Mechanical reliability design

Author: Liu Wei Xin
Publisher:
Publish Date: 2000-01-01
Features: This book systematically elaborates on the basic theories and methods of mechanical reliability design. The content includes: fundamental concepts and related terms and definitions of reliability, basic mathematical principles of reliability, principles and reliability calculation of mechanical reliability design, mechanical static strength reliability design, mechanical fatigue strength reliability design, reliability design of mechanical friction parts, system reliability design, reliability testing, reliability design of mechanical components, reliability optimization design of mechanical components, reliability design of repairable systems, etc. The book is accompanied by calculation examples and tables. It can serve as a textbook and teaching reference for senior undergraduate and graduate students majoring in mechanical engineering in higher vocational colleges, as well as for engineering technicians engaged in mechanical design, automotive design, research, manufacturing, testing, and maintenance.
Excerpt: Another important aspect of reliability design is reliability allocation, which involves reasonably distributing the allowable failure probability specified for the system to its components. In reliability design, the use of optimization methods for system reliability allocation is one of the key directions in current reliability research, known as reliability optimization design. In reliability design, redundancy design or standby methods are sometimes employed. Redundancy or standby methods involve configuring spare parts or equipment in the system to replace the original parts or equipment when they fail. Parallel redundancy, also known as parallel standby, is a method where a batch of components or equipment performing the same function work simultaneously (in parallel), and when one or some of them fail, the remaining ones still ensure the normal operation of the system. The use of standby methods in system design can significantly improve the reliability of the system. The allocation of system standby also widely employs optimization methods.
Since the specific problems handled in reliability engineering vary across different fields, the content may differ, but all approaches adopt a systematic and comprehensive methodology, with a long-term perspective to study issues. They not only emphasize technology but also management, aiming to achieve the maximum economic benefits and operational safety of the system. Mechanical reliability design, also known as mechanical probabilistic design, is one of the main components of reliability engineering and its application in mechanical design. Due to the increasing understanding of mechanical failure mechanisms, the gradual accumulation of failure probability data, and the application of probability and statistics in stress and strength analysis of mechanical components, etc., mechanical reliability design has been provided with theoretical foundations and practical experience, expanding the application of reliability theory to structural design, strength analysis, fatigue research, and other areas. Fracture mechanics analysis incorporates reliability theory, forming a branch called probabilistic fracture mechanics. It is an engineering method that uses probabilistic statistical theories and methods to analyze the occurrence of fractures in cracked materials or mechanical components.
When using traditional mechanical design methods, the probability of component failure during operation cannot be predicted. This is because the data used in the design, such as loads and material properties, are average values without considering data dispersion. Additionally, to ensure mechanical reliability, factors such as load coefficients and size coefficients are often multiplied, and a safety factor is considered. This traditional method reflects empirical estimates of these random variations. At the same time, it indicates that due to the inability to precisely calculate these random variations, the dimensions and weights of the mechanical system are empirically but inaccurately increased. Even so, traditional mechanical design methods are still unsatisfactory for product designs requiring high reliability. In contrast, the results obtained using mechanical reliability design methods are closer to actual conditions.
In mechanical reliability design, loads, material properties, strength, and component dimensions are all treated as statistical quantities belonging to a certain probability distribution. Using probability and mathematical statistics theories and strength theories, the probability formula for components not to fail under given design conditions is derived. These formulas can then be used to determine component dimensions or their safe lifespan under a given reliability level.
The characteristics of mechanical reliability design are as follows: First, it adopts reliability or other reliability indicators to ensure structural reliability, whereas traditional mechanical design uses safety factors. Therefore, the mechanical reliability design method provides a more reasonable understanding and estimation of failure possibilities. Second, in addition to introducing reliability or other reliability indicators, mechanical reliability design also statistically analyzes the safety factor, making it more realistic compared to traditional mechanical design, as it is already linked to reliability.
From the perspective of evaluating structural safety, traditional mechanical design only has one indicator: "safety factor," while mechanical reliability design has two indicators: reliability and safety factor (at a certain reliability level).
1.2 Definition of Reliability
The general understanding of reliability (Reliability) is that it represents the long-term reliability of products such as components, assemblies, parts, subassemblies, assemblies, machines, equipment, or entire systems under normal operating conditions, as well as the stability of their performance over time. This concept includes not only quantitative concepts like probability and statistics but also factors such as expected operating conditions, satisfaction with performance during normal operation, and the duration of normal operation.
The earliest definition of reliability was proposed by the U.S. AGREE in its 1957 report. In 1966, the U.S. MIL-STD-721B provided a more formal traditional or classical definition of reliability: "The ability of a product to perform its intended function under specified conditions and for a specified time." This definition has been referenced in standards worldwide, and China's GB3187-82 reliability definition is also the same. However, in practical applications, the limitations of this definition have become apparent, as it only reflects the ability to complete a mission successfully.
In 1980, the U.S. MIL-STD-785B, issued under the DODD 5000.40 directive (Defense Acquisition Directive for Critical Weapon Systems), divided reliability into mission reliability and basic reliability. The definition of mission reliability is: "The ability of a product to perform its intended function within a specified mission profile." It reflects the probability of a product successfully completing its mission, focusing only on critical failures that endanger mission success. The definition of basic reliability is: "The duration or probability of a product operating without failure under specified conditions." It includes all failures over the entire life of the product and reflects requirements for maintenance manpower and logistics support, such as MTBF (Mean Time Between Failures) and MCBF (Mean Cycles Between Failures).
The division of reliability concepts into two different purposes is a result of the U.S. Department of Defense's summary of practical experience in reliability work and deepened understanding of this issue. This is undoubtedly a significant development. China's military standard GJB450-88, issued in 1988, also adopts these two new reliability definitions.
The reliability of a product can be measured by its reliability (Reliability), which is expressed in probability and represents the degree of reliability of the product. The above definition of reliability includes the following factors:
(1) Object of Reliability Study
The research object of reliability issues is products, which are general terms that can include components, assemblies, parts, subassemblies, assemblies, machines, equipment, or even entire systems. When studying reliability, it is essential to first clarify the object. This not only involves determining the specific product but also its content and nature. If the research object is a system, it not only includes hardware but also software, as well as human judgment and operational factors. A human-machine system perspective is required to observe and analyze the issue.
(2) Operating Conditions
Operating conditions include transportation conditions, storage conditions, environmental conditions during use (such as temperature, pressure, humidity, load, vibration, corrosion, wear, etc.), usage methods, maintenance levels, and operational levels. These expected transportation, storage, and operational conditions significantly impact reliability.
(3) Specified Time
Reliability is closely related to the specified time of use, as reliability is a time-dependent definition. The time requirement must be clear. Time can be an interval (0, t) or (t, t). Sometimes, other indicators equivalent to time may be more precise for certain products, such as mileage (distance) for cars. For some products, reliability may be more appropriately defined in terms of cycles or frequency.