Fault Tolerant Systems

Another alternative derating, which was described in the previous chapter, is application of fault tolerant structures for the power converter. Fault tolerance is the property that enables a system to continue operating properly in the event of a failure of (or one or more faults within) some of its components. Fault tolerant systems are systems that can be operating after fault occurrence with no degraded performance in their basic functional requirements. This is the main difference between fault tolerant systems and derated systems. In this chapter, some methods for fault tolerance in electric power converters are presented. Fault tolerance is almost the only method for achieving a desired reliability in a converter that operates with non-zero fault possibility. There are two main approaches for this aim: re-configuration of the faulty system and using redundant systems. Redundancy is the provision of functional capabilities that would be unnecessary in a fault-free environment. Various types of redundant systems as passive and active redundancy are described and their application in power supply systems is presented. A new approach for a reliable and fault tolerant power supply is proposed and justified with experimental results. The concept of fault tolerance in electrical machines is presented.

Stress Reduction

After evaluation of reliability in the previous chapters and its consideration as a converter figure of merit, in this and the next chapters, guidelines for improvement of reliability are presented. These methods are used in both design and operation process of the converter. The focus of this chapter is on the component stress reduction in the design process. Based on background of chapter two, reliability of a converter increases if it operates at a set point with low stress. It is assumed that the converter is under design process or operates without fault. The methods for reliability improvement in faulty converters are discussed in the next chapters. In this chapter, methods for reducing electric field are described at both system and printed circuit board level. Low temperature operating conditions for an electric power converter are described and tools for this goal are presented. Series connection for voltage sharing and parallel connection for current sharing is explained. Novel control methods of power converters for reducing the complexity and reliable operation are presented. Control of inrush current as a typical transient problem in electric power converters is presented. Methods for preventing the over stress condition on the components in faulty cases are described. Techniques for reducing mechanical and environmental stress are expressed. Mechanical dampers for preventing the high amplitude vibration and insulating colors against humidity are presented. Industrial and real samples are presented to demonstrate application of the proposed methods.

Reliability Measurement

After calculation of reliability, the system is constructed. It is important to measure the calculated useful life of the system. Reliability measurement tests can also be used for the converter operating in service to estimate the remaining useful life of the converter. In this chapter, various methods of tests for this goal are presented. The main approach is accelerated aging test that reduce the time needed for failure in a system. In this method, the device is tested under conditions beyond its defined nominal specifications. Limits for this harsh condition is determined based on the calculations which are presented in chapters 3 and 4. If a problem occurs in implementing and operating process of the converter, accelerated aging tests decrease the time to failure. Theoretical concept of accelerated aging tests is described. Standard tests of electric power converters are presented. Equipment and test chambers for standard tests are explained. These tests contain all four various failure factors which are presented in chapter 2. Sample industrial examples are presented to demonstrate the procedure of the tests. Some accelerated aging tests may lead to destruction of the converter. Difference between destructive and nondestructive tests is presented. Sample devices after accelerated aging tests are shown. Measuring devices for system parameter identification are introduced. Various types of tests are expressed in details for some of the most important tests like electric withstand tests.

Condition Monitoring

Implementation of all previous methods for reliability improvement needs to have enough information about condition of the converter. This is the topic of the last chapter of this book. Condition monitoring is the process of monitoring a parameter of condition in machinery (vibration, temperature etc.), in order to identify a significant change which is indicative of a developing fault. The use of conditional monitoring allows maintenance to be scheduled, or other actions to be taken to prevent failure and avoid its consequences. In this chapter, commonly used methods for condition monitoring the converters and electric machines are presented. The aim of this task is producing an alarm in converter before failure factor damage the system. Sensor based and sensor less methods for converter and motor parameter monitoring are described. The data obtained from sensor based methods is real but sensor is a weakness point in a converter. On the other hand, sensorless methods give estimated information but they are reliable. Temperature as the most important parameter from reliability point of view is a common parameter for monitoring in all systems. Other parameters like vibration, harmonics can be used for monitoring of various faults inside the system. Many typical cases are presented to demonstrate the techniques.

Electric Power Converters

In this book, we discuss reliability in electrical energy converters. The first step is introducing these devices and recognizing their main functions as well as their importance. Electrical energy conversion systems consist of two main parts: Electrical machines and Power electronic converters. Electrical machines are used for converting electrical energy to mechanical one in the generator state and vice-versa in the motor state. To emphasize the importance of these devices, it may be noted that electrical motors consume about half of the total generated electrical energy in the world. On the other hand, power electronic converters are essential equipments which are used for electrical energy conditioning. These equipments have observed considerable growth in modern industries in recent years. Because energy conditioning allows us to use energy with higher efficiency and better performance, in this chapter, importance of electric power converters in modern industries is presented. The aim of this presentation is showing the dependence of various industrial functions to conversion of electric power. Basic relations of various electrical machines as well as power electronic converters are presented. In each section, some typical industrial examples are presented. This background will be used in the next chapters for reliability calculation and improvement. In fact, this chapter is an introduction on reasons of writing an individual book about reliability of electric power converters.

Derating

In previous chapters, we discussed the converter with or without fault. The common similarity between them is that they continue to operate without reduction of their nominal specification. In this chapter, uninterrupted operation of a faulty power conversion system with catastrophic damages in some of its parts is investigated. It is shown that a faulty electric power converter can continue to work with degraded specifications. This algorithm is named derating for accessibility. This technique can be used for both a faulty system because of its uninterrupted operation and a normal system because of extensive life time. Algorithms for derating of a faulty electric machine and a power supply are described. Derating for increasing the useful life of a motor drive system is presented. A novel method for switching frequency selection in a switching power supply is proposed based on derating concept. Derating is introduced as a technique to compensate additional losses in an electric power converter operating in harsh environment (for example: a motor drive which is supplied with a non sinusoidal voltage waveform). Industrial examples are presented in details for better understanding of the derating concept. Some of the presented examples contain novel idea for derating and others are well known in industry.

Thermal Analysis

Over temperature is one of the main reasons of failure in electric power converters. In addition, some of other failure factors such as dielectric breakdown act as over temperature in damaging process of a converter. In the previous chapter, it is emphasized that temperature factor is a key index in reliability calculation. Unlike fully electrical variables, thermal calculations require details of geometry of the system and its environment. In this chapter, thermal analysis as the most important factor in failure of converters is presented. Two main approaches for this goal are presented: numerical and lumped mode. Principles of these methods are described with various examples and a comparison is presented. Basic principles of thermal modeling are described and concept of sample node is explained. Methods for thermal management of an electric power converter are described. These methods are at both component and system levels and contain various heat transfer mechanisms like conduction and convection. Theoretical methods and practical considerations for heat sink selection and proper mounting are presented. Thermal insulation classes and various standards related to thermal management are expressed. Industrial samples are presented to show application of theoretical topics in real world.

Protection Systems

In the previous chapter, we assumed that there is no fault in the converter. To achieve a converter without failure, we presented the methods for stress reduction. However, a fault may occur in a system even operating with low stress. In the current chapter, we take one further step and assume that a fault occurs in the converter but there is a short time interval between fault occurrence and catastrophic damage to the converter. Therefore, the topic of this chapter is the methods for saving the converter in this condition. In this chapter, protection methods for saving the system against damaging faults are presented. Based on background of chapter two, protection systems should be able to bypass the effect of failure factors on electric power converter. Methods for current limiting and voltage clamping as the usual factors of failure in converters are described. Circuit diagram of a snubber is presented and its operation is described based on safe operating area of solid state power switches. Operating diagrams of fuses as emergency circuit breakers are presented. Measurement methods and devices used in protection systems are explained. Experimental samples and standard diagrams are presented to clarify the theoretical notes in all cases.

Reliability as a Figure of Merit

An electric power converter is specified by its system performance indices. Many system performance indices like efficiency are important in design process of system. However, that doesn't mean the converter performance is necessarily adequate for a practical application. At the end of design process of a converter, some desired specifications may not be achieved. In this chapter, reliability as a figure of merit in design of a system is presented and compared with other indices. We want to highlight the effect of reliability considerations on the design methodology of a power converter. The most important specification of a power supply or power converter is its robustness. Because any failure in power supply leads to failure of the whole of the system. A power converter may have poor performance but operate reliably and vice versa. In fact, this is a reliability based design approach to achieve a long useful life. It is shown that in many systems, high efficiency is not a good choice for selection of system operating point. A system can be inefficient but very reliable. Two complex examples are presented to show undesired results of neglecting reliability in design process. Methods for more reliable operation of electric power converters than high performance operation are proposed. A discussion about correct and intelligent optimization of a system parameters and operating set point is presented.

Availability

Protection methods, which were described in the previous chapter, save the converter against non-catastrophic faults. However, this method saves the converter but it also takes the converter out of the service. The subject of this chapter is converters that are not damaged but can not operate normally. In this chapter, availability of electric power converters as a most important but usually forgotten parameter is described. The concept of availability was originally developed for repairable systems that are required to operate continuously. It is explained that a system may be unavailable while none of its parts damaged. In fact, there is an important difference between reliability and availability. A converter may be highly reliable but unavailable and vice versa. One of the most important factors for this undesired state is influence of noise. In this chapter, electromagnetic interference and certain methods for reducing its undesired effects on electric power converters are presented. Electric power converters are usually the source of electromagnetic noise due to high operating voltage and/or current. Various techniques for safe operation of sensitive systems that operate close to these converters are described. In the last part of chapter, alarm management is presented based on availability concept. This method is used to prevent fast shutdown of important systems due to dispensable faults.