Disclosure of Invention
In view of the above, an objective of an embodiment of the present invention is to provide a smart bridge monitoring and early warning method and system, which loads a bridge parameter time sequence data sequence and a bridge parameter derivative data sequence into a bridge anomaly defect diagnosis network to generate a first confidence coefficient array and a second confidence coefficient array; generating an anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the first confidence coefficient array and the threshold confidence coefficient; each abnormal diagnosis sequence comprises at most a second target number of sample bridge parameter time sequence data with the confidence coefficient larger than the threshold confidence coefficient, and the second target number is positively associated with the network optimization turn of the bridge abnormal defect diagnosis network; generating a target array according to each abnormal diagnosis sequence and the bridge parameter derivative data sequence; the target array comprises prediction data of whether each two bridge parameter derived data correspond to the same bridge abnormal defect or not; generating training error parameters according to the target array and the second confidence coefficient array, updating weight information of the bridge abnormal defect diagnosis network according to the training error parameters, updating the current network optimization turn, generating a converged bridge abnormal defect diagnosis network after training is terminated, thereby improving bridge parameter time sequence data abnormal defect diagnosis of any bridge, and carrying out early warning prompt according to an abnormal defect diagnosis result.
According to one aspect of the embodiment of the invention, an intelligent bridge monitoring and early warning method and system are provided, wherein the method comprises the following steps:
loading a bridge parameter time sequence data sequence and a bridge parameter derivative data sequence corresponding to the bridge parameter time sequence data sequence into a bridge abnormal defect diagnosis network to generate a first confidence coefficient array and a second confidence coefficient array; wherein the bridge parameter time series data sequence includes a first target number of bridge anomaly defects in total, the first confidence array includes a confidence level for each bridge anomaly defect for each sample bridge parameter time series data in the bridge parameter time series data sequence, and the second confidence array includes a confidence level for each bridge parameter derivative data in the bridge parameter derivative data sequence;
generating an anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the first confidence coefficient array and the threshold confidence coefficient; each anomaly diagnostic sequence comprises at most a second target number of sample bridge parameter time sequence data with confidence degrees greater than the threshold confidence degrees, and the second target number is positively associated with a network optimization round of the bridge anomaly defect diagnostic network;
generating a target array according to each abnormal diagnosis sequence and the bridge parameter derivative data sequence; the target array comprises prediction data of whether each two bridge parameter derived data correspond to the same bridge abnormal defect or not;
generating training error parameters according to the target array and the second confidence coefficient array, updating weight information of the bridge abnormal defect diagnosis network according to the training error parameters, updating the current network optimization turn, generating a converged bridge abnormal defect diagnosis network until training is finished, performing abnormal defect diagnosis on bridge parameter time sequence data of any bridge according to the converged bridge abnormal defect diagnosis network, and performing early warning prompt according to an abnormal defect diagnosis result
An alternative embodiment, the generating an anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the first confidence array and the threshold confidence includes:
screening the confidence coefficient in the first confidence coefficient array according to the threshold confidence coefficient to generate a screened confidence coefficient larger than the threshold confidence coefficient;
generating a basic anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the bridge anomaly defect corresponding to the screening confidence and the node identification of the sample bridge parameter time sequence data corresponding to the screening confidence;
generating the second target number according to the current network optimization turn, the first target number and the number of the sample bridge parameter time sequence data;
and selecting the second target number of node identifiers with the maximum confidence according to the screening confidence corresponding to the node identifiers, and generating the abnormal diagnosis sequence formed by the second target number of node identifiers at most.
An alternative embodiment, the network optimization run corresponds to a maximum network optimization run and a minimum network optimization run, and the generating the second target number according to the current network optimization run, the first target number, and the number of sample bridge parameter timing data includes:
generating a target weight parameter according to the maximum network optimization round, the minimum network optimization round and the current network optimization round; the target weight parameter is positively associated with the current network optimization round;
and generating the second target number positively associated with the network optimization round according to the target weight parameter and a comparison parameter of the number of the sample bridge parameter time sequence data and the first target number.
In an alternative embodiment, the loading the bridge parameter time sequence data sequence and the bridge parameter derivative data sequence corresponding to the bridge parameter time sequence data sequence into the bridge abnormal defect diagnosis network, before generating the first confidence array and the second confidence array, further includes:
obtaining a bridge parameter time sequence data sequence formed by a plurality of sample bridge parameter time sequence data, respectively scrambling and deriving each sample bridge parameter time sequence data in the bridge parameter time sequence data, and generating the bridge parameter derived data sequence corresponding to the bridge parameter time sequence data sequence;
outputting all bridge parameter derivative data generated by deriving the bridge parameter time sequence data according to the same sample in the bridge parameter derivative data sequence as corresponding to the same bridge abnormal defect, and generating reference information corresponding to the bridge parameter derivative data sequence;
generating a target array according to each abnormality diagnosis sequence and the bridge parameter derivative data sequence, wherein the target array comprises the following steps:
if the current network optimization round is not greater than the minimum network optimization round, observing whether any two bridge parameter derivative data in the bridge parameter derivative data sequence correspond to the same bridge abnormal defect according to the reference information corresponding to the bridge parameter derivative data, and generating the target array;
and if the current network optimization round is heavy rain and the minimum network optimization round is heavy rain, observing whether any two bridge parameter derivative data in the bridge parameter derivative data sequence correspond to the same bridge abnormal defect according to the abnormal diagnosis sequence and the reference information corresponding to the bridge parameter derivative data, and generating the target array.
In an alternative embodiment, the bridge anomaly defect diagnosis network includes an encoder and a fully-connected output unit, and the loading the bridge parameter time-series data sequence and the bridge parameter derivative data sequence corresponding to the bridge parameter time-series data sequence into the bridge anomaly defect diagnosis network, to generate a first confidence array and a second confidence array, includes:
loading the bridge parameter time sequence data sequence and the bridge parameter derivative data sequence to the encoder to generate a first encoding vector array and a second encoding vector array; the first coding vector array comprises coding vectors corresponding to each sample bridge parameter time sequence data, and the second coding vector array comprises coding vectors corresponding to each bridge parameter derivative data;
and loading the first coding vector array and the second coding vector array to the fully-connected output unit to generate the first confidence array and the second confidence array.
In an alternative embodiment, after generating the anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the first confidence coefficient array and the threshold confidence coefficient, the method further includes:
in the first coding vector array, coding vectors corresponding to sample bridge parameter time sequence data in each abnormal diagnosis sequence are obtained, and fusion vectors corresponding to each abnormal diagnosis sequence are obtained;
generating sample bridge parameter time sequence data with the confidence coefficient larger than the threshold confidence coefficient of the first third target number with the minimum vector distance between the first target number and the fusion vector according to each abnormal diagnosis sequence and the fusion vector corresponding to the abnormal diagnosis sequence so as to update each abnormal diagnosis sequence; the third target number is smaller than the second target number.
An alternative embodiment, the generating training error parameters according to the target array and the second confidence array includes:
generating a coded vector distance array corresponding to the bridge parameter derived data sequence according to the second coded vector array, and generating a confidence associated parameter array corresponding to the bridge parameter derived data sequence according to the second confidence array;
each array member in the code vector distance array comprises a vector distance between code vectors of each two bridge parameter derived data, and each array member in the confidence coefficient associated parameter array comprises a vector distance between confidence coefficients respectively corresponding to each two bridge parameter derived data;
and generating the training error parameter according to the target array, the coding vector distance array and the confidence coefficient associated parameter array.
According to another aspect of the embodiment of the invention, there is provided a smart bridge monitoring and early warning method and system, the system comprising:
according to another aspect of an embodiment of the present invention, there is provided a computer apparatus including: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory in communication via the bus when the computer device is running, the machine-readable instructions when executed by the processor performing the steps of the image detection method according to any one of claims 1 to 7.
According to another aspect of the embodiments of the present invention, there is provided a readable storage medium having a computer program stored thereon, the computer program when executed by a processor can perform the steps of the intelligent bridge monitoring and early warning method described above.
The foregoing objects, features and advantages of embodiments of the invention will be more readily apparent from the following detailed description of the embodiments taken in conjunction with the accompanying drawings.
Detailed Description
In order to enable those skilled in the art to better understand the present invention, a technical solution of the present embodiment of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiment of the present invention, and it is apparent that the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. All other embodiments, which can be made by those skilled in the art without the benefit of the teachings of this invention, are intended to fall within the scope of the invention.
The terms first, second, third and the like in the description and in the claims and in the above drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented, for example, in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Fig. 1 shows an exemplary component diagram of a server 100. The server 100 may include one or more processors 104, such as one or more Central Processing Units (CPUs), each of which may implement one or more hardware threads. The server 100 may also include any storage medium 106 for storing any kind of information such as code, settings, data, etc. For example, and without limitation, storage medium 106 may include any one or more of the following combinations: any type of RAM, any type of ROM, flash memory devices, hard disks, optical disks, etc. More generally, any storage medium may store information using any technique. Further, any storage medium may provide volatile or non-volatile retention of information. Further, any storage medium may represent fixed or removable components of server 100. In one case, the server 100 may perform any of the operations of the associated instructions when the processor 104 executes the associated instructions stored in any storage medium or combination of storage media. The server 100 also includes one or more drive units 108, such as a hard disk drive unit, an optical disk drive unit, etc., for interacting with any storage media.
The server 100 also includes input/output 110 (I/O) for receiving various inputs (via input unit 112) and for providing various outputs (via output unit 114). One particular output mechanism may include a presentation device 116 and an associated Graphical User Interface (GUI) 118. The server 100 may also include one or more network interfaces 120 for exchanging data with other devices via one or more communication units 122. One or more communication buses 124 couple the components described above together.
The communication unit 122 may be implemented in any manner, for example, via a local area network, a wide area network (e.g., the internet), a point-to-point connection, etc., or any combination thereof. The communication unit 122 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers 100, etc., governed by any protocol or combination of protocols.
Fig. 2 is a schematic flow chart of a smart bridge monitoring and early warning method according to an embodiment of the present invention, which may be executed by the server 100 shown in fig. 1, and detailed steps of the smart bridge monitoring and early warning method are described below.
Step S110, loading a bridge parameter time sequence data sequence and a bridge parameter derivative data sequence corresponding to the bridge parameter time sequence data sequence into a bridge abnormal defect diagnosis network to generate a first confidence coefficient array and a second confidence coefficient array; wherein the bridge parameter time series data sequence includes a first target number of bridge anomaly defects in total, the first confidence array includes a confidence level for each bridge anomaly defect for each sample bridge parameter time series data in the bridge parameter time series data sequence, and the second confidence array includes a confidence level for each bridge parameter derivative data in the bridge parameter derivative data sequence;
step S120, generating an anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the first confidence coefficient array and the threshold confidence coefficient; each anomaly diagnostic sequence comprises at most a second target number of sample bridge parameter time sequence data with confidence degrees greater than the threshold confidence degrees, and the second target number is positively associated with a network optimization round of the bridge anomaly defect diagnostic network;
step S130, generating a target array according to each abnormal diagnosis sequence and the bridge parameter derivative data sequence; the target array comprises prediction data of whether each two bridge parameter derived data correspond to the same bridge abnormal defect or not;
step S140, generating training error parameters according to the target array and the second confidence coefficient array, updating weight information of the bridge abnormality defect diagnosis network according to the training error parameters and updating the current network optimization turn until training is terminated, generating a converged bridge abnormality defect diagnosis network, performing abnormality defect diagnosis on bridge parameter time sequence data of any bridge according to the converged bridge abnormality defect diagnosis network, and performing early warning prompt according to an abnormality defect diagnosis result
According to the steps, the bridge parameter time sequence data sequence and the bridge parameter derivative data sequence are loaded to a bridge abnormal defect diagnosis network to generate a first confidence coefficient array and a second confidence coefficient array; generating an anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the first confidence coefficient array and the threshold confidence coefficient; each abnormal diagnosis sequence comprises at most a second target number of sample bridge parameter time sequence data with the confidence coefficient larger than the threshold confidence coefficient, and the second target number is positively associated with the network optimization turn of the bridge abnormal defect diagnosis network; generating a target array according to each abnormal diagnosis sequence and the bridge parameter derivative data sequence; the target array comprises prediction data of whether each two bridge parameter derived data correspond to the same bridge abnormal defect or not; generating training error parameters according to the target array and the second confidence coefficient array, updating weight information of the bridge abnormal defect diagnosis network according to the training error parameters, updating the current network optimization turn, generating a converged bridge abnormal defect diagnosis network after training is terminated, thereby improving bridge parameter time sequence data abnormal defect diagnosis of any bridge, and carrying out early warning prompt according to an abnormal defect diagnosis result.
An alternative embodiment, the generating an anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the first confidence array and the threshold confidence includes:
screening the confidence coefficient in the first confidence coefficient array according to the threshold confidence coefficient to generate a screened confidence coefficient larger than the threshold confidence coefficient;
generating a basic anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the bridge anomaly defect corresponding to the screening confidence and the node identification of the sample bridge parameter time sequence data corresponding to the screening confidence;
generating the second target number according to the current network optimization turn, the first target number and the number of the sample bridge parameter time sequence data;
and selecting the second target number of node identifiers with the maximum confidence according to the screening confidence corresponding to the node identifiers, and generating the abnormal diagnosis sequence formed by the second target number of node identifiers at most.
An alternative embodiment, the network optimization run corresponds to a maximum network optimization run and a minimum network optimization run, and the generating the second target number according to the current network optimization run, the first target number, and the number of sample bridge parameter timing data includes:
generating a target weight parameter according to the maximum network optimization round, the minimum network optimization round and the current network optimization round; the target weight parameter is positively associated with the current network optimization round;
and generating the second target number positively associated with the network optimization round according to the target weight parameter and a comparison parameter of the number of the sample bridge parameter time sequence data and the first target number.
In an alternative embodiment, the loading the bridge parameter time sequence data sequence and the bridge parameter derivative data sequence corresponding to the bridge parameter time sequence data sequence into the bridge abnormal defect diagnosis network, before generating the first confidence array and the second confidence array, further includes:
obtaining a bridge parameter time sequence data sequence formed by a plurality of sample bridge parameter time sequence data, respectively scrambling and deriving each sample bridge parameter time sequence data in the bridge parameter time sequence data, and generating the bridge parameter derived data sequence corresponding to the bridge parameter time sequence data sequence;
outputting all bridge parameter derivative data generated by deriving the bridge parameter time sequence data according to the same sample in the bridge parameter derivative data sequence as corresponding to the same bridge abnormal defect, and generating reference information corresponding to the bridge parameter derivative data sequence;
generating a target array according to each abnormality diagnosis sequence and the bridge parameter derivative data sequence, wherein the target array comprises the following steps:
if the current network optimization round is not greater than the minimum network optimization round, observing whether any two bridge parameter derivative data in the bridge parameter derivative data sequence correspond to the same bridge abnormal defect according to the reference information corresponding to the bridge parameter derivative data, and generating the target array;
and if the current network optimization round is heavy rain and the minimum network optimization round is heavy rain, observing whether any two bridge parameter derivative data in the bridge parameter derivative data sequence correspond to the same bridge abnormal defect according to the abnormal diagnosis sequence and the reference information corresponding to the bridge parameter derivative data, and generating the target array.
In an alternative embodiment, the bridge anomaly defect diagnosis network includes an encoder and a fully-connected output unit, and the loading the bridge parameter time-series data sequence and the bridge parameter derivative data sequence corresponding to the bridge parameter time-series data sequence into the bridge anomaly defect diagnosis network, to generate a first confidence array and a second confidence array, includes:
loading the bridge parameter time sequence data sequence and the bridge parameter derivative data sequence to the encoder to generate a first encoding vector array and a second encoding vector array; the first coding vector array comprises coding vectors corresponding to each sample bridge parameter time sequence data, and the second coding vector array comprises coding vectors corresponding to each bridge parameter derivative data;
and loading the first coding vector array and the second coding vector array to the fully-connected output unit to generate the first confidence array and the second confidence array.
In an alternative embodiment, after generating the anomaly diagnosis sequence corresponding to each bridge anomaly defect according to the first confidence coefficient array and the threshold confidence coefficient, the method further includes:
in the first coding vector array, coding vectors corresponding to sample bridge parameter time sequence data in each abnormal diagnosis sequence are obtained, and fusion vectors corresponding to each abnormal diagnosis sequence are obtained;
generating sample bridge parameter time sequence data with the confidence coefficient larger than the threshold confidence coefficient of the first third target number with the minimum vector distance between the first target number and the fusion vector according to each abnormal diagnosis sequence and the fusion vector corresponding to the abnormal diagnosis sequence so as to update each abnormal diagnosis sequence; the third target number is smaller than the second target number.
An alternative embodiment, the generating training error parameters according to the target array and the second confidence array includes:
generating a coded vector distance array corresponding to the bridge parameter derived data sequence according to the second coded vector array, and generating a confidence associated parameter array corresponding to the bridge parameter derived data sequence according to the second confidence array;
each array member in the code vector distance array comprises a vector distance between code vectors of each two bridge parameter derived data, and each array member in the confidence coefficient associated parameter array comprises a vector distance between confidence coefficients respectively corresponding to each two bridge parameter derived data;
and generating the training error parameter according to the target array, the coding vector distance array and the confidence coefficient associated parameter array.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.