Review on condition based maintenance technology o

Summary of research on condition based maintenance technology of power equipment

1. Introduction

at present, the planned maintenance system used by most power equipment in the power system has serious defects, such as frequent temporary maintenance, insufficient or excessive maintenance, blind maintenance, etc., which makes countries around the world spend a lot of money on equipment maintenance every year. How to reasonably arrange the maintenance of power equipment, save maintenance costs, reduce maintenance costs, and ensure high reliability of the system is an important topic for system operators. With the application of sensor technology, microelectronics, computer software and hardware, digital signal processing technology, artificial neural network, expert system, fuzzy set theory and other integrated intelligent systems in condition monitoring and fault diagnosis, the research on condition based maintenance based on equipment condition monitoring and advanced diagnosis technology has developed and become an important research field in power system. The direct benefits of implementing condition based maintenance in power system include: ① saving a lot of maintenance costs; ② Improve the availability factor of the power plant; ③ Extend the service life of the equipment; ④ Increase power generation capacity; ⑤ Ensure the reliability of power generation and supply; ⑥ Reduce maintenance costs and risks. This paper mainly introduces the evolution of maintenance system, the development of condition based maintenance and the problems faced by condition based maintenance

2. Evolution of maintenance system

the evolution of maintenance concept has gone through two stages: post maintenance/fault maintenance (the first industrial revolution in the 18th century) and preventive maintenance (the second industrial revolution in the 19th century)

BM (break maintenance), also known as cm (corrective maintenance), is the earliest maintenance method. This maintenance method takes the functional failure of the equipment as the criterion, and the maintenance is carried out only when the equipment fails and cannot continue to operate. Obviously, this kind of emergency maintenance needs to pay a lot of costs and maintenance costs, which not only seriously threatens the safety of equipment or personnel, but also insufficient maintenance

during the second industrial revolution, preventive maintenance (PM) was introduced. After years of development, preventive maintenance has appeared in the following seven ways according to the different technical conditions and objectives of maintenance

(1) time based maintenance (TBM). The regular maintenance system was not transplanted from military industry to civil industry until after World War II. The periodic maintenance system of China's power industry was introduced from the Soviet Union in the 1950s. Until the 1980s, TBM was still the mainstream maintenance system. Regular maintenance does play a direct role in preventing or delaying failures in ensuring the normal operation of major mechanical equipment. However, this maintenance method, which does not disassemble the equipment to a certain extent according to the actual condition of the equipment and only according to the specified time interval, will inevitably produce excess maintenance, which will not only cause the loss of effective use time of the equipment and the waste of human, material and financial resources, but also cause maintenance failures. According to statistics, in 1996, 36%, 31% and 41% of the 100 MW, 125 MW and 200 MW thermal power units in China had to deal with the surface combination of glass fiber and plastic to plan shutdown and reduce output, respectively, due to this excess maintenance

(2) reliability centered maintenance (RCM). RCM is a maintenance method aiming at achieving the inherent reliability level of mechanical equipment at the lowest cost. This maintenance method can reasonably arrange the overhaul interval and effectively prevent the occurrence of serious faults. The research of RCM began in the late 1960s, and the power industry began in 1983. In 1984, it was used for the maintenance of nuclear power plants by the American Electric Power Research Institute (EPRI). By 1997, 68% of the top 1000 large companies in the United States had adopted RCM maintenance methods

(3) condition based maintenance (CBM) or predictive diagnosticmaintenance (PDM). This maintenance method is based on the current actual working conditions of mechanical equipment. Through high-tech condition monitoring means, it can identify the early symptoms of the fault, and judge the fault location, severity and development trend, so as to determine the best maintenance time of each machine part. Condition based maintenance began in 1970 and was first proposed by i.d.quinn of DuPont. Condition based maintenance is the current maintenance system with the lowest cost and the most advanced technology. Compared with san320 and PN (1) 17c, they need more additions. It provides reliable technical and management guarantee for the safe, stable, long-term, full performance and high-quality operation of the equipment. However, due to the large amount of monitoring content, large investment and certain risks in condition based maintenance, it still needs a long time of experience accumulation to be skillfully applied to equipment maintenance

(4) fault find (FF). This maintenance method is mainly aimed at emergency standby equipment. Start these equipment after a fixed time, and solve the problems found in time, so as to improve the availability of standby equipment

(5) RTF, run to fault. The equipment repaired in this way is not controlled to be sent for repair, and is usually used for three types of faults that do not directly endanger safety: ① accidental faults; ② Irregular fault; ③ The loss of failure is less than the maintenance cost

(6) maintenance based on life assessment. The literature believes that condition based maintenance should first estimate the service life of the equipment according to the analysis of monitoring and diagnostic data, and then determine the maintenance items, frequency and maintenance content

(7) PM (proactive maintenance). Considering many factors such as economy and service life, the focus is on the identification and elimination of mechanical faults and the analysis of fault causes, so as to obtain the maximum benefits by extending the service life of power plant machines

3. Overview of the development of condition based maintenance technology

with the development of fault diagnosis technology, condition based maintenance has gradually become practical, and has attracted extensive attention in the industry due to its huge benefits. Theoretical research and production practice are further deepened. Foreign countries have achieved relatively successful experience in the research and practical application of condition based maintenance technology. The United States, Germany, Japan and France have all reported the application of this technology. The theories and technologies that are closely related to condition based maintenance and can directly improve the quality of condition based maintenance mainly include four aspects, namely, equipment life management and prediction technology, equipment reliability analysis technology, equipment condition monitoring and fault diagnosis technology, and information management and decision-making technology

3.1 equipment life management and prediction technology

the power infrastructure in most industrialized countries has been greatly expanded between the 1960s and 1970s. Therefore, most of the main power equipment has been in service for about 25 to 30 years, and the proportion of equipment entering the aging stage is increasing. This situation forces power companies to consider how to prolong the service life of units and ensure benefits. The application of life prediction and evaluation technology in condition based maintenance is conducive to the scientific and reasonable arrangement of maintenance and the improvement of equipment availability. However, most of the benefits that power companies may obtain come from the main equipment of power plants. Therefore, all countries focus on the life prediction and evaluation of important equipment such as boilers, steam turbines, generators, transformers and high-voltage switches

(1) boiler

Japan has the most research on the residual life of boiler components in thermal power plants in the past 10 years. They adopted three representative life diagnosis techniques: stress analysis method, failure test method and non-destructive damage measurement method. Among them, the stress analysis method can evaluate the material at any position, but if the operation history or material data are inaccurate, it will lead to calculation errors, and the factor of material aging is not considered. The damage test method has higher accuracy than other methods, but it is not applicable to the parts that cannot be sampled; For this reason, Japan has developed non-destructive damage measurement methods such as micro sample method, replica metallography method, Barkhausen noise method, ultrasonic noise analysis method, etc. These methods can non destructively test the changes of physical properties of metal structures of materials while the damage of component materials progresses. The monitoring and Diagnosis Center (M & D) of the American Electric Power Research Institute (EPRI) has also developed a life management analysis software for boiler diagnosis system

(2) steam turbine generator

when carrying out condition based maintenance on steam turbine generator, we must focus on steam turbine bearing pads, blades, generator stator, rotor, shafting and other components. At present, the monitoring and Diagnosis Center (M & D) of the American Electric Power Research Institute (EPRI) has developed a blade life dynamic analysis system (blade) for steam turbine diagnosis system and a rotor crack evaluation system (safer) for generator diagnosis system, which can calculate and speculate where cracks may appear on blades and the life after cracks occur; And help engineers and technicians evaluate the remaining life of steam turbine and generator rotors and the probability of failure with operation time. Huazhong University of technology also put forward the life management system framework of steam turbine rotor, and developed a 200 MW steam turbine generator life management and fault diagnosis expert system. For the method of rotor life evaluation, there are relatively mature theories in China

for the stator of steam turbine generator, Russian researchers have formulated the main principles and basis for prolonging the service life of steam turbine generator stator on the basis of summarizing the experience of 11 different power plants in Russia. Romania has successfully developed a set of monitoring, diagnosis, data recording and prediction system for 75 MVA steam turbine generator. The prediction part mainly completes the evaluation of the remaining life of stator winding insulation and shafting

in terms of shafting, China has made some achievements in life prediction and evaluation technology. The man-disp program developed by Shanghai Jiaotong University is used to simulate the transient process of the power system under electrical disturbance, and the dynamic torsional moment of the shafting is obtained. The influence of Electrical disturbance on the fatigue life of the shafting of 300MW steam turbine generator set in 2012 is successfully evaluated. Similarly, North China Electric Power University has also studied the torsional vibration characteristics and fatigue life of 50MW Steam turbine generator that has been operating in China for nearly 30 years. Using the torsional vibration analysis model of the unit shafting with centralized parameters, and based on the on-site accident conditions, the stress-strain history and fatigue life loss of the Journal and bolts at the major shaft coupling joints of several typical accidents are simulated and calculated, so that the residual life of the unit can be properly evaluated

(3) transformer

the evaluation of transformer residual life is one of the important contents of monitoring and diagnosis. Most of the existing methods for estimating the service life of transformers only simply consider the current situation of load, temperature and insulating materials. Due to the short circuit times, overvoltage times, design weaknesses, repair and on-site transportation of transformers, the ability of transformers to function will be affected. In order to correctly estimate the service life of the transformer, it is necessary to obtain relevant operating status and historical information, and to have a deeper understanding of the technical situation of the transformer. Research and experiments show that transformers rarely quit operation due to technical or service life reasons, but are mainly limited by economic life. Therefore, in order to avoid quantitative evaluation of residual life, abb and some important European power industry departments have developed a transformer ranking method, which has done a lot of work for transformer life evaluation

(4) switch aspect

high voltage circuit breaker undertakes the dual tasks of control and protection in the power system, because it is related to the system