Fault simulation of boiler heating surface ash deposition in a power plant system

Abstract

The simulation model of a power generation system was developed based on EASY5 simulation platform. The performances of the power plant under the conditions of the furnace slagging and ash deposition of the heating surfaces in the boiler were simulated. The results show that the simulation model can reasonably reflect the characteristics of the power plant when each component is under fault conditions. Through fault simulation, the change of the performance parameters can be obtained, which can be used in fault diagnosis system as the diagnosis criterion for expert system.

Keywords

boiler / slagging / ash deposition / fault simulation

Introduction

As power equipment, the boiler has to experience operational problems such as corrosion and abrasion from gas, fire, ash, etc. It is prone to fault operation or accident such as furnace slagging, ash deposition on boiler walls. In recent years, many advances have been made in the field of coal quality, one of which has been the solution of ash-related problems in coal-fired boilers. The ash deposited on boiler walls in the radiant section of a furnace is generally referred to as slagging. Most commonly, slagging means ash fouling on water walls making up the furnace. Ash deposition on convection tube sections downstream of the furnace radiant zone is typically referred to as fouling.

The presence of ash deposits will lead to reduced heat transfer, impedance of gas flow, physical damage to pressure parts, corrosion and/or erosion of pressure parts in a boiler. In general, the ash deposition effects are interrelated. For example, ash deposition on convection tubes can reduce the cross-sectional flow area, increasing fan requirements and creating higher local gas velocities, which accelerates fly ash erosion [1,2]. These problems can result in unscheduled outages, reduced availability, costly modifications, and reduced generating capacity. A number of authors have indicated that ash deposition is a sizable problem, such as Devir [3], who states that slagging costs the global utility industry several billion dollars annually in reduced power generation and equipment maintenance.

A lot of researches have been conducted to investigate ash deposition and to control its influence. In the operation of a power plant, it would be of great importance to locate the deposition or other faults and to deal with the faults. Fault monitoring and diagnosis based on abnormal operating parameters are widely used. However, it would be expensive or difficult to obtain the operating parameters from experiments directly. In fact, it is quite difficult or dangerous to put the power generation unit in special conditions. Therefore fault simulation will be the best choice to understand the system performance under fault operation.

In this paper, the furnace slagging and ash deposition on the surface of super-heater, re-heater, economizer and air heater were calculated on the simulation platform of EASY5. Some useful criteria were proposed for the fault diagnosis in the operation of the boiler with fault simulation.

System modeling

System description

Figure 1 is a schematic diagram of the power plant system in Shanghai Baosteel Group Corporation. The steam turbine of a 350 MW power plant comprises high (HP), intermediate (IP) and low-pressure (LP) sections. In addition, the system includes eight steam extraction points and feed-water heaters to increase its thermal efficiency. Figure 2 illustrates the interaction between the major components in the boiler. The capacity of the boiler is 1160 t/h, which mainly burns coal, sometimes with byproduct of steel factory, such as blast furnace gas (BFG), coke-oven gas (COG) and so on.

Modular models

Based on the modular modeling theorem, the power plant system was divided into three subsystems: the boiler, the steam turbine, and the heat regenerative system. Several modules were built, such as a drum, an evaporator, a super-heater, a re-heater, an economizer, an air heater, a flue evaporator and a circulating water pump in the boiler subsystem. For the steam turbine subsystem, the turbine module and volume module were created [4-6]. The heat regenerative subsystem includes a condenser, a deaerator, three high pressure heaters, four low pressure heaters and a feed-water pump. In this paper, the evaporator and flue evaporator, the super-heater and re-heater, the circulating water pump and feed-water pump, the high and low pressure heater were treated as the same module based on the modular modeling theorem. Besides, the part between two steam extraction stages is seen as one turbine module, and connected by volume modules [7].

The mathematical models of the components were established based on the conservation law of mass, energy and momentum. The models for the drum, the evaporator, the super-heater/re-heater, the economizer, the pump and the deaerator can be referred to in Ref. [8]

Turbine

The same turbine model could be used in all of the cylinders in the steam turbine, and a coefficient could be introduced in the control stage of the high pressure cylinder to describe the opening level.

According to Friuli Greig formula, the mass flow in the turbine is captured as

(1)

The mass flow of the next stage after steam extraction

(2)w_{t\mathrm{2}}=w_t(\mathrm{1}-a)\mathrm{.}

The mass extracted would be

(3)w_{t\mathrm{3}}=w_{t}a,

where, a is a coefficient of steam extraction based on experimental data. Here, \mathrm{\Delta }n=\mathrm{0} for this model.

The efficiency of the turbine

(4)\eta =f(\pi ,n),

(5)\pi ={\displaystyle \frac{{\xi }_i}{{\xi }_o}}.

The outlet enthalpy

(6)h_o=h_i-\eta (h_i-h_{eb})\mathrm{.}

The export work

(7)P=w_t(h_i-h_o)\mathrm{.}

The entropy equation

(8)S=S(h_i,{\xi }_i),

(9)h_{eb}=h_{eb}(S,{\xi }_o).

Furnace

A simple model was used for the furnace applying mass and energy balance equations, ignoring the dynamic variation of the gas. The combustion process is much faster than the heat transfer between the gas and the metal wall, the wall and the steam/water; therefore static models were used. The model, with three kinds of fuels such as coal, BFG, and COG, was used in the plant [9,10].

Air flow

(10)w_a=w_{f1}{w}_{a1}+w_{f2}{w}_{a2}+w_{f3}{w}_{a3}.

The gas flow

(11)w_g=w_{f1}{\beta }_{\mathrm{1}}+w_{f2}{\beta }_{\mathrm{2}}+w_{f3}{\beta }_{\mathrm{3}}+w_a+w_{ao}.

The exit temperature of the gas

(12)T_{go}={\displaystyle \frac{E}{w_g{C}_g}},

where

(13)E=w_{f1}{Q}_{\mathrm{1}}{\xi }_{\mathrm{1}}+w_{f2}{Q}_{\mathrm{2}}{\xi }_{\mathrm{2}}+w_{f3}{Q}_{\mathrm{3}}{\xi }_{\mathrm{3}}+(w_a{T}_a+w_{ao}{T}_{ao})C_a.

Condenser

In general, the condenser works at a low pressure region, where the magnitude of derivative of the water enthalpy and density on the pressure would be great. To simplify the dynamic modeling of the condenser, the pressure of the condenser was assumed to be known.

The heat transfer equation

(14)T_{co}={\displaystyle \frac{Q}{w_w{C}_w}}+T_{ci},

(15)Q=w_i{h}_i-w_o{h}_o.

Volume model

The volume model was introduced to decouple the iteration algorithm between the pressure and the flow rate. The alteration of the pressure in the volume caused by the input/output mass was considered, and the influence of the enthalpy was ignored [6].

(16){\displaystyle \frac{dp}{dt}}={\displaystyle \frac{RT(w_i-w_o)}{V}}.

Heater

The boiler system had much greater inertia than heaters, so the quasi-static models were used for these heaters. The pressure of the heater is the steam extraction pressure.

The mass conservation

(17)w_{si}+w_{wi}=w_{so}+w_{wo}.

The energy conservation

(18)w_{si}{h}_{si}+w_{wi}{h}_{wi}=w_{so}{h}_{sb}+w_{wo}{h}_{wo}+Q.

The heat release of extracted steam

(19)Q={\displaystyle \frac{w_{si}(h_{si}-h_{wb})}{\mathrm{1}-\gamma }}.

The heat absorption of feed-water

(20)Q=w_g{C}_g(T_{go}-T_{gi}).

The pressure loss

(21)p_i=p_o+\mathrm{\Delta }p,

(22)\mathrm{\Delta }p=KC{w}_{fw}^{\mathrm{2}},

where KC is the pressure loss coefficient.

Air heater

The arithmetic means of the input and output temperature was used to calculate the heat transfer. According to the energy balance and heat transfer equation, the mathematical models can be depicted as follows.

The heat store in metal wall

(23)w_m{C}_m{\displaystyle \frac{d{T}_m}{dt}}=Q_g-Q_a.

The heat release of gas

(24){\displaystyle \frac{d{T}_{go}}{dt}}={\displaystyle \frac{w_g{C}_g(T_{gi}-T_{go})-Q_g}{w_g{C}_g}}.

The heat absorption of air

(25)Q_a=K_a{A}_a\left(T_m-{\displaystyle \frac{T_{ai}+T_{ao}}{\mathrm{2}}}\right).

The heat store in air

(26)Q_a=w_a{C}_a(T_{ao}-T_{ai}).

The heat store in gas

(27)Q_g=K_g{A}_g\left({\displaystyle \frac{T_{gi}+T_{go}}{\mathrm{2}}}-T_m\right).

System model

The module library for each component was created on EASY5 simulation platform. Based on the ready module library, the system simulation model for one power generation unit was built in Shanghai Baosteel Group Corporation, as displayed in Fig. 3. Figure 4 presents the detailed simulation model for the boiler.

Fault simulation

Based on the above simulation model, the system performance of the power plant was simulated. The simulation conditions adopted are as follows:

The power of the system was set to 350 MW. The drum water level in boiler was controlled by adjusting the feed-water flow into the boiler using a proportional-integral-derivative (PID) controller. The pressure of the drum was set to 16.92 MPa by regulating the fuel flow. The main steam and reheat steam temperature were limited below 540°C. The feeding fuel for the boiler was coal and the ambient temperature was 25°C.

The common manifestation of a deposition problem is the reduced heat transfer [1], so the fault would be simulated by changing the heat transfer coefficient. Here, the fault degree was defined as the percentage of the heat transfer coefficient change based on the normal value.

Ash deposition or slagging on single heating surface

The following results were listed when the fault occurred on one single component in the power plant. The operating parameters were obtained based on the above simulation conditions.

Figure 5 indicates that the boiler efficiency drops and the exhaust gas temperature increases with the increase of the fault degree, and it has the same trend under individual ash fouling in different heat transfer areas of boiler. It is important to note that, the efficiency dropping rate caused by furnace slagging has increased by 10% fault degree in Fig. 5(a). The reason for this is that the control system takes effect, resulting in more heat loss due to exhaust gas, as shown in Fig. 5(e), when the reheat steam temperature is overheating, as shown in Fig. 5(c).

The results also reveal that there are certain differences among the simulations of different fouling components, which have been classified and depicted in Table 1.

Furnace slagging results in the heat absorption shift to the later heating surfaces. So the steam temperatures out of the boiler get higher, and the heat absorption in the economizer increases from high gas temperature [11]. The characteristic changes of operating parameters under the fouling conditions in different components can be used as the diagnosis basis, especially, when there is not enough experimental data to build up the fault diagnosis programs.

Ash deposition or slagging on more heating surfaces

Actually, the ash deposition phenomena in the boilers occur in many tubes instead of only in one section. So it is of significance to simulate ash fouling on more heating surfaces simultaneously.

Ash deposition on reheater when superheater in fault condition (ash deposition only)

The system model was calculated under the same condition as before. The superheater was supposed to be in fault condition when the heat transfer coefficient decreased more than 10%. Figure 6 shows the results of the operating parameters’ trends with the change of re-heater heat transfer coefficient when the superheater is in fault conditions (the heat transfer coefficient decrease 0%, 10%, 20%).

Figure 6(c) indicates that the reheat steam temperature is mainly influenced by the reheater. The fouling condition of the superheater does not affect the reheat steam temperature significantly. It can be observed in Fig. 6 that the fault degree of the superheater leads other parameters to more unfavorable direction. The curves of these parameters are shifted in parallel.

Ash deposition on economizer when superheater and reheater in fault condition (ash deposition only)

The simulation condition was kept the same. The power was set to 350 MW. The steam pressure in the drum was 16.92 MPa. All the heat transfer coefficient of the reheater, the superheater and the economizer would be changed. The coefficient of the economizer was decreased by degrees when the superheater and the reheater were in fault condition (the heat transfer coefficient of super-heater and re-heater reduced 0%, 10%).

It can be observed in Fig. 7 that the feed-water temperature out of the economizer stands alone with the super-heater and re-heater condition. The two curves, before and after the super/reheater fault, coincide, as shown in Fig. 7(d). And the other parameters are similarly shifted in parallel.

Conclusions

The module library of thermal power system was created on EASY5 simulation platform, and the simulation model for one power generation unit in Shanghai Baosteel Group Corporation was built.

Slagging and fouling conditions of heating surfaces in the boiler were simulated. The model can react to the disturbance properly. The results show that typical characteristic parameters can reflect different fault conditions. Fault simulation on several faults in different heating surfaces indicates that some parameters are more unfavorable as more faults happen, while some parameters are only affected by one fault.

The parameters trends from fault simulation can be used in fault diagnosis system, providing a useful tool for establishing the expert knowledge library.

References

[1] Borio R W, Levasseur A A. Overview of coal ash deposition in boilers. In: 188th National Meeting of the American Chemical Society, Philadelphia, USA, 1984, 193-203

[2] Harding N S, O’Connor D C. Ash deposition impacts in the power industry. Fuel Processing Technology, 2007, 88(11,12): 1082-1093

[3] Devir G P, Pohl J H, Creelman R A. Development of quantitative techniques to characterize mineral matter transformations in a pulverized coal fired boiler. In: Proceedings of the, 2001 International Joint Power Generation Conference, New Orleans, LA: ASME, 2001, 1-11

[4] Qian Y, Liu Y W. Modular modeling method and its implementation on the simulation platform on EASY5. Computer Simulation, 2007, 24(8): 85-89 (in Chinese)

[5] Su M. A volume component model and its solution for power system simulation. Power Engineering, 1998, 18(2): 75-78, 82 (in Chinese)

[6] Su M, Weng S L. Modularized simulation modeling for a large steam turbine. Journals of System Simulation, 1998, (3): 30-34 (in Chinese)

[7] Fan L. Foundation and simulation of general module of steam turbine system. Dissertation for the Master’s Degree, Chongqing: Chongqing University, 2007

[8] Liu L X. Thermal system simulation and modeling technique based on general purpose platform-application in gas-steam combined cycle modular simulation. Disseration for the Master’s Degree. Shanghai: Shanghai Jiao Tong University, 2002

[9] Liu Y Q, Wang Y X, Wang H F, Li P. Experimental investigations on combustion characteristics of the blast furnace gas. In: 3rd International Conference on Bioinformatics and Biomedical Engineering (ICBBE), Beijing, 2009

[10] Flynn M E, O’Malley M J. A drum boiler model for long term power system dynamic simulation. IEEE Transactions on Power Systems, 1999, 14(1): 209-217

[11] Valero A, Cortes C. Ash fooling in coal-fired utility boilers. Monitoring and optimization of on-load cleaning. Progress in Energy and Combustion Science, 1996, 22(2): 189-200

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg