Energy Sustainability through Generation Scheduling

In a modest electrical energy sector, an economical unit cost of electricity generation is inevitable. For tropical countries like Malaysia, apart from attractive energy cost, the environmental issues due to electricity sector also play a significant role because of its tropical nature. The energy cost and its related environmental concerns are of the momentous issues of the Malaysian Government. So as to resolve the concerned issues, this research presents a direct generation scheduling strategy to match demand against power generation, to augment opportunity for energy sustainability, and to offer an attractive unit electric energy cost. Besides, the same strategy aims at minimizing emissions due to thermal power plants through generation scheduling and incorporation of renewable energy systems.


INTRODUCTION
In a modest electrical energy sector, an economical unit cost of electricity generation is inevitable. For tropical countries like Malaysia, apart from attractive energy cost, the environmental issues due to electricity sector also play a significant role because it's tropical nature. The energy cost and its related environmental concerns are of momentous issues of the Malaysian Government.
To meet the Government's vision, a method of utilizing energy effectively and economically through energy conservation has been addressed in the previous chapter. Besides, proper generation-demand matching results in the attractive unit cost of electricity and the efficient usage of the generating plants and the auxiliaries.
Hence, to achieve the research objective (a) besides energy conservation, this research presents a direct generation scheduling strategy to match demand against power generation, to augment opportunity for energy sustainability, and to offer an attractive unit electric energy cost. The same strategy aims at the research objective (b) of minimizing emissions due to thermal power plants through generation scheduling and incorporation of renewable energy systems.

GENERATION -DEMAND MATCHING
The unit cost of electricity generation is a significant index in regional and global development. In the case of fossil-fuelled power systems which is the dominant energy source, the energy tariff depends on the fuel cost that carries the maximum share of the total operation cost (Jayakumar et al., 2016;Rameshkumar et al., 2016;Saravanan et al., 2016). So as to keep electricity tariff as low as doable, fuel cost which is the highest portion of the total operating cost needs to be minimized. This is achieved through the economic operation of the power plants through generation scheduling and unit commitment (Wang et al., 2013;Sivakumar and Devaraj, 2014).
To perform economic power dispatch to attain the least cost of electricity generation, the fuel cost function of the generators becomes essential (Hong et al., 2016). This cost function is generally nonlinear and the quadratic cost representation is This Journal is licensed under a Creative Commons Attribution 4.0 International License precise and the most common one in practice where the fuel is oil, coal and gas, but also diesel generators, gas micro turbines, biomass power plants, fuel cells, etc. (Palanichamy and Babu, 2008

Objective Function
The objective function for economic dispatch to attain minimum energy cost is optimised subject to the power balance, transmission power loss and the plant's capacity constraints as given in 4, 5 and 6 (Rezaie et al., 2018;Jevtic et al., 2017).
where P D = System demand, and P Li = Transmission power loss due to generator, i.
ii. Transmission loss constraints and P Li ≤P Limax (5) iii. Plants capacity constraints Apart from these constraints, environmental restrictions also take part in the optimisation process due to large consumer receptiveness for clean electrical energy (Radosavljević, 2016). Hence, power suppliers must now control their emissions so as to meet the specified ecological requirements.

iv. Plants emission constraints
where E i : Emission from generator, i, and E Target : Hourly emission target (kg/h) The economic dispatch is very intricate to resolve because of the frequent varying system demand, huge amount of data and constraints, and the non-linear objective function. Many optimisation approaches such as integer and dynamic programming (Nemati et al., 2018;Wang et al., 2014), Genetic Algorithm (Singh et al., 2014), Simulated Annealing (He et al., 2018), hopfield neural network (Reddy and Momoh, 2015), Particle Swarm Optimization (Chen et al., 2018), Tabu Search Algorithm (Naama et al., 2013), and Grasshopper Optimization Algorithm (Suriya et al., 2018;Karthikeyan et al., 2018) are available in the market; however, each one has its own convenience and constraints.

AUXILIARY POWER CONSUMPTION (APC)
While performing generation-demand matching through economic power dispatch, the APC of the associated components of the power systems other than the generators is not usually considered (Palanichamy et al., 2015). Auxiliary systems are a significant part of a power system, regardless of whether it is of sustainable power source, fossil-fuel or nuclear energy type (ABB, 2013). Their primary purpose is to power and controls the power systems utilizing a minimum of input energy to attain most output and accessibility. They embrace all the drive control applications (pumps, fans, motors, drives), electrical stability of plant and instrumentation, management and improvement frameworks. The APC in thermal power stations is in the range of 9-10% of the power at the generator end due to the high inductive loads of motors and boiler fans (Sinha, 2015;Bhatia, 2010). For a PV plant, these auxiliaries are inverter control circuitry, transformer magnetizing circuitry, cooling fan, air conditioner, lights, computers and night time auxiliaries like street light, server, etc. The average APC is in the range of 1.5-2% of the power generated by the PV system (CERC -New Delhi, 2017). For the wind turbines, electrical energy is needed for the yaw mechanism, blade-pitch control, magnetizing the stator, heating the blade, lights, controllers, communication, sensors, metering, and data collection, etc. The auxiliary consumption for these functions exceeds even 20% of the rated capacity of the wind turbine (AWEO, 2012;Joshi, 2017;Jiang et al., 2015). Hence, due to the higher magnitude of APC, the generation scheduling to meet an attractive unit energy cost, has to accommodate the share of it in the optimization process.
The proposed generation scheduling considers the transmission power losses and the APC as well.

ECONOMIC DISPATCH WITH APC
P Gi is the net power available from the generating unit, i after the unit's APC to meet the load. So as to meet the system demand considering the APC of the unit, i its generation has to be increased depending upon the magnitude of its power consumption. Hence, the power generation of generating unit, i becomes P Gi /(1−η ai ). Due to this consideration, the generation of unit, i represented by (1) becomes where

The Coordination Equation
By making use of the Lagrange formulation, the coordination equation for economic power dispatch becomes where ∂P L /∂P Gi : Incremental transmission loss of i th generating unit (expressed in terms of transmission loss B mn coefficients), and λ: The incremental cost of received power, $/MWh.
where Taking partial derivatives of (9) and (11) with respect to i, then substituting them in (10) and applying binomial expression and simplification results in:

Generations in Terms of λ
The coordination equation represented by equation (13) is rewritten as: Equation (13) is of quadratic in nature yielding two values for the unit generations. As the unit generations can't be negative, the individual unit generations, P Gi are given by: Simplifying and rearranging of the above equation, the individual unit generations are concisely given in terms of λ as: Once the value of λ is known, the individual unit generations are readily available from (15).

The Power Balance Equation
In terms of the generations of all the participating generating units, the transmission power losses and the system demand at an instant is given by the power balance equation as: where P D = System demand, MW Substituting P Gi from (15) and P Li from (11) in (16) and simplifying, a quadratic equation in terms of λ and the system demand, P D results in as: where (17) provides two values for λ; only the positive value is considered for evaluating the individual unit generations.
Once the plant generations are known, the total fuel cost is readily available. The computational strategy is shown in Figure 1.

ENVIRONMENTAL FRIENDLY ECONOMIC DISPATCH (EFED)
The dispatch outcome of economic dispatch provides the generation of individual generating plants such that the system demand is at minimum energy cost without violating the several said constraints. Due to the recent environmental restrictions on regional and global level, the emission from fossil-fuelled power stations has to meet the stipulated specifications. For instance, if the emission level by the economic dispatch exceeds the stipulated constraints, then the generation has to be rescheduled either by reducing the plant generations or by making use of less polluting plants to generate more compared to highly polluting older power plants. In tropical countries like Malaysia, the higher percentage of humidity does not support the emissions to move up to the safe altitude; besides the haze due to the man-made forest fire also offers burden for thermal emission dispersion. Further, the day and night weather have its own influence in emission dispersal. So in energy sector the production of electricity has to be economical and environmentally friendly.
In this research, while generation scheduling to match the generation against demand, the power plant emission characteristics are amalgamated with the fuel cost equations through a price penalty factor. There are various ways of determining the price penalty factor (Palanichamy and Babu, 2008;Rao et al., 2017;Ramachandaran and Avirajamanjula, 2018). These price penalty factors result increasingly reasonable qualities just when the generating plants are working at their planned maximum capacity; for other generation levels (i.e., at less-than-full load conditions), the resulting values differ extensively from the more practical values. During partial load conditions, the heat rate requirements are higher, which makes the power plant less efficient and more polluting. Thus, in this research, a new price penalty factor appropriate for all operating load conditions is presented in the following paragraphs.
Before proceeding with the determination of the proposed price penalty factor, h, the total cost and emission equations are obtained following the coordination equation tactics with the respective cost and emission coefficients as: where P GS is the sum of the maximum generating capacity limit of all the coordinating plants.
Then the proposed price penalty factor is of the form: In (20), P Gi is the maximum generating capacity limit of the plant with lowest generating capacity among the coordinating plants.
In (21), P Gi is the maximum generating capacity limit of the plant with the largest generating capacity among the coordinating plants.   With this price penalty factor, the objective function for the EFED is presented in terms of the blended fuel cost and emission equations.

ILLUSTRATION AND DISCUSSION
The energy statistic (Statistica-2019, 2018; Malaysia Energy Information Hub, 2018), portrays the total electricity generation capacity in Malaysia as of January 2018, as by source type shown in Figure 2. In 2018, the total electricity generation capacity worked out to be 33,764 MW, and 26,492 MW came from coal, gas, and oil, which means that around 78.50% of electricity generation come from fossil-fuels.
Because of the dominance of fossil-fuels in electricity generation, the unit cost of electricity and the environment generally lie on the thermal power plants. To analyze the energy sustainability, the performance of these power plants plays a vital role apart from the      (Figure 3). The actual system data is not available; however, a standard IEEE-14 bus system which resembles the number of major thermal power stations with almost closer generating capacities of the Sarawak grid has been used for exploration. Table 1 shows the fuel cost and emission coefficients of the modified IEEE-14 bus system with three fossil-fuelled power plants along with their respective APC. These values of APCs are considered from existing power plants of similar capacities and aging (Palanichamy et al., 2015;ABB, 2013;Sinha, 2015;Bhatia, 2010). To account for the transmission losses, the loss coefficients of the test system are presented in Table 2. The loss coefficients are updateable periodically depending on the system configuration changes; however, they remain constant while performing the economic active power dispatch.

Test Data
To account for the transmission power losses, the transmission loss coefficients of the test system are offered in Table 2. These coefficients are updateable every so often subject to the system configuration changes; however, they persist constant while executing the economic power dispatch.
Apart from the thermal power plants, renewable energy generations like solar PV and small wind turbine generators are also considered following the Governments Renewable Energy Integration policy. As per the renewable energy statistics (Statistica-2019, 2018; Malaysia Energy Information Hub, 2018), solar PV is in existence and wind energy is in the exploration stage. Anticipating the future of small wind turbine in Malaysia, a small capacity of 2-3 MW generation has been considered in this work.

Economic Dispatch
The economic power dispatch has been performed for various hourly load conditions ranging from 300 MW to 700 MW without exceeding the total generating capacity of all the thermal plants. PV and wind generations are accommodated to reduce the hourly load demand to be met by thermal generators so that excess generations due to APC and pollution liberation are controllable. Following     the generation scheduling flow diagram (Figure 1), economic power dispatch has been performed and the results are presented in Tables 3-7. At every dispatch, the plant capacity constraints are duly considered and the stipulated pollution concentration has not been exceeded. The outcome of the proposed direct optimisation has been compared against a Grey Wolf Optimisation approach (Jayabarathia et al., 2016).
Among the three fossil-fuelled generating plants, Plant 1 has the uppermost APC and Plants 2 and 3 are having lesser APCs.
From the dispatch outcome, it is noticeable that for every demand varying from 300 MW to 700 MW, the excess generation needed to overcome the APC is in the range of 30.23-70.82 MW. Due to this, excess fuel cost has been incurred from a minimum of $1281.99 to a maximum of $3459.14 apart from the excess emission varying from 18.34 kg to 144.71 kg. Normally, the APC is not considered while dispatching and only the transmission power losses are considered; hence, the excess power generated, fuel cost and emission are not transparent to the utility operators and the consumers. This excess generation of power, cost and emission levels are indications for the efficient operation of power systems, and minimization of this is significant for energy sustainability.
The consolidated excess quantities are provided in Table 8. From the summary, it is evident that the total auxiliary consumption is around 10% of the hourly demand in spite of the renewable energy contribution. The excess generation and power plant emissions would have been higher if there are no renewable energy generation incorporated. An emission reduction of 137.81 kg has been resulted due to the minor renewable energy integration. The dispatch outcomes are compared with a GWO (Jayabarathia et al., 2016) and the results show the accuracy, the speed of dispatching, and the convenience of the direct method of dispatching.

EFED
The economic power dispatch offers an attractive energy cost through generation scheduling in such a way that the efficient plant (consuming less fuel) generates more than others. However, depending upon their emission characteristics and aging of the plants, the same fuel-efficient plants need not liberate less emission. The globally accepted fact that the electricity cost has to be economically and environmentally friendly to have a healthy life.
The EFED minimizes the emission level from fossil-fuelled power plants by scarifying the energy cost. Normally, both the fuel cost and emission cost coefficients are blended together with the introduction of a price penalty factor. The reduction in emission and the rise in energy cost of this approach depends on the choice of the price penalty factor. In this work, a unique penalty factor has been proposed as elaborated in Section 5. Following the proposed strategy, the price penalty factors at every load condition are determined as shown in Tables 9-13. It is worth pointing out that the price penalty factor decreases with increase in system demand.
Alike the economic dispatch, the EFED has been carried out with the same varying demand conditions using the blended cost coefficients instead of the fuel cost coefficients, and the results are presented in Tables 9-13. The dispatch outcome shows the changes in individual plant generations, transmission losses, and emission levels, which are different from the economic dispatch outcome.
The consolidated excess power generated, additional cost involved and extra emission due to APC has been shown in Table 14. The performance of economic and environmental friendly dispatches has been compared as shown in Table 15.
From Table 15, an increase of $423.92 has been noticed due to EFED, but at an advantage of 49.63 kg of emission reduction with respect to economic dispatch. So the EFED gives a comparatively clean energy at a moderate additional energy cost.

CONCLUSION
This research has offered the apprehensions of the economics and emissions controls of power systems. Two kinds of generation scheduling options are suggested -economic, and environmental friendly dispatching to improve the performance of power systems. A single direct dispatching algorithm has been proposed for both dispatch options with due consideration for APC. These two dispatching options achieve the demand matching against power generation, to augment opportunity for energy sustainability, and the minimizing emissions due to thermal power plants through generation scheduling and incorporation of renewable energy systems.
An IEEE modified 14-bus test system is used to evaluate the feasibility of the suggested algorithm. The total fuel cost, plant emissions, and transmission power loss, and the excess quantities such as generation, fuel cost and emission are the benchmarks used while performing the scheduling. Being a direct optimisation algorithm, the solution time was noticeably less than the alternative approach.