Saturday, March 9, 2019

INFLUENCE OF COOLING FLUID PARAMETER ON THE FLUID FLOW AND END PART TEMPERATURE IN END REGION OF A LARGE TURBOGENERATOR


In order to study the influence of the cooling fluid parameter on the fluid flow and end part temperature in the end region of the large turbogenerator, 330-MW water- hydrogen hydrogen-cooled turbogenerator is analyzed. The fluid velocity and pressure values from the flow network calculations are applied to the end region as boundary conditions and the losses obtained from 3-D transient electromagnetic field calculations are applied to the end parts as heat sources in the fluid and thermal coupling analysis.
After solving the fluid and thermal equations of fluid-solid conjugated heat transfer, the fluid velocity and end part temperature in the turbogenerator end region are gained under the different cooling fluid parameters. The influence of different fluid velocities and fluid temperatures in the water pipe inlet, and in the fan inlet on the fluid flow and end part temperature in the turbogenerator end region is researched. The calculation results of copper shield temperature are compared with the measured values. The calculation results coincident well with the measured values. These provide the important
reference for better cooling turbogenerator end region.

CHAPTER 1

INTRODUCTION

Turbogenerator is a kind of equipment which converts mechanical energy into electrical energy. When the turbogenerator runs, copper losses in the stator-end copper coil and current losses, in the end, parts are produced in the turbogenerator end region. These losses will be converted into the heat during the operation of the turbogenerator. On the one hand, these losses result in the temperature rise of the end parts, an, on the other hand, the heat of the end parts could be taken away by the cooling
fluid. The poor choice of cooling fluid parameters may lead to insufficient heat removal capability in the end region components. This will result in a very high temperature in the end region and shorten the service life of the turbogenerator. Therefore, it has an important engineering significance about studying the influence of cooling fluid parameter on the fluid ow and end part temperature in the end region of the large turbogenerator. Based on the results of the flow network and three-dimensional nonlinear transient electromagnetic field, the fluid and thermal equations of fluid-solid conjugated heat transfer, in the end, the region is calculated under the different cooling fluid parameters. The distribution laws of fluid velocity and end part temperature in the turbogenerator end region are researched under the different fluid velocities and fluid temperatures in the water pipe inlet and in the fan inlet.
In order to avoid confusion, fluid velocity and fluid temperature in the water pipe inlet are defined as water velocity and water temperature in the water pipe inlet. Fluid velocity and fluid temperature in the fan inlet are defined as gas velocity and gas temperature in the fan inlet in this paper. These provide an important reference for selecting a reasonable cooling fluid parameter in the larger capacity turbogenerator.

CHAPTER 2

Establishment of Flow Network and Calculation Results Based on the actual structure of a 330 MW water-hydrogen- hydrogen-cooled turbogenerator, flow network in the half of a turbogenerator is established. Fig.2.1 shows ventilation cooling system of the turbogenerator half-axial segment. Fig.2.2 shows a flow network in one half of the turbogenerator. After solving flow network equations within the half of turbogenerator under rated-load conditions, the distributions of gas rate and pressure in the turbogenerator are gained. The gas velocity in the fan inlet is 32.43 m/s. Fig.2.3 shows the pressure values in all the outlets of the end region.


                    Figure 2.1: Ventilation cooling system of a turbogenerator half axial segment

The gas velocity and pressure values from the flow network calculations are applied
to the end region of a 330 MW water- hydrogen-hydrogen cooled turbogenerator as boundary conditions. It provides an important theoretical basis for studying the influence of cooling fluid parameter on fluid velocity and end part temperature, in the end, the region of the turbogenerator.


                            Figure 2.2: Flow network in one half of the turbogenerator



                           Figure 2.3: Pressure values in all the outlets of the end region

CHAPTER 3

Fluid-Thermal Coupling Analysis Model for Turbogenerator End Region

3.1 Model for Coupling Analysis Based on the actual shape and size of the 330 MW turbogenerator end region, 3-D fluid and thermal coupling analysis model of the turbogenerator end region is es-
established. Fig.3.1 gives the turbogenerator end parts and end region inlets. Fig.3.2 shows the end region outlets.


      Figure 3.1: Turbogenerator end parts and region inlets

The gas velocity in the fan inlet of the turbogenerator is 32.43 m/s under rated-load
conditions, which is obtained from the calculated results of the flow network. The gas
temperature in the fan inlet is 41 Celsius. zone outlet between the end core and wind
board, air-gap outlet, No. 2 and No. 4 cold-gas zone outlets in the turbogenerator end


                                                     Figure 3.2: End region outlets

region are given as pressure-outlet and the pressure values are obtained from the calculated results of the flow network under rated-load conditions. The water temperature in the water pipe inlet is 40 Celsius and water velocity in the water pipe inlet is 1 m/s in the turbogenerator end region under rated-load conditions. These are applied as boundary conditions of the 3-D fluid and thermal coupling analysis model of the turbogenerator end region.

3.2 Determination of Eddy Current Losses of End Parts in the Turbogenerator End Region

Based on the classical electromagnetic theory, mathematical and geometric models of the 3-D nonlinear transient electromagnetic field are established in the turbogenerator end region. Fig.3.3 shows the 3-D electromagnetic solving region for the turbogenerator end region. The equations for the 3-D nonlinear transient field have been calculated. According to Equation, the eddy current loss in the end parts can be determined once the eddy current density has been obtained:

where Pe is the eddy current loss (in W), Tc is the period of time, k is the total number of elements in the various end parts, i is the element volume, Ji is the element eddy current density, and r is the end-part conductivity.

Fig.3.4 shows the current losses of end parts in the turbogenerator end region. It
can be seen from Fig.3.4 that the largest eddy current loss which locates copper shield



                     Figure 3.3: 3-D electromagnetic solving region of the turbogenerator end region

inner circle zone is 12872 W. However, the smallest eddy current loss which locates press finger is 946 W.


                         Figure 3.4: Current losses of end parts in the turbogenerator end region

CHAPTER 4

The Influence of Water Parameter in the Water Pipe Inlet on End Part Temperature in the Turbogenerator End Region

4.1 Influence of Water Velocity

After solving the fluid and thermal equations of fluid-solid conjugated heat transfer, the influence of water velocity in the water pipe inlet on the temperature of end parts is studied. Fig.4.1 shows the highest temperatures of end parts under the different water velocities in the water pipe inlet. Fig.4.1 shows the highest temperatures of end parts decrease as water velocity in the water pipe inlet increases. When water velocity in the water pipe inlet is between 0.5 m/s and 3 m/s, the highest temperatures of the copper shield, stator-end copper coil, press finger, and stator-end winding insulation decrease obviously as water velocity in the water pipe inlet increases. However, the highest temperatures of these end parts decrease a little after water velocity in the water pipe inlet is larger than 3 m/s (between 3 and 6 m/s). Since the stator-end copper coil is cooled by the water in the water pipe inlet, the highest temperature of the stator-end copper coil decreases obviously as water velocity in the water pipe inlet increases. The highest temperature of stator-end copper coil drops to 47.4 Celsius when water velocity in the water pipe inlet is 3 m/s. The highest temperature (when water velocity in the water pipe inlet is 3 m/s) of stator-end copper coil is 16.2 Celsius lower than that of the stator-end copper coil when water velocity in the water pipe inlet is 0.5 m/s. The highest temperature of the press plate decreases a little as water velocity in the water pipe inlet increases.

Fig.4.2 shows the highest temperature of copper shield drops to 58.2 Celsius when water velocity in the water pipe inlet is 6 m/s. The highest temperature of the copper shield is 2 Celsius lower than that of the copper shield when water velocity in the water pipe inlet is 1 m/s. Fig.4.3 shows the highest temperature and average temperature of the stator-end copper coil are 47 and 43.2 Celsius when water velocity in the water pipe inlet is 6 m/s. The highest temperature and the average temperature of the stator-end copper coil (when water velocity in the water pipe inlet is 6 m/s)


Figure 4.1: Highest temperatures of end parts under the different water velocities in
the water pipe inlet.

are 7 and 8.9 Celsius lower than those of stator-end copper coil when water velocity in the water pipe inlet is 1 m/s, respectively. As the cooling water flows in the water pipe, the water temperature gradually increases and the cooling effect of water becomes worse. It results in a higher temperature in the involute segment of stator-end copper coil.

4.2 Influence of Water Temperature

 In order to study the influence of water temperature in the water pipe inlet on end part temperature, fluid and thermal equations of fluid-solid conjugated heat transfer are calculated when water temperatures in the water pipe inlet are 35, 37.5, 40 Celsius (original scheme), 42.5, and 45 Celsius, respectively. As water temperature in the water pipe inlet increases, it has little effect on the highest temperatures and average temperatures of the copper shield, press finger, and press plate. However, the increased temperatures of the stator-end copper coil and stator-end winding insulation are almost proportional to increased water temperature in the water pipe inlet. Fig.4.4 gives the highest temperatures of stator-end copper coil and stator- end winding insulation under the different water temperatures in the water pipe inlet.



Figure 4.4: Highest temperatures of the stator-end copper coil and stator-end winding
insulation under the different water temperatures in the water pipe inlet coil


CHAPTER 5

The Influence of Gas Parameters in the Fan Inlet on End Part Temperature in Turbogenerator End Region

5.1 Influence of Gas Velocity

Based on the theory of fluid mechanics and heat transfer, the influence of gas velocity in the fan inlet on the gas velocity and end part temperature in the end region is researched: Case One, gas velocity in the fan inlet decreases by 15%; Case Two, gas velocity in the fan inlet decreases by 10%; Case Three, gas velocity in the fan inlet decreases by 5%; Case Four, gas velocity in the fan inlet increases by 5%; Case Five, gas velocity in the fan inlet increases by 10%; Case Six, gas velocity in the fan inlet increases by 15%. After solving flow network equations, the fan inlet velocities and pressure values of the end region outlet in the 3-D fluid and thermal coupling analysis model of the turbogenerator end region are obtained under the different cases, as shown in fig.5.1 and Fig.5.2. It can be seen from Fig.5.2 that all the pressure values of the end region outlet increase gradually as gas velocity in the fan inlet increases. As gas velocity in the fan inlet increases, the pressure values of No. 2 cold-gas zone outlet, No. 4 cold-gas zone outlet, and air-gap outlet increase obviously, while the pressure value of zone outlet between the end core and windboard increases a little. These provide the boundary condition for 3-D fluid and thermal coupling analysis model of the turbogenerator end the region to study the influence of gas velocity in the fan inlet on the gas velocity and end part temperature in the turbogenerator end region. 
To facilitate analysis, the ventilation ducts between the copper shield inner circle zone, copper shield transitional circle zone, copper shield external circle zone and press plate are marked by Part AB, Part BC, and Part CD, respectively. The arrow direction of the dotted line represents the direction of gas flow, as shown in Fig.5.3. 

Fig.5.4 shows average gas velocity in the ventilation duct between the copper shield



and press plate rest increases gradually and then decreases gradually along the direction of gas flow from Part AB to Part CD. As gas velocity in the fan inlet increases, average gas velocity in the ventilation duct between the copper shield and press plate increases gradually. When gas velocity in the fan inlet increases by 15%, the average gas velocities in the Part AB, Part BC, and Part CD between the copper shield and press plate reach the highest values and they are 39.75, 67.81, and 49.1 m/s, respectively. When gas velocity in the fan inlet decreases by 15%, the average gas velocities in the Part AB, Part BC and Part CD between the copper shield and press plate reach the lowest values and they are 28.98, 46.02, and 32.39 m/s, respectively.

5.2 Influence of Gas Temperature

In order to study the influence of gas temperature in the fan inlet on end part temperature, the highest temperatures of the end parts under the different gas temperatures in the fan inlet are given, as shown in Fig.5.5. The figure shows the highest temperatures of end parts increase almost linearly as the gas temperature in the fan inlet increases. As the gas temperature in the fan inlet increases, the highest temperatures of the copper shield, press plate, press finger, and stator-end winding insulation increase
obviously, especially to the highest temperature of the copper shield. The reason is that copper shield completely contacts with the cooling gas. The temperature of the cooling gas has a great effect on the temperature of the copper shield. Since the stator-end copper coil is wrapped in the stator-end winding insulation and doesn't contact with the cooling gas, the gas temperature in the fan inlet has a small effect on the stator-end copper coil. The highest temperature of the stator-end copper coil increases a little like the gas temperature in the fan inlet increases. The highest temperatures of the copper shield, press plate, press finger, stator-end winding insulation, and stator-end copper coil increase by 10, 7.5, 8.5, 5.7, and 1.7 Celsius, respectively, when the gas temperature in the fan inlet increases from 36 to 46 Celsius.
CHAPTER 6

Calculated Results and Measured Values in the Turbogenerator End Region


After solving the fluid and thermal equations of fluid-solid conjugated heat transfer, the position of the highest temperature in the turbogenerator end region is determined. The highest temperature in the turbogenerator end region appears in the copper shield inner circle zone. The highest temperature and average temperature of the copper shield inner circle zone are 60.2 and 57.3 Celsius under rated-load conditions. In order to verify the accuracy of the calculation results, the temperature of the copper shield is measured online at the power plant. To facilitate analysis, the copper shield inner circle zone is divided into three equal volumes along the generator axial direction, which is represented by M1, M2, and M3, as shown in Fig.6.1. Three temperature sensors are embedded evenly at a 120 angle along the circumference in the same radius of the copper shield inner circle zone. They are embedded in Region M1 close to Region M2. At the same load and power factor (rated power is 330 MW, rated voltage is 20 kV, rated current is 11200 A, and power factor (lagging) is 0.85), measured values and calculated results are obtained. Fig.5.2 shows measured values and calculated results in the Region M1 of the copper shield inner circle zone. The calculated results and three measured values are compared. The calculated results are closer to the measured values. In addition, it not only calculates this one geometry but also calculates lots of different geometries of practical engineering cases using the method. For example, fluid velocities and temperature distributions of parts are obtained in the wind-driven generator, permanent magnet generator, large hydro-generator, and induction motor using the method of this paper. The calculated temperature results and measured values are compared in the different geometries of these electrical machines. The errors meet the engineering requirement in these different engineering cases. It shows the calculation method is credible.
The reasons for the error between the calculated results and the measured values are that different gap distances between the water conduits of the stator-end windings and between the stator-end windings, together with the rough surface of the copper shield and the installation error, result in an asymmetric velocity distribution in the ventilation duct between the copper shield and the press plate.
CHAPTER 7

CONCLUSIONS

When the water velocity in the water pipe inlet is smaller than 3 m/s, the temperatures of the copper shield, stator-end copper coil, press finger, and stator-end winding insulation decrease obviously as water velocity in the water pipe inlet increases. However, the temperatures of these end parts decrease a little after water velocity in the water pipe inlet is larger than 3 m/s. The highest temperature (when water velocity in the water pipe inlet is 3 m/s) of stator- end copper coil is 16.22 Celsius lower than that of the stator-end copper coil when water velocity in the water pipe inlet is 0.5 m/s. When water velocity in the water pipe inlet is 3 m/s, it can reduce effectively the temperature of end parts in the turbogenerator end region. As water temperature in the water pipe inlet increases, it has little effect on the temperatures of the copper shield, press finger, and press plate. The increased temperatures of the stator-end copper coil and stator-end winding insulation are almost proportional to increased water temperature in the water pipe inlet. The water temperature in the water pipe inlet has a great effect on the temperature of stator- end copper coil. However, it has a small effect on the temperature of stator-end winding insulation. As the gas temperature in the fan inlet increases, the highest temperatures and average temperatures of end parts increase almost linearly. The temperature of end parts can be reduced obviously when the water velocity in the water pipe inlet is 3 m/s and gas velocity in the fan inlet is 37.29 m/s. This could enhance the temperature distribution in the turbogenerator end region and could provide an important reference for the cooling system design in the turbogenerator.


REFERENCES

[1] Jichao Han, Baojun Ge, Dajun Tao, and Weili Li,, Influence of cooling fluid parameter on the fluid flow and end part temperature in end region of a large turbogenerator, IEEE Trans. Energy Convers., vol. 20, no. 1, pp. 166172, Mar. 2015.

[2] R. Albanese, F. Calvano, G. D. Mut, F. Ferraioli, A. Formisano, F. Marignetti, R. Martone, A. Romano, G. Rubinacci, A. Tamburrino, and S. Ventre,, Coupled three-dimensional numerical calculation of forces and stresses on the  end windings of large turbogenerators via integral formulation, IEEE Trans. Magn., vol. 48, no. 2, pp. 875878, Feb. 2012

[3] A. Shiri and A. Shoulaie,, Design optimization and analysis of single-sided linear induction motor, considering all phenomena, IEEE Trans.Energy Convers., vol. 27, no. 2, pp. 516525, Jun. 2012.

[4] Y. Wu and Y. Li,, Diagnosis of rotor winding inter-turn short-circuit in turbine generators using virtual power, IEEE Trans. Energy Convers., vol. 30, no. 1, pp. 183188, Mar. 2015.

[5] D. Ban, D. Zarko, and I. Mandic,, Turbogenerator end-winding leakage inductance calculation using a 3-D analytical approach based on the solution of Neumann integrals, IEEE Trans. Energy Convers., vol. 20, no. 1, pp. 98105, Mar. 2005.

[6] S. Nategh, O. Wallmark, M. Leksell, and S. Zhao,, Thermal analysis of a PMaSRM using partial FEA and lumped parameter modeling, IEEE Trans. Energy Convers., vol. 27, no. 2, pp. 477488, Jun. 2012.

[7] H. Jichao, L. Weili, W. Likun, Z. Xingfu, Z. Xiaochen, and L. Yong,, Calculation and analysis of the surface heat-transfer coefficient and temperature fields on the three-dimensional complex end windings of the large turbogenerator,, IEEE Trans. Ind. Electron., vol. 61, no. 10, pp. 52225231, Oct. 2014.




No comments:

Post a Comment