CN-121997133-A - Stage suspender safety detection method and device based on feature recognition

CN121997133ACN 121997133 ACN121997133 ACN 121997133ACN-121997133-A

Abstract

The application discloses a stage suspender safety detection method and device based on feature recognition, and relates to the technical field of multi-source data processing and recognition, wherein the method comprises the steps of collecting real-time state data of a suspender; the method comprises the steps of judging whether a boom is in a static state or a moving state, calculating a displacement difference value, a strain difference value, a distance difference value and a fusion difference value in the static state, determining a static dimension positive constant according to the difference value and a static fusion threshold value, determining a static fusion safety state according to the fusion difference value and the static fusion threshold value, judging the safety state in the static state, determining a dynamic dimension positive constant according to the difference value and a dynamic threshold value in the moving state, determining a dynamic fusion safety state according to the fusion difference value and the dynamic fusion threshold value, and judging the safety state in the moving state. According to the application, the state detection is carried out from the combination of displacement, strain and distance and multiple dimensions, and the real-time performance and accuracy of the boom installation detection are effectively improved through the multi-dimensional feature recognition mode.

Inventors

HUANG PANPAN
HUANG MINZHU
SHI LIN

Assignees

西安宏源视讯设备有限责任公司

Dates

Publication Date: 20260508
Application Date: 20260123

Claims (10)

1. The stage suspender safety detection method based on the feature recognition is characterized by comprising the following steps of: acquiring real-time state data of the boom, wherein the real-time state data comprises real-time displacement data, real-time strain data, real-time distance data and real-time acceleration data; judging whether the suspender is in a static state or a moving state according to the real-time acceleration data; if the boom is in a static state, calculating a difference value between the real-time displacement data and a displacement calibration value, taking the difference value as a displacement difference value, calculating a difference value between the real-time strain data and the strain calibration value, taking the difference value as a strain difference value, calculating a difference value between the real-time distance data and a distance calibration value, taking the difference value as a distance difference value, and respectively determining a static dimension positive constant of the boom in each dimension according to the displacement difference value, the strain difference value, the distance difference value and the magnitude relation between the distance difference value and a static threshold value of the corresponding dimension; Performing weighted fusion on the real-time displacement data, the real-time strain data and the real-time distance data to obtain fusion data, calculating a difference value between the fusion data and a fusion calibration value as a fusion difference value, and determining a static fusion safety state according to the magnitude relation between the fusion difference value and a static fusion threshold value; Comprehensively judging the safety state of the boom in the static state according to the static dimension positive constant and the static fusion safety state; If the boom is in a moving state, determining a dynamic dimension positive constant of the boom in each dimension according to the displacement difference value, the strain difference value and the magnitude relation between the distance difference value and a dynamic threshold of the corresponding dimension; The dynamic fusion safety state is determined according to the magnitude relation between the fusion difference value and the dynamic fusion threshold value, the dynamic fusion threshold value is determined according to the real-time acceleration data, the dynamic threshold value is determined according to the dynamic fusion threshold value, and the safety state of the boom in the moving state is comprehensively judged according to the dynamic dimension positive constant and the dynamic fusion safety state.
2. A stage boom safety detection method based on feature recognition as claimed in claim 1, wherein the displacement calibration value, the strain calibration value, and the distance calibration value generation method comprises: the method comprises the steps of controlling a boom to be in a static state, and continuously collecting an original calibration data sequence of the boom, wherein the original calibration data sequence comprises a camera displacement sequence, a strain sensor sequence and a laser sensor sequence; denoising the original calibration data sequence; and taking an average value of the original calibration data sequence after denoising in each dimension as a calibration value of each dimension.
3. A stage boom safety detection method based on feature recognition as claimed in claim 2, wherein the method of computing the fused data comprises: extracting the maximum value in the original calibration data sequence of each dimension; calculating the ratio of the calibration value of each dimension to the corresponding maximum value; and carrying out weighted summation on the ratio of each dimension according to the weight to obtain the fusion data.
4. A stage boom safety detection method based on feature recognition as claimed in claim 2, wherein said static threshold generation method comprises: calculating corresponding standard deviation of the original calibration data sequence in each dimension; taking 3 times of standard deviation of each dimension as the static threshold of the corresponding dimension by adopting a3 times standard deviation method; the generation method of the static fusion threshold value comprises the following steps: Calculating the standard deviation of the fusion data; And adopting a 3-fold standard deviation method to take 3 times of standard deviation of the fusion data as the static fusion threshold value.
5. The stage boom safety detection method based on feature recognition according to claim 1, wherein the dynamic fusion threshold generation method comprises: smoothing the real-time acceleration data to obtain smooth acceleration; calculating mapping coefficients according to a plurality of different accelerations in a dynamic experiment: Wherein k is the mapping coefficient, Δl i is the allowable fusion deviation corresponding to each acceleration a i , Δl 0 is the static fusion threshold; Determining the dynamic fusion threshold according to the static fusion threshold, the mapping coefficient and the smooth acceleration: Wherein Δl (t) is the dynamic fusion threshold, and a smoothed (t) is the smooth acceleration.
6. A stage boom safety detection method based on feature recognition as described in claim 5, wherein said dynamic threshold is calculated according to the following equation: Wherein, the 、 And The dynamic thresholds in the displacement dimension, the strain dimension and the distance dimension respectively, 、 And The static thresholds in the displacement dimension, the strain dimension and the distance dimension, respectively.
7. The stage boom safety detection method based on feature recognition according to claim 1, wherein when a boom is in a stationary state, in each sampling period, determining the state of the boom in the stationary state as a normal state when the distance difference value, the strain difference value or the distance threshold value is smaller than the stationary threshold value of the corresponding dimension, counting the number of dimensions belonging to the normal state as the stationary dimension positive constant, determining the state of the boom in the fusion difference value is smaller than the stationary fusion threshold value as a normal state, determining the safety state of the boom in the stationary state as a normal state if the stationary dimension positive constant is equal to 3 and the stationary fusion safety state is the normal state, determining the safety state of the boom in the stationary state as a pre-warning state if the stationary dimension positive constant is equal to 2 and the stationary fusion safety state is the abnormal state, or determining the safety state of the boom in the stationary state as a dangerous state if the stationary dimension positive constant is equal to 1 or the stationary dimension positive constant is equal to 2 and the stationary fusion safety state is the abnormal state; When the boom is in a moving state, in each sampling period, determining the state of the boom as a normal state when the distance difference value, the strain difference value or the distance threshold value is smaller than the dynamic threshold value of the corresponding dimension, counting the number of dimensions belonging to the normal state as the positive constant of the dynamic dimension, determining the state of the boom as the normal state when the fusion difference value is smaller than the dynamic fusion threshold value, determining the safe state of the boom as the normal state when the dynamic dimension positive constant is equal to 3 and the dynamic fusion safe state is the normal state, determining the safe state of the boom as the normal state when the boom is in the moving state when the dynamic dimension positive constant is equal to 2 and the dynamic fusion safe state is the normal state, or determining the safe state of the boom as the early warning state when the dynamic dimension positive constant is equal to 3 and the dynamic fusion safe state is the abnormal state when the dynamic dimension positive constant is smaller than or equal to 1 or the dynamic dimension positive constant is equal to 2 and the dynamic fusion safe state is the abnormal state.
8. The stage suspension rod safety detection method based on feature recognition according to claim 7, wherein after each detection period is finished, counting the times of the suspension rod in a normal state, an early warning state and a dangerous state, calculating a safety judgment confidence, and determining a comprehensive judgment result of the suspension rod in a stationary state or a moving state according to the ratio of the safety judgment confidence to the times of the dangerous state in the total times, wherein the detection period comprises a plurality of sampling periods.
9. The stage boom safety detection method based on feature recognition according to claim 1, wherein when the boom is in a moving state, a real-time position of the boom is calculated according to an initial position, an initial speed and the real-time acceleration data of the boom, the real-time position is corrected by using the real-time distance data to obtain a corrected position of the boom, and the real-time state data of the corrected position is acquired during the movement of the boom.
10. An apparatus for applying a stage boom safety inspection method based on feature recognition as claimed in any one of claims 1-9, comprising: The sensor is used for collecting real-time state data of the suspender, wherein the real-time state data comprises real-time displacement data, real-time strain data, real-time distance data and real-time acceleration data; The system comprises a boom, a data processing device, a fusion device, a dynamic safety state determining device and a dynamic safety state determining device, wherein the data processing device is used for determining whether the boom is in a static state or a moving state according to the real-time acceleration data, calculating the difference value between the real-time displacement data and a displacement calibration value as a displacement difference value if the boom is in the static state, calculating the difference value between the real-time displacement data and the displacement calibration value as a displacement difference value, calculating the difference value between the real-time displacement data and the displacement calibration value as a strain difference value, comprehensively determining the safety state of the boom in the static state according to the static dimension positive constant and the static fusion safety state as a distance difference value, determining the static dimension positive constant of the boom in each dimension according to the displacement difference value, the strain difference value and the size relation between the distance difference value and the static threshold of the corresponding dimension respectively, weighting fusion data is carried out on the real-time displacement data and the real-time distance data to obtain fusion data, calculating the difference value between the fusion data and the fusion calibration value as a fusion difference value, determining the static fusion safety state according to the fusion state of the static safety state of the boom according to the static dimension positive constant and the dynamic safety state, and the dynamic safety state of the boom in the dynamic safety state according to the dynamic threshold, and the dynamic safety state determining the dynamic safety state of the boom in the dynamic state according to the dynamic state of the dynamic threshold and the dynamic state fusion state.

Description

Stage suspender safety detection method and device based on feature recognition Technical Field The application relates to the technical field of multi-source data processing and recognition, in particular to a method for detecting stage suspender safety by utilizing a characteristic recognition technology of multi-source data. Background The stage suspender is core bearing equipment in places such as theatres, exhibition centers and the like and is used for suspending performance equipment such as curtains, lamps and lanterns, sound equipment and the like, and the structural integrity and the installation stability of the stage suspender are directly related to performance safety. The detection method of manual visual inspection is generally adopted in the current industry, and has three major core problems that 1, the detection precision is low, potential hidden hazards such as fine deformation and hidden cracks cannot be identified, 2, the detection is not real-time, can be performed only when a performance gap or equipment is stopped, and is difficult to capture the safety risk under a dynamic working condition, 3, the subjectivity is high, the detection result depends on personnel experience, and detection omission and false detection are easy to occur. Disclosure of Invention The embodiment of the application provides a stage suspender safety detection method and device based on feature recognition, which are used for solving the problems in the prior art. In one aspect, an embodiment of the present application provides a stage suspension rod safety detection method based on feature recognition, including: Acquiring real-time state data of the boom, wherein the real-time state data comprises real-time displacement data, real-time strain data, real-time distance data and real-time acceleration data; judging whether the boom is in a static state or a moving state according to the real-time acceleration data; If the boom is in a static state, calculating a difference value between the real-time displacement data and a displacement calibration value, taking the difference value as a displacement difference value, calculating a difference value between the real-time strain data and the strain calibration value, taking the difference value as a strain difference value, calculating a difference value between the real-time distance data and a distance calibration value, taking the difference value as a distance difference value, and determining a static dimension positive constant of the boom in each dimension according to the displacement difference value, the strain difference value, the distance difference value and the magnitude relation of a static threshold value of the corresponding dimension; Carrying out weighted fusion on the real-time displacement data, the real-time strain data and the real-time distance data to obtain fusion data, calculating a difference value between the fusion data and a fusion calibration value as a fusion difference value, and determining a static fusion safety state according to the magnitude relation between the fusion difference value and a static fusion threshold value; Comprehensively judging the safety state of the boom in the static state according to the static dimension positive constant and the static fusion safety state; if the boom is in a moving state, determining a dynamic dimension positive constant of the boom in each dimension according to the displacement difference value, the strain difference value and the magnitude relation between the distance difference value and the dynamic threshold of the corresponding dimension; the dynamic fusion safety state is determined according to the magnitude relation between the fusion difference value and the dynamic fusion threshold value, the dynamic fusion threshold value is determined according to the real-time acceleration data, the dynamic threshold value is determined according to the dynamic fusion threshold value, and the safety state of the boom in the moving state is comprehensively judged according to the dynamic dimension positive constant and the dynamic fusion safety state. On the other hand, the embodiment of the application also provides a stage suspender safety detection device based on feature recognition, which comprises the following steps: The sensor is used for acquiring real-time state data of the suspender, wherein the real-time state data comprises real-time displacement data, real-time strain data, real-time distance data and real-time acceleration data; The system comprises a boom, a data processing device, a fusion device, a dynamic safety state determining device and a dynamic safety state determining device, wherein the data processing device is used for judging whether the boom is in a static state or a moving state according to real-time acceleration data, calculating the difference value between real-time displacement data and a displacement calibration value as displacement difference va