Abstract:
When combustion noise inside boilers, acoustic soot blower noise, and pipeline leakage noise coexist, a complex acoustic environment is formed, posing challenges to the accuracy and reliability of leakage acoustic detection. This paper establishes a three-dimensional acoustic model of a 600 MW power plant boiler based on COMSOL and uses a ray acoustics system to study the frequency response characteristics of the furnace wall under three types of sound sources individually, in paired coupling, and simultaneously. First, the modal distribution features and system degeneracy patterns of the boiler cavity are revealed through rectangular cavity natural frequency analysis; subsequently, the cavity impulse response is extracted, yielding a broadband reverberation time T
60 of 5.66s. The results show that:(1)The three types of acoustic sources exhibit significantly different frequency-response characteristics. Flame noise presents a broadband, medium-to-low-frequency distribution, with a peak sound pressure level of 79.87dB at 616.7Hz. Sootblower noise is dominated by low-frequency components, with a peak of 98.67dB at 210.1Hz. Leakage noise exhibits a strong broadband characteristic, with a peak sound pressure level of 105.30dB at 3929.4Hz, making it the strongest of the three sources in terms of acoustic energy.(2) The 2500-5000Hz range is identified as the optimal frequency band for leakage detection. Within this frequency range, the leakage signal is 28.9-33.3dB higher than the flame-noise background and 34.0-40.7dB higher than the sootblower-noise background. The introduction of leakage increases the overall sound pressure level by approximately 29.0-33.3dB when the sootblower is off and 27.6-34.2dB when the sootblower is on. In addition, the dominant leakage peak exhibits a broad-band characteristic spanning approximately 2000Hz, indicating that the detection system does not rely on precise identification of a narrow resonance peak.(3)The temporal evolution of the acoustic field exhibits three distinct stages:"source-driven stage → energy accumulation stage → reverberant steady-state stage." After reaching the steady state, the sound pressure level distribution over the entire furnace-wall surface becomes relatively uniform. These findings provide a theoretical basis and quantitative reference for acoustic leakage-detection strategies under multi-source operating conditions in boilers.