Dynamic Simulation of MFT and BT Processes on a 660 MW Ultra-Supercritical Circulating Fluidized Bed Boiler

Abstract

In order to study the dynamic characteristics of the 660 MW ultra-supercritical circulating fluidized bed (CFB) boiler when the main fuel trip (MFT) and boiler trip (BT) are triggered, a dynamic simulation model of the 660 MW ultra-supercritical circulating fluidized bed boiler was established on the Apros simulation platform. The model dynamically simulated the MFT and BT processes at 100% BMCR, 75% THA, and 50% THA conditions, respectively. The steady-state simulation results showed a high accuracy compared with the designed parameters. The dynamic simulation results showed that after triggering the MFT and BT, owing to the huge thermal inertia, the bed temperature and steam temperature decreased lowly. For 100% BMCR and 75% THA conditions, the moisture separator always worked in dry state during the MFT and BT processes. For the 50% THA condition, the moisture separator quickly switched from dry to wet operation after the boiler triggers MFT and BT and gradually switched from wet to dry operation after MFT and BT were reset.

1. Introduction

Owing to good fuel adaptability, high efficiency, and excellent environmental performance, circulating fluidized bed combustion technology has developed rapidly in China during the last couple of decades [1,2,3,4,5]. Larger capacity, higher parameters, and better environmental performance are the main development direction of circulating fluidized bed boilers. In 2019, there were two 660 MW ultra-supercritical circulating fluidized bed boiler unit projects approved by Chinese government [6], and the Weihe 660 MW ultra-supercritical CFB boiler project started construction in 2020. Currently, the Weihe 660 WM ultra-supercritical CFB boiler project is the boiler with the largest capacity and the most advanced performance of the circulating fluidized bed boiler in the world, which will solve the problem of efficient and clean utilization of high-sulfur anthracite [7,8].

Different from the pulverized coal fired boiler, the thermal inertia of the circulating fluidized bed boiler is much larger than that of the pulverized coal fired boiler due to a large amount of bed materials in the circulating fluidized bed boiler [9,10,11]. In case of MFT and BT accidents of a CFB boiler, although the fuel has been cut off, there are still a large amount of bed materials burning in the furnace to continue to release heat, so the furnace is more likely to be in dangerous conditions. The huge thermal inertia of the CFB boiler makes it necessary to reheat the bed material when the boiler is restarted after shutdown, so the restart time is longer than that of the pulverized coal boiler [12]. On the other hand, when emergency accidents such as MFT and BT happen to the boiler, the huge thermal inertia of the CFB boiler can also be considered as an advantage, especially in maintaining the stability of steam temperature and pressure. The huge thermal inertia can enable the steam to continue to absorb heat from the furnace after the fuel is cut off, so the temperature and pressure of the steam will not drop too fast or too low and thus can keep the steam turbine working in a safer situation.

MFT and BT are two typical accident conditions during the operations of both pulverized coal and CFB boilers [13]. MFT and BT refer to the command given to quickly cut off all fuel to the furnace and cause necessary interlocking actions when the operation conditions of the boiler cannot be met, and emergency shutdown is required so as to avoid potential hazards to the boiler and protect the safety of the boiler furnace, other equipment, and operators. For the CFB boiler, it has to cut off fuel and limestone to the furnace, and the external bed heat exchangers also need to be cut off to prevent the overheating of heat exchangers. Triggering MFT and BT of the boiler will cause shutdown of the boiler, which will affect the operation of the whole unit [14]. Therefore, it is of great engineering value to study the dynamic response of the CFB boiler after MFT and BT, especially the dynamic response of the water–steam system.

As the first 660 MW ultra-supercritical CFB boiler power plant is currently entering the construction stage in China, many works have been reported based on the 660 MW ultra-supercritical CFB boiler. Zhu et al. [15] established a simulation model to calculate the cycle efficiency of a 660 MW ultra-supercritical CFB boiler power plan. Based on their analysis, the optimized design and operation strategy were proposed. Ji et al. [16] simulated the NOx and SO₂ emissions of a 660 MW ultra-supercritical CFB boiler, which was extended from a two-dimensional combustion model of a 660 MW supercritical CFB boiler model. Additionally, some research [17,18] about water–steam system of the 660 MW ultra-supercritical CFB boiler has also been carried out.

Based on the Weihe 660 MW ultra-supercritical CFB boiler, we established the full-scale dynamic simulation model of the 660 MW ultra-supercritical CFB boiler, including the water–steam system, combustion system, air-flue gas system, and ash circulation system. The model was calibrated and verified at 100% BMCR, 75% THA, and 50% THA conditions, respectively. Based on this simulation model, the dynamic simulation analyses of MFT and BT processes were carried out at 100% BMCR, 75% THA, and 50% THA conditions, respectively. This study could provide a good reference for the operation of the 660 MW ultra-supercritical CFB boiler in the future.

2. Model Description

The dynamic simulation model established in this study is based on a 660 MW ultra-supercritical circulating fluidized bed boiler, which is under construction stage in Weihe of Guizhou Province, China. The primary air enters the furnace from the air distributor at the bottom so that the bed at the bottom of the furnace is in a fluidized state. The primary air drives the bed materials upward from the bottom of furnace. With the rise of the furnace, the upward movement speed of solids gradually decreases, and then, it move downward near the furnace wall, forming a material circulation process in the furnace. The particles at the furnace outlet are captured by the cyclone separator and then flow into the external bed heat exchanger. After heat exchange with steam in the external heat exchanger, the ash returns to furnace, which forms an external circulation of materials.

The “core-annulus” model of a gas–solid two-phase flow structure in a furnace is a simplified 2D simulation model, which is widely used to describe the dynamic characteristics of gas–solid two-phase flow in CFB boilers. The “core-annulus” model has gained much attention [19,20,21] since it was first carried out. In the radial direction of the CFB boiler, the furnace is divided into two zones: the core area and the annular area. The core area is located in the center of the furnace, where the solid particles move upward and move fast. Near the furnace wall, solid particles move downward, and solid particles gather near the wall to form a annular area with a high density. The “core-annulus” model can not only reflect the non-uniformity of gas–solid two-phase distribution in the axial direction but also reflect its change in the radial direction, so it has attracted a great deal of attention since it was first reported. The “core-annulus” model in the CFB boiler is shown in Figure 1 [8,22,23].

In the 660 MW ultra-supercritical CFB boiler simulation model, the furnace is totally divided into 13 calculation nodes in axial direction. One of nodes is the high-density bed at the bottom of furnace, and twelve of them are used to simulate the upper dilute area. In the calculation node of the upper dilute area, the secondary air, water wall, and platen superheater are simulated in different nodes based on the geometric configuration, respectively. Each node in the dilute-phase area contains a pair of “core-annulus” structures, which are used to simulate the solid distribution in the radial direction. The calculation node is shown in Figure 2.

This model was developed in a commercial software, namely Advanced Process Simulation software (APROS), which was developed by Fortum, a company in Finland, and VTT Technical Research Centre of Finland. In the APROS simulation platform, a detailed dynamic simulation model of the 660 MW ultra-supercritical CFB boiler was established in our previous work [8]; in this model, the six-equation model was used to simulate the dynamic behaviors of water–steam two-phase flow. The superheated steam system, reheated steam system, air-flue gas system, combustion system, and ash circulation loop system were simulated as well.

3. Results and Discussion

3.1. Model Validation

This study focuses on the dynamic response of the 660 MW ultra-supercritical CFB bed boiler during the MFT and BT processes at 100% BMCR condition, 75% THA condition, and 50% THA condition, respectively. The steady-state model was calibrated at 100% BMCR, 75% THA, and 50% THA conditions, and the steady-state simulation results are shown in

Table 1. Simulation results at 100% BMCR.

Parameters	Unit	Designed Value	Simulated Value	Relative Error (%)
Main steam mass flow	t/h	1902	1902	0.00
Main steam temperature	°C	605	604.3	−0.12
Main steam pressure	MPa	29.3	29.3	0.00
Reheated steam mass flow	t/h	1612	1612	0.00
Reheated steam inlet temperature	°C	366.8	366.8	0.00
Reheated steam outlet temperature	°C	623	624.4	0.22
Fuel mass flow	t/h	294.1	295.6	0.51
Water wall outlet steam temperature	°C	435	435.4	0.09
Bed temperature	°C	895	895.2	0.02
LTS outlet steam temperature	°C	473	476.2	0.68
ITS2 outlet steam temperature	°C	547	546.2	−0.15
HTS2 outlet steam temperature	°C	605	604.3	−0.12

,

Table 2. Simulation results at 75% THA.

Parameters	Unit	Designed Value	Simulated Value	Relative Error (%)
Main steam mass flow	t/h	1254	1254	0.00
Main steam temperature	°C	605.1	603.2	−0.31
Main steam pressure	MPa	24.2	24.2	0.00
Reheated steam mass flow	t/h	1064	1064	0.00
Reheated steam inlet temperature	°C	354.2	354.2	0.00
Reheated steam outlet temperature	°C	623	622.2	−0.13
Fuel mass flow	t/h	210.6	208.8	−0.85
Water wall outlet steam temperature	°C	404	420.7	4.13
Bed temperature	°C	870	873.4	0.39
LTS outlet steam temperature	°C	485	477.5	−1.55
ITS2 outlet steam temperature	°C	559	556.8	−0.39
HTS2 outlet steam temperature	°C	605	603.2	−0.30

and

Table 3. Simulation results at 50% THA.

Parameters	Unit	Designed Value	Simulated Value	Relative Error (%)
Main steam mass flow	t/h	823.1	823.1	0.00
Main steam temperature	°C	605	607.1	0.35
Main steam pressure	MPa	15.6	15.6	0.00
Reheated steam mass flow	t/h	698.8	698.8	0.00
Reheated steam inlet temperature	°C	360.4	360.4	0.00
Reheated steam outlet temperature	°C	623	623.1	0.02
Fuel mass flow	t/h	146.8	145.6	−0.82
Water wall outlet steam temperature	°C	376	376.7	0.19
Bed temperature	°C	830	851.2	2.55
LTS outlet steam temperature	°C	492	481.8	−2.07
ITS2 outlet steam temperature	°C	579	568.3	−1.85
HTS2 outlet steam temperature	°C	605	607.1	0.35

. The simulation results show that the relative errors of most parameters are less than 1%, which indicate a high simulation accuracy for all of the 100% BMCR, 75% THA, and 50% THA conditions.

Figure 3 shows the temperature and solid particle concentration distribution in the furnace under different working conditions. It can be seen that the temperature in the furnace decreases gradually with the increase of furnace height. Compared with the pulverized coal-fired boiler, the temperature distribution along the furnace is more uniform due to the circulation of solid particles in furnace. The particle concentration decreases with the increase of furnace height, and the bed density at the bottom is significantly higher than that in the upper dilute-phase zone. The density distribution of solids in furnace is similar to Qiu’s work [24], which means that the “core-annulus” two-phase flow model established in this study is reasonable.

Figure 4 shows the solid density distribution of the core and annulus area in the furnace at different conditions. In the “core-annulus” simulation model, the first core node represents the high-density bed area, so the density of the first core node is much higher than other nodes. In the dilute-phase area, solids gather near the water wall, which causes a higher solid density in the annulus than in the core area.

3.2. Simulation of Main Fuel Trip

MFT (main fuel trip) refers to the command to quickly cut off all fuel to the furnace and cause a necessary interlocking action when the safe operation conditions of the boiler cannot be met, and emergency shutdown is required so as to avoid potential hazards to the boiler and protect the safety of the boiler furnace, other equipment, and operators. After the boiler triggers MFT, the following actions will be accomplished as follows:

(1)

Cut off the fuel;

(2)

Cut off the limestone;

(3)

Trip the external bed heat exchangers;

(4)

Cut off the attemperator water;

(5)

Switch the air control model to manual;

(6)

Cut off the ash coolers.

Figure 5, Figure 6, Figure 7 and Figure 8 show the response of the main steam temperature, heated steam temperature, bed temperature, and liquid level of water storage tank when the boiler triggers MFT at 100% BMCR, 75% THA, and 50% THA, respectively. Figure 9 and Figure 10 show the response of fuel mass flow and feed water mass flow. When the boiler triggers MFT, the coal mass flow decreases to 0 rapidly. As the furnace of circulating fluidized bed boiler has extremely strong heat storage capacity, the steam still absorb heat from the furnace to maintain a certain degree of superheat after the coal feeding rate drops to zero. After the MFT is triggered, the boiler feed water decreases rapidly due to the removal of main fuel, thus maintaining the relative stability of the bed temperature and the temperature of the main steam and reheated steam. After 5 min, the MFT resets, the boiler starts the hot start-up process, the coal feed rate increases, and the ash flow of the external bed heat exchanger gradually increases as well. The temperature and mass flow rate of the main steam and reheated steam recover to 100% BMCR after about 60 min under 100% BMCR condition. The temperature and mass flow rate of the main steam and reheat steam will recover to 75% THA after 40 min under 75% THA condition. After 30 min under 50% THA condition, the moisture separator gradually switches from wet state to dry state, and the ash flow of the external bed heat exchanger gradually increases. After about 50 min, the temperature and flow of the main steam and reheat steam return to 50% THA. After the boiler triggers MFT, although the fuel is cut off, the temperature of the bed is still more than 500 °C. There is a large amount of hot solid materials in the bed, which provide huge heat storage capacity and shorten the hot start-up duration compared with a pulverized coal-fired boiler. The results show that under the given control strategy of coal and feed water supply, the CFB boiler can work at safe conditions during MFT and quickly complete the start-up process after MFT resets.

From the results under 100% BMCR, 75% THA, and 50% THA conditions, it can be seen that after triggering MFT under 50% THA condition, the steam moisture separator switches from dry state to wet state and then gradually switches from wet state to dry state after resetting MFT. Frequent dry and wet state conversion will cause great thermal stress in the moisture separator, which will affect the safety of the equipment. Therefore, the triggering of MFT under low load should be avoided as possible.

3.3. Simulation of Boiler Trip

When the boiler triggers the BT (boiler trip), it must shut down the boiler to avoid safety issues, which means all materials entering furnace must be cut off. After the boiler triggers BT, the following actions will be accomplished as follows:

(1)

Cut off the fuel;

(2)

Cut off the limestone;

(3)

Trip the external bed heat exchangers;

(4)

Shut down the attemperator water;

(5)

Switch the air control model to manual;

(6)

Cut off the ash coolers;

(7)

Shut down the primary and secondary air fan.

Figure 11, Figure 12, Figure 13 and Figure 14 show the responses of the main steam temperature, heated steam temperature, bed temperature, and liquid level of water storage tank when the boiler triggers BT at 100% BMCR, 75% THA, and 50% THA, respectively. Figure 15 and Figure 16 show the response of fuel mass flow rate and feed water mass flow rate during the BT processes. It can be seen that compared with the boiler MFT, the temperature of the furnace is higher than that of MFT since the primary air fan, secondary air fan, and induced draft air fan are shut down during BT, and the flue gas cannot take away the heat from the furnace. Five minutes after BT is triggered, the boiler starts the hot start-up process. About 60 min after BT is at 100% BMCR, the temperature of the main steam and reheated steam recover to the rated value. The temperature of the main steam and reheated steam reach the rated value after about 55 min under 75% THA condition. About 8 min after the boiler triggers BT under 50% THA condition, the moisture separator gradually switches from dry state to wet state, and the liquid level of water storage tank is controlled at 6.5 m. After that, the boiler BT is reset, and the coal mass flow, air mass flow, and feed water mass flow gradually increase. After about 25 min, the moisture separator gradually switches from wet state to dry state, and the main steam temperature, heated steam temperature, and bed temperature also increase gradually. After 40 min, the performance of the boiler gradually recovers to 50% THA.

It can be seen from the BT results under 100% BMCR, 75% THA, and 50% THA conditions that, similar to the MFT results since the load is low under 50% THA working condition, the steam separator switches from dry state to wet state after triggering BT and then gradually switches from wet state to dry state after BT is reset. It is observed that the restart-up duration during the BT process is shorter than that of the MFT process. This is caused by the removal of air supply, which causes a higher bed temperature and provides more heat during the BT process.

4. Conclusions

In this study, we established a full-scale dynamic simulation model for a 660 MW ultra-super critical CFB bed boiler based our previous work, and we expanded the work scope of the simulation model to MFT and BT processes. The model was calibrated at 100% BMCR, 75% THA, and 50% THA conditions, respectively. The steady-state simulation results of the model showed that the errors of most parameters of the model were less than 1% at 100% BMCR, 75% BMCR, and 50% BMCR conditions, which presented a good simulation accuracy. In addition, the distribution of temperature and particle concentration in the furnace showed that the model established in this paper can accurately and reasonably reflect the characteristics of gas–solid two-phase flow in the furnace.

The simulation results of MFT at different working conditions showed that after triggering MFT, although the main fuel was cut off, there was still a large amount of unburned fuel in the furnace, which continued to burn, so the steam could still absorb heat from the furnace and maintain a certain degree of superheat. As the residual fuel in the furnace was burned out, the heat in furnace was gradually reduced, so the bed temperature, main steam, and reheat steam temperature were gradually reduced. At 50% THA working condition, after the boiler triggered MFT, the moisture separator gradually switched from dry state to wet state. After MFT was reset, the coal feed rate gradually increased, so the bed temperature, main steam temperature, and reheated steam temperature gradually increased and return to the settled value as well.

Compared with results of the boiler MFT, the bed temperature after the boiler triggering BT was higher than that of MFT process because the primary air fan, secondary air fan, and induced draft air fan were shut down, and the flue gas could not take away the heat from the furnace. Similar to MFT, at 50% THA condition, the moisture separator gradually switched from dry state to wet state after the boiler triggered BT. After BT was resettled, the bed temperature, main steam temperature, and reheated steam temperature increased continuously and return to the settled value.

Combined with boiler MFT and BT working conditions, it could be found that at 50% THA working condition, the moisture water separator switched dry state to wet state, and the dry–wet conversion increases the thermal stress of the moisture separator. Therefore, it is necessary to avoid triggering boiler MFT and BT at low load to ensure the safe and stable operation of the boiler.

In this study, the dynamic characteristics during the BT and MFT process of a 660 MW ultra-supercritical CFB boiler were carried out, and the results showed high simulation accuracy compared with the designed parameters. Owing to the fact that the boiler is still in construction stage, there are no operational data to improve the accuracy of this model. Therefore, the model would be improved in further work through use of real operation data.