetta and Gilgit.
Figure 5. Average monthly earth’s surface temperature of Islamabad, Lahore, Karachi, Peshawar, Quetta and Gilgit.
Energies 10 01625 g005
Figure 6. Conventional diesel generators only system.
Figure 6. Conventional diesel generators only system.
Energies 10 01625 g006
Figure 7. (a) Isolated single generator (1Gen) and single PV (1PV) without WHR system (b) Isolated double generator (2Gen) and double PV (2PV) without WHR system (c) Isolated single generator (1Gen) and single PV (1PV) with WHR system (d) Isolated double generator (2Gen) and double PV (2PV) with WHR system (e) Isolated single generator (1Gen), single PV (1PV) and battery without WHR system (f) Isolated double generator (2Gen), double PV (2PV) and battery without WHR system (g) Isolated single generator (1Gen), single PV (1PV) and battery with WHR system (h) Isolated double generator (2Gen), double PV (2PV) and battery with WHR system.
Figure 7. (a) Isolated single generator (1Gen) and single PV (1PV) without WHR system (b) Isolated double generator (2Gen) and double PV (2PV) without WHR system (c) Isolated single generator (1Gen) and single PV (1PV) with WHR system (d) Isolated double generator (2Gen) and double PV (2PV) with WHR system (e) Isolated single generator (1Gen), single PV (1PV) and battery without WHR system (f) Isolated double generator (2Gen), double PV (2PV) and battery without WHR system (g) Isolated single generator (1Gen), single PV (1PV) and battery with WHR system (h) Isolated double generator (2Gen), double PV (2PV) and battery with WHR system.
Energies 10 01625 g007aEnergies 10 01625 g007b
Figure 8. (a) Grid-connected single generator (1Gen) and single PV (1PV) without WHR system (b) Grid-connected double generator (2Gen) and double PV (2PV) without WHR system (c) Grid-connected single generator (1Gen) and single PV (1PV) with WHR system (d) Grid-connected double generator (2Gen) and double PV (2PV) with WHR system (e) Grid-connected single generator (1Gen), single PV (1PV) and battery without WHR system (f) Grid-connected double generator (2Gen), double PV (2PV) and battery without WHR system (g) Grid-connected single generator (1Gen), single PV (1PV) and battery with WHR system (h) Grid-connected double generator (2Gen), double PV (2PV) and battery with WHR system.
Figure 8. (a) Grid-connected single generator (1Gen) and single PV (1PV) without WHR system (b) Grid-connected double generator (2Gen) and double PV (2PV) without WHR system (c) Grid-connected single generator (1Gen) and single PV (1PV) with WHR system (d) Grid-connected double generator (2Gen) and double PV (2PV) with WHR system (e) Grid-connected single generator (1Gen), single PV (1PV) and battery without WHR system (f) Grid-connected double generator (2Gen), double PV (2PV) and battery without WHR system (g) Grid-connected single generator (1Gen), single PV (1PV) and battery with WHR system (h) Grid-connected double generator (2Gen), double PV (2PV) and battery with WHR system.
Energies 10 01625 g008aEnergies 10 01625 g008b
Figure 9. Homer Pro flow chart.
Figure 9. Homer Pro flow chart.
Energies 10 01625 g009
Figure 10. TNPC comparison.
Figure 10. TNPC comparison.
Energies 10 01625 g010
Figure 11. COE comparison.
Figure 11. COE comparison.
Energies 10 01625 g011
Figure 12. GHG emissions comparison.
Figure 12. GHG emissions comparison.
Energies 10 01625 g012
Figure 13. WHR comparison.
Figure 13. WHR comparison.
Energies 10 01625 g013
Figure 14. Grid sales comparison.
Figure 14. Grid sales comparison.
Energies 10 01625 g014
Table 1. Costs.
Table 1. Costs.
ComponentCapital CostReplacement CostO&M CostLife Time
PV module3000 $/kW2500 $/kW10 $/year20 years
Power converter800 $/kW600 $/kW5 $/year15 years
Battery300 $/kWh250 $/kWh10 $/year12,600 kWh
Diesel generator400 $/kW300 $/kW0.25 $/h15,000 h
Table 2. Technical details.
Table 2. Technical details.
Technical DetailsValue
Derating factor80%
Ground reflection20%
Converter efficiency90%
Fuel cost0.75 $/L
Annual nominal interest rate8%
Project lifetime25 years
Emissions penalty20 $/ton
Capacity shortage penalty20 $/kWh
Table 3. Sizes.
Table 3. Sizes.
ComponentRange
PV0–4000 kW
Diesel Generator0–2000 kW
Battery0–200 string size
Converter0–5000 kW
Table 4. Optimal configurations for Islamabad, Lahore and Karachi.
Table 4. Optimal configurations for Islamabad, Lahore and Karachi.
CitySr. No.DGs’ ConfigurationDGen1 (kW)DGen2 (kW)PV1 (kW)PV2 (kW)Battery (kWh)Converter (kW)
All CitiesCONVENTIONAL DIESEL ONLY SYSTEM
1Diesel only system1570-----
IslamabadISOLATED SYSTEM
11Gen + 1PV + Batt (without WHR)750-1388-21121179
22Gen + 2PV + Batt (without WHR)30055050891816720
31Gen + 1PV + Batt (with WHR)600-1065-1536889
42Gen + 2PV + Batt (with WHR)30050062.5910480696
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (without WHR)1050-2766--1733
6Grid + 2Gen + 2PV (without WHR)5013506441405-1343
7Grid + 1Gen + 1PV (with WHR)850-2366--1521
8Grid + 2Gen + 2PV (with WHR)50100010531269-1492
9Grid + 1Gen + 1PV + Batt (without WHR)1050-2698-17041727
10Grid + 2Gen + 2PV + Batt (without WHR)5501550102313082401110
11Grid + 1Gen + 1PV + Batt (with WHR)900-2238-12241393
12Grid + 2Gen + 2PV + Batt (with WHR)550500436138415361201
LahoreISOLATED SYSTEM
11Gen + 1PV + Batt (without WHR)750-1502-21121184
22Gen + 2PV + Batt (without WHR)250550124860864700
31Gen + 1PV + Batt (with WHR)600-1146-1560914
42Gen + 2PV + Batt (with WHR)35050096.9928480698
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (without WHR)1050-2889--1733
6Grid + 2Gen + 2PV (without WHR)5013007291386-1307
7Grid + 1Gen + 1PV (with WHR)850-2354--1468
8Grid + 2Gen + 2PV (with WHR)5010009761484-1515
9Grid + 1Gen + 1PV + Batt (without WHR)750-1839-2184809
10Grid + 2Gen + 2PV + Batt (without WHR)400850319145122081098
11Grid + 1Gen + 1PV + Batt (with WHR)800-2133-14161501
12Grid + 2Gen + 2PV + Batt (with WHR)250850230626510321444
KarachiISOLATED SYSTEM
11Gen + 1PV + Batt (without WHR)700-1448-18961233
22Gen + 2PV + Batt (without WHR)300550501012672743
31Gen + 1PV + Batt (with WHR)550-1169-1608870
42Gen + 2PV + Batt (with WHR)300500160946456750
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (without WHR)1000-2930--1733
6Grid + 2Gen + 2PV (without WHR)5012501316880-1368
7Grid + 1Gen + 1PV (with WHR)800-2313--1425
8Grid + 2Gen + 2PV (with WHR)1009009191533-1483
9Grid + 1Gen + 1PV + Batt (without WHR)750-1863-2352724
10Grid + 2Gen + 2PV + Batt (without WHR)4501550132712392401177
11Grid + 1Gen + 1PV + Batt (with WHR)850-2280-10801423
12Grid + 2Gen + 2PV + Batt (with WHR)500500804119811921154
Table 5. Optimal configurations for Peshawar, Quetta and Gilgit.
Table 5. Optimal configurations for Peshawar, Quetta and Gilgit.
CitySr. No.DGs’ ConfigurationDGen1 (kW)DGen2 (kW)PV1 (kW)PV2 (kW)Battery (kWh)Converter (kW)
PeshawarISOLATED SYSTEM
11Gen + 1PV + Batt (without WHR)750-1332-21601187
22Gen + 2PV + Batt (without WHR)300550250671744724
31Gen + 1PV + Batt (with WHR)650-1075-1776932
42Gen + 2PV + Batt (with WHR)300500123818552684
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (without WHR)1050-2704--1724
6Grid + 2Gen + 2PV (without WHR)5012509081023-1322
7Grid + 1Gen + 1PV (with WHR)850-2356--1530
8Grid + 2Gen + 2PV (with WHR)50105011401242-1532
9Grid + 1Gen + 1PV + Batt (without WHR)1000-2746-20641762
10Grid + 2Gen + 2PV + Batt (without WHR)350155058617302401123
11Grid + 1Gen + 1PV + Batt (with WHR)850-1899-10321197
12Grid + 2Gen + 2PV + Batt (with WHR)4005005481251624957
QuettaISOLATED SYSTEM
11Gen + 1PV + Batt (without WHR)700-1356-19201087
22Gen + 2PV + Batt (without WHR)250550150850672767
31Gen + 1PV + Batt (with WHR)600-1089-1632941
42Gen + 2PV + Batt (with WHR)35050067.3937504700
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (without WHR)1000-2827--1731
6Grid + 2Gen + 2PV (without WHR)5012505711503-1280
7Grid + 1Gen + 1PV (with WHR)800-2313--1468
8Grid + 2Gen + 2PV (with WHR)509509361399-1463
9Grid + 1Gen + 1PV + Batt (without WHR)950-2657-12241570
10Grid + 2Gen + 2PV + Batt (without WHR)450155068816882401148
11Grid + 1Gen + 1PV + Batt (with WHR)850-2493-10321573
12Grid + 2Gen + 2PV + Batt (with WHR)4501500163119792401735
GilgitISOLATED SYSTEM
11Gen + 1PV + Batt (without WHR)750-1150-21601189
22Gen + 2PV + Batt (without WHR)400550106749672731
31Gen + 1PV + Batt (with WHR)650-937-1704937
42Gen + 2PV + Batt (with WHR)30050095.5823432750
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (without WHR)1050-2386--1735
6Grid + 2Gen + 2PV (without WHR)5013506801160-1457
7Grid + 1Gen + 1PV (with WHR)850-2025--1510
8Grid + 2Gen + 2PV (with WHR)5010001431709-1574
9Grid + 1Gen + 1PV + Batt (without WHR)1050-2225-14641186
10Grid + 2Gen + 2PV + Batt (without WHR)5001650101710532401265
11Grid + 1Gen + 1PV + Batt (with WHR)850-1867-13201372
12Grid + 2Gen + 2PV + Batt (with WHR)6006501075103815841105
Table 6. Multiple objectives for Islamabad, Lahore and Karachi.
Table 6. Multiple objectives for Islamabad, Lahore and Karachi.
CitySr. No.DGs’ ConfigurationsTNPC (million$)COE ($/kWh)GHG Emissions (tons/year)WHR (kWh/year)Grid Sale (kWh/year)
All CitiesCONVENTIONAL DIESEL ONLY SYSTEM
1Diesel only system50.40.97003595.017--
IslamabadISOLATED SYSTEM
11Gen + 1PV + Batt (Without WHR)16.40.315886.650--
22Gen + 2PV + Batt (Without WHR)10.80.2092694.932--
31Gen + 1PV + Batt (With WHR)13.80.3171014.785904,483-
42Gen + 2PV + Batt (With WHR)9.420.2142112.3801,056,847-
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (Without WHR)19.40.1642145.700-6,239,867
6Grid + 2Gen + 2PV (Without WHR)7.250.0566902.004-7,289,230
7Grid + 1Gen + 1PV (With WHR)15.60.1551905.514976,2285,245,573
8Grid + 2Gen + 2PV (With WHR)6.290.0555078.6121,676,6916,255,967
9Grid + 1Gen + 1PV + Batt (Without WHR)20.30.1802010.973-5,677,536
10Grid + 2Gen + 2PV + Batt (Without WHR)8.950.0706844.752-7,114,622
11Grid + 1Gen + 1PV + Batt (With WHR)16.30.1612068.991987,3665,259,435
12Grid + 2Gen + 2PV + Batt (With WHR)8.770.0973860.6611,211,3174,188,724
LahoreISOLATED SYSTEM
11Gen + 1PV + Batt (Without WHR)16.60.321875.618--
22Gen + 2PV + Batt (Without WHR)11.00.2112712.651--
31Gen + 1PV + Batt (With WHR)13.90.3191000.214904,691-
42Gen + 2PV + Batt (With WHR)9.620.21902102.3631,056,003-
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (Without WHR)20.00.1692154.424-6,231,266
6Grid + 2Gen + 2PV (Without WHR)7.560.0596783.642-7,081,981
7Grid + 1Gen + 1PV (With WHR)16.10.1621935.756976,4755,140,511
8Grid + 2Gen + 2PV (With WHR)6.700.0585069.3491,677,0876,258,396
9Grid + 1Gen + 1PV + Batt (Without WHR)20.60.2781608.812-2,079,787
10Grid + 2Gen + 2PV + Batt (Without WHR)9.160.0895361.747-6,769,393
11Grid + 1Gen + 1PV + Batt (With WHR)16.90.1831860.616961,1114,310,902
12Grid + 2Gen + 2PV + Batt (With WHR)8.890.0834817.3341,538,4875,688,099
KarachiISOLATED SYSTEM
11Gen + 1PV + Batt (Without WHR)16.90.325925.864--
22Gen + 2PV + Batt (Without WHR)11.10.2142692.043--
31Gen + 1PV + Batt (With WHR)14.20.3251024.433900,601-
42Gen + 2PV + Batt (With WHR)9.810.2232108.4101,061,929-
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (Without WHR)20.20.1752115.538-5,941,478
6Grid + 2Gen + 2PV (Without WHR)8.030.0646708.314-6,969,520
7Grid + 1Gen + 1PV (With WHR)16.40.1731893.449965,7634,733,953
8Grid + 2Gen + 2PV (With WHR)7.260.0674838.5881,591,4495,736,755
9Grid + 1Gen + 1PV + Batt (Without WHR)20.90.2781688.857-2,197,825
10Grid + 2Gen + 2PV + Batt (Without WHR)9.640.0746944.456-7,296,741
11Grid + 1Gen + 1PV + Batt (With WHR)17.00.1761976.607970,7294,358,460
12Grid + 2Gen + 2PV + Batt (With WHR)9.570.1073840.0751,210,0314,071,456
Table 7. Multiple objectives for Peshawar, Quetta and Gilgit.
Table 7. Multiple objectives for Peshawar, Quetta and Gilgit.
CitySr. No.DGs’ ConfigurationsTNPC (million $)COE ($/kWh)GHG Emissions (tons/year)WHR (kWh/year)Grid Sale (kWh/year)
PeshawarISOLATED SYSTEM
11Gen + 1PV + Batt (Without WHR)16.40.316905.329--
22Gen + 2PV + Batt (Without WHR)10.80.2082689.675--
31Gen + 1PV + Batt (With WHR)13.80.3181005.361910,638-
42Gen + 2PV + Batt (With WHR)9.400.2132116.1301,056,336-
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (Without WHR)19.20.1622151.258-6,222,348
6Grid + 2Gen + 2PV (Without WHR)7.150.0576680.821-6,909,362
7Grid + 1Gen + 1PV (With WHR)15.40.1531901.308976,3485,260,739
8Grid + 2Gen + 2PV (With WHR)6.200.0525210.1331,724,2416,547,619
9Grid + 1Gen + 1PV + Batt (Without WHR)20.00.1791979.710-5,564,821
10Grid + 2Gen + 2PV + Batt (Without WHR)8.670.0686833.651-6,738,256
11Grid + 1Gen + 1PV + Batt (With WHR)16.20.1712008.673968,1854,119,405
12Grid + 2Gen + 2PV + Batt (With WHR)8.410.0993744.6821,199,1423,658,048
QuettaISOLATED SYSTEM
11Gen + 1PV + Batt (Without WHR)16.40.317924.143--
22Gen + 2PV + Batt (Without WHR)10.90.2092663.564--
31Gen + 1PV + Batt (With WHR)13.90.3191011.226905,825-
42Gen + 2PV + Batt (With WHR)9.540.2172112.7111,059,623-
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (Without WHR)19.30.1672064.567-5,966,449
6Grid + 2Gen + 2PV (Without WHR)7.440.0596693.132-9,879,361
7Grid + 1Gen + 1PV (With WHR)15.70.1631841.852965,2244,852,982
8Grid + 2Gen + 2PV (With WHR)6.590.0594939.3421,634,3525,962,893
9Grid + 1Gen + 1PV + Batt (Without WHR)20.00.1821801.340-5,405,521
10Grid + 2Gen + 2PV + Batt (Without WHR)9.120.0696917.252-7,241,042
11Grid + 1Gen + 1PV + Batt (With WHR)16.60.1742004.305987,8944,592,684
12Grid + 2Gen + 2PV + Batt (With WHR)8.920.0696041.4822,040,2828,322,268
GilgitISOLATED SYSTEM
11Gen + 1PV + Batt (Without WHR)16.30.314959.355--
22Gen + 2PV + Batt (Without WHR)10.70.2072669.885--
31Gen + 1PV + Batt (With WHR)13.70.3151040.364910,025-
42Gen + 2PV + Batt (With WHR)9.280.2112107.5381,055,539-
GRID-CONNECTED SYSTEM
5Grid + 1Gen + 1PV (Without WHR)19.20.1632214.816-6,178,188
6Grid + 2Gen + 2PV (Without WHR)6.600.0506922.720-7,330,439
7Grid + 1Gen + 1PV (With WHR)15.20.1521969.277975,3595,193,023
8Grid + 2Gen + 2PV (With WHR)5.790.0495116.6361,673,3696,337,434
9Grid + 1Gen + 1PV + Batt (Without WHR)20.00.2051936.572-4,230,165
10Grid + 2Gen + 2PV + Batt (Without WHR)8.570.0627096.481-7,514,326
11Grid + 1Gen + 1PV + Batt (With WHR)15.90.1681952.439962,8083,895,477
12Grid + 2Gen + 2PV + Batt (With WHR)8.300.0884349.3941,329,1604,719,079
Table 8. Most optimal values of each objective in different optimal configurations.
Table 8. Most optimal values of each objective in different optimal configurations.
CitySr. No.Optimal ConfigurationObjectives
TNPC (million$)COE ($/kWh)GHG Emissions (tons/year)WHR (kWh/year)Grid Sales (kWh/year)
Gilgit1Grid + 2Gen + 2PV (With WHR)5.790.0495116.6361,673,3696,337,434
Lahore21Gen + 1PV + Batt (With WHR)13.90.3191000.214904,691-
Quetta3Grid + 2Gen + 2PV + Batt (With WHR)8.920.0696041.4822,040,2828,322,268

Share and Cite

MDPI and ACS Style

Waqar, A.; Shahbaz Tanveer, M.; Ahmad, J.; Aamir, M.; Yaqoob, M.; Anwar, F. Multi-Objective Analysis of a CHP Plant Integrated Microgrid in Pakistan. Energies 2017, 10, 1625. https://doi.org/10.3390/en10101625

AMA Style

Waqar A, Shahbaz Tanveer M, Ahmad J, Aamir M, Yaqoob M, Anwar F. Multi-Objective Analysis of a CHP Plant Integrated Microgrid in Pakistan. Energies. 2017; 10(10):1625. https://doi.org/10.3390/en10101625

Chicago/Turabian Style

Waqar, Asad, Muhammad Shahbaz Tanveer, Jehanzeb Ahmad, Muhammad Aamir, Muneeb Yaqoob, and Fareeha Anwar. 2017. "Multi-Objective Analysis of a CHP Plant Integrated Microgrid in Pakistan" Energies 10, no. 10: 1625. https://doi.org/10.3390/en10101625

Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop