I. Background With the deepening of power system reform, power plants are increasingly stringent in controlling power generation costs. How to reasonably reduce maintenance costs while effectively improving operational safety has become a top priority. While main equipment such as steam turbines and boilers are critical components, their manufacturing technology is relatively mature, and monitoring technology is also quite advanced, resulting in high reliability. However, due to the complexity of thermal power plant systems, some auxiliary equipment is often a weak link in equipment condition monitoring, and is one of the main causes of unplanned unit shutdowns. Ensuring the safe operation of auxiliary equipment is an important aspect of daily maintenance and repair in power plants. Furthermore, any system or major auxiliary equipment failure will affect the economics of the power plant, increasing power generation costs. Therefore, carrying out condition monitoring of auxiliary equipment in thermal power plants to ensure the good operating condition of major auxiliary equipment and achieve optimized maintenance is of great significance. In recent years, the development of condition monitoring and diagnostic technologies for auxiliary components has been rapid, and condition monitoring technologies for auxiliary components (motors and rotating parts, etc.) are now mature. The main technologies include: 1. Vibration diagnostic technology; 2. Oil analysis technology; 3. Infrared equipment diagnostic technology; 4. Ultrasonic leak monitoring technology. Vibration monitoring technology mainly uses online and portable vibration monitoring instruments to continuously or frequently detect the vibration spectrum of equipment to analyze its vibration characteristics, determine trends in operating conditions, and provide information for equipment operation and maintenance. Oil analysis mainly detects the composition, contamination level, and wear condition of lubricating oil to understand its deterioration and determine trends in wear conditions, providing information for equipment operation and maintenance. Infrared equipment diagnostic technology mainly uses portable infrared detectors to detect overheating of motor casings to identify overheated areas and take timely repair measures. Ultrasonic leak monitoring devices also utilize the characteristics of ultrasound to detect minute leaks emitted by equipment to locate leaks and take timely repair measures. The development of auxiliary component condition monitoring technology abroad is mature, and the monitoring devices and analysis software are relatively advanced, leading to increasingly widespread application in domestic power plants. However, in application, it was found that these monitoring technologies are often independent, mainly targeting the condition of specific components, and cannot comprehensively monitor the condition of auxiliary equipment systems; they generally cannot comprehensively analyze equipment change trends, i.e., they lack comprehensive fault diagnosis capabilities. Further research is needed on how to provide overall equipment condition diagnosis results to offer more comprehensive support for maintenance decisions. II. Introduction to Key Technologies for Auxiliary Equipment Condition-Based Maintenance This research project is a sub-project of the State Power Corporation's condition-based maintenance project, and was listed as a research project of Fujian Power Corporation in 2000 as a collaborative project with Fujian Provincial Power Company, Fujian Provincial Electric Power Research Institute, and Xiamen Huaxia International Power Development Co., Ltd. The main research contents include: exploration of auxiliary equipment condition-based maintenance models; selection and implementation of auxiliary equipment condition monitoring technologies; development of system safety monitoring technologies; development of system operation economic monitoring technologies; development of a comprehensive auxiliary equipment condition diagnosis system; implementation in power plants; and, after three years of effort, the Fujian implementation project has been basically completed and passed the appraisal organized by the Fujian Provincial Science and Technology Commission. The Taicang Power Plant implementation project is still underway. 1. Discussion on the Basic Mode of Auxiliary Equipment Condition-Based Maintenance Research shows that auxiliary equipment maintenance mainly includes three aspects: maintenance due to equipment failure leading to functional degradation, maintenance due to decreased system safety, and maintenance due to decreased system performance (economy). Previous monitoring technologies primarily focused on the state changes of auxiliary equipment components, neglecting changes at the system level, which is clearly unreasonable. Currently, auxiliary equipment vibration condition monitoring methods implemented in China include online and offline methods. Online methods are expensive but provide a large amount of information and have been adopted in some power plants, such as those in Shandong. Offline monitoring methods have long been widely used in power plants. In recent years, with the improvement of monitoring instrument performance, the accuracy of offline monitoring has become quite high, fully meeting the needs of equipment condition monitoring. Therefore, there is no need to use online methods, as satisfactory results can be achieved without them. To address this, the Thermal Power Research Institute designed a new model combining offline and online monitoring of auxiliary equipment, safety monitoring and economic monitoring, and equipment monitoring and system monitoring. This model integrates offline equipment condition monitoring, online system safety monitoring, online system operation economic monitoring, and comprehensive fault diagnosis and maintenance decision support. This model fully considers the operating conditions of auxiliary equipment in Chinese power plants and the technical needs of condition-based maintenance, aiming to provide a complete overall solution for condition-based maintenance of auxiliary equipment in Chinese power plants. 2. Selection and Application of Auxiliary Equipment Condition Monitoring Technology This project was piloted on the main auxiliary equipment of the 300MW Units 1 and 2 of Xiamen Huaxia International Power Company. It adopted various mature monitoring technologies from abroad, such as vibration monitoring, oil analysis, motor monitoring, and infrared thermal imaging, to periodically monitor the condition of the main auxiliary equipment (rotating machinery) of the power plant offline, including forced draft fans, induced draft fans, primary air fans, feedwater pumps, condensate pumps, and circulating water pumps. The main monitoring contents include the vibration of auxiliary equipment, lubricating oil quality, motor operating status, rotor cage bar breakage, mechanical eccentricity between stator and rotor, and thermal images (temperature distribution maps) of the equipment. After more than two years of joint efforts from all parties, the monitoring work has gradually become standardized and achieved phased results. In terms of vibration monitoring, when the monitoring of the 1a induced draft fan began, the axial vibration of bearing No. 1 (motor extension end) and bearing No. 2 (motor coupling end) gradually increased, exceeding the qualified value of 4.5 mm/s, with maximum values ​​of 10.13 mm/s and 5.52 mm/s, respectively. In particular, the vibration of bearing No. 1 approached the danger value, seriously affecting the safe operation of the unit. According to the analysis, the vertical and horizontal vibration of bearing No. 1 were within the qualified range, at 1.2 mm/s and 3.3 mm/s, respectively, indicating that the excessive axial vibration was not caused by a large excitation force. Analysis of its spectrum diagram showed that the main components were the 3rd and 5th harmonics, and bearing No. 2 had the same problem. The preliminary analysis is that the abnormal vibration was caused by excessive working clearance of the fan rotor thrust bearing. Because the bearings of the No. 1 induced draft fan had not been replaced for five years since its commissioning, it was decided during a minor overhaul in April 2002 to disassemble and inspect bearings No. 1 and No. 2, as well as the thrust bearing of the fan. It was confirmed that the thrust bearing had excessive clearance. After replacing bearings No. 1 and No. 2 and adjusting the thrust bearing clearance, the fault was eliminated, and the vibration was within acceptable limits. In May 2001, a motor fault diagnosis instrument was used to monitor the auxiliary equipment, successfully diagnosing a cage bar breakage fault in the electric feedwater pump motor of Unit 2. Based on this, the power plant carried out timely maintenance on the motor, preventing further deterioration of the accident. On November 5th and December 10th, 2001, sampling and analysis were conducted on the bearing lubrication oil systems of 14 pieces of equipment in the auxiliary equipment of Unit 1 of the power plant, including the induced draft fan lubrication and hydraulic system, primary air fan, forced draft fan, condensate pump, steam-driven feedwater pump, electric feedwater pump, and circulating water pump. A large number of visible suspended hard particles were found in the lubrication oil tanks of the induced draft fan motors (1a and 1b). Furthermore, the circulating water pumps (1a and 1b) had not been thoroughly cleaned after the thrust bearing failure, leaving behind a large amount of wear particles. The particle size detection results all exceeded NAS12 level. Because a large number of particles exceed the filter element's precision, it will cause filter element failure and damage. Simultaneously, filter element blockage will cause unstable oil supply, affecting the thickness of the oil film on the bearing rotating surface and leading to poor lubrication. In addition, large particles entering the bearing rotating surface will cause abrasive cutting wear, accelerating bearing wear, shortening service life, and affecting the operational stability of the auxiliary equipment. Moreover, due to the extremely high particle size, it will not only mask the detection of minor wear but also clog sensors and damage instruments. Therefore, a solution was promptly proposed to the power plant. Oil tank filtration was performed to track changes in internal particle size. Infrared monitoring was used to monitor the bearings of the main auxiliary motors. Shortly after a major overhaul in May 2001, it was discovered that the bearing temperature of the 1a induced draft fan was too high. Inspection revealed that improper bearing orientation caused axial displacement of the shaft, leading to severe wear between the oil guide ring and the oil slinger ring. The bearing was replaced during a minor overhaul of the unit in April 2002, and the fault was resolved. By November 2002, the bearing temperature of the 1a induced draft fan had decreased. [b]3. Development of System Safety Monitoring Technology[/b] The safety of auxiliary systems is a crucial concern for power plants. Therefore, safety monitoring systems for flue gas systems and pump sets were developed. For example, power plant fans, especially axial flow fans, have a large stall zone. When the fan operates in this zone, the airflow pressure fluctuations within the fan are severe. When the frequency of the airflow pressure fluctuations is an integer multiple of the blade's natural frequency, it can easily cause blade resonance and breakage. Simultaneously, it also causes severe fluctuations in primary and secondary air pressure and furnace negative pressure, affecting combustion and potentially causing the unit to trip. Different types of fans have different stall zones due to their different blade shapes. We calibrate them through cold-state tests and establish a real-time stall alarm system so that measures can be taken in advance when the operating point approaches the stall zone. 4. Development of System Operation Economic Monitoring Technology The actual operating status of power plant fans reflects the flue gas resistance characteristics of boiler operation. The resistance characteristics of the boiler's flue gas system change with the extension of the unit's operating time. The actual operating parameters of the power plant fans can be used to depict the constantly changing flue gas resistance characteristics of the boiler, while also showing the changes in fan operating efficiency and the deviation between the detection dial opening and the actual opening, providing a basis for boiler overhaul and fan modification. The characteristic parameters reflecting pump unit performance mainly include temperature, pressure, flow rate, power, current, voltage, and speed. The collected state parameters are analyzed and calculated to give the pump unit's performance parameters, such as efficiency and head, and compared with the design parameters to analyze the main reasons for poor performance and point out the methods and steps for operation adjustment. [b]5. Development of an Auxiliary Equipment Condition Comprehensive Diagnosis System[/b] This system comprises two parts: a power plant flue gas system fault diagnosis system and a power plant pump set fault diagnosis system. The key to predicting, diagnosing, and maintaining power plant fan faults lies in quickly and accurately diagnosing the cause of vibration when its vibration level exceeds a set alarm value, and providing corresponding solutions based on comprehensive analysis results. Vibration faults in power plant fans mainly manifest as: bearing damage, mass imbalance, bent shafts, misaligned couplings, and mechanical loosening. Pump set fault diagnosis primarily includes shaft vibration, bearing temperature, and oil analysis. Using state parameters such as shaft vibration, bearing temperature, and hydraulic coupling working oil temperature, the system analyzes and evaluates the pump set's operating level, predicts and diagnoses pump set faults, eliminates potential hazards in a timely manner, and improves equipment availability. The Thermal Engineering Research Institute has developed a general diagnostic platform and, based on this, constructed auxiliary equipment fault diagnosis software. This software enables comprehensive fault analysis and diagnosis, including vibration, and provides solutions. Experts can establish diagnostic rules through the diagnostic platform and use these rules to simulate expert thinking, achieving condition diagnosis of the equipment. These rules can be easily revised within the power plant. The system consists of three parts: knowledge acquisition, system diagnosis, and interface design. Its main features include: It reflects the characteristics of power plant auxiliary equipment monitoring, fills some functional blind spots in auxiliary equipment operation status monitoring within the power plant's DCS and MIS systems, enhances system safety and economic monitoring capabilities, and provides decision support for maintenance and safe equipment operation; Based on the power plant equipment categories, it incorporates the necessary calculation formulas and analysis models, integrates a knowledge base of power experts, possesses diagnostic functions, and has certain configuration capabilities; It adopts advanced multi-layer distributed software development technology to improve software operating speed; The system is easy to implement, stable and reliable, easy to operate, highly scalable, has a user-friendly interface, and requires minimal maintenance. Simultaneously, the developed fault diagnosis and maintenance decision support system has remote diagnostic capabilities, allowing for a two-tier management model of local + remote management. A primary status monitoring workstation is established at the power plant to perform specific monitoring work based on different equipment and monitoring technologies, and the collected offline data is input into the fault diagnosis and maintenance decision support system. This work is completed by trained power plant inspection personnel. A remote equipment condition monitoring center is established, accessing the power plant's condition monitoring workstations remotely via a wide area network to remotely monitor the operating status of auxiliary equipment. A fault analysis and diagnosis system is used to analyze and diagnose abnormal equipment data, determine the development trend of equipment status, and regularly submit short-, medium-, and long-term trend analysis and diagnostic reports to the power plant. III. Conclusion Through three years of research and development, the Thermal Power Research Institute has achieved breakthroughs in key technologies for auxiliary equipment condition-based maintenance, mainly in the following aspects: 1. Through practical application, a new model for implementing auxiliary equipment condition-based maintenance in Chinese power plants has been proposed and established; 2. Multiple monitoring technologies, such as vibration monitoring, oil analysis, motor monitoring, and infrared thermal imaging, are integrated to achieve comprehensive offline monitoring of the operating status of major auxiliary equipment, which is more effective and less expensive than online monitoring. 3. The developed system safety monitoring system monitors the overall safety of auxiliary equipment online, broadening the monitoring scope, compensating for the shortcomings of monitoring individual devices, and achieving simultaneous monitoring of both hard and soft faults, demonstrating innovation. 4. The developed system economy monitoring system monitors the overall performance of auxiliary equipment online, establishing a method for improving maintenance decisions through monitoring economy, and realizing a new model of integrated safety and economy monitoring to rationally arrange maintenance time and cycles, demonstrating innovation. 5. The developed general-purpose diagnostic platform software is advanced, suitable for building diagnostic software for main and auxiliary equipment, meeting the needs of predictive maintenance, and providing remote diagnostic functions. 6. A remote diagnostic center was established, and an auxiliary equipment condition monitoring database was built, integrating various monitoring data under a unified database for easy data management and application. This achieves a two-tier management model between the power plant and the research institute.