Using Simulation Modelling for Designing Optimal Strategies of Fuel Mix to Comply for SOx and NOx Emission Standards in Industrial Boilers

Abstract

SOx and NOx emissions originating from the combustion of fuels during the operation of industrial boilers are one of the primary contributors to regional and local air pollution, which have major effects on human well-being and ecosystems. As governmental bodies attempt to regulate and enforce emission norms, the industries operating the boilers are presented with an opportunity to optimise their fuel mix configurations to achieve a reduction in SOx, NOx, and CO₂ emissions while ensuring the financial sustainability of plant operations. In this study, a System Dynamic simulation model is constructed from the perspective of an individual plant to facilitate the analysis of SOx, NOx, and CO₂ emissions and the expenditure incurred through energy use and pollution control systems under different fuel-mix configurations, market conditions, and policy scenarios. The model facilitates the identification of alternative fuel-mix configurations to replace existing single-fuel firing systems while also reducing both emissions and expenditure. To demonstrate the model, exemplar data based on prevalent market and policy conditions in India are used to identify alternative fuel-mixes for typical 40TPH boiler. The experiments additionally highlight the significance of having an appropriate penalty or tax on emissions to foster compliance with emission norms in the industry without adversely impacting their financial sustainability.

1. Introduction

Industrial boilers are closed vessels in which liquids such as water are heated to generate vapour or steam, which is then utilised in various physiochemical processes. For supplying the required thermal energy for the conversion of liquid into vapour, fuels such as coal, natural gas, fuel oil, etc. are combusted through a fuel burner system attached to the boiler. The combustion of these fuels leads to generation of several compounds including carbon-di-oxide (CO₂), Nitrous Oxides (NO_X), and Sulphur Oxides (SO_X), which have an adverse impact on the environment either through the greenhouse effect or through acid rain. Industrial boilers with their fuel combustion system are one of the leading sources of anthropogenic SOx and NOx emissions. As a result of which, governmental bodies across the globe have been introducing norms and penalties to persuade industries to take measures for reducing SOx and NOx emissions. Developing countries such as China and India have introduced new norms for restricting SOx and NOx emissions to 100 mg/NM3 in 2012 and 2015 respectively [1,2]. As the new norms are being enforced across the various industrial domains, the plants are exploring various mitigation strategies to meet the emission standards including Flue Gas Desulphurisation (FGD) systems and Selective Non-Catalytic Reduction (SNCR) systems. As the majority of these emissions in the plants are from the combustion of fuels in the fuel burner system, the amount of emissions varies depending on the fuel used. Typically, the majority of the burner systems are designed to utilise a single type of fuel, but they can be upgraded to a multi-fuel burner system which allows them to combust multiple fuels simultaneously. This presents the plants with an opportunity to choose a fuel mix to optimise their SOx, NOx, and CO₂ emissions, thereby improving their compliance with local emission norms as well as financial sustainability. While there are existing calculation methods to determine the spot SOx emissions for different fuel mixes, many firing systems attached to boilers continue to operate on a single fuel, typically coal in developing countries such as India and China. As an alternative, simulation, modelling can be employed to determine the cumulative performance of various fuel mixes, both in terms of emission reduction and plant expenditure, which could act as an effective decision-making tool for the plant stakeholders. Such a tool could enable the stakeholders to quickly determine the performance of different fuel mixes specific to their plant and the local market conditions and compare them to choose the optimal option that would aid them to comply with emission norms while ensuring the financial sustainability of the pathway.

The benefits of co-firing systems has been previously explored in studies such as by Milićević et al. [3], in which the SOx and NOx emissions are analysed when co-firing coal and biomass, with 10% of thermal energy derived from biomass. Wils et al. [4] studied the improvement in fuel-side costs when co-firing woody biomass alongside coal, in which it was determined that co-firing could lead to a sizeable reduction in plant expenditure on fuels. McIlveen-Wright et al. [5] assessed the reduction of CO₂ emissions when co-firing biomass with coal, as well as the associated economic benefits under various carbon trading schemes. Lintunen et al. [6] studied the impact of subsidy policies on biomass as a co-combustion fuel in the electricity sector. While the aforementioned studies analyse the effects of co-firing systems under specific conditions, the findings are not universally applicable to all the plants operating a boiler due to differences in boiler size, fuel characteristics and availability, local market tariffs, and policy conditions. As such, there is a utilitarian need for a decision support tool that would assist the plant stakeholders to assess the long-term impacts of using various combinations of fuel-mix under varying market and policy conditions on cumulative emissions (SOx, NOx, and CO₂) and as well as the plant expenditure.

System Dynamics (SD) is a computer-aided simulation modelling technique, which is suitable for studying the long-term consequences of decision-making related to fuel-mix choice and emission compliance in holistic systems with complex interconnections and cause-and-effect relationships between the various parameters within the system [7]. In the domain of electricity generation, Mwanza and Ulgen [8] have utilised the SD approach to determine the optimal electricity generation mix in Zambia based on parameters such as CO₂ emissions, air pollution reduction, capital expenditure, etc. Similarly, the impact of the electricity generation mix in captive power plants in the cement domain was explored using the SD modelling approach [7]. Romagnoli et al. [9] utilised the SD approach to determine the impact of replacing a portion of fossil fuels used for district heating in Latvia with biomass sourced locally in terms of CO₂ emissions and expenditure under various policy schemes. Pamik et al. [10] studied the impact of various SOx and NOx emission reduction technologies applicable to marine diesel engines using the SD modelling approach.

Taking into consideration the above-mentioned gaps, the following study aims to utilise the SD modelling approach to build a simulation model to aid in determining the optimal fuel mix for individual boiler and fuel-burner systems to improve emission compliance while minimising expenditure from the perspective of plant stakeholders. The study also aims to build the model targeting individual plants that are operating a boiler since the fuel availability and tariffs of raw materials in the local markets varies from plant to plant. Additionally, the model aims to allow policymakers to determine appropriate penalties for non-compliant industrial plants while ensuring the cost of compliance through either optimising fuel-mix or implementing emission reduction systems such as FGD and SNCR is lower than the afflicted penalties. This would be relevant in countries such as India where the penalties for non-compliance are significantly lower than the cost of compliance [11].

2. Methodology

For achieving the objectives described in the previous section, an SD model would be built encompassing three additional sub-modules focusing on SOx, NOx, and CO₂ emissions respectively, which are further described in Section 2.1 and Section 2.2. The SD models are typically represented in either Causal Loop Diagrams (CLD) or stock-and-flow diagrams. The causal loop diagrams consist of varying combinations of reinforcing and corrective loops which facilitate the description of complex systems with interrelated variables and the causal relationships between them. The stock-and-flow diagrams are the integral equations that are used to represent the accumulation and movement of resources between the various entities of a complex system over a period of time.

2.1. Model Conceptualisation

The emissions from the burner system are dependent on the type of fuel mix utilised for meeting the thermal energy requirement of the boiler for converting liquid to vapour. The SOx emissions are calculated on the basis of the sulphur content of the fuel, which further varies depending on the procurement source. The calculation of NOx emissions is more nuanced as it depends on multiple factors, such as (i) the nitrogen content of the fuel mix and (ii) the chemical formation of NO in the boiler from nitrogen and oxygen present in the air at temperatures over 1400 °C [12]. The carbon emissions are calculated on the basis of the carbon intensity of the fuel, i.e., the amount of CO₂ generated per unit of fuel combusted. Figure 1 depicts the causal loop diagram of the proposed model, in which the plant would procure the raw materials needed for the fuel mix depending on the available financial reserves, from which the amount of SOx, NOx, and CO₂ emissions are calculated. Depending on the calculated emissions, the expenditure through emission compliances is calculated based on the (i) penalties incurred due to excess emissions released over permissible limits and (ii) capital and operational costs incurred through the use of pollution control systems such as FGD, SNCR, and carbon capture. Additionally, the revenue through the sale of by-products generated during the process of FGD and carbon capture systems is appended back into the available plant financial reserves, thus completing the feedback loop.

The CLD of the FGD system is depicted in Figure 2, in which the SOx emissions are calculated on the basis of sulphur content in the fuel used and the amount of thermal energy required for the operation of the boiler. The dry FGD process featured in this module utilises limestone to convert SOx into Gypsum as shown in Equations (1)–(3). In this process, Gypsum is generated as a by-product that could be sold in the local markets for generating additional income for the plant.

CaCO₃(s) + SO₂(g) → CaSO₃(s) + CO₂(g)

(1)

Ca(OH)₂(s) + SO₂(g) → CaSO₃(s) + H₂O(l)

(2)

CaSO₃(aq) + 2H₂O(l) + ½O₂(g) → CaSO₄·2H₂O(s) (Gypsum)

(3)

The CLD of the SNCR system is depicted in Figure 3, in which the NOx emissions in weight is determined on the basis of existing plant readings and the amount of fuel used in the burner system (for the calculation of the volume of flue gas generated).

The SNCR process featured in this module utilises Urea as a reactant that is injected into the boiler at a specific location depending on the temperature. It results in a chemical redox reaction in which Nitrous Oxide is converted into Nitrogen with CO₂ released as a by-product, as depicted in Equations (4) and (5). In this module, the amount of Urea required in the plant is calculated using stochiometric analysis, which depends on the amount of NOx emissions that are to be reduced. Similarly, the amount of CO₂ released as a by-product is also calculated. The additional CO₂ generated as a by-product would then be utilised for calculating any applicable carbon tax, which then impacts the available financial reserves of the plant, thereby completing the loop.

2NO(g) + CO(NH₂)₂(s) + 1/2O₂(g) → 2N₂(g) + CO₂(g) + 2H₂O(l)

(4)

2NO₂(g) + 2CO(NH₂)₂(s) +O₂(g) → 3N₂(g) + 2CO₂(g) + 4H₂O(l)

(5)

The CLD of the carbon capture system is depicted in Figure 4, which features two different capture techniques, i.e., indirect carbonation method using Sodium or Barium hydroxide, and the Carbon Capture and Storage (CCS) method. In the case of the indirect carbonation method, the process involves cooling the exhaust gases from the fuel burner system to 50 °C and removing any water vapour present in it [13]. Then the dry, cool exhaust gas is sent to crystallisers containing either Sodium Hydroxide or Barium Hydroxide which then react with CO₂ to form their respective carbonates, i.e., Sodium Carbonate and Barium Carbonate, which can be sold back in the local market for additional income to the plant. In the case of CCS, the CO₂ in the exhaust gases is captured, compressed, transported, and stored in designated locations where it is stored underground, either on-shore or off-shore for foreseeable future. In locations where a carbon tax is applicable, this system could lead to further savings, thereby positively impacting the available financial reserves of the plant.

2.2. Model Construction

The proposed concepts from Section 2.1 are expanded and realised as a deterministic model with modules for FGD, SNCR, and carbon capture. The model is built using a web tool, SilicoAI, which facilities SD modelling through a graphical interface and allows for the construction of the modules through stock-and-flow diagrams [14]. The stock-and-flow diagrams of other modules and the description of all the variables used in the modules are attached to this study as Appendix A.

• Primary Module: Figure 5 depicts the stock-and-flow diagram of the proposed model, in which the input parameters from the plant are utilised along with the fuel mix to calculate the quantity of the fuels required as well as the relevant expenditure depending on the local market prices. The quantity of the fuel utilised along with other input plant parameters are then utilised in dedicated modules for carbon capture, FGD, and SNCR, which compute the amount of CO₂, SOx, and NOx emissions respectively. The variables and the formulas used in the stock-and-flow diagram of the primary model are depicted in

  Table A1. Description of the variables used in the proposed model.

  Nomenclature	Description	Equation	Units/Timestep
  Input parameters and exogenous variables
  Steam enthalpy	It is a measure of the total energy in form of heat in the steam at a reference temperature and pressure	Input parameter	Kcal/kg
  Efficiency	Boiler efficiency is the ratio of heat input through combustion of fuels to heat output to steam	Input parameter	%
  Boiler capacity	Refers to the amount of steam (in weight) which is outputted by the boiler	Input parameter	Tons/hr
  Feedwater enthalpy	It is a measure of the total energy in form of heat in feedwater at a reference temperature and pressure	Input parameter	Kcal/kg
  Toggle: PETCOKE	For setting the amount of heat energy that is to be supplied through PETCOKE	Input parameter	%
  Toggle: Coal	For setting the amount of heat energy that is to be supplied through coal	Input parameter	%
  Toggle: Natural Gas	For setting the amount of heat energy that is to be supplied through natural gas	Input parameter	%
  Toggle: Biomass	For setting the amount of heat energy that is to be supplied through biomass	Input parameter	%
  Toggle: RDF	For setting the amount of heat energy that is to be supplied through RDF	Input parameter	%
  Coal tariffs	Cost of procuring 1 ton of coal	Input dataset	INR
  PETCOKE tariffs	Cost of procuring 1 ton of PETCOKE	Input dataset	INR
  Natural Gas tariffs	Cost of procuring 1 ton of natural gas	Input dataset	INR
  Biomass tariffs	Cost of procuring 1 ton of biomass	Input dataset	INR
  RDF tariffs	Cost of procuring 1 ton of RDF	Input dataset	INR
  Coal GCV	The quantity of heat liberated by the combustion of 1 unit of Coal	Input parameter	Kcal/kg
  PETCOKE GCV	The quantity of heat liberated by the combustion of 1 unit of PETCOKE	Input parameter	Kcal/kg
  Natural Gas GCV	The quantity of heat liberated by the combustion of 1 unit of natural gas	Input parameter	Kcal/kg
  Biomass GCV	The quantity of heat liberated by the combustion of 1 unit of biomass	Input parameter	Kcal/kg
  RDF GCV	The quantity of heat liberated by the combustion of 1 unit of RDF	Input parameter	Kcal/kg
  Carbon tax	The amount of tax levied per ton of CO2 emitted from plant processes	Input dataset	INR/tCO2
  Emission penalty–SOx	The amount of penalty levied per ton SOx emissions above permissible levels	Input dataset	INR/ton
  Emission penalty–NOx	The amount of penalty levied per ton NOx emissions above permissible levels	Input dataset	INR/ton
  Carbon intensity–Coal	Amount of CO2 emissions released through combustion of 1 unit of coal	Input parameter	tCO2/ton
  Carbon intensity–PETCOKE	Amount of CO2 emissions released through combustion of 1 unit of PETCOKE	Input parameter	tCO2/ton
  Carbon intensity–NG	Amount of CO2 emissions released through combustion of 1 unit of natural gas	Input parameter	tCO2/ton
  Carbon intensity–Biomass	Amount of CO2 emissions released through combustion of 1 unit of biomass	Input parameter	tCO2/ton
  Carbon intensity–RDF	Amount of CO2 emissions released through combustion of 1 unit of RDF	Input parameter	tCO2/ton
  Toggle: Multifuel combustion system	For enabling multi-fuel combustion system for utilising a combination of fuels in the boiler	Input parameter	-
  Discount rate	It is the interest rate used to determine the present value of future cash flows	Input parameter	%
  Dynamic variables
  Energy requirement for steam generation	The amount of energy required by the boiler for generating steam	“Boiler capacity”*1000*(“Steam enthalapy”-”Feedwater enthalapy”)	Kcal/hr
  Thermal energy requirement	The amount of energy required per month for steam generation	(“Energy requirement for steam generation”/(“Efficiency”/100))*24*30	Kcal
  CAPEX: Multifuel combustion system	The capital expenditure required for setting up the multifuel combustion system including burners and respective fuel handling systems for additional fuels	if “time” = 0 and “Toggle: Multifuel combustion system” = 1 then select case “Boiler capacity” ≤ 10: 7,000,000 case “Boiler capacity”>10 and “Boiler capacity” ≤ 30: 8,000,000 case “Boiler capacity”>30 and “Boiler capacity” ≤ 50: 10,000,000 case “Boiler capacity”>50 and “Boiler capacity” ≤ 70: 15,000,000 case “Boiler capacity”>70 and “Boiler capacity” ≤ 100: 20,000,000 default: 0 else 0	INR
  Amount of Coal required	For calculating the amount of coal required for supplying the required energy to the boiler, based on the input parameters	(((“Toggle: Coal”/100)*”Thermal energy requirement”)/”Coal GCV”(“time”))/1000	Tons
  Amount of PETCOKE required	For calculating the amount of PETCOKE required for supplying the required energy to the boiler, based on the input parameters	(((“Toggle: PETCOKE”/100)*”Thermal energy requirement”)/”PETCOKE GCV”(“time”))/1000	Tons
  Amount of Natural Gas required	For calculating the amount of natural gas required for supplying the required energy to the boiler, based on the input parameters	(((“Toggle: Natural Gas”/100)* ”Thermal energy requirement”)/”Natural Gas GCV”(“time”))/1000	Tons
  Amount of Biomass required	For calculating the amount of biomass required for supplying the required energy to the boiler, based on the input parameters	(((“Toggle: Biomass”/100)*”Thermal energy requirement”)/”Biomass GCV”(“time”))/1000	Tons
  Amount of RDF required	For calculating the amount of RDF required for supplying the required energy to the boiler, based on the input parameters	(((“Toggle: RDF”/100)*”Thermal energy requirement”)/”RDF GCV”(“time”))/1000	Tons
  Cost from coal use	Amount of cost incurred through use of coal for generating thermal energy required in the boiler	“Amount of Coal required”*”Coal tariffs”(“time”)	INR
  Cost from PETCOKE use	Amount of cost incurred through use of PETCOKE for generating thermal energy required in the boiler	“Amount of PETCOKE required”*”PETCOKE tariffs”(“time”)	INR
  Cost from Biomass use	Amount of cost incurred through use of biomass for generating thermal energy required in the boiler	“Amount of Biomass required”*”Biomass tariffs”(“time”)	INR
  Cost from RDF use	Amount of cost incurred through use of RDF for generating thermal energy required in the boiler	“Amount of RDF required”*”RDF tariffs”(“time”)	INR
  Cost from Natural gas use	Amount of cost incurred through use of natural gas for generating thermal energy required in the boiler	“Amount of Natural Gas required”*”Natural Gas tariffs”(“time”)	INR
  List of stocks
  Net Expenditure	Holds the total expenditure of the system including cost of procurement of raw materials, CAPEX, OPEX, emission penalties, emission abatement systems, at the end of the simulation run. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly expenditure \left(s\right)\right)ds+Net expenditure \left(t0\right)$	INR
  CO2 emissions	Holds the total amount of CO2 emissions released through plant operations. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly C{O}_2 emissions \left(s\right)-Monthly C{O}_2 emissions captured \right)ds+C{O}_2 emissions \left(t0\right)$	Tons
  CO2 emissions reduced	Holds the total amount of CO2 emissions captured. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly C{O}_2 emissions reduced\left(s\right)\right)ds+C{O}_2 emissions reduced \left(t0\right)$	Tons
  SOx emissions	Holds the total amount of SOx emissions released through combustion of fuel. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly SOx emissions\left(s\right)-Monthly SOx emissions reduced\left(s\right)\right)ds+SOx emissions\left(t0\right)$	Tons
  SOx emissions reduced	Holds the total amount of SOx emissions reduced through FGD system. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly SOx emisssions reduced \left(s\right)\right)ds+SOx emissions reduced\left(t0\right)$	Tons
  NOx emissions	Holds the total amount of NOx emissions released through combustion of fuels. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly NOx emissions\left(s\right)-Monthly NOx emissions reduced\left(s\right)\right)ds+NOx emissions\left(t0\right)$	Tons
  NOx emissions reduced	Holds the total amount of NOx emissions reduced through SNCR system. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly NOx emissions reduced\left(s\right)\right)ds+NOx emissions reduced\left(t0\right)$	Tons
  Net Present Value	The value of all future cash flows over a specific duration, discounted to the present value. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Discounted cash flow\left(s\right)\right)ds+Net Present Value\left(t0\right)$	INR
  List of flows
  Monthly expenditure	Calculated monthly expenditure of the proposed system	“Cost from RDF use” + ”Cost from PETCOKE use” + ”Cost from Natural gas use” + ”Cost from coal use” + ”Cost from biomass use” + ”CAPEX: Multifuel combustion system” + ”[Shadow variable] Monthly expenditure” + “Carbon capture system”->“Monthly expenditure” + “Flue gas desulphurisation (FGD) system”->“Monthly expenditure” + “Selective non-catalytic reduction (SNCR) System”->“Monthly expenditure”	INR
  Monthly CO2 emissions	Calculated monthly CO2 emissions of the proposed system	“Amount of Coal required”*”Carbon intensity-Coal” + “Amount of Biomass required”*”Carbon intensity-Biomass” + “Amount of Natural Gas required”*”Carbon intensity-NG” + “Amount of PETCOKE required”*”Carbon intensity-PETCOKE” + “Amount of RDF required”*”Carbon intensity-RDF” + “Carbon capture system”->“Emissions from electricity consumption”	Tons
  Monthly CO2 emissions captured	Calculated monthly CO2 emissions captured in the proposed system	“Carbon capture system”->“Monthly emissions reduction”	Tons
  Monthly SOx emissions	Calculated monthly SOx emissions in the proposed system	“Flue gas desulphurisation (FGD) system”->“Amount of SO2 (in weight)”	Tons
  Monthly SOx emissions reduced	Calculated monthly SOx emissions reduced in the proposed system	“Flue gas desulphurisation (FGD) system”->“SO2-reduction required (weight)”	Tons
  Monthly NOx emissions	Calculated monthly NOx emissions in the proposed system	“Selective non-catalytic reduction (SNCR) System”->“Amount of NOx emissions (weight)”	Tons
  Monthly NOx emissions reduced	Calculated monthly NOx emissions reduced in the proposed system	“Selective non-catalytic reduction (SNCR) System”->“NOx-reduction required (weight)”	Tons
  Discounted Cash Flow	Current value of the monthly expenditure in future	if “time” = 0 then “Monthly expenditure” else “Monthly expenditure” * ((1 + ((“Discount rate”/100)/12))^(12*1))	INR

  .
• Flue Gas Desulphurisation (FGD) Module: The stock-and-flow diagram for the FGD module is depicted in Figure A1, in which the sulphur content of the fuel used is provided as an input value, as it changes depending on the source from which the fuel is procured. From the primary module, the quantity of fuel combusted is taken to calculate the amount of flue gas generated for determining the SOx emissions in terms of weight per volume of flue gas. Based on the amount of SOx emissions that are to be reduced, the raw material needed in the FGD process, i.e., limestone, is calculated with the expenditure associated with it. Depending on the weight of SOx emissions reduced, the quantity of the by-product produced, i.e., Gypsum, is calculated. The cost recovered through either the sale of Gypsum in the local market or by reusing it within the plant is then calculated on the basis of prevailing market conditions at the location of the plant, which is taken as an input parameter. The variables used in this module are described in

  Table A2. Description of the variables used in the FGD Module.

  Nomenclature	Description	Equation	Units/Timestep
  Input parameters and exogenous variables
  Sulphur % (Wt.) in Coal	For setting the % of sulphur content in Coal utilised in the burner system	Input parameter	%
  Sulphur % (Wt.) in PETCOKE	For setting the % of sulphur content in PETCOKE utilised in the burner system	Input parameter	%
  Sulphur % (Wt.) in NG	For setting the % of sulphur content in Natural gas utilised in the burner system	Input parameter	%
  Sulphur % (Wt.) in Biomass	For setting the % of sulphur content in biomass utilised in the burner system	Input parameter	%
  Sulphur % (Wt.) in RDF	For setting the % of sulphur content in RDF utilised in the burner system	Input parameter	%
  Limestone tariff	Cost of procuring 1 ton of limestone from the local market, for use in the FGD system	Input dataset	INR/ton
  Gypsum tariff	Cost of 1 ton of Gypsum in the local market	Input dataset	INR/ton
  OPEX-FGD	Operational expenditure of the FGD system	Input parameter	INR
  Desired Output (SOx emissions)	For setting the desired output of SOx emissions for the FGD system	Input parameter	mg/NM3
  Toggle: FGD System	For enabling and disabling the FGD system	Input parameter	-
  Dynamic variables
  Amount of flue gas generated (in weight)	Weight of the flue gas generated through combustion of fuels	(“Amount of Coal used”*9.25) + (“Amount of Natural Gas used”*11) + (“Amount of PETCOKE used”*13) + (“Amount of RDF used”*7.5) + (“Amount of Biomass used”*7.5)	Tons
  Amount of flue gas generated (in volume)	Volume of the flue gas generated through combustion of fuels at 0 °C	(“Amount of flue gas generated (in weight)”*1000)/1.29	NM3
  Amount of SO2 (in weight)	Weight of SOx generated through combustion of fuels	((“Amount of Coal used”*”Sulphur % (Wt.) in Coal”/100) + (“Amount of Biomass used”*”Sulphur % (Wt.) in Biomass”/100) + (“Amount of PETCOKE used”*”Sulphur % (Wt.) in PETCOKE”/100) + (“Amount of RDF used”*”Sulphur % (Wt.) in RDF”/100) + (“Amount of Natural Gas used”*”Sulphur % (Wt.) in NG”/100))*2	Tons
  Amount of SO2 emission	SOx emissions generated through combustion of fuels at 0 °C	(“Amount of SO2 (in weight)”*1,000,000,000)/”Amount of flue gas generated (in volume)”	mg/NM3
  SO2 emission–reduction required	Amount of SOx emissions that are to be reduced to obtain the desired output at 0 °C	(“Amount of SO2 emission”-” Desired output (SO2 emissions)”)* ”Toggle: FGD System”	mg/NM3
  SO2–reduction required (weight)	Amount of SOx emissions in weight that are to be reduced to obtain the desired emission output	“SO2 emission-reduction required” ”Amount of flue gas generated (in volume)”/1,000,000,000	Tons
  Amount of Gypsum produced	Amount of Gypsum produced, which is the by-product of the FGD system	if “SO2-reduction required (weight)” > 0 then “SO2-reduction required (weight)” * 1.16 else 0	Tons
  Amount of limestone required	Amount of limestone required, which is the raw material used in the FGD process	if “SO2-reduction required (weight)” > 0 then “SO2-reduction required (weight)”*1.817 else 0	Tons
  CAPEX–FGD	Capital expenditure required for setting up the FGD system for reducing SOx emissions	if “time” = 0 and “Toggle: FGD System” = 1 then select case “Multi-fuel burners”->“Boiler capacity” ≤ 10: 14,000,000 case “Multi-fuel burners”->“Boiler capacity” > 10 and “Multi-fuel burners”->“Boiler capacity” ≤ 30: 16,000,000 case “Multi-fuel burners”->“Boiler capacity” > 30 and “Multi-fuel burners”->“Boiler capacity” ≤ 50: 18,000,000 case “Multi-fuel burners”->“Boiler capacity” > 50 and “Multi-fuel burners”->“Boiler capacity” ≤ 70: 20,000,000 case “Multi-fuel burners”->“Boiler capacity” > 70 and “Multi-fuel burners”->“Boiler capacity” ≤ 100: 25,000,000 default: 0 else 0	INR
  Income through the sale of by-products	Income generated through sale of by-products generated during the FGD process	“Amount of Gypsum produced”*”Gypsum tariff”(“time”)	INR
  Expenditure from raw material sourcing	Expenditure incurred through procurement of raw materials required during the FGD process	“Amount of limestone required”*”Limestone tariff”(“time”)	INR
  List of stocks
  Net expenditure [FGD System]	Holds the total expenditure of the FGD system. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly expenditure\left(s\right)\right)ds+Net expenditure\left(t0\right)$	INR
  List of flows
  Monthly expenditure	Calculated monthly expenditure of the FGD system	“Expenditure from raw material sourcing”-”Income through sale of by-products” + ”OPEX-FGD”(“time”) + ”CAPEX-FGD”	INR

  along with the equations utilised.
• Selective Non-Catalytic Reduction (SNCR) Module: The stock-an-flow diagram of the SNCR system is depicted in Figure A2, in which the measured NOx emissions from the existing plant are taken as input as the NOx emissions cannot be calculated based on fuel alone. Depending on the amount of NOx emissions to be reduced, the quantity of Urea is calculated along with the expenditure incurred for procuring the raw material from the local market. This method releases CO₂ as a by-product which is calculated on the basis of the amount of Urea utilised and appended to the primary module. The variables utilised in this module are then described in

  Table A3. Description of the variables used in the SNCR Module.

  Nomenclature	Description	Equation	Units/Timestep
  Input parameters and exogenous variables
  Measured NOx emissions	For inputting the amount of NOx emissions generated during combustion of fuels in the burner system	Input parameter	mg/NM3
  Desired output (NOx emissions)	For inputting the desired output of NOx emissions generated during combustion of fuels in the burner system	Input parameter	mg/NM3
  Toggle: SNCR System	For enabling or disabling the SNCR system	Input parameter	-
  OPEX–SNCR	Operational expenditure of the SNCR system	Input parameter	INR
  Urea tariff	Cost of procuring 1 ton of Urea from the local markets for use in the SNCR system	Input dataset	INR/ton
  Dynamic variables
  Amount of NOx emissions (weight)	Amount of NOx emissions in weight, generated through combustion of fuels in the burner system	(“Measured NOx emissions”* ”Multi-fuel burners”->“Flue gas desulphurisation (FGD) system”->“Amount of flue gas generated (in volume)”)/1,000,000	Tons
  NOx emission–reduction required	Amount of reduction required in NOx emissions	“Measured NOx emissions”-” Desired output (NOx emissions)”	mg/NM3
  NOx–reduction required (weight)	Amount of reduction required in NOx emissions, in weight	“NOx emission-reduction required”*” Amount of flue gas generated (in volume)”/1,000,000,000	Tons
  Amount of flue gas generated (in weight)	Amount of flue gas generated through combustion of fuels in the burner system, in weight	(“Amount of Coal used”*9.25) + (“Amount of Natural Gas used”*11) + (“Amount of PETCOKE used”*13) + (“Amount of RDF used”*7.5) + (“Amount of Biomass used”*7.5)	Tons
  Amount of flue gas generated (in volume)	Amount of flue gas generated through combustion of fuels in the burner system, in volume	(“Amount of flue gas generated (in weight)”*1000)/1.29	NM3
  Amount of Urea required	Amount of Urea required, which is the raw material used in the SNCR process	“NOx-reduction required (weight)”*1	Tons
  CAPEX–SNCR	Capital expenditure required for setting up the SNCR system for reducing NOx emissions	if “time” = 0 and “Toggle: SNCR System” = 1 then select case “Multi-fuel burners”->“Boiler capacity” ≤ 10: 13,000,000 case “Multi-fuel burners”->“Boiler capacity” > 10 and “Multi-fuel burners”->“Boiler capacity” ≤ 30: 15,000,000 case “Multi-fuel burners”->“Boiler capacity” > 30 and “Multi-fuel burners”->“Boiler capacity” ≤ 50: 17,000,000 case “Multi-fuel burners”->“Boiler capacity” > 50 and “Multi-fuel burners”->“Boiler capacity” ≤ 70: 20,000,000 case “Multi-fuel burners”->“Boiler capacity” > 70 and “Multi-fuel burners”->“Boiler capacity” ≤ 100: 25,000,000 default: 0 else 0	INR
  Expenditure from raw material sourcing	Cost incurred for procurement of raw materials required in the SNCR process	“Amount of Urea required”*”Urea tariff”(“time”)	INR
  Amount of CO2 generated as by-product	Amount of CO2 emissions released as a by-product of SNCR process	“Amount of Urea required”*0.73	Tons
  List of stocks
  Net expenditure [SNCR system]	Holds the total expenditure of the SNCR system. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly expenditure\left(s\right)\right)ds+Net expenditure\left(t0\right)$	INR
  List of flows
  Monthly expenditure	Calculated monthly expenditure of the SNCR system.	“Expenditure from raw material sourcing” + ”OPEX-SNCR”(“time”) + ”CAPEX-SNCR”	INR

  .
• Carbon Capture Module: The stock-and-flow diagram of the carbon capture system is depicted in Figure A3 in which the amount of CO₂ emissions is calculated on the basis of quantity of fuels combusted and their associated carbon intensity. The module allows for 2 different methodology for carbon capture, (i) indirect carbonation method, in which the quantity of raw material needed, i.e., either Sodium Hydroxide or Barium Hydroxide, is calculated based on the amount of CO₂ emissions generated through the fuel-burner system. The by-products of the chemical reaction, Sodium Carbonate or Barium Carbonate is also calculated along with additional income obtained through sale in the local market. The variables used in this module are described in

  Table A4. Description of the variables used in the Carbon Capture Module.

  Nomenclature	Description	Equation	Units/Timestep
  Input parameters and exogenous variables
  Toggle: Operation Mode	Toggle for setting the mode of operation of this module: 0—None 1—Sodium Carbonate 2—Barium Carbonate 3—CO2 compression and storage	Input parameter	-
  Raw material tariff: Sodium Hydroxide	Cost of procuring 1 ton of Sodium Hydroxide from the local market, for use in the indirect carbonation process	Input dataset	INR/ton
  Raw material tariff: Barium Hydroxide	Cost of procuring 1 ton of Barium Hydroxide from the local market, for use in the indirect carbonation process	Input dataset	INR/ton
  OPEX: Sodium Hydroxide	Cost of operating the equipment necessary for indirect carbonation per ton of Sodium Hydroxide	Input dataset	INR/ton
  OPEX: Barium Hydroxide	Cost of operating the equipment necessary for indirect carbonation per ton of Barium Hydroxide	Input dataset	INR/ton
  CAPEX	Total capital investment required for setting up the necessary equipment to facilitate indirect carbonation in the cement plant	Input parameter	INR
  By-product tariff: sodium carbonate	Income through the sale of 1 ton of sodium carbonate, which is a by-product of the indirect carbonation process	Input dataset	INR/ton
  By-product tariff: Price of Barium carbonate	Income through the sale of 1 ton of Barium Carbonate, which is a by-product of the indirect carbonation process	Input dataset	INR/ton
  Electricity requirement for compression	Amount of electrical energy required for compressing 1 ton of flue gas	Input parameter	kWh/ton
  Electricity requirement for capture	Amount of electrical energy required for capturing 1 ton of flue gas	Input parameter	kWh/ton
  Capture and storage: OPEX	Cost of operating the equipment necessary for facilitating capture and compression of 1 ton of flue gas	Input dataset	INR/ton
  Monthly CO2 emissions from plant processes	Total amount of CO2 emissions generated through combustion of fuels in the burner system	Input from primary module	Tons
  Grid Emission Factor	Average amount of CO2 generated per unit of electricity generated in the local grid	Input dataset	kgCO2/kWh
  Electricity tariff	Cost per unit of electricity purchased from the local grid	Input dataset	INR/kWh
  Dynamic variables
  CO2 emissions mitigated	Total amount of emissions mitigated through various CO2 capture techniques employed in this module	(“CO2 emissions from fuel combustion”*(“Capture efficiency”/100))	Tons
  Capture efficiency	Efficiency of the CO2 capture techniques employed in this module, i.e., efficiency of x% indicates that only x% of the CO2 emissions will be captured.	select case “Toggle: Operation Mode” = 1: 98 case “Toggle: Operation Mode” = 2: 65 case “Toggle: Operation Mode” = 3: 90 default: 0	%
  Raw material requirement: Sodium Hydroxide	Calculates the amount of Sodium Hydroxide required by the process to generate the required amount of Sodium carbonate	“Amount of Sodium Carbonate generated”*1.325	Tons
  Amount of Sodium Carbonate generated	Calculates the amount of Sodium Carbonate generated during the process of indirect carbonation depending on the capture efficiency.	if “Toggle: Operation Mode” = 1 then ((“Monthly CO2 emissions from plant processes”*(“Capture efficiency”/100))*1.37) else 0	Tons
  Sodium Hydroxide: Cost of procurement	Calculates the total cost of procurement of Sodium Hydroxide required by the indirect carbonation process	“Raw material requirement: Sodium Hydroxide”*”Raw material tariff: Sodium Hydroxide”(“time”)	INR
  Barium Hydroxide: Cost of procurement	Calculates the total cost of procurement of Barium Hydroxide required by the indirect carbonation process	“Raw material requirement: Barium Hydroxide”*”Raw material tariff: Barium Hydroxide”(“time”)	INR
  Raw material requirement: Barium Hydroxide	Calculates the amount of Barium Hydroxide required by the process to generate the required amount of Barium carbonate	“Amount of Barium Carbonate generated”*1.1517	Tons
  Amount of Barium Carbonate generated	Calculates the amount of Barium Carbonate generated during the process of indirect carbonation depending on the capture efficiency.	if “Toggle: Operation Mode” = 2 then ((“Monthly CO2 emissions from plant processes”*(“Conversion efficiency”/100))*4.467) else 0	Tons
  Sodium Carbonate: Revenue from sales	Calculates the total revenue generated from the sale of the by-product, Sodium Carbonate	“Amount of Sodium Carbonate generated”*”By-product tariff: sodium carbonate”(“time”)	INR
  Carbonation methods: Operational expenditure	Calculates the total operational expenditure of the indirect carbonation method per timestep	select case “Toggle: Operation Mode” = 1: “OPEX: Sodium Hydroxide”(“time”) case “Toggle: Operation Mode” = 2: “OPEX: Barium Hydroxide”(“time”) default: 0	INR
  Green subsidy	Calculates the applicable green subsidy for capture of CO2 emissions in this module	“CO2 emissions mitigated”*”Subsidy rate for carbon capture”(“time”)	INR
  Barium Carbonate: Revenue from sales	Calculates the total revenue generated from the sale of the by-product, Barium Carbonate	“Amount of Barium Carbonate generated”*”By-product tariff: Price of Barium carbonate”(“time”)	INR
  CO2 captured and concentrated for storage	Calculates the total amount of CO2 captured and concentrated for storage	“Monthly CO2 emissions from plant processes”*(“Conversion efficiency”/100)	Tons
  CO2 capture and storage: Expenditure	Calculates the total cost for capture and concentration of CO2 in the flue gases	“Capture and storage: OPEX”*” CO2 captured and concentrated for storage”	INR
  Emissions from electricity consumption	Calculates the total amount of CO2 emissions generated as result of electricity consumption in carbon capture system	((“Electricity requirement for capture” + ”Electricity requirement for compression”)*” CO2 captured and concentrated for storage”)*”Grid Emission Factor”(“time”)	tCO2
  Cost from electricity consumption	Calculates the total amount of cost incurred from purchase of electricity from local grid for use in carbon capture system	(“Electricity requirement for capture” + ”Electricity requirement for compression”)*” CO2 captured and concentrated for storage”*”Electricity tariff”(“time”)	INR
  List of stocks
  Net expenditure [Carbon capture module]	Holds the total expenditure of the carbon capture system. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly expenditure\left(s\right)\right)ds+Net expenditure\left(t0\right)$	INR
  Net emission reduction	Holds the total amount of CO2 emissions reduced in the carbon capture system. Initialised to 0.	${{\displaystyle \int }}_{t0}^t\left(Monthly emissions reduction\left(s\right)\right)ds+Net emission reduction\left(t0\right)$	tCO2
  List of flows
  Monthly emissions reduction	Calculated monthly emissions reduced in the carbon capture system	“CO2 emissions mitigated”	tCO2
  Monthly expenditure	Calculated monthly expenditure of the carbon capture system	“Barium Hydroxide: Cost of procurement” +”Sodium Hydroxide: Cost of procurement” +” CO2 capture and storage: Expenditure” +”Carbonation methods: Operational expenditure” +”CAPEX” -”Barium Carbonate: Revenue from sales” -”Sodium Carbonate: Revenue from sales” -”Green subsidy” +”Cost from electricity consumption”	INR

  along with all the equations used.

2.3. Model Validation

The model is validated through sensitivity analysis and dimensional consistency tests before running the experiments under scenarios described in Section 2.4. The data used for the purpose of demonstrating the model’s utility is described in Section 2.5 and attached as Supplement dataset S1. The model is constructed using SilicoAI, a web-based simulation tool, and each simulation is run for a period of 10 years, with timesteps of 1 month. The units of measure are then checked among the variables for dimensional consistency.

2.3.1. Boundary Adequacy

The current model is designed to represent individual industrial plants that utilise a steam boiler. As such, the datasets for parameters such as fuel, raw materials, and by-product tariffs are prepared exogenously based on the existing tariffs in the applicable local markets and are escalated at each time step based on the existing trends for consumer price index in the geographical region. The endogenously calculated metrics in the model include the monthly expenditures for each module as well as the emissions generated and reduced or captured for SOx, NOx, and CO₂.

2.3.2. Sensitivity Analysis

The sensitivity analysis is then conducted on the model to verify if changes in core input parameters reflect relatable shifts in the output provided by the system. The expenditure for operating a boiler is directly dependent on the cost of procurement of the respective fuels that are to be utilised in the burner system. As such, the sensitivity of the system is tested in the following scenarios:

• Base case, in which the fuel tariff remains constant throughout the simulation period.
• Moderate increment case, in which the fuel tariff linearly increases by 10% by the end of the simulation run.
• High increment case, in which the fuel tariff linearly increases by 100% by the end of the simulation run.

The changes in the net expenditure of the system in Indian Rupees (INR) for all three aforementioned cases is depicted in Figure 6. The moderate increment case and high increment case have an equivalent impact on the cumulative expenditure of the plant, thereby the model is deemed sensitive to the changes in the core input variables.

2.4. Scenarios Considered

For demonstrating the model, a boiler of size 40 Tons Per Hour (TPH) is considered for all the applicable scenarios tested in this study, as it is the typical size of boilers used for steam generation in various industrial sectors such as textile and paper. Based on the parameters applicable to the aforementioned boiler, the energy requirement in each time step is computed and is fulfilled through user-set values of fuel mix and pollution control modules such as FGD, SNCR, and carbon capture. For testing the expenditure and emissions, the following fuel mix configurations are considered for the purpose of this study:

• 100% coal
  • With SNCR and FGD systems
  • Without SNCR and FGD systems
• 70% coal and 30% PETCOKE
  • With SNCR and FGD systems
  • Without SNCR and FGD systems
• 70% coal and 30% Biomass (cotton stalk)
  • With SNCR and FGD systems
  • Without SNCR and FGD systems
• 70% coal and 30% RDF
  • With SNCR and FGD systems
  • Without SNCR and FGD systems
• 50% coal, 40% PETCOKE, and 10% Biomass (cotton stalk)
  • With SNCR and FGD systems
  • Without SNCR and FGD systems

In combination with the above-mentioned configurations, the following scenarios are tested, which are:

• Business as usual (BAU), in which current prevailing conditions in India for non-compliance penalties for SOx and NOx emissions are considered. Currently, there are no penalties for excess SOx and NOx emissions for plants in India, as the deadlines for compliance had been already extended three times in the last five years [15]. As such, this scenario attempts to deduce the benefits of alternative fuel-mix configurations and implementation of pollution control systems, if any, if the current circumstances (as of October 2022) are prevalent throughout the duration of the simulation period.
• Low Mitigation Effort (LME) scenario, in which a base penalty of INR 8000 per ton of excess NOx and SOx emissions over permissible limits is considered. The monthly adjustment factor is then considered for calculating the penalty during the duration of the simulation period, which factors in the changes in the consumer price index as well as an annual increment multiplier of 5%. Additionally, the tariffs on fossil fuels (Coal, Natural Gas, and PETCOKE) are considered to increase by an additional 10% each year when compared to BAU, in order to reflect the local governmental bodies’ efforts in encouraging renewable alternatives. The penalty at each time step is then calculated as depicted in Equations 6 and 7. This scenario attempts to deduce the benefits of using alternative fuel-mix configurations and implementing pollution control systems under market conditions in which the governmental bodies are taking a conservative approach to enforcing emission compliances through penalties and taxes.

Penalty = 5000 × Monthly Adjustment Factor × tons of excess SOx and NOx

(6)

Carbon tax = 1500 × Monthly Adjustment Factor × tons of CO₂ emissions

(7)

c.

High Mitigation Effort (HME) scenario, in which a base penalty of INR 25,000 per ton of excess NOx and SOx emissions over permissible limits is considered with an annual increment of 15%. Additionally, the tariffs on fossil fuels (Coal, Natural Gas, and PETCOKE) are considered to increase by an additional 35% each year when compared to BAU, in order to reflect the local governmental bodies’ efforts in encouraging renewable alternatives. The penalty at each time step is then calculated as depicted in Equations (8) and (9). This scenario attempts to deduce the benefits of using alternative fuel-mix configurations and implementing pollution control systems under market conditions in which the governmental bodies are prioritising the adaptation of pollution control measures in industries, by imposing significant penalties and taxes for non-compliance.

Penalty = 25,000 × Monthly Adjustment Factor × tons of excess SOx and NOx

(8)

Carbon tax = 10,000 × Monthly Adjustment Factor × tons of CO₂ emissions

(9)

Table 1. Summary of the chosen scenarios for experimentation.

Scenario	Fossil Fuel Tariffs	Penalties
Business As Usual (BAU)	Existing trends will continue	No penalties imposed for non-compliance of emission norms. No carbon tax is levied upon CO2 emissions.
Low Mitigation Effort (LME)	Additional 10% increase annually when compared to BAU	INR 5000 per ton (USD 61) penalty imposed on NOx and SOx emissions above permissible limits, which is then set to increment at each time step. A carbon tax, starting at INR 1500 per ton (USD 18.42) is levied for CO2 emissions at each time step.
High Mitigation Effort (HME)	Additional 35% increase annually when compared to BAU	INR 25,000 per ton (USD 307) penalty imposed on NOx and SOx emissions above permissible limits, which is then set to increment at each time step. A carbon tax, starting at INR 10,000 per ton (USD 123) is levied for CO2 emissions at each time step.

depicts the changes in major input parameters under the three aforementioned scenarios, i.e., a, b, and c.

2.5. Dataset Preparation

The dataset used in the simulation is attached as Supplementary dataset S1, which is prepared as described in

Table 2. Description of data used for experiments in this study.

Data	Description
Coal tariffs	The coal index is calculated based on linear extrapolation of historic data sourced from Ministry of Coal, Government of India [16].
Natural Gas tariffs	The NG price index is determined based on extrapolation of historic price data from Federal Reserve Economic Data [17].
Consumer Price Index (India)	The Consumer Price Index is calculated based on linear extrapolation of historic data sourced from Ministry of Labour and Employment, Government of India [18].
PETCOKE tariffs	The price tariffs of various raw materials and by-products relevant to the model is calculated based on the current spot prices accessible on IndiaMart and calculating the future values on the basis of consumer price index forecast [19].
Refuse Derived Fuels tariffs
Cotton Stalk (Biomass) tariffs
Limestone tariffs
Gypsum tariffs
Urea tariffs
Barium Hydroxide tariffs
Sodium Hydroxide tariffs
Sodium Carbonate tariffs
Barium Carbonate tariffs
Electricity tariffs	The electricity tariffs are calculated based on extrapolation of historic price tariffs sourced from Central Electricity Authority, Government of India, and consumer price index forecast [20].
Grid Emission Factor	The grid emission factor is calculated based on the extrapolation of historic trends sourced from Central Electricity Authority, Government of India [21].

. Additionally, the typical sulphur content in fuels that were utilised in the simulation experiments is presented in Appendix B.

3. Results

The constructed model is utilised to run under different fuel-mix configurations described in Section 2.4, under the three different penalty/tax scenarios, i.e., BAU, LME, and HME and the results are subsequently compared.

Figure 7 depicts the cumulative expenditure at the end of the simulation run under the BAU scenario for all five fuel-mix configurations, in which utilising Coal is determined to be the most expensive option. Using a mixture of 70% Coal and 30% biomass was determined to be the least expensive option, i.e., 20.9% cheaper than using coal exclusively at the end of the 10-year simulation period.

The fuel-mixes are then compared with and without the utilisation of pollution control systems, i.e., FGD and SNCR in Figure 8. In the BAU scenario, no penalties were considered for excess SOx and NOx emissions over permissible norms, and it resulted in the cumulative expenditure of configurations that include FGD and SNCR to be more expensive at the end of the simulation run when compared to configurations that do not include any pollution control systems as visualised in Figure 9.

The impact of using pollution control systems under LME and HME scenarios which penalise excess SOx and NOx emissions, is depicted in Figure 10. However, the penalties tested in this simulation experiment resulted in an increase of 1% in cumulative expenditure at the end of the simulation run for equipping the plant with FGD and SNCR systems in the LME scenario. In the HME scenario, the penalties resulted in a decrease of 1% in cumulative expenditure at the end of the simulation run.

The amount of cumulative SOx emissions generated under each of the five fuel-mix configurations considered in this study is depicted in Figure 11. A combination of 70% Coal and 30% Biomass led to the least amount of SOx emissions among the configurations tested, which is 23.4% lower than using 100% coal.

The impact of using a carbon capture system, specifically with the indirection carbonation method using Sodium Hydroxide, under LME and HME scenarios, is depicted in Figure 12. The capital expenditure of setting up a carbon capture facility in the plant is prohibitively high due to it being a relatively new application in the domain of emissions control and it would not be financially viable without introducing very high penalties for carbon emissions as it can be visualised in the figure. In the case of the LME scenario, the cumulative expenditure at the end of the simulation run with a carbon capture system is 80.3% higher than the cumulative expenditure without a carbon capture system. However, the higher carbon taxes in HME scenario meant having a carbon capture system resulted in 16% lower cumulative expenditure at the end of the simulation run.

4. Discussion

System dynamic simulation modelling allowed for the analysis of different fuel-mix configurations under varying market conditions and policy scenarios. By considering cumulative emissions and expenditures, the long-term impact of different fuel mixes could be analysed and compared with each other. As various governmental bodies in developing countries such as India attempt to enforce tighter regulations to curb SOx and NOx emissions, the simulation model constructed in this study allows for individual industrial plants that operate a boiler to find an optimal fuel mix configuration under varying market conditions, in terms of both tariffs and availability. The simulation experiments in this study were run using example data applicable for 40 TPH boilers operating under market conditions prevalent in India. Figure 9 shows that having low penalties for non-compliance provides no incentive for the stakeholders to adopt FGD or SNCR instead of paying the penalties.

Table 3. Observations from the experiment results.

Scenario	Summary
BAU	This scenario offers no incentive or reason for industries to adopt pollution control measures for reducing SOx, NOx, and CO2 emissions as there are no penalties for non-compliance, and adopting these measures only leads to an increase in expenditure for the plant.
LME	The parameters chosen in this scenario for penalties for non-compliance are insufficient as there is minimal incentive for industries to adopt pollution control measures as the difference in expenditure at the end of simulation run is minimal.
HME	While more effective than LME, the parameters chosen in this scenario could be further improved to widen the gap in cumulative expenditure and incentivise the adoption of pollution control measures.

describes the observations from the BAU, LME, and HME scenarios from the experiments conducted in the previous section.

Under the BAU scenario, exclusively utilising fossil fuels such as coal and PETCOKE (fuel-mix configurations i and ii, as described in Section 2.4) led to the highest cumulative expenditure at the end of the simulation run due to increasing price trends of non-renewable fuels. Utilising a fuel mix of 70% coal and 30% biomass (cotton stalk) led to a significant reduction in cumulative expenditure due to the lower cost of procurement of cotton stalk from the local market and comparable GCV. As indicated in Figure 10, the penalties for emissions considered in scenarios LME and HME could be further tweaked to offer a significant advantage to utilising pollution control systems as opposed to paying the penalties for emissions. In terms of SOx emissions, a fuel mix of 50% coal, 40% PETCOKE, and 10% biomass led to the highest cumulative emissions, which is 53% higher than the configuration with the least cumulative emissions, i.e., 70% coal and 30% biomass. While utilising PETCOKE offers reduces the cumulative expenditure of fuel usage when compared to exclusively coal, it also increases the SOx emissions due to typically higher concentration of sulphur content in PETCOKE.

Based on the results obtained in Section 3, it is evident that it is important to ensure the cost of implementation of pollution control systems such as FGD and SNCR should be markedly lower than the cost of penalties in order to encourage the stakeholders in the industrial domains to adopt compliance measures. Similarly, the cost of technologies for carbon capture, especially in developing countries such as India, is prohibitively high and would require significantly high taxation for CO₂ emissions, which also may adversely impact the financial sustainability of the plant. Based on the experiment results, it can be determined that replacing a portion of Coal with either biomass (cotton stalk in the current experiment) or RDF could lead to both the reduction in SOx emissions while also the operational expenditure of the plant. Additionally, in case of unavailability of either RDF or biomass at the plant location, PETCOKE could also be used as a substitute to reduce operational expenditure, but at a cost of additional Sox emissions, which could be mitigated using an additional FGD system. The model could also be utilised by policymakers to study the impact of various taxes and penalties for non-compliance on individual plants in order to determine an optimal balance between fostering faster adoption of pollution control measures and the financial sustainability of the local industry.

4.1. Limitations

The exogenous datasets used for running the experiments in this study are based on either published forecast data or linearly extrapolated historic data, thereby the accuracy of the quantitative results is dependent on the accuracy of the forecasts and extrapolated trends. Therefore, the quantitative results discussed in the study are meant to demonstrate the application of the model constructed in this study and should only be considered as an approximation of future trends. Additionally, the SNCR module featured in this model does not compute the NOx emissions in the outlet flue gas due to the complexity of the calculations and would require the plant stakeholders using this model to input the data from their existing measurement devices. As such, the module may only be applicable to existing plants that are already operational and have a pollution measurement sensor installed and not new plants that are yet to be constructed or commissioned.

4.2. Considerations for Future Work

The model can be further improved by embedding relevant, multi-regional, accurate datasets that would further improve the ease of using the model by the stakeholders in the industry to find an optimal fuel mix configuration for their specific plant. This would facilitate quicker adoption of pollution control measures in the industries. Additionally, the SNCR module could be improved to include calculations of NOx emissions without the use existing sensor data to improve applicability of the model to even new plants that are to be commissioned. Selective Catalytic Reduction (SCR) system could also be appended to the model, which while are more expensive to implement than SNCR, it could be relevant in large boilers where an SNCR system might not be sufficient to meet the emission norms for NOx emissions.

5. Conclusions

An SD model for assessing different fuel-mix configurations for industrial boilers in terms of SOx, NOx, and CO₂ emissions as well as expenditure is developed in this study. For demonstrating the utility of the model, exemplar data based on prevalent market conditions in India was utilised and simulated under different scenarios for a period of 10 years. The results from the simulation revealed the significance of having appropriate emission penalties to foster the adoption of pollution control measures in industries that utilise boilers in their production processes. Furthermore, the model could be adapted to any specific plant that utilises an industrial boiler by modifying the relevant input datasets and parameters applicable to the plant and the local market conditions. The models could also be additionally utilised by the policymakers to design penalties and taxes that would hasten adoption of pollution control systems without adversely impacting the financial sustainability of the local industry. The study concludes with recommendations for improvements to the current model in future research as well as the inclusion of more pollution control systems that would further boost the applicability of the current model as a decision-making tool for choosing fuel-mix configurations for optimising emissions and expenditure.