Current Status and Development Trends of Fault Diagnosis for Power Plant Equipment

  The main equipment of large thermal power plants includes boilers, steam turbines, and generators, which complete the conversion process from thermal energy to mechanical energy and then to electrical energy. The coupling between equipment, the complexity of the system, and the special working environment of high temperature, high pressure, and high-speed rotation determine that thermal power plants are production sites with high failure rates and significant fault hazards. These faults will cause significant economic losses and social consequences. Therefore, it is necessary to monitor and analyze equipment status parameters through advanced technical means to determine whether there are abnormalities or faults in the equipment, the location and cause of the faults, and the deterioration trend of the faults, so as to determine the appropriate maintenance timing.

    1. Typical faults and diagnostic methods of main equipment in thermal power plants

    1. Main faults and diagnostic methods of boilers

    Main faults

    (1) Superheater leakage. The leakage and tube burst area of the superheater is concentrated on the outer ring of the lower bend of the high-temperature superheater facing the fire side. The main reason is that the furnace height is relatively low, causing overheating at that location, and there are also issues of insufficient material selection margin and welding quality.

    (2) Economizer leakage. The main cause is the thinning of the tube wall due to fly ash erosion, especially at wall-penetrating tubes, furnace wall air leakage, and bends.

    (3) Water wall leakage. The main causes are local overheating and corrosion. Local overheating is caused by water circulation disruption and internal tube scaling, while flame deflection or excessive flue gas temperature in the combustion zone leads to high-temperature corrosion of the water wall.

    (4) Dust collector faults. The main reasons include excessive flue gas flow velocity, larger ash particle size, high dust concentration, and exhaust gas temperature below the dew point temperature.

    Diagnostic methods

    In boiler fault diagnosis, physical diagnostic methods include: infrared temperature measurement technology, with specific application ranges including identification and control of boiler flame and combustion state, fatigue damage of thermal equipment, thermomechanical characteristics of thermal equipment, diagnosis and evaluation of heat leakage and insulation in thermal systems, and boiler thermal pollution control; ultrasonic diagnostic methods, which can be used to monitor the flue gas temperature in the upper furnace area, determine when to perform soot blowing operations, maintain good boiler operating performance, monitor flue gas temperature near each burner area of the furnace, help identify and eliminate abnormal combustion conditions caused by burner faults, and optimize control of temperatures that significantly affect pollutant formation to achieve clean combustion; non-destructive testing technology refers to non-destructive inspection of materials and components to detect surface and internal defects. Mathematical diagnostic methods include fault tree diagnosis and fuzzy diagnosis methods. In terms of diagnostic systems, the main ones include the remote monitoring and fault diagnosis system for large utility boilers developed by Tsinghua University, and the online monitoring and condition diagnosis expert system for circulating fluidized bed boilers developed by Huazhong University of Science and Technology.

    2. Main faults and diagnostic methods of steam turbine units

    Main faults of steam turbines

    (1) Unbalance. Mainly caused by machining errors and permanent bending of the rotor, blade shedding, coupling misalignment, poor manufacturing of the extended shaft, thermal deflection due to uneven heating or material defects.

    (2) Misalignment. Mainly caused by sticking of the sliding pin system, excessive temperature difference in the cylinder, too fast unit speed-up, excessive vacuum drop, uneven foundation settlement, etc., resulting in poor expansion of the unit and uneven expansion of bearing housings, as well as coupling eccentricity.

    (3) Rotor rubbing. Main causes include improper design and adjustment of steam seal clearances, poor cylinder expansion, excessive thermal deformation of the cylinder, excessive unit vibration, and increased axial displacement.

    (4) Blade shedding. Mainly due to insufficient design strength, large fluctuations in steam parameters, long-term operation under large grid frequency changes and low load, and prolonged residence time at a speed below rated speed.

    (5) Steam excitation. Mainly due to the relatively low critical speed of the high and intermediate pressure rotors, large deflection changes of the high and intermediate pressure rotors in the hot state, large changes in bearing elevation, poor processing and installation quality of the extended shaft, excessive radial or axial error in the coupling center, and deterioration of shafting balance and alignment, causing instability of the high and intermediate pressure rotors.

    Main faults of generators

    (1) Stator winding faults. Mainly including insulation aging, moisture, and damage caused by mechanical stress, etc.

    (2) Rotor winding faults. Mainly including inter-turn short circuits, grounding faults, and poor contact.

    (3) Bearing faults. Mainly including bearing wear, overheating, and damage.

    Diagnostic methods

    In steam turbine fault diagnosis, vibration analysis is the most important method. By analyzing the vibration signals of the unit, the type and severity of faults can be determined. Common vibration analysis methods include spectrum analysis, waterfall plots, axis trajectory analysis, and wavelet analysis. In addition, oil analysis can detect wear of bearings and gears, and temperature monitoring can detect overheating faults. In recent years, intelligent diagnostic methods such as neural networks and expert systems have also been applied to steam turbine fault diagnosis.

    3. Main faults and diagnostic methods of generators

    Main faults

    (1) Stator winding faults. Mainly including insulation aging, moisture, and damage caused by mechanical stress, etc.

    (2) Rotor winding faults. Mainly including inter-turn short circuits, grounding faults, and poor contact.

    (3) Bearing faults. Mainly including bearing wear, overheating, and damage.

    Diagnostic methods

    Generator fault diagnosis mainly uses electrical tests, insulation monitoring, and vibration analysis. Electrical tests include insulation resistance measurement, DC resistance measurement, and dielectric loss testing. Insulation monitoring can detect insulation aging and moisture. Vibration analysis can detect mechanical faults such as bearing wear and rotor unbalance. In addition, partial discharge detection is an effective method for diagnosing insulation faults.

    2. Development trends of fault diagnosis technology

    (1) Intelligentization. With the development of artificial intelligence technology, fault diagnosis is moving towards intelligence. Intelligent diagnostic methods such as expert systems, fuzzy logic, neural networks, and genetic algorithms are increasingly applied in fault diagnosis. These methods can automatically identify fault types and predict fault development trends, improving diagnostic accuracy and efficiency.

    (2) Integration. Future fault diagnosis systems will integrate multiple diagnostic technologies, such as vibration analysis, oil analysis, temperature monitoring, and electrical tests, to achieve comprehensive and accurate diagnosis. At the same time, they will be integrated with the unit's control system and management system to achieve real-time monitoring and diagnosis.

    (3) Networking. With the development of Internet technology, remote fault diagnosis has become possible. Through the network, diagnostic experts can remotely monitor and diagnose equipment in different locations, achieving resource sharing and improving diagnostic efficiency.

    (4) Visualization. Visualization technology will be applied to fault diagnosis, displaying equipment status and fault information in an intuitive way, helping operators and maintenance personnel to understand and handle faults more conveniently.

    (5) Standardization. The development of fault diagnosis technology requires unified standards and specifications. Standardization will help promote the application and development of fault diagnosis technology, improving its reliability and interoperability.

    In summary, with the continuous advancement of technology, fault diagnosis technology for power plant equipment is developing towards intelligence, integration, networking, visualization, and standardization. These developments will further improve the safety and reliability of power plant equipment, reduce maintenance costs, and provide strong support for the stable operation of the power system.