An edge gateway for lightweight bridge health monitoring

Abstract

The present invention discloses an edge gateway for realizing lightweight monitoring of bridge health. The present invention relates to the technical field of bridge health monitoring, and solves the problem that when data is unified in format, a better processing method is not set, resulting in a relatively slow unification efficiency. The present invention collects periodic data generated in an associated period, and then performs format unification processing on the data in the periodic data, and preferentially confirms the conversion format of the corresponding data. The conversion format of the data has a corresponding data conversion rate in the conversion process. When performing numerical confirmation, the rate range associated with the corresponding stage can be locked, and then based on different rate ranges, the best format to be converted is finally confirmed. The determined best format to be converted is a group of conversion formats with the fastest conversion rate in the entire data conversion process, which can fully reduce the specific conversion time to achieve the best conversion processing effect.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge health monitoring, and in particular to an edge gateway for realizing lightweight monitoring of bridge health. Background Art

[0002] With the development of modern science and technology and the continuous growth of transportation demand, the scale of highway bridges has also expanded year by year. The mainstream bridge health monitoring solutions on the market currently include sensor installation, data acquisition instrument installation, industrial computer installation, data acquisition and transmission system construction, data analysis and processing system construction, early warning and alarm system construction, monitoring center construction, cloud platform construction, operation and maintenance, etc. These usually involve expensive equipment and complex technologies, which may be limited by the site and environment, and are difficult to be flexibly applied in various scenarios. They are not suitable for all bridge health monitoring scenarios. The lightweight monitoring solution focuses on cost-effectiveness, uses relatively simple and economical equipment and methods to reduce monitoring costs, and usually has a smaller size and weight, which is easy to install and implement. Therefore, considering the construction cost, monitoring flexibility, and convenient deployment, lightweight bridge monitoring technology can be used to improve monitoring efficiency and accuracy, and provide more reliable protection for the structural safety of bridges.

[0003] The application with publication number CN115314528A discloses a bridge structure health monitoring system, method and platform. The bridge structure health monitoring system includes: a front-end sensor network and a cloud big data platform; the front-end sensor network includes a variety of multi-sensory dimension sensors and data acquisition instruments, and an integrated gateway device. The front-end sensor network is used to measure the front-end original perception data of the target bridge and access the cloud big data platform according to the Internet of Things protocol; the cloud big data platform adaptively defines the edge analysis mode to support edge analysis based on the front-end original perception data, extract valid data, and realize comprehensive intelligent evaluation of bridge health based on multi-modal combination analysis of deep learning neural networks, and promote the aggregation, governance and sharing of bridge structure health big data.

[0004] When its edge gateway processes related data of different types, it generally stores the corresponding data in real time based on its generation frequency. However, in the actual processing process, when different data from sensors from different sources are managed in a unified manner, different databases have different data storage methods, which leads to slow storage optimization. In the actual processing process, the overall monitoring and collection methods of different data are not in the best state, and better monitoring and collection effects cannot be achieved. When the data is formatted in a unified manner, a better processing method is not set, resulting in slow unification efficiency. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides an edge gateway for realizing lightweight monitoring of bridge health, which solves the problem that when the data is formatted in a unified manner, a better processing method is not set, resulting in a relatively slow unification efficiency.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an edge gateway for realizing lightweight monitoring of bridge health, comprising:

[0007] The frequency data acquisition terminal confirms the data acquisition frequency of several groups of monitoring sensors associated with the edge gateway, and transmits the confirmed data acquisition frequency to the periodic data acquisition terminal;

[0008] The periodic data acquisition end determines a group of associated periods based on the different data acquisition frequencies confirmed by different monitoring sensors, acquires the periodic data monitored by different monitoring sensors within the associated period, generates a periodic data packet, and transmits the generated periodic data packet to the data format unified end. The specific method is as follows:

[0009] The different data acquisition frequencies confirmed by different monitoring sensors are calibrated as P i , where i represents different monitoring sensors, and P i =1 and P i =0 to eliminate, and then confirm the remaining data acquisition frequency P i The common multiples of the data are marked as the correlation period. If there are no other data acquisition frequencies P left, i , then directly set 24h as the corresponding association period;

[0010] Based on the confirmed association cycle, starting from the time when the edge gateway intervenes, the monitoring data generated by different monitoring sensors are acquired, and the monitoring data generated by different monitoring sensors are combined into data packets to confirm the periodic data packets;

[0011] The historical data analysis end, based on the different data formats associated with different monitoring data in the periodic data packet, confirms the data conversion rates between different data formats from the historical data, and then performs numerical analysis on the confirmed several different data conversion rates to determine the best format to be converted. The specific method is as follows:

[0012] Confirm different data formats associated with different monitoring data in the periodic data packet, mark the confirmed different data formats as the main format, and randomly select a group of data formats from the main format as the formats to be converted;

[0013] The data conversion rate generated during the conversion from the main format to the format to be converted is identified from the historical data, and the data conversion rates of several groups of data when the single main format is converted to the format to be converted are numerically confirmed. Based on the confirmed minimum data conversion rate and maximum data conversion rate, the rate range associated with the corresponding single main format is confirmed. Then, based on the different rate ranges corresponding to different main formats, the minimum values ​​of several groups of different rate ranges are calibrated as QJ. k min, mark the maximum value as QJ k max, where k represents the different rate intervals associated with different main formats, and then determines several different minimum values ​​QJ k The equilibrium value H1 between min will be less than QJ of H1 k min is marked as the lower value, and QJ greater than H1 k min is calibrated as the upper value, the sum of the differences between H1 and several lower values ​​is equal to the sum of the differences between the upper value and H1, and then the same processing method is used to confirm several different maximum values ​​QJ k max balance value H2, based on the balance values ​​H1 and H2, determine the characteristic rate range of the format to be converted. If H1=H2, directly determine the characteristic parameters of the corresponding format to be converted, and its characteristic parameters are the same as H1 and H2;

[0014] Then, other main formats are taken as formats to be converted in turn, and the characteristic rate range or characteristic parameter associated with the corresponding formats to be converted is determined by the same processing method as above;

[0015] Based on different characteristic rate intervals or characteristic parameters confirmed by different formats to be converted, the middle value of the characteristic rate interval is used as the characteristic parameter associated therewith, and then the format to be converted associated with the maximum characteristic parameter is selected as the best format to be converted, and the confirmed best format to be converted is transmitted to the data format unified terminal;

[0016] The data format unification end converts the monitoring data in the periodic data packet that is in a different format from the best format to be converted based on the confirmed periodic data packet and the best format to be converted, so that the data format of the monitoring data in the periodic data packet is converted to the best format to be converted;

[0017] The process adjustment and optimization end monitors the data format conversion process existing in the data format unification end to confirm the real-time conversion rate, and then confirms the abnormal process based on the characteristic parameters associated with the corresponding conversion process, and optimizes and adjusts the data conversion process based on the specific results of the real-time confirmation. The specific sub-steps are:

[0018] The conversion rate of different data format conversion processes monitored in real time is calibrated as V q, where q represents different data format conversion processes, and determines the characteristic parameters associated with the data format conversion process, and marks the associated characteristic parameters as T q ;

[0019] If (T q -V q )>Y1, where Y1 is a preset value, the currently associated data format conversion process is marked as an abnormal process. If there are other processes marked as abnormal processes within a certain period after the marking time, the corresponding abnormal process is marked as a synchronous abnormal process, and the certain period is the preset period;

[0020] For the confirmed synchronous abnormal process, the data conversion frequency associated with the corresponding abnormal process is confirmed, and the process sequence is determined by sorting several groups of abnormal processes in ascending order according to the value, and the data conversion frequencies corresponding to the two adjacent groups of abnormal processes in the process sequence are reversely adjusted until the T associated with the synchronous abnormal process is q and V q Satisfaction (T q -V q )≤Y1;

[0021] For the confirmed abnormal process, after a certain period of time, the CPU utilization of the abnormal process is increased to the associated T q and V q Satisfaction (T q -V q )≤Y1.

[0022] Preferably, it also includes:

[0023] The data encryption storage end encrypts and stores the periodic data packets after the format is unified. There are two ways to encrypt and store the data packets, namely, using RC4 encryption to generate binary data or encrypting the request parameters by sorting the request parameters in alphabetical order to generate a request parameter string.

[0024] The data encryption storage end generates binary data by using RC4 encryption, including: using the protobuf protocol format for data transmission, and using the CRC16-CCITT algorithm for verification, generating a 2-byte checksum, splicing it into the message body, and then encrypting it with RC4, and then sending the generated binary data to the MQTT server of the upper-level Internet of Things platform;

[0025] The method for the data encryption storage end to sort the request parameters in alphabetical order to generate a request parameter string includes: sorting the request parameters in alphabetical order to generate a request parameter string; converting the request parameters into a key-value array, and sorting them in ascending order by key. = & = & = ......The value is returned in the form of a numeric value. After the request parameters are sorted, the timestamp is concatenated to the end, and then the SM3 algorithm is used to use the key to the parameter string. =1& =2& =3×tamp=1713166038000 signature, get the signed encrypted data, the signed encrypted data is the request header parameter, parameter name: Authorization, the parameter value is divided into 3 parts, including signature algorithm, signature string, timestamp, format: SM3signature=signature,timestamp=171316603800.

[0026] The present invention provides an edge gateway for realizing lightweight monitoring of bridge health. Compared with the prior art, it has the following beneficial effects:

[0027] The present invention confirms the data acquisition frequency of different monitoring sensors, confirms the corresponding associated period based on the confirmed data acquisition frequency, collects the periodic data generated in the associated period, and then performs format-unified processing on the data in the periodic data, and gives priority to confirming the conversion format of the corresponding data. The conversion format of the data has a corresponding data conversion rate in the conversion process. When confirming the value, the rate interval associated with the corresponding stage can be locked, and then the best format to be converted can be finally confirmed based on different rate intervals. The best format to be converted is a set of conversion formats with the fastest conversion rate in the entire data conversion process, which can fully reduce the specific conversion time to achieve the best conversion processing effect.

[0028] Monitor the data conversion process to effectively ensure that the conversion rates associated with multiple data format processes are all guaranteed, so as to comprehensively optimize the process, improve the comprehensive optimization rate of several data, and reduce the conversion time;

[0029] Based on the generated periodic data, different encryption methods are used to encrypt it, so that the security of the monitored data can be fully guaranteed and achieve better security protection effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic diagram of the principle framework of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] First embodiment

[0033] See also Figure 1 , the present application provides an edge gateway for realizing lightweight monitoring of bridge health, including a frequency data acquisition terminal, a periodic data acquisition terminal, a historical data analysis terminal, a data format unification terminal, a process adjustment and optimization terminal, and a data encryption storage terminal;

[0034] The frequency data acquisition end is electrically connected to the input node of the periodic data acquisition end, and the periodic data acquisition end and the historical data analysis end are both electrically connected to the input node of the data format unification end, and the process adjustment and optimization end is bidirectionally connected to the data format unification end, wherein the data format unification end is electrically connected to the input node of the data encryption storage end;

[0035] Among them, the frequency data acquisition end confirms the data acquisition frequency of several groups of monitoring sensors associated with this edge gateway. The data acquisition frequency is the interval time of data generation, such as once every 5 seconds or once every 6 seconds, which means that the corresponding monitoring sensor monitors and generates a group of data within the corresponding interval time. When different monitoring sensors perform data monitoring, they do not generate corresponding data in real time, but have periodic data generation. If the data monitored by the corresponding monitoring sensor is continuously generated, then the corresponding interval time is 0 seconds, and the confirmed data acquisition frequency is transmitted to the periodic data acquisition end;

[0036] Among them, the periodic data acquisition end determines a group of associated periods based on different data acquisition frequencies confirmed by different monitoring sensors, acquires the periodic data monitored by different monitoring sensors within the associated period, generates a periodic data packet, and transmits the generated periodic data packet to the data format unified end. The specific method of generating the periodic data packet is:

[0037] The different data acquisition frequencies confirmed by different monitoring sensors are calibrated as P i , where i represents different monitoring sensors, and P i =1 and P i =0 to eliminate, and then confirm the remaining data acquisition frequency P i The common multiples of the data are marked as the correlation period. If there are no other data acquisition frequencies P left, i, then directly set 24h as the corresponding association period, specifically, set the remaining data acquisition frequency P i For {0, 1, 5, 6, 10}, remove the corresponding 0 and 1, and from the three groups of data of 5, 6 and 10, confirm that the common multiple of the three groups of data is 30, so the determined 30 is the corresponding correlation period;

[0038] Based on the confirmed association cycle, starting from the moment the edge gateway intervenes, the monitoring data generated by different monitoring sensors are acquired, and the monitoring data generated by different monitoring sensors are merged into data packets to confirm the periodic data packets, and the different monitoring data generated by different monitoring sensors in the periodic data packets are all marked with different tags.

[0039] The historical data analysis end, based on the different data formats associated with different monitoring data in the periodic data packet, confirms the data conversion rate between different data formats from the historical data, and then performs numerical analysis on the confirmed several different data conversion rates to determine the best format to be converted. Specifically, when the data is processed in a unified format, the conversion format of the corresponding data needs to be confirmed. The conversion format of the data has a corresponding data conversion rate in the conversion process. When the numerical confirmation is performed, the rate range associated with the corresponding stage can be locked, and then based on different rate ranges, the best format to be converted is finally confirmed. The specific method for determining the best format to be converted is:

[0040] Confirm different data formats associated with different monitoring data in the periodic data packet, mark the confirmed different data formats as the main format, and randomly select a group of data formats from the main format as the formats to be converted;

[0041] The data conversion rate generated during the conversion from the main format to the format to be converted is identified from the historical data, and the data conversion rates of several groups of data when the single main format is converted to the format to be converted are numerically confirmed. Based on the confirmed minimum data conversion rate and maximum data conversion rate, the rate range associated with the corresponding single main format is confirmed. Then, based on the different rate ranges corresponding to different main formats, the minimum values ​​of several groups of different rate ranges are calibrated as QJ. k min, mark the maximum value as QJ k max, where k represents the different rate intervals associated with different main formats, and then determines several different minimum values ​​QJ k The equilibrium value H1 between min will be less than QJ of H1 k min is marked as the lower value, and QJ greater than H1 k min is calibrated as the upper value, the sum of the differences between H1 and several lower values ​​is equal to the sum of the differences between the upper value and H1 (the corresponding equilibrium value H1 can be determined), and then the same processing method is used to confirm several different maximum values ​​QJk max, and confirm the characteristic rate interval of the format to be converted based on the balance values ​​H1 and H2. If H1=H2, directly determine the characteristic parameters of the corresponding format to be converted, and its characteristic parameters are the same as H1 and H2. Specifically, the main formats to be converted are A, B, C and D, respectively. According to the corresponding processing process, the main format A is preferentially used as the corresponding format to be converted, and the specific data conversion rates of BA, CA and DA can be confirmed from the historical data, so that the corresponding characteristic rate can be confirmed based on the corresponding data conversion rate, so as to perform relevant confirmation of the data rate interval, and thus lock the corresponding characteristic rate interval;

[0042] Then, other main formats are taken as formats to be converted in turn, and the characteristic rate range or characteristic parameter associated with the corresponding formats to be converted is determined by the same processing method as above;

[0043] Based on different characteristic rate intervals or characteristic parameters confirmed by different formats to be converted, the middle value of the characteristic rate interval is used as the characteristic parameter associated therewith, and then the format to be converted associated with the maximum characteristic parameter is selected as the best format to be converted, and the confirmed best format to be converted is transmitted to the data format unified terminal;

[0044] Specifically, the determined optimal format to be converted is a set of conversion formats with the fastest conversion rate in the entire data conversion process, which can fully reduce the specific conversion time to achieve the best conversion processing effect.

[0045] Among them, the data format unification end, based on the confirmed periodic data packet and the optimal format to be converted, converts the format of the monitoring data in the periodic data packet that is different from the optimal format to be converted, so that the data format of the monitoring data in the periodic data packet is converted to the optimal format to be converted.

[0046] Second embodiment

[0047] When performing data format conversion, there are different data format conversion processes. Since the conversion processes interfere with each other, the specific data conversion rate will be affected. Therefore, data optimization is performed through the corresponding process adjustment optimization end.

[0048] Among them, the process adjustment and optimization end monitors the data format conversion process existing in the data format unification end to confirm the real-time conversion rate, and then confirms the abnormal process based on the characteristic parameters associated with the corresponding conversion process, and optimizes and adjusts the data conversion process based on the specific results of the real-time confirmation to ensure the conversion rate during the data conversion process. The specific sub-steps of optimization and adjustment are:

[0049] The conversion rate of different data format conversion processes monitored in real time is calibrated as V q, where q represents different data format conversion processes, and determines the characteristic parameters associated with the data format conversion process, and marks the associated characteristic parameters as T q ;

[0050] If (T q -V q )>Y1, where Y1 is a preset value, and its specific value is determined by the operator based on experience. The currently associated data format conversion process is marked as an abnormal process. If there are other processes marked as abnormal processes within a certain period after the calibration time, the corresponding abnormal process is marked as a synchronous abnormal process. The certain period is a preset period, which is generally 5 seconds or even shorter.

[0051] For the confirmed abnormal process, after a certain period of time, the CPU utilization of the abnormal process is increased to the associated T q and V q Satisfaction (T q -V q )≤Y1;

[0052] For the confirmed synchronous abnormal process (representing the situation that two groups of processes interfere with each other), confirm the data conversion frequency associated with the corresponding abnormal process, sort several groups of abnormal processes according to the value from small to large to determine the process sequence, and reversely adjust the data conversion frequencies corresponding to the two adjacent groups of abnormal processes in the process sequence (that is, when the frequency of one abnormal process increases, the frequency of the other abnormal process decreases. The frequency is different from the rate. The frequency represents the number of processing times within the corresponding time, and the rate represents the corresponding processing volume. When the frequencies of the two groups of processes are superimposed, mutual interference will occur), until the T associated with the synchronous abnormal process q and V q Satisfaction (T q -V q )≤Y1;

[0053] Specifically, by adopting this process optimization processing method, the conversion rates associated with multiple data format processes can be effectively guaranteed, thereby performing comprehensive optimization of the process, improving the comprehensive optimization rate of several data, and reducing the conversion time.

[0054] The data encryption storage end encrypts and stores the periodic data packets after the format is unified. There are two ways to encrypt and store the data packets, namely, using RC4 encryption to generate binary data or encrypting the request parameters by sorting the request parameters in alphabetical order to generate a request parameter string.

[0055] The method of using RC4 encryption to generate binary data includes: using the protobuf protocol format for data transmission, and using the CRC16-CCITT algorithm for verification, generating a 2-byte checksum, splicing it into the message body, and then encrypting it with RC4, and then sending the generated binary data to the MQTT server of the upper-level IoT platform;

[0056] Sorting the request parameters in alphabetical order to generate a request parameter string includes: Sorting the request parameters in alphabetical order to generate a request parameter string, example: = & = & = ; After converting the request parameters into a key-value array, sort them in ascending order by key. = & = & = ......The value is returned in the form of a numeric value. After the request parameters are sorted, the timestamp is concatenated to the end. Example: =1& =2& =3×tamp=1713166038000, and then use the national encryption sm3 algorithm and the key (the key in the access information automatically generated by the system after the device is registered) to encrypt the parameter string =1& =2& =3×tamp=1713166038000 signature, get the signed encrypted data, the signed encrypted data is the request header parameter, parameter name: Authorization, the parameter value is divided into 3 parts, including signature algorithm, signature string, timestamp, format: SM3signature=signature,timestamp=171316603800.

[0057] Furthermore, in order to achieve persistent storage of bridge health monitoring data, the data can be stored in a cloud server or a local storage device. Cloud servers can provide larger storage space and stronger data processing capabilities, but network security and privacy protection issues need to be considered. Local storage devices can provide faster data access speeds and higher data security, and at the same time, regular data backup and maintenance policy settings need to be made.

[0058] The databases included in its cloud server are: time series database, relational database, memory database and distributed database. The time series database supports data storage middleware such as ElasticSearch and InfluxDB, the relational database supports data storage middleware such as Mysql, SqlServer, Sqlite, the memory database supports the cache-type memory database Reids, and the distributed database supports the distributed database MongoDB.

[0059] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0060] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An edge gateway for implementing lightweight monitoring of bridge health, characterized in that: include: The frequency data acquisition terminal confirms the data acquisition frequency of several groups of monitoring sensors associated with the edge gateway, and transmits the confirmed data acquisition frequency to the periodic data acquisition terminal; The periodic data acquisition end determines a group of associated periods based on different data acquisition frequencies confirmed by different monitoring sensors, acquires the periodic data monitored by different monitoring sensors within the associated period, generates a periodic data packet, and transmits the generated periodic data packet to the data format unified end; The historical data analysis end, based on the different data formats associated with different monitoring data in the periodic data packet, confirms the data conversion rates between different data formats from the historical data, and then performs numerical analysis on the confirmed several different data conversion rates to determine the best format to be converted. The specific method is as follows: Confirm different data formats associated with different monitoring data in the periodic data packet, mark the confirmed different data formats as the main format, and randomly select a group of data formats from the main format as the formats to be converted; The data conversion rate generated during the conversion from the main format to the format to be converted is identified from the historical data, and the data conversion rates of several groups of data when the single main format is converted to the format to be converted are numerically confirmed. Based on the confirmed minimum data conversion rate and maximum data conversion rate, the rate range associated with the corresponding single main format is confirmed. Then, based on the different rate ranges corresponding to different main formats, the minimum values ​​of several groups of different rate ranges are calibrated as QJ. k min, mark the maximum value as QJ k max, where k represents the different rate intervals associated with different main formats, and then determines several different minimum values ​​QJ k The equilibrium value H1 between min will be less than QJ of H1 k min is marked as the lower value, and QJ greater than H1 k min is calibrated as the upper value, the sum of the differences between H1 and several lower values ​​is equal to the sum of the differences between the upper value and H1, and then the same processing method is used to confirm several different maximum values ​​QJ k max balance value H2, based on the balance values ​​H1 and H2 to determine the characteristic rate range of the format to be converted, if H1=H2, then directly determine the characteristic parameters of the corresponding format to be converted, Its characteristic parameters are the same as those of H1 and H2; Then, other main formats are taken as formats to be converted in turn, and the characteristic rate range or characteristic parameter associated with the corresponding formats to be converted is determined by the same processing method as above; Based on different characteristic rate intervals or characteristic parameters confirmed by different formats to be converted, the middle value of the characteristic rate interval is used as the characteristic parameter associated therewith, and then the format to be converted associated with the maximum characteristic parameter is selected as the best format to be converted, and the confirmed best format to be converted is transmitted to the data format unified terminal; The data format unification end converts the monitoring data in the periodic data packet that is in a different format from the best format to be converted based on the confirmed periodic data packet and the best format to be converted, so that the data format of the monitoring data in the periodic data packet is converted to the best format to be converted; The process adjustment and optimization end monitors the data format conversion process existing in the data format unification end to confirm the real-time conversion rate, and then confirms the abnormal process based on the characteristic parameters associated with the corresponding conversion process, and optimizes and adjusts the data conversion process based on the specific results of real-time confirmation.

2. The edge gateway for realizing lightweight monitoring of bridge health according to claim 1, characterized in that: The specific method of generating the periodic data packet at the periodic data acquisition end is as follows: The different data acquisition frequencies confirmed by different monitoring sensors are calibrated as P i , where i represents different monitoring sensors, and P i =1 and P i =0 to eliminate, and then confirm the remaining data acquisition frequency P i The common multiples of the data are marked as the correlation period. If there are no other data acquisition frequencies P left, i , then directly set 24h as the corresponding association period; Based on the confirmed association cycle, starting from the moment the edge gateway intervenes, the monitoring data generated by different monitoring sensors are acquired, and the monitoring data generated by different monitoring sensors are merged into data packets to confirm the periodic data packets.

3. The edge gateway for realizing lightweight monitoring of bridge health according to claim 1, characterized in that: The specific sub-steps of optimizing and adjusting the data conversion process at the process adjustment and optimization end are as follows: The conversion rate of different data format conversion processes monitored in real time is calibrated as V q , where q represents different data format conversion processes, and determines the characteristic parameters associated with the data format conversion process, and marks the associated characteristic parameters as T q ; If (T q -V q )>Y1, where Y1 is a preset value, the currently associated data format conversion process is marked as an abnormal process. If there are other processes marked as abnormal processes within a certain period after the marking time, the corresponding abnormal process is marked as a synchronous abnormal process, and the certain period is the preset period; For the confirmed synchronous abnormal process, the data conversion frequency associated with the corresponding abnormal process is confirmed, and the process sequence is determined by sorting several groups of abnormal processes in ascending order according to the value, and the data conversion frequencies corresponding to the two adjacent groups of abnormal processes in the process sequence are reversely adjusted until the T associated with the synchronous abnormal process is q and V q Satisfaction (T q -V q )≤Y1.

4. The edge gateway for realizing lightweight monitoring of bridge health according to claim 3, characterized in that: For the confirmed abnormal process, after a certain period of time, the CPU utilization of the abnormal process is increased to the associated T q and V q Satisfaction (T q -V q )≤Y1.

5. The edge gateway for realizing lightweight monitoring of bridge health according to claim 1, characterized in that: Also includes: The data encryption storage end encrypts and stores the periodic data packets after the format is unified. There are two ways of encryption storage, namely, using RC4 encryption to generate binary data or encrypting the request parameters by sorting them in natural alphabetical order to generate a request parameter string.

6. The edge gateway for realizing lightweight monitoring of bridge health according to claim 5, characterized in that: The data encryption storage end generates binary data using RC4 encryption, including: using the protobuf protocol format for data transmission, and using the CRC16-CCITT algorithm for verification, generating a 2-byte check code, splicing it into the message body, and then encrypting it with RC4, and then sending the generated binary data to the MQTT server of the upper-level Internet of Things platform.

7. The edge gateway for implementing lightweight monitoring of bridge health according to claim 5, characterized in that: The method for the data encryption storage end to sort the request parameters in alphabetical order to generate a request parameter string includes: sorting the request parameters in alphabetical order to generate a request parameter string; converting the request parameters into a key-value array, and sorting them in ascending order by key. = & = & = ......The value is returned in the form of a numeric value. After the request parameters are sorted, the timestamp is concatenated to the end, and then the SM3 algorithm is used to use the key to the parameter string. =1& =2& =3×tamp=1713166038000 signature, get the signed encrypted data, the signed encrypted data is the request header parameter, parameter name: Authorization, the parameter value is divided into 3 parts, including signature algorithm, signature string, timestamp, format: SM3signature=signature,timestamp=171316603800.

Images