WebFrame a Multi-Phase Reconfigurable Middleware Platform

WebFrame: a Multi-Phase Reconfigurable Middleware Platform Fan Guochuang,Chen An, Zhang Wenbo,Huang Tao

Technology Center of Software Engineering

Institute of Software, Chinese Academy of Sciences

Beijing, 100080, P. R. China

fanguo@https://www.wendangku.net/doc/682366515.html,

Abstract

Web application servers (WAS s) are middleware platforms for deployment and execution of component-based web applications. To cater for an increasingly diverse range of QoS demands, WAS must be capable

of adaptation during execution to modify itself and to respond to changing conditions in its external environment. To accommodate such changes, WAS should provide both deployment-time configurability and run-time reconfigurability. Unfortunately, most of the mainstream web application servers adopt a monolithic architecture and “black box” philosophy to their design, and fail to properly address such requirements. In our point of view, adaptation and reconfigurability of WAS s should be available at any time of the whole lifecycle. In this paper, a middleware architecture (WebFrame) that supports multi-phase adaptation using computational reflection, microkernel, and component techniques is proposed for web application servers. The architecture is structured into five layers. Both deployment-time configuration and run-time reconfiguration at multiple layers is supported in this architecture. The key insight

to this work is the MS ervice Reconfiguration design pattern, which provides dynamic adaptation at run time by swapping in/out the optional middleware components. The comparative evaluation of the performance impact of reflection and multi-phase reconfigurability on systems are given.

1.Introduction

Web application servers (WASs), which are considered to be one of the most exciting enterprise technologies today [1], have become very popular in the last few years [2, 3]. They provide middleware platforms that offer basic services to support for the component-based web applications [4~6]. Recent advances in distributed, mobile, and ubiquitous systems are creating new web computing environments that are characterized by a high degree of dynamism. As the new generation middleware platforms, WASs should address an increasingly diverse range of changing QoS requirements [7~9]. These changes may occur from the design-time, deployment-time to run time. We distinguish two types of changes: programmed changes are foreseen and anticipated by the system designer, while evolutionary changes are unanticipated and become necessary over the execution lifetime of the system. However, it is impossible to construct a ready-made platform for cater for all those changing QoS needs.

To achieve the accommodation of those changing QoS requirements, WAS must be adaptable and flexible enough during execution to modify itself and to respond to changing conditions. WASs should be capable of both deployment-time configurability and run-time reconfigurability [8] so that their service implementation can be reconfigured dynamically at run time to adapt to changes in the environment and provide support for fine-grain resource management. Unfortunately, the current generation of mainstream WASs such as BEA Weblogic and IBM Websphere, which typically adopt a monolithic architecture and “black-box” philosophy to their design[11,12], fails to fully support for (re)configurability and open engineering[13]. In Iona’s Orbix2000, there have been some efforts to introduce reconfigurability but not completely. Despite the usefulness of some reflective features, the degree of support for customization and dynamic adaptation is only partial, and the extent to which elements of the design can be opened and exposed to the programmer is severely limited[13]. In our point of view, a reconfigurable architecture covering all aspects of the design and the different phases of a WAS life cycle is needed, and the architecture supports dynamic customization and reconfigurability at multiple layers.

The paper proposes an approach to the design of a reconfigurable and open WAS (referred to as WebFrame) based on the concept of reflection, supplemented by the marriage of microkernel technology and component technology. The approach followed for the design of WebFrame is driven by two principles: clear separation of concerns and What You Need Is What You Get . The first principle requires a clear and easy to understand decomposition of the web application server into modules. The second principle, WYNIWYG, requires that only the required functionality is present. Whenever more functionality

is required, it can be easily added either statically or

dynamically. When some functionality is no longer

required therefore, it can be removed.

In this paper, a middleware architecture

(WebFrame) that supports multi-phase adaptation

using computational reflection, microkernel, and

component techniques is proposed for web application

servers. The architecture supports both deployment-

time configuration and run-time reconfiguration.

Dynamic reconfiguration allows modifying the

architecture of WASs and Web applications at run-time

and therefore it is possible to choose at any time only

the functionality that is required. The key insight to this work is the MService Reconfiguration pattern,

which provides dynamic adaptation at run time by swapping in/out the optional middleware components. The rest of this paper is structured as follows. Firstly, the reflective microkernel structure for WASs is described in section 2. In section 3, five-layer structure of WASs and a multi-phase reconfigurable architecture are presented. The service reconfigurable framework is described in Section 4. Section 5 illustrates the implementation of runtime reconfiguration of services and components. Section 6 presents preliminary performance results. Section 7 discusses the related work. Section 8 concludes the paper lastly. 2.The Structure of Reflective Microkernel A web application server provides two categories of the services: i ) Basic services for component-based transactional web applications. For example, transaction service, component container, messaging service, naming service, and hot deploy service are basic services. ii ) Common services that provide the most essential services for those basic services. One common solution, which has been widely used in “black box” design, is to hard-code them in the system at compile time. To provide the capability of customization, flexibility and reconfigurability, the Microkernel architectural pattern [15] is applied. As illustrated in Figure 1, the basic services are separated from the common services by the microkernel. The common services are reserved in the kernel while the basic services that can be reconfigured or customized by the application programmer are moved out from the kernel. The microkernel also serves as a socket for plugging in these services and coordinating their collaboration.

Figure 1. The structure of reflective microkernel All services are managed by the MS ervice S erver .Every service binds itself a Meta-interface, which is implemented by a class called MService (Meta Service ),which is the most basic unit of services. It not only provides Meta information and the interfaces that control the life cycle of a service itself, but also offers additional interfaces that tailor non-functional QoS demands. MService provides reflective access to the representation of a particular service in terms of its set of methods and associated attributes. Information about the internal structure of a service can be obtained from the interfaces of MService. System and application code may use MService to inspect the

internal configuration and behavior of a service and, if

needed, reconfigure it to adapt to changes in the

environment and, therefore changing its own behavior.

The microkernel consists of these following common

services:

z S erver Configurator , which is started first, reads the basic configuration information from the system configuration description file (.XML) and loads all component loaders. z Loader Repository contains many Service Loaders, which can load physical resources such as component binary file and physical devices. A service loader can inspect the component type of a service and adapt to

change the right manner to load that service.

z MS ervice Repository maintains all services that are registered successfully in the

MService Server. Each service’s current state

is maintained in the Service Repository, as

well.

z ervice Dependency Manager provides the management of the dependency relationship

between services.

z MService Communication Manager deals with communication protocols, message arrivals,

dispatching, and marshalling/unmarshalling

etc. Clients communicate with services using

adapters provided by the microkernel.

z MService Server is the server of all MServices.

Client can interact with MServices though the

invoke method of MService Server.

Since the microkernel is able to observe and manipulate itself through these services, we refer to the concept of reflective microkernel.

3. A Multi-phase Reconfi

g urable

Architecture for Web Application Server

Separation of concerns is an essential issue in reflective systems. The usual approach is to regard web applications as the base-level and basic services and common services as their meta-level. An alternative approach is to consider the meta-level as orthogonal plane to the middleware and application layers. In this approach, both applications and middleware have their functionality defined at the base-level, whereas the entities at meta-level provide the facilities to reify, inspect and control such functionality. Although viewing reflective middleware in this way enables a cleaner and unified model for reflection programming,

it is not enough to support reconfiguration and customization at multiple layers during the whole lifecycle of WASs. Therefore, a multi-layer reflective architecture for WASs is required.

To achieve this, a reflective tower of meta-level should be created. As shown in Figure 2, the architecture is usually structured into five layers: Kernel Layer, Reconfiguration Layer, Meta S ervice Layer, Basic S ervice Layer, and Application Layer. The bottom layer is the microkernel, and the topmost layer is the applications. Each layer is concerned with the representation and manipulation of the layer above

it (which is its relative base-level). Each layer means different reconfiguration level. In this architecture, one can statically configure the Reconfigurator and Service Controller via specifying the URL location of their binary package and class name in the system configuration file at deployment-time. After Reconfigurator service has been started successfully, one can reconfigure WASs to adapt to the changing requirements at any time from the design-time to the run-time. Basic services or upper components can be added, removed, and replaced at run-time. WASs can be reconfigured at multiple phases of their lifecycle from the design-time, deployment-time to run-time. The dynamic reconfiguration entities can be common services, basic services and application components.

3.1. Kernel Layer

The most essential services are reserved in the microkernel. MService Server is started after the Server Configurator dynamically loads the basic configuration information from the system configuration description file at system startup time. All the services in the reconfiguration layer are configured and started by the microkernel after kernel services are started. To meet the particular requirements of applications, the Reconfiguration Layer can be statically customized and configured in system configuration file.

3.2. Reconfiguration layer

All the basic services outside of kernel are configured and Loaded by the Reconfiguration Layer services, instead of by WASs directly as “Black Box” WASs. As presented in Figure 2, there are three services at Reconfiguration Layer: Service Configurator, Service Controller, and Reconfigurator.

Service Configurator first reads that configuration information from a service descriptor file and gets the component binary file and resource for a service, and then instantiates the service.

Service Controller is used to control the life cycle of

a service by invoking the MService meta-interface using reflection. Services can be executed in several execution models: single threaded model, multi-threaded model, and multi-process model. Every service exposes a uniform set of life cycle management operations such as initialize, start, suspend, resume, stop and destroy.

Reconfigurator is used to dynamically reconfigure basic services at run-time, while introducing little (or ideally no) impact on the system’s execution. In this way, the system does not have to be taken off-line to accommodate changes. Operations on basic services can be replacement, addition, and removal. The reconfiguration process is triggered when some internal states have been changed or reconfiguration script is executed. In addition, Reconfigurator preserves system consistency to make sure that the

Figure 2

. The multi-phase reconfigurable architecture

WASs under reconfiguration must be left in a “correct” state after reconfiguration (a more detailed discussion will be presented in another paper). 3.3. Meta Service Layer Each service of the Basic Service Layer has a Meta Service, which can be considered as meta-objects of a basic service. Every service may have different implementation strategies. A service instance can be replaced by another instance that implements the same functionality, but with a different strategy. For example, in a large-scale dynamic environment, the

transaction services should also be reflective to dynamically provide the necessary support for various

transaction models in accordance to user requirements, changing environments, and so on[17]. Load balancing service should be adaptive to the change of server’s load and select different request scheduling strategies [10, 16]. In addition, Meta Service provides the lifecycle interface and Meta interfaces which allow structure reflection and behavior reflection. 3.4. Basic Service Layer The Basic Service Layer provides a series of basic services including Container, Transaction, Naming,

Messaging, Security, and Hot Deploying Service etc.

To meet the requirements of “What You Need is What

You Get”, our solution is a dynamically customized

WAS that allows changing its whole functionality or

selecting different service implementations provided by the third party. The possible configurations range from minimal functionality versions with minimal memory and resource requirements to fully functional versions. All those services are configured and loaded by the Reconfiguration Layer as described above. 3.5. Application Layer

Users can develop and deploy applications or server-side components in the Application Layer. These components typically contain the business logic that should be quickly adapted to the changes that the

market brings. To meet the new requirements for the changing market, the Basic Service Layer provides the deploy service to deploy, undeploy, and redeploy different types of applications and components at run-

time.

Component is a most essential unit of web

applications. The reconfiguration of those applications

is accomplished through the dynamic reconfiguration of component. Each component has a container, which acts as a meta-object. Component reconfiguration is described as an example in section 5.3.

Figure 3. The structure of MService Pattern 4.Service Reconfiguration Framework

Service Reconfiguration Framework consists of both reconfiguration management components and concrete basic services. The former includes Reconfigurator, Service Configurator, Service Controller, Service Repository, and Service Dependency Manager. The later implements the concrete services that perform application-specific processing and define a uniform interface that can be used to (re)configure or alter the functionality of the concrete service. To optimize, control, and reconfigure the behavior of a service at run-time, decoupling the implementation of a service from its configuration makes it possible to fine-tune certain implementation or configuration parameters of services. Moreover, to meet advanced QoS requirements, some concrete services may provide meta-services to inspect the internal information and reconfigure the behavior of them. For instance, Container is a typical Meta-object that can be reconfigurable.

We extend the Service Configurator pattern in [14].The structure of the extended Service Configurator pattern, called MService Reconfiguration Pattern, is illustrated in Figure 3:

The key participants in the MService pattern including:

z IMServiceLifeCycle specifies the interface containing hook methods (such as init, start,

pause, resume, stop, and destroy) which are

invoked by Service Configurator and Service

Controller to dynamically (re)configure the

services at run-time.

z IMServiceMetaInterface inherits from the IMServiceLifeCycle and provides access to the

operations and attributes declared to be reflective

by the services. Moreover, the interface defines a

series of standard methods to support the

capability of reflection. All meta-level access to a

service must be done via operations defined in the

interface. The invoke method lets application or

system code call any of the operations exposed by

the service.

z ConcreteService implements the IMServiceMeta-Interface that provides meta-level functionality

and other service-specified base-level functionality. Each concrete service inherits such

a meta-interface whose name is formed by adding

the MetaInterface suffix to the Service’s name.

For example, the service named “Container”

should implement the interface namely

“ContainerMetaInterface”.

z ServiceRepository maintains all services that are registered successfully in the MService Server.

This allows administrative entities to centrally

manage and control the behavior of the

configured services.

Figure 4 depicts the collaboration diagram between the components of Service Reconfiguration Framework. MServices have four phases: configuration stage, running stage, reconfiguration stage, and termination stage.

5.Implementation

5.1. Deployment-time Configuration

The section explains the steps to configure Reconfiguration Layer in WebFrame at deployment-time.

z Implement the reconfiguration interface: IServiceReconfigurator, IServiceConfigurator,

and IServiceController.

z Define the MService Reconfiguration Layer boot-strap configuration file as depicted in

table 1.

After Microkernel is started, the configuration information of Reconfiguration Layer is loaded. Microkernel invokes the start method on each instance

of the service in Reconfiguration Layer. After these services are started successfully, Meta Services can be started either at deployment-time or at run-time.

Figure 4. Interaction diagram for MService Pattern

5.2. Run-time Reconfiguration of Services Run-time reconfiguration allows modifying system

architecture to accommodate the changing requirements of applications. Dynamic reconfiguration operations on services can be replacement, addition, and removal. Those steps are examined briefly below: z Define a new Interface that inherits IMServiceMetaInterface. z Implement the new interface in a Meta Service class. z Write the reconfiguration script in a file. z Deploy the reconfiguration script. The reconfiguration criteria can be specified in

various ways: on the command line, through a user

interface, or in a reconfiguration script file. As illustrated in table 2, the example script presents how to use reconfiguration operations of add, remove, and replace.In the script file, MailService is added, and TransactionManager is removed. LoadBalancing -Service is replaced by AdaptiveLoadBalancingService .5.3. Run-time reconfiguration of components WebFrame is a web application server that supports J 2EE specifications. Containers provide run-time environments for EJ B components. EJ B components never interact directly with other J 2EE application components, so does J

ava client. They use the protocols and methods of the container for interacting

with each other and with platform services. In the

multi-phrase reconfigurable architecture, containers are

considered to be meta-objects of EJB components.

Table 1. Reconfiguration layer configuration file

WebFrame provides a customizable and reconfigurable container, which supports run-time reconfiguration on a per method invocation basis. Here, run-time reconfiguration refers to mechanisms that allow the implementation of the container to be dynamically adapted to cater for changing environmental conditions.

The structure of a customizable and reconfigurable

container is illustrated in Figure 5. The Interceptor

Pattern is applied in the structure. The basic idea is to

enable the container to intercept all method invocations made on the server components within it and make the request go through all the responding metaobjects before being delegated to the target component. However, Metaobjects does not implement the

Figure 5

. The structure of customizable and reconfigurable container

function directly themselves, and they are responsible for selecting the implementation strategies to deal with an invocation. As shown in the Figure 5, the InvocationDispatcher intercepts invocations from client and implements a general framework for providing different types of Invokers for containers that can be accessed using different protocols such as J RMP, RMI-IIOP, and SOAP. InvocationDispatcher inspects the type of the accessing protocol and matches with the policy to reconfigure the InvokerMetaObject to the selected Invoker. After an invoker is selected, all method invocations are dispatched to it directly. However, instead of implementing the function directly inside the selected invoker, it is divided into a number of smaller tasks that are implemented in InterceptorMetaObjects .As shown in Figure5, ConcurrentMetaObject and TransactionMetaObejct are examples of InterceptorMetaObjects. The invoker is responsible for coordinating the InterceptorMetaObjects to fulfill the whole function that the user specifies. To provide flexibility and adaptability, users can plug metaobjects in the form of MService before or/and after method execution on the target component. Each task can be implemented in a number of ways in different

metaobjects, thus users can customize container to suit

the application by using the most appropriate Metaobjects.

ContextObject allows InterceptorMetaObjects to access and control aspects of the container’s internal state and behavior in response to certain events. 6.Preliminary Performance Evaluation 6.1. Scope of Evaluation A number of preliminary tests have been carried out in order to measure the performance impact of reflection and multi-phase reconfigurability on the web application server by comparing the performance between WebFrame and BEA Weblogic 7.0, the latter is one of the mainstream web application servers and compatible with J2EE1.3 specifications.

We are primarily interested in the performance measure: Response Time per Invocation . It is the mean total elapsed time from the instant the client starts its request cycle to the time the final reply is received from the server used in the cycle. The parameter is a measure of the latency for the system.

For testing purpose, some small applications consisting of Java clients and Enterprise Java Beans of different categories have been built. Users deploy some E Bs into the target application server, and then invocate the methods of those EJ Bs. BEA Weblogic uses the default settings, and WebFrame uses an optimized implementation of persistence strategy adaptively via PersisteneceMetaObject.

The hardware consists of a dedicated application server with 1 Intel CPU 1.70 GHz and 780 MB of RAM and one client with 1 Intel CPU 1.70 GHz and 512 MB of RAM, interconnected in a 100Mbit LAN. The database is Cloudscape provide by SUN. 6.2. Experimental Results

Our hope and expectation is that Webframe should perform as well as existing web application servers without multi-phase reconfigurable characteristics. As

illustrated in Figure 6, the response time per invocation

of WebFrame is 1.6 milliseconds longer than that of

BEA Weblogic partly due to the use of reflection, and

the number of invocations of per second is less than

that of Bea Weblogic with the same reason. However,

the Bean overhead of reflection is very lower

compared with the cost of business methods of a session bean in real applications, so the design of the

container has little effect on the performance of session

beans although the cost of reflection affects

performance. 6Figure 7 and Figure 8 report the results using entity

beans with CMP (Container-Managed Persistence) and BMP (Bean-Managed Persistence).The number of invocations per second for CMP of WebFrame is between 6 and 9 times more than that of Weblogic. In the case of entity beans with BMP, the number of invocations per second is about 1.5 times more than that of Weblogic. It is the optimized implementation of persistence strategy that improves the performance of entity beans, and the reason is that the improvement is far more than the cost of the extensible use of reflection and multi-phase architecture. The performance may be further improved after reconfiguring the WebFrame to adapt to other more optimized persistence strategies although there are overheads to achieve the capability of openness and flexibility. In summary, the cost of the reflection and multi-phase reconfigurable architecture may affect performance of web application servers, but the significant performance improvement achieved by reconfiguring system effectively to adapt to changes through the multi-phase reconfigurable architecture is far more than the cost of the reconfiguration activity itself. Figure 8. Comparison results for BMP Bean

7.Related work This discussion is organized accordingly to the related fields that our work touches: z Web Application Servers Most of the mainstream WASs (such as Iona

Orbix2000, BEA Weblogic, and IBM Websphere)

typically adopt a monolithic architecture and “black-

box” philosophy to their design. Despite the usefulness

of some reflective features, the degree of support for customization and dynamic adaptation is only partial,

and the extent to which elements of the design can be opened and exposed to the programmer is severely limited. While WebFrame provides a reconfigurable

architecture covering all aspects of the design and the different phases of a WAS life cycle .In this

architecture, both deployment-time configuration and

run-time reconfiguration is supported at multiple

layers.

Boss, an open-source 2EE application server,

employs the microkernel architectural pattern to their design. Java Management Extensions (JMX) is used as a reflective microkernel at the very basis of the whole server in JBOSS. However, JMX should be used as a means to instrument selected parts of the application server and make them manageable through MX-compliant clients. It is too big and complicated to be a microkernel. While WebFrame’s microkernel is based on MService model. It is more lightweight than that of JBoss. One of MService’s benefits is that it can reduce the footprint size of middleware platform for use, especially in resource-constrained embedded system. In WebFrame, users can choose MX module if needed. In addition, WebFrame provides Reconfigurator to deal with unanticipated changes, and services can be added, removed, and replaced using reconfiguration script at run-time while preserving system’s consistency. z Reflective middleware and Dynamic Reconfiguration

A reflective architecture for configurable middleware platforms is presented in [8]. The approach relies on reflective techniques for reconfiguring the OR

B to the desired behavior. Reconfiguration is done by configuring the meta-level of the reflective architecture, while the ORB implementation is residing at the base-level. However, reconfiguration is limited to application wide concerns and cannot be addressed on a per remote method invocation basis. Universal Interoperable Core (UIC, formerly LegORB) is a reflective middleware infrastructure that is customizable to ubiquitous computing scenarios. It can be configured at run-time so that it loads just enough components to provide the middleware services required by its applications. .It also supports “What You Need Is What You Get”. Run-time configuration is achieved by defining pre-existing hooks for changing strategies related to concurrency, security, monitoring, and the like. This implies that only anticipated non-functional requirements can be

meet whereas our approach is able of coping with

changing non-functional aspects as well. Due to multi-phrase dynamic reconfiguration design, WebFrame can

add or remove unanticipated functionalities at multi-layers from deployment-time to run-time. DynamicTAO is a reflective CORBA ORB, built as an extension of TAO. Some reflective features are added to make the system more flexible, dynamic, and customized. Conversely, WebFrame focuses on a novel middleware architecture where reconfiguration from the kernel layer to application layer is supported inherently from the scratch. ZEN is a highly configurable real-time COBRA ORB implemented using J ava. ZEN’s ORB architecture is based on the concept of layered pluggability where various components of the middleware may be plugged or unplugged on an as-needed basis. WebFrame has some similarity in the micro-ORB design. ZEN has five generations of ORB design. The current research work focuses on dynamic micro-ORB which does not use reflection to provide feedback to allow run-time reconfiguration to customize ZEN. While WebFrame provides reflection capability at multiple layers of the reflection tower, thus, internal states can be inspected, and WebFrame can be adaptive to adjust structure and behavior of itself at run-time. z Design Patterns for configuration Design patterns for configuration issues include Service Configurator, Component Configurator, and Virtual Component pattern. Service Configurator pattern decouples the implementation of services from their configuration. It provides the ability to configure a service without modifying, recompiling, or statically relinking existing code. Component Configurator pattern promotes a decoupling between components and components connection, aiming at supporting ad-hoc dynamic reconfiguration and the migration of component with state transferring. Virtual Component pattern is applied throughout ZEN to decompose and factor most of the unused or rarely used component out of memory. This pattern provides an application-transparent way of loading and unloading components that implementing middleware software functionality. MService Reconfiguration design pattern can be viewed as a compound pattern that combines elements of the Service Configurator and Component Configurator patterns. What is different from other configuration design patterns is that MService Reconfiguration design pattern support reflection mechanism. It decomposes meta-level functionality of services from base-level functionality. It means that the services in MService design pattern are open and reflective.

8.Conclusions and future work

In this paper, a middleware architecture (WebFrame) that supports multi-phase adaptation using computational reflection, microkernel, and component techniques is proposed for web application servers. The architecture is usually structured into five layers: Kernel Layer, Reconfiguration Layer, Meta Service Layer, Basic Service Layer, and Application Layer, Both deployment-time configuration and run-time reconfiguration at multiple layers is supported in this architecture. Our approach overcomes the limitations of monolithic architecture and “black box” philosophy by offering a reflective microkernel, and the basic services are separated from the common services. All basic services can be run-time reconfigured in application-specific manner with the support of Service Reconfiguration Framework. The reconfiguration process can be executed in script manner. To validate the multi-phase reconfigurable architecture, we develop a J2EE WAS prototype. It is called https://www.wendangku.net/doc/682366515.html,paring the performance between WebFrame and BEA Weblogic 7.0, preliminary experimental results show that the significant performance improvement achieved by run-time reconfiguration may be far more than the cost of the reconfiguration activity itself. Here are some of our future research work on WebFrame: (1) To minimize the cost of the use of multi-phase reconfigurable architecture; (2) To carry out the further development of other reconfigurable services (such as transaction service, caching service etc) to achieve great improvements of performance. (3)

To apply the multi-phase adaptation architecture to other distributed systems. Acknowledgments The research described in this paper is funded by the National 863 Program of China (No.2001AA113010, 2001AA414020, and 2001AA414310), and by the 973 Program of China (No 2002CB312005). References [1]Ron Copeland. “Web Application Servers”, https://www.wendangku.net/doc/682366515.html,/704/04iuapp.htm, 1999. [2]Mike Ricciuti, ”Application Server Eludes Definition”, https://www.wendangku.net/doc/682366515.html,/2100-1001_3-214783.html,1998. [3] C.Mohan. “Tutorial: Application Servers and Associated Technologies”, Proceedings of the 2002

ACM SIGMOD international conference on

Management of data. New York, NY, USA:ACM

Press,2002, 636-636.

[4] C.Mohan. “Application Servers:Born-Again TP

Monitors For The Web”,Proceedings of the 2001 ACM

SIGMOD international conference on Management of

data. New York :ACM Press,2001, 622-622.

[5]Fan GC, Zh H, Huang T, Feng YL. “A Survey of Web

Application Servers”. J ournal of Software (Chinese),

2003,14(1):1278~1239.

[6]Scott Dietzen. “Web Application Server Architecture”,

Ninth International Workshop on High Performance

Transaction Systems (HPTS), 2001.

[7]J ohn Bates, “The State of The Art In Distributed and Dependable Computing”, A CaberNet Sponsord report,

1998.https://www.wendangku.net/doc/682366515.html,/bates98state.html.

[8]Blair, G.S., Coulson, G., Robin, and P., Papathomas,

M., "An Architecture for Next Generation

Middleware", Proc. IFIP International Conference on

Distributed Systems Platforms and Open Distributed

Processing (Middleware'98), Springer, 1998, pp. 191-

206.

[9]Geoff Coulson and Gordon S. etc., "The design of a

configurable and reconfigurable middleware

platform",Distributed Computing, vol 15(2):109-

126,2002.

[10]O. Othman, C. O'Ryan, and D. C. Schmidt, "An

Efficient Adaptive Load Balancing Service for

CORBA," IEEE Distributed Systems Online, vol. 2,

2001.https://www.wendangku.net/doc/682366515.html,/othman01efficient.html.

[11]Gordon Blair Geoff Coulson, Fabio Costa ,and Hector

A.Duran. “On The Design Of Reflective Middlewaare

Platforms”.Workshop on Reflective middleware,

Proc.,IFIP/ACM Middlware’ 2000.

[12]Marc Fleury and J uha lindfors,”Enabling Component

Architecture with J MX”,the O’Reilly Network,2001.

https://www.wendangku.net/doc/682366515.html,/pub/a/onjava/2001/02/01/jmx.h

tml.

[13]Gordon S. Blair, Geoff Coulson, “The Case For Reflective Middleware:Buidling Middleware that is Flexible,Reconfigurable And Yet Simple To Use”,

2002.https://www.wendangku.net/doc/682366515.html,p.br/~kon/papers/cacm02.pdf.

[14]Prashant J ain and Douglas C. Schmidt,”Dynamically

Configuring Communication Services with the Service Configurator Pattern”. C++ Report,9(5),J

une 1997. https://www.wendangku.net/doc/682366515.html,/article/jain97dynamically.ht ml.

[15]Buschmann, F., R. Meunier, H. Rohnert, P. Sommerlad, and M. Stal. …Pattern-Oriented Software

Architecture: A System Of Patterns”. West Sussex,

England: John Wiley & Sons Ltd., 1996.

[16]Othman, C. O’Ryan, and D. C. Schmidt, “Strategies for

CORBA Middleware-Based Load Balancing,” IEEE

Distributed Systems Online, Volume 2, Number 3 ,March 2001.

[17]Weihai Yu and Randi Karlsen,”Reflective Transaction

Service”,Proc., IFIP/ACM Middlware’2000,2000.

https://www.wendangku.net/doc/682366515.html,/423363.html.

变电所母线桥的动稳定校验

变电所母线桥的动稳定校验 随着用电负荷的快速增长，许多变电所都对主变进行了增容，并对相关设备进行了调换和校验，但往往会忽视主变母线桥的动稳定校验，事实上此项工作非常重要。当主变增容后，由于阻抗发生了变化，短路电流将会增大许多，一旦发生短路，产生的电动力有可能会对母线桥产生破坏。特别是户内母线桥由于安装时受地理位置的限制，绝缘子间的跨距较长，受到破坏的可能性更大，所以应加强此项工作。 下面以我局35kV/10kv胡店变电所#2主变增容为例来谈谈如何进行主变母线桥的动稳定校验和校验中应注意的问题。 1短路电流计算 图1为胡店变电所的系统主接线图。(略) 已知#1主变容量为10000kVA，短路电压为7.42%，#2主变容量为12500kVA，短路电压为7.48%(增容前短路电压为7.73%)。 取系统基准容量为100MVA，则#1主变短路电压标么值 X1＝7.42/100×100×1000/10000＝0.742， #2主变短路电压标么值 X2＝7.48/100×100×1000/12500＝0.5984 胡店变电所最大运行方式系统到35kV母线上的电抗标么值为0.2778。 ∴#1主变与#2主变的并联电抗为： X12=X1×X2/(X1+X2)＝0.33125； 最大运行方式下系统到10kV母线上的组合电抗为： X＝0.2778+0.33125＝0.60875

∴10kV母线上的三相短路电流为：Id=100000/0.60875*√3*10.5，冲击电流：I sh=2.55I =23032.875A。 d 2动稳定校验 (1)10kV母线桥的动稳定校验： 进行母线桥动稳定校验应注意以下两点： ①电动力的计算，经过对外边相所受的力，中间相所受的力以及三相和二相电动力进行比较，三相短路时中间相所受的力最大，所以计算时必须以此为依据。 ②母线及其支架都具有弹性和质量，组成一弹性系统，所以应计算应力系数，计及共振的影响。根据以上两点，校验过程如下： 已知母线桥为8×80mm2的铝排，相间中心线间距离为210mm，先计算应力系数： ∵频率系数N f＝3.56，弹性模量E＝7×10.7 Pa，单位长度铝排质量M＝1.568kg/m，绝缘子间跨距2m，则一阶固有频率: f’=(N f/L2)*√(EI/M)=110Hz 查表可得动态应力系数β＝1.3。 ∴单位长度铝排所受的电动力为： f ph＝1.73×10-7I sh2/a×β＝568.1N/m ∵三相铝排水平布置，∴截面系数W＝bh2/6＝85333mm3，根据铝排的最大应力可确定绝缘子间允许的最大跨距为： L MAX=√10*σal*W/ f ph=3.24m ∵胡店变主变母线桥绝缘子间最大跨距为2m，小于绝缘子间的最大允许跨距。

英国高中课程(A-Level)介绍

英国高中课程（A-Level）介绍 英国高中课程（A-Level）是英国的全民课程体系,也是英国大学的入学考试课程,其课程证书被几乎所有英语授课的大学作为招收新生的入学标准。因此,英国高中课程（A-Level）被国际教育界誉为“金牌”教育课程和全球大学入学的“黄金”标准。 英国高中课程（A-Level）是英国学生中学五年级义务教育结束以后,可以就读的一种2年制的课程体系。课程本身并不对应任何特定年龄,但英国本土及国际学生一般要满16周岁,年龄段向上没有限制。从不同课程体系之间的横向比较看,英国高中课程（A-Level）最好被看作是英国的预科体系里的学术性课程体系, 并且是主流预科课程体系。 若从学程/学龄段上进行类比,英国高中课程（A-Level）大体相当中国的高三和大学一年级。但若从课程的结构方面进行比较,英国高中课程（A-Level）在程度上都远远超过中国的高中。以数学这一门课程为例,它可以包括纯数学、工程数学、统计学、决策数学,仅纯数学部分就包括微分方程、线性代数、向量代数等内容；若从整体课程体系的方面看,英国高中课程（A-Level）提供60多门课程给学生选择,学生可以任意选择其中的3-4门课程进行学习,包括经济学、商业学等中国高中课程体系中根本不存在的部分。因此,如果中国学生选择高中阶段去英国留学,尤其是高二或高二之前,学习成绩属于中等以上,那么他们留学英国 的最佳选择就非英国高中课程（A-Level）莫属。 英国高中课程（A-Level）介绍

英国高中课程（A-Level）的考试将由所属的考试委员会分别于每年5-6月和 10-11月在全球统一组织,成绩分别在8月和次年 2月公布。成绩分为A、B、C、D、E、U六个等级,A为最优,E为通过,U为不及格。如果学生对某门课的成绩不满意,可以选择重考,最终成绩以最好的一次为准。学生达到及格成绩,就可以申请英国或其他国外的大学。 具体的等级与百分制之间的关系如下表所示： 在入学要求方面,英国及世界较好的大学通常要求学生3门课的成绩均应达到C 以上；而对于世界一流大学,如牛津、剑桥、哈佛大学等名校,则要求申请学生3门课的成绩达到 AAA或AAB。 文章来源：启德教育

母线电动力及动热稳定性计算

母线电动力及动热稳定性计算 1 目的和范围 本文档为电气产品的母线电动力、动稳定、热稳定计算指导文件，作为产品结构设计安全指导文件的方案设计阶段指导文件，用于母线电动力、动稳定性、热稳定性计算的选型指导。 2 参加文件 表1 3 术语和缩略语 表2 4 母线电动力、动稳定、热稳定计算 4.1 载流导体的电动力计算 4.1.1 同一平面内圆细导体上的电动力计算

? 当同一平面内导体1l 和2l 分别流过1I 和2I 电流时（见图1），导体1l 上的电动力计 算 h F K I I 4210 π μ= 式中 F ——导体1l 上的电动力（N ） 0μ——真空磁导率，m H 60104.0-?=πμ； 1I 、2I ——流过导体1l 和2l 的电流（A ）； h K ——回路系数，见表1。 图1 圆细导体上的电动力 表1 回路系数h K 表 两导体相互位置及示意图 h K 平 行 21l l = ∞=1l 时，a l K h 2= ∞≠1l 时，?? ? ???-+=l a l a a l K h 2)(12 21l l ≠ 22 2) ()(1l a m l a l a K h ++-+= 22)()1(l a m +-- l a m =

? 当导体1l 和2l 分别流过1I 和2I 电流时，沿1l 导体任意单位长度上各点的电动力计 算 f 124K f I I d μ= π 式中 f ——1l 导体任意单位长度上的电动力（m N ）； f K ——与同一平面内两导体的长度和相互位置有关的系数，见表2。 表2 f K 系数表

4.1.2 两平行矩形截面导体上的电动力计算 两矩形导体（母线）在b ＜＜a ，且b ＞＞h 的情况下，其单位长度上的电动力F 的 计算见表3。 当矩形导体的b 与a 和h 的尺寸相比不可忽略时，可按下式计算 712 210x L F I I K a -=? 式中 F -两导体相互作用的电动力，N ； L -母线支承点间的距离，m ； a -导体间距，m ； 1I 、2I -流过两个矩形母线的电流，A ； x K -导体截面形状系数； 表3 两矩形导体单位长度上的电动力 4.1.3 三相母线短路时的电动力计算

解读英国私立高中排名

解读英国私立高中排名 核心提示：本文作者系上海华申英国留学高级专家沈凡老师，96年开始从事留学咨询工作，有深厚的海外学习工作阅历，读到的简介，详尽为大家独家揭秘英国私立高中排名之奥秘。 英国私立高中排名是留英选校的一项重要参考指标，但上海华申英国留学资深专家沈凡老师说，因为英国与中国的学制和评分标准有所不同，很多家长和学生在接触华申之前，并不清楚英国私立高中排名的性质和特点，或者干脆置之不理，或者盲目推崇，都无法充分利用高中排名对于选校的价值。为了知己知彼，选到理想的学校，有必要在选校前对英国私立高中排名的规律和特征进行充分的了解。 英国私立高中并没有一个绝对权威性的官方排名。英国的几家知名报纸，如《泰晤士报》和《卫报》，每年都会推出各自的私立高中排名，而像BEST UK SCHOOLS这种公信力比较高的教育网站，也会发表自己制作的排名。但无论哪家机构，排名的依据都是各个高中的全英会考A-Level成绩，按照得分为A*-A或A-B的学生占全校学生的百分比排列。比如很喜欢招收中国学生的Concord College，按A*-A排名得分是78.54，说明该校当年参加A-Level课程的学生中，有78.54%的学生得分超过A，这个水平在全部学校中可以排到第22位。而排名第一的Cardiff Sixth Form College，得分A*-A的学生可以达到93.85%之多。 在英国2700多所私立高中里，400位以后的英国私立高中一般不计入排名。学校参加考试学生人数在20以下的，一般也不会进入各家媒体所给出的排名榜单。还有80所左右实施国际会考课程（IB）的学校可以单独按照会考的平均成绩排名。大多数英国私立高中通常每年都会公布本校成绩，知道了排名标准后，我们自己可以根据不同的需要，订制私立高中的排名。如英国高中分男校（Boys Only）、女校（Girls Only）和合校（Co-ed），各自分别有相应的排名，家长可以根据孩子的具体情况，参考不同类别的排名进行选校方面的考虑。 必须清楚的是，英国私立高中分为走读（Day School）和寄宿（Boarding School），有

高压电缆热稳定校验计算书

筠连县分水岭煤业有限责任公司 井 下 高 压 电 缆 热 稳 定 性 校 验 计 算 书 巡司二煤矿 编制：机电科 筠连县分水岭煤业有限责任公司

井下高压电缆热稳定校验计算书 一、概述： 根据《煤矿安全规程》第453条及456条之规定，对我矿入井高压电缆进行热稳定校验。 二、确定供电方式 我矿高压供电采用分列运行供电方式，地面变电所、井下变电所均采用单母线分段分列供电方式运行，各种主要负荷分接于不同母线段。 三、井下高压电缆明细： 矿上有两趟主进线，引至巡司变电站不同母线段，一趟931线，另一趟925线。井下中央变电所由地面配电房10KV输入。 入井一回路：MYJV22-8.7/10KV 3*50mm2--800m(10KV) 入井二回路：MYJV22-8.7/10KV 3*50mm2--800m(10KV) 四、校验计算 1、井下入井回路高压电缆热稳定性校验 已知条件：该条高压电缆型号为，MYJV22-8.7/10KV 3*50mm2 ,800m，电缆长度为800m=0.8km。 （1）计算电网阻抗 查附表一，短路电流的周期分量稳定性为 电抗：X=0.072*0.8=0.0576Ω； 电阻：R=0.407*0.8=0.3256 Ω； (2)三相短路电流的计算

A Z I 5.174693305 .0310000 3v 3=?== ∞ （3）电缆热稳定校验 由于断路器的燃弧时间及固有动作时间之和约为t=0.05S; 查附表二得热稳定计算系数取K=142; 故电缆最小热值稳定截面为 23mm 51.2705.0142/5.17469t )/(min ===∞）（K I S Smin<50mm 2 故选用 MYJV 22 -8.7/10KV 3*50 电缆热稳定校验合格，符合要求。 附表一：三相电缆在工作温度时的阻抗值（Ω/Km ） 电缆截面S （mm 2 ） 4 6 10 16 2 5 35 50 70 95 120 150 185 240 交联聚乙烯 R 4.988 3.325 2.035 1.272 0.814 0.581 0.407 0.291 0.214 0.169 0.136 0.11 0.085 X 0.093 0.093 0.087 0.082 0.075 0.072 0.072 0.069 0.069 0.069 0.07 0.07 0.07 附表二 不同绝缘导体的热稳定计算系数 绝缘材料 芯线起始温度（° C ） 芯线最高允许温度（°C ） 系数K 聚氯乙烯 70 160 115（114） 普通橡胶 75 200 131 乙丙橡胶 90 250 143（142） 油浸纸绝缘 80 160 107 交联聚乙烯 90 250 142

最新整理英国高中Alevel课程介绍培训资料

英国高中Alevel课程介绍 一、课程体系 英国高中课程(General Certificate of Education Advanced Level)简称A-Level课程，于1951年在英国正式启用。它是英国的全民课程体系，也是英国学生的大学入学考试课程，A-Level课程证书被几乎所有英语授课的大学作为招收新生的入学标准。因此，A-Level课程被国际教育界誉为“金牌”教育方式（British Golden Education System）。英国学生一般在16岁或稍大一些开始学习这种课程，圆满修完这种课程即可进入大学就读。这种课程可以简单理解为中国的高中。能否上英国大学特别是好的名牌大学，A Level成绩是关键。 二、课程的权威性 该课程体系的教学大纲、课程设置及其考试分别由英国四个主要考试 局Cambridge International Examinations,简称 CIE, Oxford Cambridge and RSA Examinations 简称 OCR, Assessment and Qualifications Alliance 简称AQA 和EDEXEL设计并组织，其权威性得到了国际上的广泛认可。迄今为止，全球已有5000多个教育机构开设了英国高中课程，每年有数百万学生参加由这些考试局组织的统一考试。由于该课程的科学性和权威性，新加坡甚至直接将该课程考试作为大学入学的全国统一考试。 三、教学质量控制 英国国家考试局对每一个开设A-Level的高级附属中心都进行严格的教学质量控制措施。每一位教授A-Level课程的老师都经过严格的筛选和测评。同时，老师们也可以通过多种渠道得到英国国家考试局的协助和培训。为了监控分布在世界上150多个国家教育中心的教学质量，这些考试局已经在全球各地建立了完善的组织网络；通过其批准的高级附属中心，不定期地对教学质量进行检查和评定。从而保证每一个就读A-Level课程的学生能接受到高质量的教育。 四、A-Level课程的优点 除大学预科课程之外，A-Level课程是中国学生入读英国大学的最佳途径。这是因为A-Level课程要大大优于大学基础课程，其优点有以下三方面： 第一、中国学生在国内的高一或高二的在校学生可以赴英国学习A-Level课程。这样，他们再经过两年的学习就可以进入英国大学就读，而不是像在中国那样，高中毕业然后还要学习一年的大学预科课程，才能入读英国大学。 第二、这种课程是为中国学生进入英国大学做准备的理想课程。无论是在学业方面还是在语言方面都会高于大学基础课程可达到的程度。中国学生要用英语学习各门课程，亲自体验新的教学方法，其英语也会达到相当熟练的程度。 第三、牛津、剑桥、帝国理工和伦敦大学学院这样的名牌英国大学几乎是从不录取大学预科的学生，而只录取A-Level毕业证书或同等学历的学生。同样，如果没有A-Level毕业证书或同等学历，要想就读像医科或医疗卫生领域的某些课程

热稳定性校验(主焦

井下高压开关、供电电缆动热稳定性校验 一、-350中央变电所开关断路器开断能力及电缆热稳定性校验 1 23 G 35kV 2 Uz%=7.5△P N.T =12kW △P N.T =3.11kW S N.T =8MVA 6kV S1点三相短路电流计算： 35kV 变压器阻抗： 2 22.1. u %7.5 6.30.37()1001008z N T N T U Z S ?===Ω? 35kV 变压器电阻：2 22.1.22. 6.30.0120.007()8 N T N T N T U R P S =?=?=Ω 35kV 变压器电抗：10.37()X = ==Ω 电缆电抗：02(x )0.415000.08780 0.66()1000 1000i L X ??+?== =Ω∑ 电缆电阻：02(x )0.11815000.118780 0.27()1000 1000 i L R ??+?== =Ω∑ 总阻抗： 21.370.66) 1.06( Z ==Ω S1点三相短路电流：(3)1 3.43()d I KA === S2点三相短路电流计算： S2点所用电缆为MY-3×70+1×25，长400米，变压器容量为500KV A ，查表的：(2)2d I =2.5KA

S2点三相短路电流：32 d d =2.88I I KA = 1、架空线路、入井电缆的热稳定性校验。已知供电负荷为3128.02KV A ，电压为6KV ，需用系数0.62，功率因数cos 0.78φ=，架空线路长度1.5km ，电缆长度780m (1)按经济电流密度选择电缆，计算容量为 3128.020.62 2486.37cos 0.78 kp S KVA φ?= ==。 电缆的长时工作电流Ig 为239.25 Ig === A 按长时允许电流校验电缆截面查煤矿供电表5-15得MYJV42-3×185-6/6截面长时允许电流为479A/6kV 、大于239.25A 符合要求。 (2)按电压损失校验，配电线路允许电压损失5%得 60000.1300Uy V ?=?=，线路的实际电压损失 109.1L U COS DS φφ?====，U ?小于300V 电压损失满足要求 (3)热稳定性条件校验，短路电流的周期分量稳定性为 电缆最小允许热稳定截面积： 3 2min d =S I mm 其中：i t ----断路器分断时间，一般取0.25s ； C----电缆热稳定系数，一般取100，环境温度35℃，电缆温升不超过120℃时，铜芯电缆聚乙烯电缆熔化温度为130℃，电

2020英国高中排名

2020英国高中排名 英国众多高中的性质类型各不同，给大家提供英国高中排名前50名校榜单，供申请英国高中的童鞋们参考。最新英国高中排名如下，包括学校名称，学校性质，所在地，学校类型，平均每科得分。 1 圣保罗女子学校St Paul‘s Girls’ School 私立伦敦地区女校 267.4 2 威斯敏斯特学校Westminster School 私立伦敦地区混校266.3 3 皮尔斯女子学校Perse School for Girls 私立剑桥郡-伦敦北80km 混校 266 4 吉尔福德高中 Guildford High School 私立萨里-伦敦西南边女校 265.1 5 威辛顿女子学校Withington Girls‘ School 私立曼彻斯特地区,奥尔德姆女校 265 6 莫德林学院学校 Magdalen College School 私立牛津郡男校 265 7 圣保罗学校St Paul’s School 私立伦敦地区泰晤士河畔里士满男校 264.7 8 伦敦城市女子学校City of London School for Girls 私立伦敦,格林威治女校 264.2 9 威克姆阿贝学校Wycombe Abbey School 私立白金汉郡女校

264.1 10 皇家文理学校Royal Grammar School 私立萨里-伦敦西南边男校 263.9 11 牛津中学Oxford High School GDST 私立牛津郡女校263.8 12 詹姆斯艾伦女子学校James Allen‘s Girls’ School 私立伦敦地区女校 263.1 13 北伦敦学院学校North London Collegiate School 私立大伦敦哈罗区女校 262.7 14 南汉普斯德高级中学 South Hampstead High School 私立伦敦卡姆登女校 262.3 15 伊顿公学Eton College 私立温莎和梅登黑德男校 261 16 乔丁汉女子学院The Cheltenham Ladies‘ College 私立格鲁斯特郡女校 261 17 温切斯特学院Winchester College 私立汉普郡男校260.8 18 哈博戴斯阿斯克男校Haberdashers’ Aske‘s Boys’School 私立赫特福德郡-伦敦北边男校 260.7 19 圣玛丽学校·阿斯科特St Mary‘s School Ascot 私立温莎和梅登黑德女校 260.5 20 哈博戴斯阿斯克女子学校Haberdashers’ Aske‘s School for Girls 私立赫特福德郡-伦敦北边女校 260.5 21 协和学院Concord College 私立萨罗普郡-英西部混校260.4

上海理工大学高等传热学试题及答案

1.试求出圆柱坐标系的尺度系数，并由此导出圆柱坐标系中的导热微分方程。 2 .一无限大平板，初始温度为T 0；τ>0时，在x = 0表面处绝热；在x = L 表面以对流方式向温度为t f 的流体换热。试用分离变量法求出τ>0时平板的温度分布（常物性）。（需求出特征函数、超越方程的具体形式，范数（模）可用积分形式表示）。（15分） , 3.简述近似解析解——积分法中热层厚度δ的概念。 答：近似解析解：既有分析解的特征：得到的结果具有解析函数形式，又有近似解的特征：结果只能近似满足导热解问题。在有限的时间内，边界温度 的变化对于区域温度场的影响只是在某一有限的范围内，把这个有限的范围定义为热层厚度δ。 4.与单相固体导热相比，相变导热有什么特点 答：相变导热包含了相变和导热两种物理过程。相变导热的特点是 1.固、液两相之间存在着 移动的交界面。 2.两相交界面有潜热的释放（或吸收） | 对流部分（所需量和符号自己设定） 1 推导极坐标系下二维稳态导热微分方程。 2 已知绕流平板流动附面层微分方程为 y u y u V x u u 22??=??+??ν 取相似变量为： x u y νη∞ = x u f νψ∞= 写出问题的数学模型并求问题的相似解。 3 已知绕流平板流动换热的附面层能量积分方程为： ?=∞?? =-δ00)(y y t a dy t t u dx d 当Pr<<1时，写出问题的数学模型并求问题的近似积分解及平均Nu （取三次多项式）。 4 ] O x

5写出常热流圆管内热充分发展流动和换热问题的数学模型并求出速度和温度分布及Nu x.辐射 1．请推导出具有n个表面的净热流法壁面间辐射换热求解公式，并简要说明应用任一种数值方法的求解过程。 2．试推导介质辐射传递方程的微分形式和积分形式，要求表述出各个步骤和结果中各个相关量的含义。 3．根据光谱辐射强度表示下面各量：1）光谱定向辐射力；2）定向辐射力；3）光谱辐射力；4）辐射力；5）辐射热流量。要求写清各量的符号、单位。 4．说明下列术语（可用数学表达式）（每题4分） a)光学厚度 b)漫有色表面 c)? d)兰贝特余弦定律 e)光谱散射相函数 f)定向“灰”入射辐射

高压电缆热稳定校验计算书

*作品编号：DG13485201600078972981* 创作者：玫霸* 筠连县分水岭煤业有限责任公司 井 下 高 压 电 缆 热 稳 定 性 校 验 计 算 书 巡司二煤矿

编制：机电科 筠连县分水岭煤业有限责任公司 井下高压电缆热稳定校验计算书 一、概述： 根据《煤矿安全规程》第453条及456条之规定，对我矿入井高压电缆进行热稳定校验。 二、确定供电方式 我矿高压供电采用分列运行供电方式，地面变电所、井下变电所均采用单母线分段分列供电方式运行，各种主要负荷分接于不同母线段。 三、井下高压电缆明细： 矿上有两趟主进线，引至巡司变电站不同母线段，一趟931线，另一趟925线。井下中央变电所由地面配电房10KV输入。 入井一回路：MYJV22-8.7/10KV 3*50mm2--800m(10KV) 入井二回路：MYJV22-8.7/10KV 3*50mm2--800m(10KV) 四、校验计算 1、井下入井回路高压电缆热稳定性校验 已知条件：该条高压电缆型号为，MYJV22-8.7/10KV 3*50mm2 ,800m，电缆长度为800m=0.8km。 （1）计算电网阻抗 查附表一，短路电流的周期分量稳定性为

电抗：X=0.072*0.8=0.0576Ω； 电阻：R=0.407*0.8=0.3256 Ω； (2)三相短路电流的计算 （3）电缆热稳定校验 由于断路器的燃弧时间及固有动作时间之和约为t=0.05S; 查附表二得热稳定计算系数取K=142; 故电缆最小热值稳定截面为 Smin<50mm2故选用 MYJV22 -8.7/10KV 3*50 电缆热稳定校验合格，符合要求。 附表一：三相电缆在工作温度时的阻抗值（Ω/Km）

2016年最新英国高中alevel排名

https://www.wendangku.net/doc/682366515.html, 排名学校名称所在地学校 类型 参加 考试 人数 获 得 成 绩 单 科 A 的 科 目 获得 成绩 单科 B的 科目 单科A 或B成 绩的科 目 参考 总科 目 优秀成绩 比例 1 St Paul's Girls' School London 女校 100 301 38 339 34 2 99.12% 2 King Edward VI High Sch for Girls W Mids 女校78 197 57 254 258 98.45% 3 City of London School For Girls London 女校71 193 33 226 232 97.41% 4 Westminster School London 合校 172 58 5 85 670 689 97.24% 5 North London Collegiate School - The L ondon 女校 112 311 50 361 373 96.78% 6 St Helen & St Katharine - The School of Oxon 女校77 178 53 231 239 96.65% 7 Haberdashers' Aske's Sch for Girls H erts 女校 113 272 63 335 347 96.54% 8 Withington Girls' School Manch 女校67 159 42 201 209 96.17% 9 Channing School London 女校46 96 41 137 143 95.80% 10 St Paul's School London 男校 168 432 105 537 561 95.72% 11 Downe House Berks 女校73 167 51 218 228 95.61% 12 Wycombe Abbey School Bucks 女校81 205 43 248 260 95.38% 13 Magdalen College School Oxon 男校73 205 36 241 253 95.26% 14 Eton College Berks 男校 264 679 193 872 922 94.58% 15 Godolphin & Latymer School - The London 女校 100 225 75 300 319 94.04% 16 Haberdashers' Aske's Boys' School - The Herts 男校 159 362 105 467 497 93.96% 17 Royal Grammar School Surrey 男校 112 279 68 347 370 93.78%

最新英国中学排名

最新英国高中排名，包括学校名称，学校性质，所在地，学校类型，平均每科得分 1 圣保罗女子学校St Paul‘s Girls’ School 私立伦敦地区女校 267.4 2 威斯敏斯特学校Westminster School 私立伦敦地区混校 266.3 3 皮尔斯女子学校Perse School for Girls 私立剑桥郡-伦敦北80km 混校 266 4 吉尔福德高中 Guildford High School 私立萨里-伦敦西南边女校 265.1 5 威辛顿女子学校Withington Girls‘ School 私立曼彻斯特地区,奥尔德姆女校 265 6 莫德林学院学校 Magdalen College School 私立牛津郡男校 265 7 圣保罗学校St Paul’s School 私立伦敦地区泰晤士河畔里士满男校 264.7 8 伦敦城市女子学校City of London School for Girls 私立伦敦,格林威治女校 264.2 9 威克姆阿贝学校Wycombe Abbey School 私立白金汉郡女校 264.1 10 皇家文理学校Royal Grammar School 私立萨里-伦敦西南边男校 263.9 11 牛津中学Oxford High School GDST 私立牛津郡女校 263.8 12 詹姆斯艾伦女子学校James Allen‘s Girls’ School 私立伦敦地区女校 263.1 13 北伦敦学院学校North London Collegiate School 私立大伦敦哈罗区女校 262.7 14 南汉普斯德高级中学 South Hampstead High School 私立伦敦卡姆登女校 262.3

传热学上海理工大学硕士研究生入学考试试题

2004年上海理工大学硕士研究生入学考试试题考试科目：传热学准考证号：得分： 一、问答题(每题5分) 1. 一无内热源平板沿厚度x方向发生一维稳态导热，其一侧表面上的温度梯度 =30 ℃/m，导热系数λ1=40W/(m.℃)，如果其另一侧表面上的导热系数λ2=50W/(m.℃)，问这一侧表面上的温度梯度是多少? 2. 解释毕渥准则数Bi的物理含义，并说明为什么用Bi判别非稳态导热问题能否采用集总参数法求解。 3. 图1.1示出了常物性、有均匀内热源、二维稳态导热问题局部边界区域的网格配置，试用元体平衡法建立节点0关于温度t的有限差分方程式（设 ，所需参数的符号自己设定）。 4. 当条件相同时，物体在空气中冷却快还是在水中冷却快？这一现象说明对流换热与什么因素相关？ 5. 试用简图表示流体沿平板流动时速度边界层的发展并说明速度边界层内分成哪些区域？ 6. 试解释普朗特数Pr的物理意义,并示意性的画出Pr>1时的速度边界层和热边界层厚度沿板长的变化(速度边界层和热边界层要画在同一图上以便比较)。 7. 说明温度附面层的概念及附面层能量微分方程在物理上忽略了哪部分换热。 8. 在应用管内旺盛紊流实验关联式时，当流体与换热壁面温差较大时需要对计算结果修正，为什么? 9. 试说明为什么一个细长圆柱水平放置时自然对流换热一般大于竖直放置时的自然对流换热？ 10.在稳定膜态沸腾过程中，为什么换热系数随 增加而迅速上升？

11.试说明大气中CO2含量增高为什么会出现大气温室效应？ 二、计算题 1. （10分）一直径为5cm的钢球，其初始温度为500℃，突然被置于温度为 30℃的空气中。设钢球表面与周围环境的对流换热系数为10 W/m2℃，试计算钢球非稳态导热的时间常数及其被冷却到300℃所需的时间。已知钢球的比热为c=0.48kJ/kg℃, ρ=7753kg/m3, λ=33W/m℃。 2. （20分）长10m、外径133mm的水平管道通过一大房间，房间壁面及其内 的空气温度均为30℃。若管道表面温度为90℃、黑度为0.9，求管道的散 热量（自然对流换热的努塞尔特数用下式计算）。3. （22分）如图2所示为一半径R=1m的半球，球冠3绝热。底面1和2的 温度分别为500℃和100℃，黑度都为0.9，求底面1和2间的辐射散热量。 4. （23分）温度为95℃的热空气流经一内径100mm、厚度6mm的圆管，管 壁导热系数为22 W/m℃。管外环境温度为30℃，管外壁与环境的总换热系数为10 W/m2℃。若管内空气质量流量为407kg/h，求管出口空气温度降低到65℃时的管长(不需考虑修正)。 三、理论题 1.（8分）一厚度为2δ的无内热源薄平板，其导热系数和初始温度分别为 λ和t0，突然被插在温度为t f的流体中。平板表面与流体的换热系数为h，给出问题的完整数学描述。 2. （12分）绕流平板换热的附面层积分方程为： 平板温度为t W，来流速度和温度分别为u∞和t∞，若Pr<<1，可以忽略速

案例--变电所母线桥的动稳定校验

案例--变电所母线桥的动稳定校验 朱时光修改 下面以35kV/10kv某变电所#2主变增容为例来谈谈如何进行主变母线桥的动稳定校验和校验中应注意的问题。 1短路电流计算 图1为某变电所的系统主接线图。(略) 已知#1主变容量为10000kVA，短路电压为7.42%，#2主变容量原为1000为kVA 增容为12500kVA，短路电压为7.48%。 取系统基准容量为100MVA，则#1主变短路电压标么值 X1＝7.42/100×100×1000/10000＝0.742， #2主变短路电压标么值 X2＝7.48/100×100×1000/12500＝0.5984 假定某变电所最大运行方式系统到35kV母线上的电抗标么值为0.2778。 ∴#1主变与#2主变的并联电抗为： X12=X1×X2/(X1+X2)＝0.33125； 最大运行方式下系统到10kV母线上的组合电抗为： X＝0.2778+0.33125＝0.60875 ∴10kV母线上的三相短路电流为：Id=100000/0.60875*√3*10.5=9.04KA，冲击电流：I s h=2.55I d=23.05KA。 2动稳定校验

(1)10kV母线桥的动稳定校验： 进行母线桥动稳定校验应注意以下两点： ①电动力的计算，经过对外边相所受的力，中间相所受的力以及三相和二相电动力进行比较，三相短路时中间相所受的力最大，所以计算时必须以此为依据。 ②母线及其支架都具有弹性和质量，组成一弹性系统，所以应计算应力系数，计及共振的影响。 根据以上两点，校验过程如下： 已知母线桥为8×80mm2的铝排，相间中心线间距离A为210mm，先计算应力系数： 6Kg/Cm2， ∵频率系数N f＝3.56，弹性模量E＝0.71×10 -4kg.s2/cm2，绝缘子间跨距2m， 单位长度铝排质量M＝0.176X10 截面惯性矩J=bh3/12=34.13c m4或取惯性半径（查表）与母线截面的积， ∵三相铝排水平布置，∴截面系数W＝bh2/6＝8.55Cm3， 则一阶固有频率: f0=(3.56/L2)*√(EJ/M)=104(Hz) 查表可得动态应力系数β＝1.33。 ∴铝母排所受的最大机械应力为： σMAX＝1.7248×10-3I s h2（L2/Aw）×β＝270.35 kg/c m2<σ允许=500 根据铝排的最大应力可确定绝缘子间允许的最大跨距为：（简化公式可查表） L MAX=1838√a/ I s h=366(c m) ∵某变主变母线桥绝缘子间最大跨距为2m，小于绝缘子间的最大允许跨距。

2020年传热学考研大纲——上海理工大学材料科学与工程学院

2020年传热学考研大纲——上海理工大学材料科 学与工程学院 传热学A《传热学》杨世铭、陶文铨，高等教育出版社，2006年 二、基本要求 1.掌握热量传递的三种方式(导热、对流和辐射)的基本概念和基本定律; 2.能够对常见的导热、对流、辐射换热及传热过程进行定量的计算，并了解其物理机理和特点，进行定性分析; 3.对典型的传热现象能进行分析，建立合适的数学模型并求解; 4.能够用差分法建立导热问题的数值离散方程，并了解其计算机求解过程。 三、主要知识点 第一章绪论：热量传递的三种基本方式;导热、对流和热辐射的基本概念和初步计算公式;热阻;传热过程和传热系数。 第二章导热基本定律和稳态导热：温度场、温度梯度;傅里叶定律和导热系数;导热微分方程、初始条件与边界条件;单层及多层平壁的导热;单层及多层圆筒壁的导热;通过肋端绝热的等截面直肋的导热;肋效率;一维变截面导热;有内热源的一维稳态导热。 第三章非稳态导热：非稳态导热的基本概念;集总参数法;描述非稳态导热问题的数学模型(方程和定解条件); 第四章导热问题的数值解法：导热问题数值解法的基本思想;用差分法建立稳态导热问题的数值离散方程。 第五章对流换热：对流换热的主要影响因素和基本分类、牛顿冷却公式和对流换热系数的主要影响因素;速度边界层和热边界层的概念;横掠平板层流换热边界层的微分方程组;横掠平板层流换热边界

层积分方程组;动量传递和热量传递比拟的概念;相似的概念及相似 准则;管槽内强制对流换热特征及用实验关联式计算;绕流单管、管 束对流换热特征及用实验关联式计算;大空间自然对流换热特征及对流换热特征及用实验关联式计算。 第六章凝结与沸腾换热：凝结与沸腾换热的基本概念;珠状凝结与膜状凝结特点;膜状凝结换热计算;影响膜状凝结的因素;大容器饱和沸腾曲线;影响沸腾换热的因素。 第七章热辐射基本定律及物体的辐射特性：热辐射的基本概念;黑体、白体、透明体;辐射力与光谱辐射力;定向辐射强度;黑体辐射基本定律：普朗克定律，维恩定律，斯忒藩—玻尔兹曼定律，兰贝 特定律;实际固体和液体的辐射特性、黑度;灰体、基尔霍夫定律。 第八章辐射换热的计算：角系数的概念、性质、计算;两固体表面组成的封闭系统的辐射换热计算;表面热阻;空间热阻;多表面系统辐射换热的网络法计算;辐射换热的强化与削弱、遮热板;辐射换热 系数和复合换热表面传热系数;气体辐射特点。 第九章传热过程分析与换热器计算：传热过程及传热系数的计算;临界绝热直径;换热器型式及对数平均温差;用平均温差法进行换热 器的热计算;换热器效能ε的概念和定义;强化传热。

铜排动热稳定校验

都是需要考虑的，特别是母桥距离比较长的时候。需要计算出现短路故障时的电动力，绝缘子类固定件的安装距离、绝缘子安装件的抗屈服力等。不很少有人会特别计算，我感觉是大家都自觉不自觉的把母线规格放大了，所以基本上不用计算。 4 母线的热效应和电动力效应 4.1母线的热效应 4.1.1母线的热效应是指母线在规定的条件下能够承载的因电流流过而产生的热效应。在开关设备和控制设备中指在规定的使用和性能条件下，在规定的时间内,母线承载的额定短时耐受电流（IK）。 4.1.2根据额定短时耐受电流来确定母线最小截面 根据GB3906-1991《3.6-40.5kV交流金属封闭开关设备和控制设备》[附录F]中公式：S=（I/a）(t/△θ)1/2来确定母线的最小截面。 式中： S—母线最小截面,mm2； I--额定短时耐受电流，A； a—材质系数，铜为13，铝为8.5； t--额定短路持续时间，s； △θ—温升（K）,对于裸导体一般取180K，对于4s持续时间取215K。 如对于31.5kA/4S系统,选用铜母线最小截面积为： S=（31500/13）×（4/215）1/2=330 mm2 铝母线最小截面积与铜母线最小截面积关系为： SAl=1.62SCu 式中, SAl为铝母线的最小截面积；SCu为铜母线的最小截面积。 如对于31.5kA/4S系统,铝母线最小截面积为： SAl=1.62×330 =540 mm2 根据DL404-1997《户内交流高压开关柜订货技术条件》中7.4.3条规定，接地汇流排以及与之连接的导体截面，应能通过铭牌额定短路开断电流的87%，可以计算出各种系统短路容量下（短路时间按4S）的接地母线最小截面积。 如对于31.5kA/4S系统,接地铜母线最小截面积为： S=330×86.7% =287mm2 根据以上公式计算，对应各种额定短时耐受电流时，开关设备和控制设备中对应几种常用的额定短时耐受电流，母线最小截面及所用铜母线和铝母线的最小规格见表1： 表1 母线kA/4s 25 31.5 40 63 80 设备中铜母线规格50×6 60×6 80×6或60×8 80×10 100×10 接地铜母线规格50×5 50×6 50×8 80×8 80×10 设备中铝母线规格80×6或60×8 80×8 100×8或80×10 设备中铜母线 最小截面（mm2）260 330 420 660 840 设备中铝母线 最小截面（mm2）425 540 685 1075 1365 4.2 母线的电动力效应 母线是承载电流的导体，当有电流流过时势必在母线上产生作用力。母线受电流的作用力与

2004年上理传热学研究生考试

2004年上海理工大学硕士研究生入学考试试题 考试科目：传热学准考证号：得分： 一、问答题(每题5分) 1. 一无内热源平板沿厚度x方向发生一维稳态导热，其一侧表面上的温度梯度=30 ℃/m，导热系数λ1=40W/(m.℃)，如果其另一侧表面上的导热系数λ2=50W/(m.℃)，问这一侧表面上的温度梯度是多少? 2. 解释毕渥准则数Bi的物理含义，并说明为什么用Bi判别非稳态导热问题能否采用集总参数法求解。 3. 图1.1示出了常物性、有均匀内热源、二维稳态导热问题局部边界区域的网格配置，试用元体平衡法建立节点0关于温度t的有限差分方程式（设，所需参数的符号自己设定）。 4. 当条件相同时，物体在空气中冷却快还是在水中冷却快？这一现象说明对流换热与什么因素相关？ 5. 试用简图表示流体沿平板流动时速度边界层的发展并说明速度边界层内分成哪些区域？ 6. 试解释普朗特数Pr的物理意义,并示意性的画出Pr>1时的速度边界层和热边界层厚度沿板长的变化(速度边界层和热边界层要画在同一图上以便比较)。 7. 说明温度附面层的概念及附面层能量微分方程在物理上忽略了哪部分换热。

8. 在应用管内旺盛紊流实验关联式时，当流体与换热壁面温差较大时需要对计算结果修正，为什么? 9. 试说明为什么一个细长圆柱水平放置时自然对流换热一般大于竖直放置时的自然对流换热？ 10.在稳定膜态沸腾过程中，为什么换热系数随增加而迅速上升？ 11.试说明大气中CO2含量增高为什么会出现大气温室效应？ 二、计算题 1. （10分）一直径为5cm的钢球，其初始温度为500℃，突然被置于温度为30℃的空 气中。设钢球表面与周围环境的对流换热系数为10 W/m2℃，试计算钢球非稳态导热 的时间常数及其被冷却到300℃所需的时间。已知钢球的比热为c=0.48kJ/kg℃, ρ =7753kg/m3, λ=33W/m℃。 2. （20分）长10m、外径133mm的水平管道通过一大房间，房间壁面及其内的空气温 度均为30℃。若管道表面温度为90℃、黑度为0.9，求管道的散热量（自然对流换 热的努塞尔特数用下式计算）。 3. （22分）如图2所示为一半径R=1m的半球，球冠3绝热。底面1和2的温度分别为 500℃和100℃，黑度都为0.9，求底面1和2间的辐射散热量。 4. （23分）温度为95℃的热空气流经一内径100mm、厚度6mm的圆管，管壁导热系数 为22 W/m℃。管外环境温度为30℃，管外壁与环境的总换热系数为10 W/m2℃。若

英国高中alevel排名

排名学校名称所在地 学 校 类 型 参加 考试 人数 获 得 成 绩 单 科 A 的 科 目 获 得 成 绩 单 科 B 的 科 目 单科A 或B成 绩的 科目 参考 总科 目 优秀成 绩比例 1 St Paul's Girls' School London 女 校 100 301 38 339 342 99.12% 2 King Edward VI High Sch for Girls W Mids 女 校 78 197 57 254 258 98.45% 3 City of London School For Girls London 女 校 71 193 33 226 232 97.41% 4 Westminster School London 合 校 172 585 85 670 689 97.24% 5 North London Collegiate School - The London 女 校 112 311 50 361 373 96.78% 6 St Helen & St Katharine - The School of Oxon 女 校 77 178 53 231 239 96.65% 7 Haberdashers' Aske's Sch for Girls Herts 女 校 113 272 63 335 347 96.54% 8 Withington Girls' School Manch 女 校 67 159 42 201 209 96.17% 9 Channing School London 女 校 46 96 41 137 143 95.80% 10 St Paul's School London 男 校 168 432 105 537 561 95.72% 11 Downe House Berks 女 校 73 167 51 218 228 95.61% 12 Wycombe Abbey School Bucks 女 校 81 205 43 248 260 95.38% 13 Magdalen College School Oxon 男 校 73 205 36 241 253 95.26% 14 Eton College Berks 男 校 264 679 193 872 922 94.58% 15 Godolphin & Latymer School - The London 女 校 100 225 75 300 319 94.04% 16 Haberdashers' Aske's Boys' School - The Herts 男 校 159 362 105 467 497 93.96%