CN115905979A - Bridge structure abnormity online diagnosis method based on beam body longitudinal displacement monitoring - Google Patents
Bridge structure abnormity online diagnosis method based on beam body longitudinal displacement monitoring Download PDFInfo
- Publication number
- CN115905979A CN115905979A CN202211512683.XA CN202211512683A CN115905979A CN 115905979 A CN115905979 A CN 115905979A CN 202211512683 A CN202211512683 A CN 202211512683A CN 115905979 A CN115905979 A CN 115905979A
- Authority
- CN
- China
- Prior art keywords
- displacement
- bridge
- longitudinal displacement
- sampling data
- monitoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to the technical field of bridge structure monitoring data processing and abnormal state identification, in particular to a bridge structure abnormal online diagnosis method based on beam longitudinal displacement monitoring, which comprises the following steps: acquiring longitudinal displacement monitoring data of a plurality of beam bodies; preprocessing the monitoring data of the longitudinal displacement of the plurality of beam bodies to obtain sampling data of the longitudinal displacement of the beam bodies; acquiring a bridge longitudinal displacement characteristic vector according to the beam longitudinal displacement sampling data; and inputting the longitudinal displacement characteristic vector of the bridge into the bridge structure abnormity diagnosis network model to obtain the bridge structure state. The characteristic vectors reflecting the abnormal state of the beam structure are constructed by utilizing the plurality of beam longitudinal displacement monitoring data, the abnormal state of the beam structure is diagnosed on line through the trained bridge structure diagnosis network model, and a method for rapidly determining the abnormal state of the beam structure is provided for bridge management personnel, so that the diagnosis efficiency is improved, and further the risk early warning can be timely carried out.
Description
Technical Field
The invention relates to the technical field of bridge structure monitoring data processing and abnormal state identification, in particular to a bridge structure abnormality online diagnosis method based on beam longitudinal displacement monitoring.
Background
The longitudinal displacement of a beam body is one of important monitoring items in a bridge structure health monitoring system, bridge management personnel usually lay sensors at preset positions on a bridge for monitoring, and then judge whether the bridge structure is abnormal or not according to monitoring data. In the prior art, bridge management personnel determine whether the bridge structure is abnormal or not by combining monitoring data with expert experience, structural calculation and the like, and can determine the abnormal type of the bridge body structure by combining field manual inspection under the condition of abnormal structure. The mode efficiency of confirming bridge structures abnormal state like this is not high, and has the problem that professional experience dependence is strong, risk early warning is untimely.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a bridge structure abnormity on-line diagnosis method based on beam body longitudinal displacement monitoring, and the diagnosis efficiency is improved.
The invention adopts the technical scheme that the method is a bridge structure abnormity on-line diagnosis method based on beam body longitudinal displacement monitoring.
In a first implementation manner, a bridge structure abnormity online diagnosis method based on beam body longitudinal displacement monitoring comprises the following steps:
acquiring longitudinal displacement monitoring data of a plurality of beam bodies;
preprocessing the monitoring data of the longitudinal displacement of the plurality of beam bodies to obtain sampling data of the longitudinal displacement of the beam bodies;
acquiring a bridge longitudinal displacement characteristic vector according to the beam longitudinal displacement sampling data;
and inputting the longitudinal displacement characteristic vector of the bridge into the bridge structure abnormity diagnosis network model to obtain the bridge structure state.
With reference to the first implementable manner, in a second implementable manner, acquiring longitudinal displacement monitoring data of a plurality of beam bodies includes:
at least one temperature monitoring point and at least two displacement monitoring points are respectively arranged at the left beam end and the right beam end of the bridge; temperature monitoring data are acquired in real time through temperature sensors arranged at all temperature monitoring points, and displacement monitoring data are acquired through displacement sensors arranged at all displacement monitoring points.
With reference to the first implementable manner, in a third implementable manner, the method for preprocessing the monitoring data of the longitudinal displacement of the plurality of beams to obtain the sampling data of the longitudinal displacement of the beams includes:
adopting a pulse coupling neural network to carry out self-adaptive noise reduction processing on the monitoring data of the longitudinal displacement of the plurality of beam bodies to obtain alternative data of the longitudinal displacement of the beam bodies;
and carrying out simultaneous same-frequency processing on the beam longitudinal displacement alternative data to obtain beam longitudinal displacement sampling data.
With reference to the third implementable manner, in a fourth implementable manner, simultaneous common-frequency processing is performed on the beam longitudinal displacement candidate data to obtain beam longitudinal displacement sampling data, including:
sampling alternative data of the longitudinal displacement of the beam body in the same acquisition time period at the same frequency to obtain two groups of sampling data; each group of sampling data comprises a plurality of beam displacement sampling values or a plurality of temperature sampling values;
obtaining effective monitoring values of each group of sampling data;
and respectively determining beam displacement sampling data and temperature sampling data according to each effective monitoring value, taking the beam displacement sampling data and the temperature sampling data as beam longitudinal displacement sampling data, and taking the acquisition time as a label of the beam longitudinal displacement sampling data.
With reference to the first implementable manner, in a fifth implementable manner, obtaining a characteristic vector of longitudinal displacement of the bridge according to the sampling data of longitudinal displacement of the beam body includes:
acquiring displacement change rate of a beam monitoring point according to the beam longitudinal displacement sampling data;
acquiring displacement change acceleration of a beam monitoring point according to the longitudinal displacement change rate of the beam;
acquiring longitudinal sliding characteristics of a plurality of beam bodies according to the longitudinal displacement sampling data of the beam bodies, the displacement change rate of the beam body monitoring points and the displacement change acceleration of the beam body monitoring points;
acquiring a plurality of beam body corner features according to the beam body longitudinal displacement sampling data, the beam body monitoring point displacement change rate and the beam body monitoring point displacement change acceleration;
acquiring the clamping characteristic of the expansion joint according to the longitudinal displacement sampling data of the beam body;
and forming a bridge longitudinal displacement characteristic vector by the longitudinal sliding characteristics of the plurality of beam bodies, the corner characteristics of the plurality of beam bodies and the clamping characteristics of the expansion joints.
Combining the fifth realizable mode, in the sixth realizable mode, obtaining a plurality of beam longitudinal sliding characteristics according to the beam longitudinal displacement sampling data, the beam monitoring point displacement change rate and the beam monitoring point displacement change acceleration, and including:
respectively acquiring a first displacement average value of the left end of the bridge and a second displacement average value of the right end of the bridge according to the beam longitudinal displacement sampling data, and taking the difference between the first displacement average value and the second displacement average value as a first beam longitudinal sliding characteristic;
respectively acquiring a first average speed value at the left end of the bridge and a second average speed value at the right end of the bridge according to the displacement change rate of the monitoring point of the beam body, and taking the difference between the first average speed value and the second average speed value as the longitudinal sliding characteristic of the second beam body;
and respectively acquiring a first acceleration average value at the left end of the bridge and a second acceleration average value at the right end of the bridge according to the displacement change acceleration of the monitoring point of the beam body, and taking the difference between the first acceleration average value and the second acceleration average value as the longitudinal slippage characteristic of the third beam body.
Combining the fifth realizable mode, in the seventh realizable mode, obtaining a plurality of beam corner features according to the beam longitudinal displacement sampling data, the beam monitoring point displacement change rate, the beam monitoring point displacement change acceleration, including:
acquiring a first displacement absolute difference value of the left end of the bridge and a second displacement absolute difference value of the right end of the bridge according to the beam longitudinal displacement sampling data, and taking the average value of the first displacement absolute difference value and the second displacement absolute difference value as a first beam corner characteristic;
respectively acquiring a first speed absolute difference value at the left end of the bridge and a second speed absolute difference value at the right end of the bridge according to the displacement change rate of the monitoring point of the beam body, and taking the average value of the first speed absolute difference value and the second speed absolute difference value as a second beam body corner characteristic;
and respectively acquiring a first acceleration absolute difference value at the left end of the bridge and a second acceleration absolute difference value at the right end of the bridge according to the displacement change acceleration of the monitoring point of the beam body, and taking the average value of the first acceleration absolute difference value and the second acceleration absolute difference value as the corner characteristic of the third beam body.
With reference to the fifth implementable manner, in the eighth implementable manner, the expansion joint locking feature is obtained according to the longitudinal displacement sampling data of the beam body, including:
performing second-order derivation according to the displacement sampling data and the temperature sampling data of each monitoring point to obtain the blocking characteristics of a plurality of alternative expansion joints;
and taking the sum of the absolute values of the clamping characteristics of the multiple alternative expansion joints as the clamping characteristics of the expansion joints.
With reference to the first implementable manner, in a ninth implementable manner, the bridge structure abnormality diagnosis network model includes an input layer, a membership function layer, a fuzzy inference layer, a normalization layer, and an output layer:
the input layer inputs the longitudinal displacement characteristic vectors of the bridge into the network model, and the number of the input layer nodes corresponds to the number of the characteristic vectors;
the membership function layer maps a plurality of eigenvectors onto corresponding vocabulary variables through a cloud model normal distribution membership function to obtain a plurality of membership values, and the number of the vocabulary variables corresponds to the number of the bridge structure states;
the fuzzy inference layer comprises a plurality of neurons, each neuron represents a fuzzy rule and is used for matching the fuzzy rules with all membership values and calculating the fitness value of each rule;
the normalization layer is used for normalizing the membership values of all the fuzzy rules in the front, and the number of nodes of the normalization layer is the same as that of the nodes of the fuzzy inference layer;
and the output layer acquires the occurrence probability of different bridge structure state types by using the probability value obtained by normalization and outputs the bridge structure state type corresponding to the maximum probability value.
With reference to the ninth implementable manner, in a tenth implementable manner, the membership degree is calculated by:
wherein n represents the number of neurons, u ij The ith Membership Function (MF), c) expressed as the jth variable ij Representing the height, σ, of the membership function ij Width, x, of function of degree of membership i Expressed as the ith eigenvector, and m and n are positive integers.
According to the technical scheme, the beneficial technical effects of the invention are as follows:
1. the characteristic vectors reflecting the abnormal state of the beam structure are constructed by utilizing the plurality of beam longitudinal displacement monitoring data, the abnormal state of the beam structure is diagnosed on line through the trained bridge structure diagnosis network model, and a method for rapidly determining the abnormal state of the beam structure is provided for bridge management personnel, so that the diagnosis efficiency is improved, the risk early warning can be timely carried out, the state of the bridge structure can be diagnosed on line, and the real-time monitoring of the abnormal structure of the bridge is realized.
2. Compare among the prior art and directly adopt monitoring data to acquire bridge structural state, this scheme carries out the preliminary treatment to roof beam body longitudinal displacement monitoring data, after obtaining roof beam body longitudinal displacement sampling data, again according to roof beam body longitudinal displacement sampling data acquisition bridge longitudinal displacement eigenvector, like this, carry out the preliminary treatment back to monitoring data, the eigenvector of acquireing more can represent the structural feature of bridge, and then acquire bridge structural state according to bridge longitudinal displacement eigenvector, the rate of accuracy is higher, and reduced the reliance to the manual work, the intellectuality has been improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. Throughout the drawings, like elements or portions are generally identified by like reference numerals. In the drawings, elements or portions are not necessarily drawn to scale.
FIG. 1 is a schematic diagram of a bridge structure abnormality online diagnosis method based on beam longitudinal displacement monitoring provided by the invention;
fig. 2 is a schematic layout view of a bridge monitoring point provided by the invention.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and therefore are only examples, and the protection scope of the present invention is not limited thereby.
It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which the invention pertains.
Referring to fig. 1, the present embodiment provides a bridge structure abnormality online diagnosis method based on beam longitudinal displacement monitoring, including:
s01, acquiring longitudinal displacement monitoring data of a plurality of beam bodies;
s02, preprocessing the monitoring data of the longitudinal displacement of the plurality of beam bodies to obtain sampling data of the longitudinal displacement of the beam bodies;
s03, acquiring a longitudinal displacement feature vector of the bridge according to the longitudinal displacement sampling data of the beam body;
and S04, inputting the longitudinal displacement characteristic vector of the bridge into a bridge structure abnormity diagnosis network model to obtain a bridge structure state.
The method comprises the steps of utilizing longitudinal displacement monitoring data and temperature monitoring data of a plurality of beam bodies to construct a characteristic vector reflecting abnormal states of the beam body structure, diagnosing the abnormal states of the bridge structure on line through a trained bridge structure diagnosis network model, wherein the diagnosable types of the bridge structure diagnosis model comprise: the beam body longitudinal sliding, the beam body corner, the beam body expansion joint blocking and the like, provides a method for rapidly determining the abnormal state of the beam body structure for bridge management personnel, and improves the diagnosis efficiency.
Optionally, acquiring longitudinal displacement monitoring data of a plurality of beams includes: at least one temperature monitoring point and at least two displacement monitoring points are respectively arranged at the left beam end and the right beam end of the bridge; temperature monitoring data are acquired in real time through temperature sensors arranged at all temperature monitoring points, and displacement monitoring data are acquired through displacement sensors arranged at all displacement monitoring points.
In some embodiments, temperature and humidity sensors are arranged at each temperature monitoring point to monitor the temperature, and temperature monitoring data is obtained.
In some embodiments, as shown in fig. 2, the monitoring points are round points at two beam ends of the bridge, and the bridge deck of each beam end is provided with a temperature measuring point and two beam end displacement measuring points. The four beam end displacement measuring points are symmetrically distributed on two sides of the bridge floor and are respectively a left displacement monitoring point s1, a left displacement monitoring point s2, a right displacement monitoring point s3 and a right displacement monitoring point s4. The two temperature measuring points are distributed at two beam ends and are respectively a left temperature monitoring point and a right temperature monitoring point THM1, the temperature THM1 of the left temperature monitoring point is used as the common temperature of a left displacement monitoring point s1 and a left displacement monitoring point s2, and the temperature THM2 of the right temperature monitoring point is used as the common temperature of a right displacement monitoring point s3 and a right displacement monitoring point s4.
Optionally, the preprocessing is performed on the monitoring data of the longitudinal displacement of the plurality of beams to obtain sampling data of the longitudinal displacement of the beams, and the preprocessing includes: adopting a pulse coupling neural network to carry out self-adaptive noise reduction processing on the monitoring data of the longitudinal displacement of the plurality of beam bodies to obtain alternative data of the longitudinal displacement of the beam bodies; and carrying out simultaneous same-frequency processing on the beam longitudinal displacement alternative data to obtain beam longitudinal displacement sampling data.
Optionally, the beam longitudinal displacement monitoring data is obtained through a formula dis = dis T +dis L +dis tra +dis r Represents; wherein dis is beam longitudinal displacement monitoring data, dis T Is a temperature effect influence value expressed in the form of Gaussian noise dis L Is the change value of the bridge structure, dis tra With respect to the monitored dataThen the effect of high frequency noise, dis r Is a random effect influence.
In some embodiments, because the bridge body longitudinal displacement monitoring data is mainly influenced by environment temperature and humidity, traffic load, random effect and the like, burrs and mutation points exist on data performance, the pulse coupled neural network well filters superimposed mixed noise, and after the Pulse Coupled Neural Network (PCNN) is adopted to carry out self-adaptive noise reduction processing on the beam end longitudinal displacement and temperature monitoring data, the influence of Gaussian noise and the influence of high-frequency noise are reduced, so that dis = dis T +dis L +dis' tra +dis' r (ii) a Wherein, dis' tra Is residual high frequency noise influence, dis' r The influence of the residual amount of the Gaussian noise and the high-frequency noise on the monitoring result is small and can be directly ignored. Therefore, the monitoring data only remain the temperature influence and the change value of the bridge structure, namely dis = dis T +dis L Interference data in the monitoring data are greatly reduced, and therefore accuracy of follow-up analysis is improved.
Optionally, the simultaneous same-frequency processing is performed on the beam longitudinal displacement candidate data to obtain beam longitudinal displacement sampling data, including: sampling the alternative data of the longitudinal displacement of the beam body in the same acquisition time period at the same frequency to obtain two groups of sampling data; each group of sampling data comprises a plurality of beam displacement sampling values or a plurality of temperature sampling values; obtaining effective monitoring values of each group of sampling data; and respectively determining beam displacement sampling data and temperature sampling data according to each effective monitoring value, taking the beam displacement sampling data and the temperature sampling data as beam longitudinal displacement sampling data, and taking the acquisition time as a label of the beam longitudinal displacement sampling data.
In some embodiments, since the bridge structure is slowly changing, the displacement data may be considered to be less changing in a short time; according to the specification requirements of JT/T1037-2022, the sampling frequency table of the structural response monitoring content specifies that the dynamic sampling frequency and the static sampling frequency of the displacement sampling frequency are 20Hz and 1Hz respectively. The displacement of the beam end belongs to a displacement monitoring item, and static sampling with the frequency of 1Hz is generally adopted. The temperature sampling frequency is less than or equal to 1/600Hz. Therefore, beam displacement monitoring data and temperature monitoring data in the beam longitudinal displacement candidate data need to be processed at the same time, so that other interference data can be eliminated, beam displacement sampling data and temperature sampling data at the same acquisition time are obtained, and correlation characteristics between the beam displacement sampling data and the temperature sampling data are established.
In some embodiments, within a preset acquisition time period, sampling is performed on the beam displacement monitoring data at a frequency of 1/60Hz, and sampling is performed on the temperature monitoring data at a frequency of 1/60Hz, so that the uniform sampling time scale of the monitoring data is 10 min/time, and the monitoring result can be approximately considered to be unchanged.
Optionally, obtaining valid monitoring values of each set of sample data includes: sequencing the sampling values in each group of sampling data according to the sequence from large to small; removing the maximum sampling value and the minimum sampling value in each group of sampling data, and remaining a plurality of sampling values; and acquiring the average value of the residual sampling values, and taking the average value as an effective monitoring value.
In some embodiments, the preset acquisition time period is 10min, the beam displacement monitoring data are sampled strictly according to 1/60Hz, k sampling values are obtained in each group, 10 sampling times in 10min are not practical due to the limitation of hardware and transmission conditions, and finally, the sampling data are mostly 10 sampling times in 10min for 8-9 times, so k is less than 10. Sequencing the k sampling values from large to small to obtain a sequence x 1 ,x 2 ,…,x k . And calculating by the following formula to obtain an effective monitoring value.
Alternatively, data = (sum (x) 1 ,x 2 ,…,x k )-x min -x max )/(k-2)
In the above formula, data is effective monitoring data of the first acquisition time period, x min Is the minimum sample value, x max Is the maximum sample value, x k Is the kth sample value.
Optionally, the number of minutes of the acquisition time is subjected to forward alignment processing, and the processed acquisition time is used as a label of the beam displacement sampling data and the temperature sampling data. In some embodiments, performing the forward alignment on the number of minutes includes processing a number between 0 and 10 to 0, processing a number between 10 and 20 to 10, processing a number between 20 and 30 to 20, processing a number between 30 and 40 to 30, processing a number between 40 and 50 to 40, and processing a number between 50 and 60 to 50. That is, if the data of the acquisition time is 4:10, if the acquisition time after the forward alignment processing is 4.
Optionally, obtaining a bridge longitudinal displacement feature vector according to the beam longitudinal displacement sampling data includes: acquiring displacement change rate of a beam monitoring point according to the beam longitudinal displacement sampling data; acquiring displacement change acceleration of a monitoring point of the beam body according to the longitudinal displacement change rate of the beam body; acquiring a plurality of beam body longitudinal sliding characteristics according to the beam body longitudinal displacement sampling data, the beam body monitoring point displacement change rate and the beam body monitoring point displacement change acceleration; acquiring a plurality of beam body corner features according to the beam body longitudinal displacement sampling data, the beam body monitoring point displacement change rate and the beam body monitoring point displacement change acceleration; acquiring the clamping characteristic of the expansion joint according to the longitudinal displacement sampling data of the beam body; and forming a bridge longitudinal displacement characteristic vector by using the longitudinal sliding characteristics of the plurality of beam bodies, the corner characteristics of the plurality of beam bodies and the clamping characteristics of the expansion joint.
Optionally, obtaining displacement change rate of the beam monitoring point according to the beam longitudinal displacement sampling data includes: by calculation of
Obtaining the displacement change rate of each monitoring point; wherein v is sq For the rate of change of displacement of the qth monitoring point, dis sq For the shifted sample data of the qth monitoring point, q = {1, 2, 3, 4}, and/or>
Representing the derivation.
Optionally, obtaining displacement change acceleration of the beam monitoring point according to the longitudinal displacement change rate of the beam, including: by calculation of
Obtaining the displacement change rate of each monitoring point; wherein v is sq The displacement change rate of the qth monitoring point, a sq The changing acceleration of the qth monitoring point.
Optionally, according to roof beam body longitudinal displacement sampling data, roof beam body monitoring point displacement rate of change, roof beam body monitoring point displacement acceleration of change obtain a plurality of roof beam body longitudinal slip characteristics, include: respectively acquiring a first displacement average value of the left end of the bridge and a second displacement average value of the right end of the bridge according to the beam longitudinal displacement sampling data, and taking the difference between the first displacement average value and the second displacement average value as a first beam longitudinal sliding characteristic; respectively acquiring a first average speed value at the left end of the bridge and a second average speed value at the right end of the bridge according to the displacement change rate of the monitoring point of the beam body, and taking the difference between the first average speed value and the second average speed value as the longitudinal sliding characteristic of the second beam body; and respectively acquiring a first acceleration average value at the left end of the bridge and a second acceleration average value at the right end of the bridge according to the displacement change acceleration of the monitoring point of the beam body, and taking the difference between the first acceleration average value and the second acceleration average value as the longitudinal slippage characteristic of the third beam body.
Optionally, the first beam longitudinal slippage characteristic is obtained by the following formula:
in the above formula, Q 1 Is a longitudinal sliding feature of the first beam body, dis s1 ,dis s2 ,dis s3 ,dis s4 Respectively are displacement sampling data of the beam body longitudinal displacement measuring points s1, s2, s3 and s4.
Optionally, the longitudinal sliding characteristic of the second beam body is obtained by the following formula:
in the above formula, Q 2 Is a secondLongitudinal sliding characteristic of beam body, v s1 ,v s2 ,v s3 ,v s4 The displacement change rates of the beam body longitudinal displacement measuring points s1, s2, s3 and s4 are respectively.
Optionally, the third beam longitudinal sliding characteristic is obtained by the following formula:
in the above formula, Q 3 Is a longitudinal sliding characteristic of the third beam body, a s1 ,a s2 ,a s3 ,a s4 The change accelerations of the beam longitudinal displacement measuring points s1, s2, s3 and s4 are respectively.
Optionally, according to roof beam body longitudinal displacement sampling data, roof beam body monitoring point displacement rate of change, roof beam body monitoring point displacement variation acceleration obtain a plurality of roof beam body corner characteristics, include: acquiring a first displacement absolute difference value of the left end of the bridge and a second displacement absolute difference value of the right end of the bridge according to the beam longitudinal displacement sampling data, and taking the average value of the first displacement absolute difference value and the second displacement absolute difference value as a first beam corner feature; respectively acquiring a first speed absolute difference value at the left end of the bridge and a second speed absolute difference value at the right end of the bridge according to the displacement change rate of the monitoring point of the beam body, and taking the average value of the first speed absolute difference value and the second speed absolute difference value as a second beam body corner characteristic; and respectively acquiring a first acceleration absolute difference value at the left end of the bridge and a second acceleration absolute difference value at the right end of the bridge according to the displacement change acceleration of the monitoring point of the beam body, and taking the average value of the first acceleration absolute difference value and the second acceleration absolute difference value as the corner characteristic of the third beam body.
Optionally, the first beam body corner feature is obtained by the following formula:
in the above formula, Q 4 Is a first beam body corner feature, dis s1 ,dis s2 ,dis s3 ,dis s4 Are respectively beamsAnd displacement sampling data of the body longitudinal displacement measuring points s1, s2, s3 and s4.
Optionally, the second beam corner characteristic is obtained by the following formula:
in the above formula, Q 5 Is a second beam body corner feature, v s1 ,v s2 ,v s3 ,v s4 The displacement change rates of the beam body longitudinal displacement measuring points s1, s2, s3 and s4 are respectively.
Optionally, the third beam body corner feature is obtained by the following formula:
in the above formula, Q 6 Is a corner feature of the third beam body, a s1 ,a s2 ,a s3 ,a s4 The change accelerations of the beam longitudinal displacement measuring points s1, s2, s3 and s4 are respectively.
Optionally, the dead characteristic of expansion joint card is obtained according to roof beam body longitudinal displacement sampling data, includes: performing second-order derivation according to the displacement sampling data and the temperature sampling data of each monitoring point to obtain the blocking characteristics of a plurality of alternative expansion joints; and taking the sum of the absolute values of the clamping characteristics of the multiple alternative expansion joints as the clamping characteristics of the expansion joints.
Optionally by calculating
And acquiring the clamping characteristic of the alternative expansion joint, wherein dis is displacement sampling data, and T is temperature sampling data.
Optionally, the dead-locking feature of the expansion joint is calculated by the following formula:
Q 7 ＝|y s1 |+|y s2 |+|y s3 |+|y s4 |
in the above formula, Q 7 Is a characteristic of dead locking of the expansion joint y s1 Is a first alternative expansion joint cardDeath feature, y s2 A second alternative expansion joint jamming feature, y s3 A third alternative expansion joint jamming feature, y s4 The fourth alternative expansion joint jamming feature.
In some embodiments, the beam configuration failure types are mainly: the longitudinal sliding of the beam body, the corner of the beam body and the blocking of the expansion joint of the beam body. The method comprises the steps that a first beam body longitudinal sliding characteristic, a second beam body longitudinal sliding characteristic and a third beam body longitudinal sliding characteristic are used for evaluating whether longitudinal sliding occurs to the bridge or not, a first beam body corner characteristic, a second beam body corner characteristic and a third beam body corner characteristic are used for evaluating whether corners occur to the bridge beam body or not, and an expansion joint clamping characteristic is used for evaluating whether expansion joint clamping occurs to the bridge or not. And the longitudinal sliding characteristic of the first beam body, the longitudinal sliding characteristic of the second beam body, the longitudinal sliding characteristic of the third beam body, the corner characteristic of the first beam body, the corner characteristic of the second beam body, the corner characteristic of the third beam body and the blocking characteristic of the expansion joint form a longitudinal displacement characteristic vector of the bridge, namely [ Q ] 1 Q 2 … Q 7 ]. Therefore, the characteristics and the bridge structural state are not in a simple one-to-one relation, and the structural state type of the bridge can be reflected on a plurality of characteristics, so that the structural state type of the bridge is evaluated by adopting the plurality of characteristics, and the accuracy is improved.
Optionally, before inputting the feature vector of the longitudinal displacement of the bridge into the network model for diagnosing abnormality of the bridge structure, the network model for diagnosing abnormality of the bridge structure is constructed by the following method, including:
s11, acquiring longitudinal displacement monitoring data of a bridge body under various structural states of the bridge, wherein the types of the monitoring data comprise displacement monitoring data and temperature monitoring data under structural states of normal bridge, longitudinal sliding of the bridge, torsion of the bridge body, clamping of an expansion joint of the bridge and the like;
s12, respectively constructing monitoring data in different structural states according to the acquired monitoring data, taking the monitoring data as a sample, taking the structural state corresponding to the monitoring data as a label of the monitoring data, wherein the number of types in each structural state is more than or equal to 10000, and dividing the monitoring data into an alternative training set and an alternative testing set according to the proportion of 9:1;
s13, carrying out self-adaptive noise reduction and simultaneous same-frequency processing on sample data of the alternative training set to obtain a training set, and carrying out self-adaptive noise reduction and simultaneous same-frequency processing on the sample data of the alternative test set to obtain a test set;
s14, acquiring a bridge longitudinal displacement feature vector for training according to the training set, and acquiring a bridge longitudinal displacement feature vector for testing according to the testing set;
s15, sending the bridge longitudinal displacement feature vectors used for training into a fuzzy neural network model for training to obtain an alternative bridge structure abnormity diagnosis network model;
s16, inputting the longitudinal displacement characteristic vectors of the bridge for testing into an alternative bridge structure abnormity diagnosis network model to obtain the testing precision of various bridge structure states; and if the test precision does not reach 99%, continuing to add the sample corresponding to the bridge structure state, returning to the step S15 to carry out training until the model training requirement is met, and if the test precision reaches 99%, taking the alternative bridge structure abnormity diagnosis network model as a final bridge structure abnormity diagnosis network model.
Optionally, the bridge structure anomaly diagnosis network model includes an input layer, a membership function layer, a fuzzy inference layer, a normalization layer, and an output layer: the input layer inputs the longitudinal displacement characteristic vectors of the bridge into the network model, and the number of the input layer nodes corresponds to the number of the characteristic vectors; the membership function layer maps a plurality of eigenvectors onto corresponding vocabulary variables through a cloud model normal distribution membership function to obtain a plurality of membership values, and the number of the vocabulary variables corresponds to the number of the bridge structure states; the fuzzy inference layer comprises a plurality of neurons, each neuron represents a fuzzy rule and is used for matching the fuzzy rules with all membership values and calculating the fitness value of each rule; the normalization layer is used for normalizing the membership values of all the fuzzy rules in the front, and the number of nodes of the normalization layer is the same as that of the nodes of the fuzzy inference layer; and the output layer acquires the occurrence probability of different bridge structure state types by using the probability value obtained by normalization and outputs the bridge structure state type corresponding to the maximum probability value.
Optionally, the degree of membership is calculated by:
wherein n represents the number of neurons, u ij The ith Membership Function (MF), c) expressed as the jth variable ij Representing the height, σ, of the membership function ij Width, x, of function of degree of membership i Expressed as the ith eigenvector, and m and n are positive integers.
Optionally, the fuzzy inference layer is represented by formula B i ＝u i1 (x 1 ),u i2 (x 2 ),…,u in (x m ) Calculating the fitness value of each rule; wherein, B i Representing the degree of membership of each i fuzzy rules.
Optionally, the normalization layer is formulated by
Normalizing the membership values of all the fuzzy rules; wherein it is present>
Normalized confidence corresponding to each i fuzzy rules.
Optionally, the output layer is formulated by
Obtaining the probability of each bridge structure state type; wherein, w is a weight value, and a back propagation algorithm is adopted for updating.
In some embodiments, the input layer will correspond to a bridge structure feature vector [ Q [ ] 1 Q 2 … Q 7 ]And transmitting the data into a network model, wherein the number of the nodes is 7 and corresponds to the number of the feature vectors. The membership function layer maps 7 features to 4 vocabulary variables, the 4 vocabulary variables correspond to 4 bridge structure states, 7 groups of membership values are obtained, and each group has 4 modulesFuzzy subsets, each fuzzy subset corresponding to 1 degree of membership. Each neuron in the fuzzy inference layer represents a fuzzy rule and is used for matching the fuzzy rule, and the fuzzy inference layer calculates the fitness value of each rule according to 7 groups of membership values. The normalization layer normalizes membership values of all the fuzzy rules in front to obtain the normalized credibility, namely a probability value, of each fuzzy rule. The output layer obtains the probability accumulation sum of different fuzzy rules by using the probability value obtained by normalization, namely the result of different structural state types, and outputs the bridge structural state type corresponding to the maximum probability value. Therefore, by utilizing the monitoring data and the temperature monitoring data of the longitudinal displacement of the plurality of beam bodies, the characteristic vector reflecting the abnormal state of the beam body structure is constructed, the abnormal state of the bridge structure is diagnosed on line through the trained bridge structure diagnosis network model, and the diagnosable type of the bridge structure diagnosis model comprises the following steps: the method for rapidly determining the abnormal state of the beam structure is provided for bridge management personnel by adopting the types of longitudinal sliding of the beam body, corner of the beam body, blocking of expansion joints of the beam body and the like.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the present invention, and they should be construed as being included in the following claims and description.
Claims (10)
1. A bridge structure abnormity on-line diagnosis method based on beam body longitudinal displacement monitoring is characterized by comprising the following steps:
acquiring longitudinal displacement monitoring data of a plurality of beam bodies in real time;
preprocessing the monitoring data of the longitudinal displacement of the plurality of beam bodies to obtain sampling data of the longitudinal displacement of the beam bodies;
acquiring a bridge longitudinal displacement characteristic vector according to the beam longitudinal displacement sampling data;
and inputting the longitudinal displacement characteristic vector of the bridge into a bridge structure abnormity diagnosis network model to obtain the bridge structure state.
2. The method of claim 1, wherein acquiring a plurality of beam longitudinal displacement monitoring data comprises:
at least one temperature monitoring point and at least two displacement monitoring points are respectively arranged at the left beam end and the right beam end of the bridge; temperature monitoring data are acquired in real time through temperature sensors arranged at all temperature monitoring points, and displacement monitoring data are acquired through displacement sensors arranged at all displacement monitoring points.
3. The method of claim 2, wherein preprocessing the plurality of beam longitudinal displacement monitoring data to obtain beam longitudinal displacement sampling data comprises:
adopting a pulse coupling neural network to carry out self-adaptive noise reduction processing on the monitoring data of the longitudinal displacement of the plurality of beam bodies to obtain alternative data of the longitudinal displacement of the beam bodies;
and carrying out simultaneous same-frequency processing on the beam longitudinal displacement alternative data to obtain beam longitudinal displacement sampling data.
4. The method according to claim 3, wherein the simultaneous co-frequency processing of the beam longitudinal displacement candidate data to obtain beam longitudinal displacement sampling data comprises:
sampling the alternative data of the longitudinal displacement of the beam body in the same acquisition time period at the same frequency to obtain two groups of sampling data; each group of sampling data comprises a plurality of beam displacement sampling values or a plurality of temperature sampling values;
obtaining effective monitoring values of each group of sampling data;
and respectively determining beam displacement sampling data and temperature sampling data according to each effective monitoring value, taking the beam displacement sampling data and the temperature sampling data as beam longitudinal displacement sampling data, and taking the acquisition time as a label of the beam longitudinal displacement sampling data.
5. The method of claim 1, wherein obtaining a bridge longitudinal displacement feature vector from the beam longitudinal displacement sampling data comprises:
acquiring displacement change rate of a monitoring point of the beam body according to the longitudinal displacement sampling data of the beam body;
acquiring displacement change acceleration of a beam monitoring point according to the longitudinal displacement change rate of the beam;
acquiring a plurality of beam body longitudinal sliding characteristics according to the beam body longitudinal displacement sampling data, the beam body monitoring point displacement change rate and the beam body monitoring point displacement change acceleration;
acquiring a plurality of beam corner characteristics according to the beam longitudinal displacement sampling data, the beam monitoring point displacement change rate and the beam monitoring point displacement change acceleration;
acquiring the clamping characteristic of the expansion joint according to the longitudinal displacement sampling data of the beam body;
and forming a bridge longitudinal displacement characteristic vector by the longitudinal sliding characteristics of the plurality of beam bodies, the corner characteristics of the plurality of beam bodies and the clamping characteristics of the expansion joints.
6. The method according to claim 5, wherein obtaining a plurality of beam longitudinal slip characteristics according to the beam longitudinal displacement sampling data, the beam monitoring point displacement change rate and the beam monitoring point displacement change acceleration comprises:
respectively acquiring a first displacement average value of the left end of the bridge and a second displacement average value of the right end of the bridge according to the beam longitudinal displacement sampling data, and taking the difference between the first displacement average value and the second displacement average value as a first beam longitudinal sliding characteristic;
respectively obtaining a first average speed value at the left end of the bridge and a second average speed value at the right end of the bridge according to the displacement change rate of the monitoring point of the beam, and taking the difference between the first average speed value and the second average speed value as the longitudinal sliding characteristic of the second beam;
and respectively acquiring a first acceleration average value at the left end of the bridge and a second acceleration average value at the right end of the bridge according to the displacement change acceleration of the monitoring point of the beam body, and taking the difference between the first acceleration average value and the second acceleration average value as the longitudinal slippage characteristic of the third beam body.
7. The method of claim 5, wherein obtaining a plurality of beam corner features according to the beam longitudinal displacement sampling data, the beam monitoring point displacement change rate and the beam monitoring point displacement change acceleration comprises:
acquiring a first displacement absolute difference value of the left end of the bridge and a second displacement absolute difference value of the right end of the bridge according to the beam longitudinal displacement sampling data, and taking the average value of the first displacement absolute difference value and the second displacement absolute difference value as a first beam corner feature;
respectively acquiring a first speed absolute difference value at the left end of the bridge and a second speed absolute difference value at the right end of the bridge according to the displacement change rate of the beam monitoring point, and taking the average value of the first speed absolute difference value and the second speed absolute difference value as a second beam corner characteristic;
and respectively acquiring a first acceleration absolute difference value at the left end of the bridge and a second acceleration absolute difference value at the right end of the bridge according to the displacement change acceleration of the beam monitoring point, and taking the average value of the first acceleration absolute difference value and the second acceleration absolute difference value as the corner characteristic of the third beam.
8. The method as claimed in claim 5, wherein obtaining the clamping characteristic of the expansion joint according to the longitudinal displacement sampling data of the beam body comprises:
performing second-order derivation according to the displacement sampling data and the temperature sampling data of each monitoring point to obtain the blocking characteristics of a plurality of alternative expansion joints;
and taking the sum of the absolute values of the clamping characteristics of the multiple alternative expansion joints as the clamping characteristics of the expansion joints.
9. The method of claim 1, wherein the bridge structure anomaly diagnosis network model comprises an input layer, a membership function layer, a fuzzy inference layer, a normalization layer and an output layer:
the input layer inputs the longitudinal displacement characteristic vectors of the bridge into the network model, and the number of the input layer nodes corresponds to the number of the characteristic vectors;
the membership function layer maps a plurality of eigenvectors onto corresponding vocabulary variables through a cloud model normal distribution membership function to obtain a plurality of membership values, and the number of the vocabulary variables corresponds to the number of the bridge structure states;
the fuzzy inference layer comprises a plurality of neurons, each neuron represents a fuzzy rule and is used for matching the fuzzy rules with all membership values and calculating the fitness value of each rule;
the normalization layer is used for normalizing the membership values of all the fuzzy rules in the front, and the number of nodes of the normalization layer is the same as that of the nodes of the fuzzy inference layer;
and the output layer acquires the occurrence probability of different bridge structure state types by using the probability value obtained by normalization and outputs the bridge structure state type corresponding to the maximum probability value.
10. The method of claim 9, wherein the degree of membership is calculated by:
wherein n represents the number of neurons, u ij The ith Membership Function (MF), c) expressed as the jth variable ij Representing the height, σ, of the membership function ij Width, x, of function of degree of membership i Expressed as the ith eigenvector, and m and n are positive integers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211512683.XA CN115905979A (en) | 2022-11-28 | 2022-11-28 | Bridge structure abnormity online diagnosis method based on beam body longitudinal displacement monitoring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211512683.XA CN115905979A (en) | 2022-11-28 | 2022-11-28 | Bridge structure abnormity online diagnosis method based on beam body longitudinal displacement monitoring |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115905979A true CN115905979A (en) | 2023-04-04 |
Family
ID=86492288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211512683.XA Pending CN115905979A (en) | 2022-11-28 | 2022-11-28 | Bridge structure abnormity online diagnosis method based on beam body longitudinal displacement monitoring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115905979A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117216701A (en) * | 2023-09-13 | 2023-12-12 | 广州桐富科技发展有限公司 | Intelligent bridge monitoring and early warning method and system |
-
2022
- 2022-11-28 CN CN202211512683.XA patent/CN115905979A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117216701A (en) * | 2023-09-13 | 2023-12-12 | 广州桐富科技发展有限公司 | Intelligent bridge monitoring and early warning method and system |
| CN117216701B (en) * | 2023-09-13 | 2024-03-15 | 华夏安信物联网技术有限公司 | Intelligent bridge monitoring and early warning method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108038300B (en) | Optical fiber state evaluation method based on improved membership function combined with neural network | |
| JP4914457B2 (en) | Automatic virtual measurement system and method | |
| US6892163B1 (en) | Surveillance system and method having an adaptive sequential probability fault detection test | |
| Wang et al. | On line tool wear monitoring based on auto associative neural network | |
| CN113435644B (en) | Emergency prediction method based on deep bidirectional long-short term memory neural network | |
| CN111008363A (en) | Multivariable causal-driven complex electromechanical system service safety situation evaluation method | |
| CN115905979A (en) | Bridge structure abnormity online diagnosis method based on beam body longitudinal displacement monitoring | |
| CN106934242B (en) | The health degree appraisal procedure and system of equipment under multi-mode based on Cross-Entropy Method | |
| CN110866448A (en) | Flutter signal analysis method based on convolutional neural network and short-time Fourier transform | |
| Kärkkäinen et al. | Robust formulations for training multilayer perceptrons | |
| CN115510950A (en) | Aircraft telemetry data anomaly detection method and system based on time convolution network | |
| Wu et al. | A neurofuzzy network structure for modelling and state estimation of unknown nonlinear systems | |
| EP0416370A2 (en) | Method and device for detecting and indentifying sensor errors | |
| Cateni et al. | A fuzzy logic-based method for outliers detection. | |
| CN110673568A (en) | Method and system for determining fault sequence of industrial equipment in glass fiber manufacturing industry | |
| CN112333147B (en) | Nuclear power plant DCS platform network operation situation sensing method and system | |
| Yang et al. | A structure optimization algorithm of neural networks for large-scale data sets | |
| Erginel | Fuzzy individual and moving range control charts with α-cuts | |
| CN113724211B (en) | Fault automatic identification method and system based on state induction | |
| AlRababah | Neural networks precision in technical vision systems | |
| CN114154266B (en) | Gas turbine fault prediction method based on bias rank correlation flow causal structure learning | |
| Sobchuk et al. | Evaluation of efficiency of application of functionally sustainable generalized information system of the enterprise | |
| CN116386128A (en) | Tunnel worker construction state detection method, system, medium and equipment | |
| Sum et al. | Extended Kalman filter–based pruning method for recurrent neural networks | |
| Rotshtein | Algebra of algorithms and fuzzy logic in system reliability analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |