CN-122014656-A - Fault early warning method for high-speed exhaust system

CN122014656ACN 122014656 ACN122014656 ACN 122014656ACN-122014656-A

Abstract

The invention relates to the technical field of fault prediction and health management, in particular to a fault early warning method for a high-speed exhaust system, which comprises the following steps of S1, setting a fan speed-up stage as a power rise period, collecting power and vibration data and generating a corresponding trend chart; the method comprises the steps of S2, fitting a curve to a trend graph, counting historical data according to power characteristic points to obtain a comprehensive radial reference curve and an axial reference curve, S3, generating upper and lower limit threshold points based on a statistical standard deviation, fitting to obtain dynamic upper and lower limit curves corresponding to power, S4, comparing real-time vibration with the dynamic threshold values, judging that no fault exists when the two directions are normal, and S5, judging working conditions through vibration variance in a power stabilizing period. The method is used for constructing a dynamic power-vibration curve aiming at the fan speed-up stage, breaks through the limitation of traditional steady-state monitoring, can realize accurate identification of early hidden faults, adopts radial and axial modeling and dynamic threshold intervals respectively, effectively reduces erroneous judgment and missed judgment, and improves fault early warning precision and system operation reliability.

Inventors

YUAN SIKUN
SHI HAIYING
SHI HAIQI

Assignees

广东生泰通风环保设备有限公司

Dates

Publication Date: 20260512
Application Date: 20260325

Claims (8)

1. The fault early warning method for the high-speed exhaust system is characterized by comprising the following steps of: The method comprises the steps of recording the process of increasing the power of a fan to rated power as a power increasing period, obtaining the running power, the radial vibration speed and the axial vibration speed of the fan in the power increasing period, generating a power-radial trend chart by taking the running power as an abscissa and the radial vibration speed as an ordinate, and generating a power-axial trend chart by taking the running power as an abscissa and the axial vibration speed as an ordinate.
2. The fault pre-warning method for a high-speed exhaust system according to claim 1, wherein a corresponding radial curve equation C1 (x) in a "power-radial" trend chart is fitted, and a corresponding axial curve equation C2 (x) in a "power-axial" trend chart is fitted; Presetting a power interval p, setting a power characteristic point every interval power p, and executing the following operations based on a power-radial trend chart: acquiring historical running records of the fan, and averaging radial vibration speeds of N radial curve equations on the same power characteristic points Wherein N represents the number of radial curve equations, a represents the operation power corresponding to the power characteristic points, and C1 n (a) represents the radial curve equation corresponding to the nth operation record of the fan; Generating radial average points (a, C1 avg (a)), repeating the steps to generate radial average points corresponding to all power characteristic points, generating a comprehensive radial curve equation C1_sta (x) through radial average point fitting, and generating a comprehensive axial curve equation C2_sta (x) in the same way.
3. The method for pre-warning faults in a high-speed exhaust system according to claim 2, characterized in that, on the same power characteristic points, the standard deviation s of the radial vibration speeds of N radial curve equations is calculated, and radial upper threshold points (a, C1 avg (a) +s) and radial lower threshold points (a, C1 avg (a) -s) are generated, and the above steps are repeated for all the power characteristic points; generating a radial upper limit curve equation C1_up (x) through radial upper threshold point fitting, and generating a radial lower limit curve equation C1_down (x) through radial lower threshold point fitting; the axial upper limit curve equation c2_up (x) and the axial lower limit curve equation c2_down (x) are generated in the same manner.
4. The fault early warning method for a high-speed exhaust system according to claim 1, wherein when a fan is started and is in a power rising period, acquiring the running power b of the fan and the radial vibration speed C1_b corresponding to the running power b in real time, if C1_down (b) is less than or equal to C1_b and less than or equal to C1_up (b), marking the radial vibration as normal, and otherwise marking the radial vibration as abnormal; And when the radial vibration and the axial vibration are normal, recording that the power rising period is normal, otherwise recording that the power rising period is abnormal, and prompting the working personnel that the exhaust system is in fault.
5. The method for pre-warning a fault in a high-speed exhaust system according to claim 4, wherein when the power rising period is normal, and the fan reaches the target power and stably operates for a preset period of time, the fan is recorded as a power stabilization period; when the power is in the stable period, the vibration sensor is arranged at a radial preset position of the driving end of the fan, the axial vibration amplitude is collected, and the preset collection frequency is 1kHz and the collection duration is longer; calculating the variance of the collected axial vibration amplitude, if the variance is larger than or equal to a preset vibration limit value, indicating that the exhaust system has faults, prompting a worker to shut down the fan and overhaul, and if the variance is smaller than the preset vibration limit value, indicating that the exhaust system is normal.
6. The fault early warning method for a high-speed exhaust system according to claim 1, wherein a power rising period corresponding to 0 to 0.5% of rated power of a fan is eliminated, and the subsequent calculation is not participated.
7. The fault pre-warning method for a high-speed exhaust system according to claim 1, wherein the power rising period acquisition frequency is not lower than 10Hz.
8. The method for fault early warning of a high speed exhaust system according to claim 1, wherein invalid records of shutdown, power cut-off, manual power adjustment are removed.

Description

Fault early warning method for high-speed exhaust system Technical Field The invention relates to the technical field of fault prediction and health management, in particular to a fault early warning method for a high-speed exhaust system. Background The high-speed exhaust system is widely applied to the scenes of rail transit, large industrial plants, underground buildings and the like, and is a key device for guaranteeing ventilation and safe smoke discharge. In the actual running process, the working condition of the fan is changed severely from the starting to the speed-up stage reaching rated power, and the impeller, the bearing and the transmission mechanism are easy to generate early hidden faults, and if the early hidden faults are not recognized and early-warned in time, the early hidden faults are easy to develop into excessive vibration, component damage and even system shutdown, so that the running stability and the use safety of the ventilation system are seriously affected. At present, fault monitoring for a high-speed exhaust system is mostly focused on a fan steady-state operation stage, attention to dynamic characteristics in a power rising process is insufficient, effective early warning is difficult to realize in a fault germination stage, and an obvious monitoring blind area exists. The existing fan fault early warning method mostly adopts a fixed threshold judgment mode, does not combine with the power change process to establish a dynamic judgment standard matched with the power change process, is easily influenced by factors such as load fluctuation, environmental interference and the like, and causes insufficient early warning accuracy. Meanwhile, most monitoring schemes do not conduct differential analysis and statistical modeling on radial and axial vibration, cannot distinguish fault sensitivity characteristics of vibration in different directions, and are prone to misjudgment or missed judgment. In addition, the prior art lacks effective invalid record rejection rules and sample statistics optimization means in the data processing process, the utilization rate of historical operation data is low, the constructed judgment standard is difficult to truly reflect the normal operation interval of the equipment, and the reliability of fault identification is further reduced. Overall, the prior art has shortcomings in early warning timeliness, judging accuracy and system applicability, and is difficult to meet long-term stable, safe and reliable operation requirements of a high-speed exhaust system. Disclosure of Invention The invention aims to provide a fault early warning method for a high-speed exhaust system, which solves the technical problems. The aim of the invention can be achieved by the following technical scheme: a fault early warning method for a high-speed exhaust system comprises the following steps: The method comprises the steps of recording the process of increasing the power of a fan to rated power as a power increasing period, obtaining the running power, the radial vibration speed and the axial vibration speed of the fan in the power increasing period, generating a power-radial trend chart by taking the running power as an abscissa and the radial vibration speed as an ordinate, and generating a power-axial trend chart by taking the running power as an abscissa and the axial vibration speed as an ordinate. Fitting a corresponding radial curve equation C1 (x) in the power-radial trend chart and fitting a corresponding axial curve equation C2 (x) in the power-axial trend chart; Presetting a power interval p, setting a power characteristic point every interval power p, and executing the following operations based on a power-radial trend chart: acquiring historical running records of the fan, and averaging radial vibration speeds of N radial curve equations on the same power characteristic points Wherein N represents the number of radial curve equations, a represents the operation power corresponding to the power characteristic points, and C1 n (a) represents the radial curve equation corresponding to the nth operation record of the fan; Generating radial average points (a, C1 avg (a)), repeating the steps to generate radial average points corresponding to all power characteristic points, generating a comprehensive radial curve equation C1_sta (x) through radial average point fitting, and generating a comprehensive axial curve equation C2_sta (x) in the same way. Calculating standard deviation s of radial vibration speeds of N radial curve equations on the same power characteristic points to generate radial upper threshold points (a, C1 avg (a) +s) and radial lower threshold points (a, C1 avg (a) -s), and repeating the steps for all the power characteristic points; generating a radial upper limit curve equation C1_up (x) through radial upper threshold point fitting, and generating a radial lower limit curve equation C1_down (x) through radial lower threshold point fitting