Thermalanalysisofahigh-speedgenerator.pdf资源-CSDN文库

需积分: 9 26 浏览量 2021-02-18 11:10:35 上传评论收藏 455KB PDF 举报

### 高速发电机热分析的关键知识点 #### 一、引言与背景本文主要探讨了在结合微型涡轮机的联合热电生产（CHP）系统中，高速永磁同步发电机的热分析问题。该研究既包括理论计算也包含实验测量，旨在验证并优化发电机在高速运行时的性能及热稳定性。 #### 二、研究对象概述 1. **目标功率与转速**：本研究所涉及的发电机目标输出功率为110kW，设计转速为70000rpm。 2. **微涡轮机与CHP单元**：该发电机计划与微型涡轮机集成，作为CHP单元的一部分，以实现高效能源利用。微涡轮机的优势在于体积小、安装灵活，适用于偏远地区或电网承载能力有限的区域。 3. **热分析的重要性**：对于此类高速运转的设备而言，热管理至关重要。过高的温度不仅会影响发电机效率，还会缩短其使用寿命。 #### 三、发电机结构与特点 1. **集成式设计**：该发电机与微型涡轮机紧密集成，减少了额外轴承的需求，降低了整体重量与尺寸。 2. **永磁转子**：采用永磁材料制成的转子，具有较高的能量密度和良好的机械强度，适合高速运行。 3. **低损耗铁心与利兹线圈**：为了降低铁损和铜损，采用薄片低损耗电工钢制造定子铁心，并使用利兹线圈以提高高频特性。 4. **水冷系统**：虽然发电机本身效率较高，但由于高速运行产生的热量较大，因此采用了水冷系统来确保温控效果。 #### 四、热分析方法 1. **损失分布计算**：通过理论计算方法确定发电机内部各部件的能量损失分布情况，如铁损、铜损等。 2. **热量计测量**：进行实际测量以验证理论计算结果，确保数据的准确性。 3. **温度测量**：在发电机运行过程中记录关键部位的温度变化，以便于评估其热稳定性。 4. **铜屏蔽层的作用**：研究表明，在转子中引入铜屏蔽层可以有效降低损耗至可接受水平，这对于提高发电机的整体效率和稳定性至关重要。 #### 五、结论与展望 1. **理论与实践相结合**：本研究通过理论计算与实验验证相结合的方式，为高速发电机的设计提供了重要参考。 2. **优化建议**：通过引入铜屏蔽层来降低损耗，是一种有效的优化措施，有助于提高发电机的工作效率和热稳定性。 3. **未来发展方向**：随着技术的进步，进一步探索更高效的冷却技术和新材料的应用将成为未来研究的重点。高速永磁同步发电机的热分析是确保其稳定高效运行的关键环节。通过对损失分布、温度变化以及铜屏蔽层效应的研究，不仅可以提升发电机的设计水平，还能够推动微涡轮机CHP系统整体性能的提升。

资源推荐

资源详情

资源评论

Thermal analysis of a high-speed generator

O. Aglén

Dept. of Electrical Engineering

KTH The Royal Institute of Technology

SE 100 44, Stockholm, Sweden

Å. Andersson

ABB Automation Technologies

Motors & Machines / LV Motors

SE 721 70, Västerås, Sweden

Abstract

- The machine under study is a permanent magnet

synchronous generator intended for operation at high speeds

together with a micro-turbine in a Combined Heat and Power

(CHP) unit. Target power is 110 kW at 70 000 rpm. The study is

both theoretical and experimental since it includes calculation of

loss distribution and thermal analysis together with calorimetric

measurements of losses and temperature measurements. In this

paper the thermal model is described. Using results from loss

calculation and calorimetric measurements a thermal analysis is

carried out. It is shown that introducing a copper screen in the

rotor reduces the losses to acceptable values. A limit for the

allowable rotor losses is given.

I. INTRODUCTION

In areas where the grid has reached its power limit, in

remote areas or in countries with low transmission

capabilities, installing small power units, where they are

needed, can be more cost efficient than building large power

plants and power lines, at least it is much quicker. A compact

microturbine Combined Heat and Power (CHP) unit can offer

low emissions, fuel flexibility and high efficiency. The

system is described further in [1]. It uses a high-speed

generator operating in the speed range 50 000 rpm to 70 000

rpm. A thermal analysis based on results from calculations

and measurements of one generator prototype is carried out.

The thermal model is described in this paper.

II. T

HE GENERATOR

The electric power is generated in a high-speed permanent

magnet synchronous generator, which is integrated with the

microturbine. The rotor is suspended by one bearing on each

side of the permanent magnet rotor and there are no

additional bearings on the turbine shaft. The stator core is

made with thin laminations of low-loss electrical steel and the

winding is made with Litz-wire to achieve good high-

frequency characteristics. The four-pole design contributes to

short end-windings that make possible a short distance

between bearings. Although the efficiency is high, water-

cooling is used to keep the temperature low in the winding as

well as in the magnets. Together with vacuum pressure

impregnation and thicker than standard insulation this ensures

a long lifetime of the winding.

The rotor consists of a magnetic steel body with surface

mounted permanent magnets as shown in Fig. 1. A carbon-

fiber bandage retains the magnets even at 70000 rpm.

TABLE I GENERATOR DATA

Stator core outer diameter 135 mm

Stator core length 160 mm

Rotor outer diameter 66 mm

Rated voltage 500 V

Rated speed 70 000 rpm

Rated power 110 kW

An important issue is to ensure that the rotor never reaches

the temperature that would de-magnetize the magnets or

drastically reduce the lifetime of the carbon-fiber bandage.

This is done both by reducing the rotor losses and by

providing efficient cooling in the air-gap. Two main factors,

air friction and asynchronous mmf-waves in the air-gap cause

a temperature rise in the rotor. The latter are mainly caused

by stator current time harmonics. Since the carbon-fiber

bandage acts like a thermal insulator, the rotor is more

sensitive to current harmonics than a rotor in a conventional

machine.

Conceptual methods of rotor loss reduction are filters,

special modulation patterns in the power electronics and

multilevel converters. These methods reduce the effect of

harmonics in the stator current. In principal the problem can

be solved with a filter only. If the output current is perfectly

sinusoidal the rotor losses due to time harmonics are zero.

Although possible it is not always the most cost efficient

method.

The machine under study was made to operate together

with a, more or less, standard three-phase converter without a

filter for current harmonics.

Magnetic shaft

Screen

(optional)

NdFeB magnet

Pole separator

(could be part of

magnet or shaft)

Center hole

(optional)

Retaining cylinder of

composite bandage

Magnetic shaft

Screen

(optional)

NdFeB magnet

Pole separator

(could be part of

magnet or shaft)

Center hole

(optional)

Magnetic shaft

Screen

(optional)

NdFeB magnet

Pole separator

(could be part of

magnet or shaft)

Center hole

(optional)

Retaining cylinder of

composite bandage

Fig. 1. Cut-away view of the generator rotor

547

To keep the losses in the converter at a reasonable value,

switching frequency was limited to 15 kHz, although not

fixed at that value due to hysteresis band current control. The

focus has therefore been on rotor loss reduction and improved

cooling. Methods include:

•Shield the magnets

•Increased transient inductance

•Increased air-gap length

•Coolant flow (air-gap, bearings, stator)

•Cooling surfaces

•Magnet geometry and resistance

It is shown that introducing a copper screen in the rotor

reduces the losses to acceptable values. This is in line with

results from earlier investigations [2]. The losses are located

in the screen and losses in the magnets are negligible. This

way the geometry of the magnets can be optimized with

respect to manufacturing and not to loss reduction.

Consequently the magnet geometry was not investigated with

respect to losses in this study.

Loss calculation is only briefly described here. A more

thorough explanation was given in a previous paper [3].

III. L

OSS CALCULATION

A. Air friction losses

Air friction was calculated using equations found in [4].

The power to overcome drag resistance of a rotating cylinder

lrCP

πρω

= (1)

where ρ is the density of the fluid, in this case air, ω is the

angular velocity, l is the length of the cylinder and r is the

radius. C

, the friction coefficient, depends on the structure of

the surface. In this case the surface is assumed to be smooth.

A rough surface gives a higher friction coefficient.

The friction loss at the rotor ends is calculated using

()

553

, aendfendf

rrCP −=

ρω

(2)

The equation calculates the sum of the losses at both rotor

ends since it is valid for a disc rotating in free space

Since the calculation of air friction losses is very difficult a

separate test was made to determine the friction losses in the

generator. This method was described in a previous paper [5]

and the result showed that calculations were reasonably

accurate. In Table II the air friction losses are presented in

subdivided form, calculated with 10 g/s cooling flow

corresponding to test conditions. The largest portion is found

in the air gap. Section A refers to the part of the rotor surface

that is outside the air gap.

TABLE II. SUBDIVIDED AIR FRICTION LOSSES IN WATTS

Losses 51000 rpm 61000 rpm 70000 rpm

Air gap 349 512 690

Section A 94 153 227

Rotor ends 59 92 139

Total 524 757 1056

Test 540 820 1150

The most probable reason for the difference between

measured and calculated values of air friction could be that

the carbon fiber surface is not as smooth as the surface of the

cylinder used to obtain the friction loss coefficients.

The axial flow rate has an important influence on the

losses. In Table III one can see that to improve efficiency the

flow of cooling air should be as low as possible. However

that means also the rotor losses have to be kept low.

TABLE III. AIR FRICTION LOSSES IN THE AIR GAP IN WATTS

Axial flow 51000 rpm 61000 rpm 70000 rpm

0 g/s 171 264 384

10 g/s 349 512 690

20 g/s 524 818 1087

B. Bearing losses

The bearings are oil-lubricated and oil-cooled rolling

element bearings. The authors have not calculated the friction

losses in them but according to the manufacturer the losses

should be about 1 kW at 70000 rpm. About the same value is

obtained from measurements.

C. Iron losses

Iron losses in the rotor due to fundamental flux are

assumed to be zero. In the stator they were calculated using

the measured no-load losses as a base value and then they

were scaled with B

. As a measure of the flux density B the

air gap voltage is often used. Because of large leakage

reactance in the generator, using the air gap voltage would

over estimate the losses in motor operation and under

estimate in generator operation. How to compensate for this

is shown in [3]

D. Stator winding losses

Stator winding losses were obtained by simple I

R-

calculation with the nominal resistance corrected for

temperature. For frequencies below 12 kHz the resistance

factor R

/ R

is less than 1.01 so the influence of skin effect

can be neglected for the fundamental. At 50 kHz frequency

the factor is 1.16.

E. Harmonic losses in the rotor

The harmonic losses in the rotor are mainly located in the

Cu-screen. Using the measured values of stator current

harmonics the rotor losses were calculated using the Finite

Element Method (FEM). In Table IV the present rotor design

(Design A) is compared with one design without the screen

548

剩余7页未读，继续阅读

评论收藏

内容反馈

bigface1

粉丝: 0
资源: 10

Thermal analysis of a high-speed generator.pdf

最新资源

Thermal analysis of a high-speed generator.pdf

ISO 13400-4-2016_Part 4：Ethernet-based high-speed data link connector.pdf

hsdpa-hsupa-for-umts-high-speed-radio-access-for-mobile-communications.pdf

clement-sd-mmc-high-speed-support-in-linux-kernel.pdf

Study on the saturation characteristics of high-speed uni-traveling-carrier photodiodes based on field screening analysis

High-speed

Highspeed

mybatis-plus最新代码生成器项目源码 ：mybatis-plus-generator.zip

mybatis-plus-generator-3.2.0.jar

pg057-fifo-generator.pdf

mybatis-plus-code-generator-3.5.4.1-1.jar

mybatis-generator-core-1.3.5.jar

Performance analysis of an integrated scheme in optical burst switching high-speed networks

Design of high speed Energy-Efficient SAR ADC_劉純成.pdf

tractable stochastic analysis in high dimensions via robust optimization.pdf

s7300_sm335_high_speed_analog_io_module_manual_en-US_en-US.pdf

mybatis-generator-core-1.3.7-API文档-中文版.zip

loadrunner-11-load-generator.iso

activiti-image-generator-5.21.0-API文档-中文版.zip

activiti-image-generator-5.21.0-API文档-中英对照版.zip

activiti-image-generator-5.22.0.jar.rar

PCIE 4.0 connector high speed electrical test procedure.pdf

generator-mybatis-generator-1.3.6.tar.gz

generator-mybatis-generator-1.3.0.tar.gz

generator-mybatis-generator-1.4.0.tar.gz

generator-mybatis-generator-1.3.3.tar.gz

mybatis-generator+generator.xml

Ansible-ansible-openwisp2-imagegenerator.zip

mybatis-generator.jar

最新资源

mybatis-plus最新代码生成器项目源码：mybatis-plus-generator.zip