Operation optimization of regional integrated energy systems

To study the role of energy system demand‐side regulation capacity on carbon emission reduction, we proposed a regional integrated energy system (RIES) optimal dispatch model considering carbon trading cost and dynamic demand response (DR). First, based on the framework of the RIES, the impact of DR of transferable load, reducible load, and substitutable load demand on system operation, considering the response characteristics of electrical and thermal load, was analyzed. Second, the carbon quota of the system was analyzed, and the effect of carbon emissions and carbon capture technology on system operation was considered. Finally, a low carbon optimized operation model of RIES was established with system operating cost and operating efficiency as the target, and system capacity and energy storage power as the constraints. The effectiveness of the optimized operation of the improved intelligent evolutionary algorithm nondominated sorting genetic algorithm II (NSGA‐II) was verified as a research method. The results show that the energy efficiency of the system can be improved by tuning the DR of load demand with the application of the optimal operation model, and a coordination optimization on the economic and low carbon in RIES operation can also be achieved by the improved NSGA‐II algorithm.


| INTRODUCTION
2][3] European Green Deal, as a guiding agreement on carbon neutrality goal for the European Union before 2050, has been announced.In the agreement, addressing climate change and sustainable development transformation have been argued as two core substances. 4To ensure the implementation of the agreement, the Sustainable European Investment Plan and European Climate Act were released in 2020 and 2021, respectively, which aimed the carbon emissions reduction goal, that is, a carbon emissions reduction of 45% relative to the 1990' before 2030, and an achievement of carbon neutrality before 2050. 5Rebuild Better Act was also published in 2021 by the United States.In the act, related developments such as wind power, photovoltaic, energy storage, and new energy vehicles, and carbon emissions reduction have been proposed to address climate change. 6In 2020, a commitment of carbon neutrality before 2060 was published by China, which was recognized and welcomed by major countries. 7The carbon trading market in some regions has been gradually built up for the goal of low-carbon development in China.The traditional carbon trading strategy has been improved, and a ladder carbon trading cost model, has been proposed 8 ; however, this model does not consider the effect of demand response (DR) on the operation of the energy system.A two-level optimal energy dispatching strategy is proposed, which uses the information gap decision theory (IGDT) to model the uncertain factors in the day ahead scheduling plan, and the model predictive control (MPC) method to carry out the rolling optimization of the day ahead scheduling in the lower level. 9A universal optimal programming model based on price-based demand response and incentive-based demand response is proposed to evaluate the economic and environmental benefits of capacity allocation in-comprehensive energy systems and to analyze the economic and environmental benefits of demand response. 10A consistency control strategy has been proposed to solve the distributed coordinated control problem among multiple energy storage units, which has certain advantages in improving the operational efficiency of energy storage systems and provides a certain reference idea for improving the energy utilization efficiency of integrated energy system (IES). 11An IES low-carbon economic scheduling model considering multitime scale carbon quota allocation has been proposed.The upper layer adopts a scenario-based stochastic optimization method, and the lower layer adopts a second-order distributed robust optimization model based on Wasserstein distance.The results show that this model can achieve reasonable carbon quota allocation and reduce annual operating cost. 12In a previous study, the diversity of industrial park load demand was considered, the load was subdivided according to the energy grade, and a cascade optimization scheduling method for RIESs was proposed based on dynamic energy efficiency to solve the problem of low energy efficiency in the industrial park. 13 The investment in China's power grid has been studied based on the transmission and distribution electricity price policy, and the optimal coordination strategy between capacity and demand has been determined. 14The price elasticity matrix has been proposed to describe the demand behavior, and DR has been found to be effective for the system's peak shaving response [15][16][17] Price-based DR offers certain advantages, such as reduced peak-valley difference of the load and improved absorption of renewable energy in the power system. 18,19The demand model of electricity and gas load has been established using the price elasticity matrix method, and the thermal load demand model considering the fuzzy perception and delay of the thermal load has been established; these models improve energy utilization. 20The total operating cost and the amount of wind and solar discarded by the system under different electrical and thermal loads have been compared; the complement of multiple energy sources reduces the operating cost and improves the consumption of new energy. 21However, only interruptible and transferable loads have been simplified in the electrical load demand.In a previous study, 22 the operation mechanism of the electricity market and carbon trading market was discussed, carbon emissions from the power generation of renewable energy such as wind energy and photovoltaic energy were determined, demand-side response plan of the regionally integrated energy system (RIES) for the operation of cogeneration units and energy storage control strategy were proposed, and the economic operation of the RIES in the industrial park was verified.However, a suitable load-side model was not developed.
To sum up, a lot of research on RIES has been performed, which is mainly focused on the planning and operation based on the mechanistic model, however, a little work of policy analysis on the impact of RIES system operation.In this paper, an optimal operation model on RIES scheduling is obtained, and the related coordination optimization of carbon emissions on RIES have been investigated based on the model, as follows: (1) In the model, an energy system framework is proposed.The energy transfer relationship among production capacity, energy storage, and energy consumption are briefly analyzed in the framework, and a mathematical model of capacity planning on RIES is obtained.(2) The DR of mechanism and strategies in investigation on the loads such as transferable load, reducible load, and substitutable load is analyzed in RIES.(3) Considered the role of carbon trading models in RIES, combined with local government market policies, and the effect of carbon trade cost on the system economy.(4) Under the constraint conditions such as output power of production capacity equipment, storage power of energy storage system, and system energy balance, two objective functions on operating cost and energy efficiency are obtained, and an improved intelligent evolutionary nondominated sorting genetic algorithm II (NSGA-II) algorithm are proposed to resolve the objective functions.
TANG ET AL.
| 4543 What's left of this study is organized as follows: Section 2 established a typical comprehensive energy system framework.First, a detailed mathematical model of each operating device was analyzed; Second, the internal connections between the three dynamic response mechanisms of system load demand were explained; in addition, the general mathematical expression of the carbon trading model was also analyzed.In Section 3, a comprehensive energy system capacity planning optimization model was established, taking into account the system's production capacity, power, and storage power as constraints; In Section 4, a parklevel comprehensive energy system was taken as the research object, and its operational planning was discussed and analyzed using the previously established optimization model.Section 5 summarizes the study and points out future research directions.

| Overall framework
Traditional energy systems are limited to a single form of energy (e.g., electrical energy or thermal energy) and do not take advantage of the complementary advantages of different energy sources.RIESs can realize the coordination and complementarity of electrical energy and thermal energy, improve the total utilization efficiency of energy, meet the power load demand of users for multiple energy cascade uses, and ensure continuous and reliable energy supply.
The RIES studied in this paper, 23 are illustrated in Figure 1.The RIES include a power system and a thermal system to realize energy synergy and complementarity between electrical energy and thermal energy.In this system, electrical energy can be supplied by renewable energy units such as wind turbines and photovoltaic panels as well as the power grid, whereas thermal energy is supplied by a natural gas combustion device.Natural gas is supplied to cogeneration devices and gas-fired boilers.The generated electrical energy is supplied to the user side, and the surplus electrical energy generated outputed to the power grid in the reverse direction to realize the dynamic adjustment of energy.The energy coupling equipment in the system includes a cogeneration device, a heat pump, and a gas boiler for realizing the two-way flow of electrical energy.The cogeneration unit includes a micro gas turbine, a waste heat boiler, a low-temperature waste heat power generation unit.The operation mode is electrothermal decoupling, which can adapt to different operating conditions.The heat pump and the gas-fired boiler absorb wind power and bear part of the thermal load.The use of the DR strategy in the system can suppress the fluctuation of the load curve; realize the interactive coupling of electrical heating, peak shaving, and valley filling; and reduce operating costs and carbon emissions.

| Energy output and conversion model
The RIES consists of modules such as the capacity side, transmission side, energy storage side, and energy use side modules.The physical model of each module is described as follows.
F I G U R E 1 Architecture of the regional integrated energy system.

| Wind power model
The power output of the wind turbine mainly depends on the wind speed, which has high randomness.The power output of the wind turbine depends on the cut-in speed and cut-out speed of the natural wind.
where v i , v, and v o represent the cut-in speed, actual speed, and cut-out speed, respectively (m/s).

| Photovoltaic model
Solar photovoltaic generators employ photovoltaic panels to absorb energy from solar radiation to generate electrical energy.The power generated mainly depends on the intensity of light radiation and light incident angle.The expression for photovoltaic generator power generation as follows:

| Cogeneration plant
The cogeneration device is a comprehensive energy system that includes the integration of heating and power generation processes. 24It mainly comprises gas turbines and waste heat boilers and can realize the cascade utilization of energy from different sources.The cogeneration device not only improves energy utilization but also reduces the emissions of carbides and harmful gases.The generating power of the microgas turbine can be obtained as follows: The residual heat generated by the microgas turbine is recovered by the waste heat boiler and provided to users to meet the thermal load.The thermal power model can be expressed as follows:

| Heat pump model
A heat pump is a device used to transfer thermal energy from a low-temperature heat source to a hightemperature heat source.It uses electrical energy to realize the exchange of the compressor and water circulation system with the outside energy to achieve the heating effect.Its model can be expressed as follows: HP HP i HP (5)

| Power model of energy storage equipment
The energy storage system is an important part of the RIES and is used to realize the storage and release of electrical energy and thermal energy.The working process of the system energy storage device can be mathematically expressed as follows:

| DR model
DR is an important way of demand management in RIESs.In this paper, we discussed the load DR based on the load response characteristics of the user side.The user adjusts the energy consumption behavior of the load side according to the time-of-use price or incentive mechanism and participates in power grid energy interactions.The load is classified as uncontrollable load and controllable load.The controllable load can be further classified as transferable load, reducible load, and energy-replaceable load. 25,26

| Transferable load DR
The energy demand side transfers the load energy consumption during the period of high energy price to the period of low energy price according to the change in the electricity price at different periods and keeps the overall load unchanged.This is known as price-based DR.The DR characteristics can be described using the price elasticity matrix, which can be expressed using the ratio of the time-sharing price of electrical energy to the initial price: is the elasticity coefficient of the load at time i to the electricity price at time j, P i Tran,0 is the initial load at time i, P Δ i Tran is the change in load transfer at time i after DR, ν j 0 is the initial energy price at time j, and ν Δ j is the change in the energy price at time j after DR.
To ensure that the total load before and after the transfer remains unchanged and to avoid causing energy fluctuations in the original energy system, the following energy constraints must be adhered to:

| Reduced load DR
The energy demand side reduces the pressure on the RIES energy supply by reducing the energy consumption of the electrical load at peak power.The reducible price elasticity matrix can be expressed as follows:

| Alternative load DR
For loads that can be supplied by thermal energy or electrical energy, the strategy of consuming electrical energy during low electricity price periods and directly consuming thermal energy during high electricity price periods can be adopted to realize the replacement of some loads by other forms of energy.The alternative load model can be mathematically expressed as follows: where the negative sign indicates that a reduction in the replaceable electrical load corresponds to the increase in the substituted thermal load.For alternative loads, the maximum alternative load constraint (Equation 11) must be considered:

| Carbon capture model
The carbon source in RIES comes from coal-fired power plants and gas combustion, and the capture device 27,28 can absorb a certain amount of system carbon emissions.Its working principle can be expressed as follows:

| Carbon trading model
The carbon trading mechanism uses carbon emission rights as commodities to trade in the carbon trading market, thus achieving carbon emissions reduction.
Enterprises can sell excess carbon emission indicators when the carbon emissions generated during actual production are lower than the stipulated quota, and they can buy excess carbon emission indicators when the carbon emissions generated are higher than the stipulated quota.

| Carbon emission quota
In the carbon trading mechanism 29 first, the carbon emission quota of the system is determined.Two commonly used carbon emission quota allocation methods are free distribution and paid distribution. 30Free allocation refers to the preallocation of a certain amount of free carbon emission quota to the system.Paid distribution requires the system to pay a certain fee for its own carbon emissions.Currently, in China, some urban energy systems are provided carbon emission quotas by free allocation and based on the baseline method.For the RIES with DR established in this study, The carbon emission quota of the system is determined based on the government's carbon emission quota allocation plan, using the historical method. 31The carbon emission quota of the system at time t can be expressed as follows:

| Carbon emission cost
The actual carbon emissions CE ac t , of the energy system at time t is the sum of the carbon emissions generated by the micro gas turbine, gas boiler, power purchase and the carbon emissions absorbed by the carbon capture equipment.In this study, we assumed that the actual carbon emissions of the unit is proportional to the output of the unit, Accordingly, the actual carbon emissions of the system at time t can be expressed as follows: To encourage active participation in the carbon trading market, when the actual carbon emissions are less than the carbon emission quota, the system can sell the remaining carbon emission quota at the market price to earn a profit.On the contrary, it is required to buy the excess carbon emission quota from the market.The carbon transaction cost at time t can be described as follows: 3 | OPTIMAL OPERATION MODEL OF THE REGIONAL INTEGRATED ENERGY SYSTEM

| Objective function
The RIES optimization model considering DR under the carbon trading mechanism is economical and meets the system operation constraints.
where C Buy is the minimum sum of energy purchase cost, C cost is the carbon transaction cost, and C om is the operation and maintenance cost.These parameters are used as objective functions to obtain the total system operation cost F.
(1) Energy purchase cost The energy system can trade electricity with the superior power grid.When the system generates less power than the demand, it needs to purchase electricity from the superior power grid.Similarly, when the system generates surplus electricity, it can sell the surplus electricity to the superior power grid.In addition to the required electrical energy, the system needs to purchase natural gas from the gas network to maintain the operation of cogeneration devices, gas-fired boilers, and other devices.The total energy purchase cost of the system can be obtained as follows: (2) Carbon trading cost The carbon trading cost of an operation cycle is the sum of all the cost at all times (3) Operation and maintenance cost   where i = 1, 2, …, 7, representing the wind turbine, photovoltaic device, cogeneration unit, heat pump, gas cooker, electricity storage unit, and heat storage transposition, respectively; ω i is the operation and maintenance coefficient of equipment I and P i t , is the output power of device i.
For the normal operation of thermoelectric load, the energy efficiency objective function of RIES can be expressed using the below equation to reflect the lowcarbon nature of the overall operation: where the system energy consumption loss ratio is the ratio of the system function to the system carbon consumption.

| System constraints
The operation constraints of the RIES considering DR under the carbon trading mechanism include renewable energy output constraints, energy balance constraints, equipment energy conversion constraints, and energy storage equipment constraints.
(1) Energy balance constraint In the RIES studied in this paper, the energy flow includes electrical energy flow, thermal energy flow, and gas energy flow, all of which need to meet the energy balance constraints, which can be described using the following equation (2) Renewable energy output constraint Clean energy on the energy supply side mainly includes wind power and photovoltaic power.Due to the uncertainty associated with renewable energy output, power grid transmission capacity, and other factors, the system is often unable to absorb the entire wind power and photovoltaic power.That is, the actual output is less than the predicted output: (3) Cogeneration plant constraints Cogeneration plant constraints include the power generation and heat generation constraints of the cogeneration device system, wherein the power generation constraints include the power generation constraints of the microgas turbine and the power generation constraints of the low-temperature waste heat power generation device, whereas the heat generation constraints include the waste heat boiler's heat generation constraints, gas-to-electricity constraints, and gas-to-heat constraints: CHP

| Solution method
In this study, a mathematical model is established based on the RIES framework.First, price based demand response and alternative demand response are analyzed, and the load curve after demand response is obtained; Next, introduce a carbon trading model; Then the integer linear programming model is improved with the input conditions of electricity, natural gas price, system output power, and so on.NSGA-II 32 is an excellent multiobjective evolutionary optimization algorithm.With the results of linear programming as the input of improved NSGA-II, the system operation cost as the first objective, and the system operation energy efficiency as the second objective, the results of system optimal scheduling are comprehensively analyzed.The calculation process involved is shown in Figure 2.
As shown in Figure 2, the improved NSGA-II algorithm may be seen primarily in two areas: ① Choosing the Jiaer set to initialize the population improves the diversity and coverage of the population, reduces the number of similar individuals in the population, and thus avoids the risk of the algorithm falling into local optima.
② Introduced nonuniform mutation operators to improve algorithm search, as shown in Table 1.

| Research data
A complete simulation cycle of about 24 h was performed.The product capacity equipment mainly consists of renewable energy system microgas turbines in the cogeneration plant, heat pump, and gas boiler.The renewable energy system includes wind turbines with a wind power of 2081.94 kW and photovoltaic cells, with a photovoltaic power of 1683.33 kW.The hourly output power curve on renewable energy is shown in Figure 3.The installation capacity of the microgas turbine is about 5500 kW, with an electrical coefficient of 0.3 and a heating coefficient of 0.4, respectively.An installation capacity of 1300 kW and a heating coefficient of 3.16 are exhibited in the heat pump, respectively.An installation capacity of 427.78 kW and a heating coefficient of 0.92 are shown in the gas boiler, respectively.The storage energy systems are composed of a heat storage system and an electricity storage system.A maximum capacity of 600 kW, a storage coefficient of 0.93, and a discharge coefficient of 0.89 are indicated in the heat storage system, respectively.The electricity storage system shows a maximum capacity of 800 kW, a storage coefficient of 0.95, and a discharge coefficient of 0.91, respectively.In the paper, the parameters of population number and maximum iteration times in the optimization algorithm are set to 80 and 150, respectively.The daily electricity prices are shown in Table 2.
The price of metered heat in the area is 0.78 ￥/kWh, and the price of natural gas is 3.25 ￥/m³.By using this regional carbon emissions trading center data as a guide, T A B L E 1 Induced nonuniform crossover operator process.; 4: the evolutionary direction of designing offspring: the carbon trading benchmark price was set at 48 ￥/t, and the growth rate of trading price was set as 25%.

| Analysis of result
By analyzing an example based on the bi-objective function optimization model, the Pareto front set results are obtained as shown in Figure 4. Compared with the original solver, the proposed solver has a running time as shown in Figure 5, and the proposed method takes less time.The optimal scheduling of the system is achieved when the total cost is 22435￥ and the system efficiency reaches 86.4%.

| Economic feasibility and low carbon effectiveness analyses from the perspective of load DR
To verify the effectiveness of the optimal distribution strategy considering DR proposed in this paper, three operation scenarios were used to simulate the operating conditions before and after the system participates in DR to examine the effect of different tariff mechanisms on the economic feasibility and energy efficiency of the RIES and to perform a conduct comparative analysis.
(3) Scenario 3: Carbon trading mechanism and grid DR are considered.
The cost for each scenario and the actual carbon emissions from the simulation are presented in Table 3.
As can be seen in Table 3, the carbon transaction cost of Scenario 3 was 12.1% lower than that of Scenario 2, and the carbon emissions of the system were 405.8 kg lower in Scenario 3 than that in Scenario 2. This is because Scenario 3 considers the carbon emission model; as such, the system has the initial carbon emission allowance, which can offset part of the carbon emission cost, and the carbon emissions is used as a constraint in the actual operation to give priority to the effect of carbon emissions.Scenario 3 exhibited a 10.19% reduction in system energy purchase cost and an 182.3 kg reduction in system carbon emissions compared to Scenario 1.This is because Scenario 3 considers DR by shifting part of the load to low energy price periods during high energy price periods and reducing part of the load during high energy price periods so that the system can select a more economical energy purchase method.Comparing Scenario 1 and Scenario 2, in Scenario 2, the system operated with lower energy purchase cost but higher carbon trading cost and actual carbon emissions, which shows that considering the carbon trading model   is beneficial in reducing carbon emissions.In Scenario 1, the system effectively reduced carbon emissions but exhibited the limitations of higher operating cost lower economic feasibility.In Scenario 3, part of the load was transferred from the high tariff period to the low tariff period, and the load energy consumption was partially reduced; however, this requires the use of electricity and heat on the demand side of customers.By comparing the cost of electricity and gas purchases and equipment output at different times, the system can choose the more economical and less carbon-emitting mode of operation, and the RIES can strive for a balance between high economic feasibility and low carbon emissions.Scenario 3 electrical heating load response variation is shown in Figure 6A,B the high electricity price periods (11:00-15:00 and 21:00-22:00) to the low electricity price periods (00:00-07:00 and 23:00-24:00) by reducing the load during the high electricity price period and increasing the load during the low electricity price period.
The alternative response was to convert part of the thermal load into the electrical load at (8:00-10:00 and 15:00-19:00) and part of the electrical load into the thermal load at (08:00-10:00 and 15:00-20:00).Furthermore, at the time of high tariff price heat source (11:00-14:00), the thermal load was reduced, and the three DRs were coupled, making the load curve smooth, realizing peak shaving and valley filling, and effectively optimizing the electrical load of the system in each period through the electrical load DR characteristics.
The energy scheduling results of each set of equipment in Scenario 3 are illustrated in Figure 7A,B.As can be seen from Figure 7A when the power load demand of users was small, and there was no light, the power load demand of the system was mainly met by wind power and cogeneration devices, and the surplus power was stored by the power storage system.During the peak period of power load demand of users and under the condition of sufficient illumination, the power load demand was met by purchasing power from the grid, photovoltaic output, and discharge of the power storage system.
When there is surplus electricity in the RIES, part of the electricity can be sold to the grid.As can be seen from Figure 7B, when the user's thermal load demand was small, the thermal load demand was mainly met by the heat generated by the cogeneration equipment, heat pumps, and gas-fired boilers, and the heat storage device stored the excess heat.When the user's heat demand was large, the user's thermal load demand was met by the heat released by the heat storage system.The RIES optimization model can meet the overall thermal load dynamic balance through DR adjustment.

| Influence of carbon trading model on system optimization operation
Carbon transaction cost is one of the influencing factors of the cost target function, and directly affects the carbon emissions and operation cost; in addition, the energy efficiency loss target function has a certain relationship with carbon emissions.Therefore, the analysis of the carbon trading model is of great significance for the development of economically feasible systems with low carbon emissions.The following will be analyzed from carbon trading unit price, carbon emissions, and so on.
As can be seen from the proportion of system cost displayed in Figure 8, the carbon transaction cost was relatively small.This is because, under the comprehensive constraints of the system, low carbon emissions were mainly considered, thus highlighting the characteristics of energy saving and carbon saving of the system.
As can be seen from Figure 9, with the gradual increase in the carbon trading unit price, the carbon trading cost first increased and then decreased.This is because when the carbon trading price was low, the actual carbon emission remained unchanged, and the | 4553 carbon trading cost increased with the increase in the carbon trading price.When the carbon trading price was increased, the actual carbon emissions of the system reduced considerably; the reduction range was greater than the price increase.With the increase in the carbon trading price, when the output of cogeneration equipment increased, the system's gas purchase volume increased accordingly.The system operation cost exhibited a trend of rapid growth and then tended to slow down with a further increase in the carbon trading price.This is because the carbon trading price constitutes a considerable part of the total cost.When the carbon trading price was low, the carbon trading cost accounted for a small proportion of the total cost, and the total system operation cost increased with the increase in the energy purchase cost.With the increase in the carbon trading price, the carbon emissions of the system remained almost unchanged, and the carbon trading cost reached the minimum when the carbon price was 80 ￥/t.Thus, the establishment of a reasonable carbon trading price can promote the economic and low-carbon optimal operation of the RIES.

| CONCLUSIONS
In this study, we comprehensively considered the effect of the carbon trading model and DR, on the RIES, analyzed and compared the operation of three typical scenarios, and studied the effect of carbon trading price on system operation.The conclusions can be shown in the following: (1) An improvement of energy efficiency and a reduction of operating cost and carbon emissions have been achieved with the comprehensive consideration of load DR and the carbon trading strategy.(2) Based on the carbon trading strategy, the carbon trade cost shows an increase at the beginning, and then a decrease with the increase of the carbon trade price, and the total operating cost in RIES also indicates a similar change tendency.It is shown that carbon trading price plays a key role in the operating cost and carbon emission of the system, and a reasonable carbon trade price favors a reduction of operating costs and carbon emissions.
In future studies, the carbon trading strategy can be optimized further, the ladder carbon trading mechanism can be improved, and the effect of the load DR on system operation can be considered.

d
coefficient of load at time i to electricity price at time j, P i Red,0 is the initial load at time i, and P Δ iRe is the load reduction change at time i after DR.

Algorithm 1 :
Induced nonuniform crossingInput: p 1 : parent individual 1; p 2 : parent individual 2; bounds: boundary conditions of parameters;Output: c 1 : offspring 1; c 2 : offspring 2;1: preparation; 2: differences in paternal genes △ p p return offsprings F I G U R E 3 Renewable energy forecast output curves.T A B L E 2 Regional time-of-use electricity price.

F
I G U R E 4The biobjective optimization results.

F I G U R E 5
Comparison of algorithm time consumption.TA B L E 3 Running results of three scenarios.
. As shown in the illustration, transferable power DR transferred part of the load during F I G U R E 6 Load response curves: (A) Electrical load curve before and after DR; (B) Thermal load curve before and after DR.DR, demand response.TANG ET AL. | 4551 F I G U R E 7 Electrothermal load dynamic response curve: (A) Dynamic balance of electrical load; (B) dynamic balance of thermal load.
g GT natural gas combustion rate of gas turbine (kJ/m³)F I G U R E 9 C-costs curve.