A Method and System for Dynamic Settlement Control of Soft Soil Culvert Bridges Based on Internet of Things Monitoring

Abstract

This invention discloses a method and system for dynamic settlement control of soft soil culvert bridges based on Internet of Things (IoT) monitoring, belonging to the field of IoT monitoring technology. The method includes the following steps: constructing a settlement motion state machine for each settlement monitoring point; constructing a sliding verification window for each settlement monitoring point; calculating the average rate of the window; comparing the average rate of the window with the average rate of the previous window corresponding to the steady state within the settlement motion state machine by a factor of change to determine if it is a false rate; if determined to be a false rate, forcibly setting the settlement motion state machine to a transient state, and outputting the arithmetic average rate of the independent sliding verification window to the early warning control logic to remove the degradation restriction. This invention, by constructing a settlement motion state machine, a sliding verification window, and performing degradation processing, solves the problem in existing technologies where supplementary data is misjudged as instantaneous high rates, leading to unnecessary triggering of load limits or pole drops.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) monitoring technology, and in particular to a method and system for dynamic control of settlement of soft soil culvert bridges based on IoT monitoring. Background Technology

[0002] In the operation of culvert bridges in soft soil areas, solar power combined with batteries is often used to power settlement monitoring equipment at the bridge site, while data transmission relies on narrowband IoT or low-power wide-area networks. Due to continuous rainy weather, voltage fluctuations can easily lead to random loss of sampling time at the data acquisition terminal; in addition, the soil and structures inside the culvert and near the abutment can obstruct the wireless signal, causing frequent reconnections and packet loss. Even with local buffering, retransmissions can still result in irregular and dense arrivals.

[0003] Existing monitoring systems typically calculate the settlement rate directly based on a fixed time window and then compare the rate with a preset threshold to trigger a load limit or pole drop command.

[0004] In practical applications, when communication is interrupted for several days and then resumed, multiple data packets arrive simultaneously within a very short period. The system may misinterpret the small amount of settlement accumulated over several days as a sudden high rate, thus erroneously triggering orange or even red alerts, causing electric barriers to fall and block traffic without warning. On-site personnel only realize the false alarm after verification, but unnecessary traffic disruptions and operational losses have already occurred. Therefore, a dynamic settlement control method is needed that can identify and suppress false high rates caused by data discontinuity without affecting the actual hazard response. Summary of the Invention

[0005] This application provides a method and system for dynamic control of settlement of soft soil culvert bridges based on Internet of Things monitoring. It solves the problem in the prior art where data transmission is misjudged as instantaneous high rate due to power fluctuations and communication interruptions, thus triggering unnecessary load limits or even lowering barriers to block traffic. It achieves the technical effect of accurately identifying and suppressing false settlement rates and avoiding ineffective control actions under data interruption conditions.

[0006] This application provides a method for dynamic control of settlement of soft soil culvert bridges based on Internet of Things monitoring, including: constructing a settlement motion state machine for each settlement monitoring point and setting the initial state of all settlement monitoring points to a steady state;

[0007] A sliding verification window for settlement monitoring points is constructed. The local timestamps of the current data point and the previous data point in the database are extracted, and consecutive valid sampling points are selected and filled into the sliding verification window.

[0008] When the sliding verification window is filled with a preset number of valid sampling points, the average window rate is calculated.

[0009] The average rate of the window is compared with the average rate of the previous window corresponding to the steady state in the settling motion state machine by a factor of change. If the factor of change exceeds the set factor of change, the settling motion state machine is switched to the transition state and the inertial rejection logic is triggered.

[0010] Extract the first-order difference sequence of historical valid sampling points from the historical database and calculate the historical maximum rate to determine whether it is a false rate;

[0011] If the rate is determined to be false, the settlement motion state machine is forcibly set to transient, and replacement values ​​and discontinuity correction flags are generated to perform downgrade processing on the data of multiple consecutive sliding verification windows of the settlement monitoring point.

[0012] Output the arithmetic average rate of the independent sliding verification window to the early warning and control logic to remove the downgrade restriction.

[0013] Furthermore, the steps for selecting consecutive valid sampling points to fill the sliding verification window include:

[0014] Receive current settlement monitoring data points collected by sensors and extract the latest local timestamp information;

[0015] Access the historical database to query the last local timestamp information of the previous data point that has been stored in the database for the same settlement monitoring point;

[0016] The actual data reception time interval is obtained by calculating the difference between the latest local timestamp information and the previous local timestamp information.

[0017] Extract the preset maximum normal sampling time interval parameter and multiply it by the tolerance factor to obtain the maximum allowable discontinuity time threshold;

[0018] If the actual data reception time interval is determined to be no greater than the maximum discontinuity time threshold, the current data point is marked as a valid sampling point and sequentially filled into the internal storage queue of the sliding verification window;

[0019] If the actual data reception time interval is determined to be greater than the maximum discontinuity time threshold, the current data point is marked as an invalid sampling point.

[0020] Furthermore, when the sliding verification window is filled with a preset number of valid sampling points, the step of calculating the window average rate includes:

[0021] Continuously monitor the cumulative number of valid sampling points in the internal storage queue of the sliding verification window;

[0022] When the cumulative amount is identified to have precisely reached the preset full load threshold, the initial cumulative settlement value and the initial local timestamp record contained in the first valid sampling point of the time series in the storage queue are located and extracted.

[0023] Extract the cumulative settlement value and local timestamp record at the end of the last valid sampling point in the time series.

[0024] Subtract the initial cumulative settlement value from the cumulative settlement value at the end to obtain the absolute cumulative settlement difference within the span;

[0025] Subtract the initial local timestamp record from the end local timestamp record to obtain the absolute time elapsed difference within the span;

[0026] The window average rate is generated by dividing the absolute cumulative settlement difference by the absolute time elapsed difference; it is then determined whether the current settlement motion state machine is in a steady state.

[0027] If the system is in a steady state, extract the average rate of the previous window corresponding to the previous sampling period, and calculate the absolute change factor of the average rate of the window compared to the average rate of the previous window as the basis for control.

[0028] Furthermore, the steps of extracting the first-order difference sequence composed of historical valid sampling points from the historical database and calculating the historical maximum rate include:

[0029] When the comparison reveals that the absolute change factor parameter exceeds the system's set safe floating factor threshold, the settlement motion state machine is updated from steady state to a transitional state flag awaiting verification, and the inertial veto logic for abnormal settlement rate mutations is activated simultaneously.

[0030] Extract a historical time period of a preset length based on the start time of the current sliding verification window;

[0031] Scan and extract all historical valid sampling point datasets with valid labels stored within the historical time period and count the total number of data entries;

[0032] If the total number of data entries is less than the minimum required number, it is determined that there are insufficient historical samples, the inertial rejection logic is terminated, the state machine is brought back to a steady state, and the current window average rate is adopted.

[0033] If the total number of data entries reaches the minimum required number, sort all extracted historical valid sampling points according to the time series.

[0034] Calculate the rate of settlement change between adjacent historical valid sampling points, construct a set of first-order difference sequences that reflect the historical settlement fluctuation pattern, and screen out the historical maximum rate extreme value in the set of sequences.

[0035] Furthermore, the steps of extracting the first-order difference sequence composed of historical valid sampling points from the historical database and calculating the historical maximum rate to determine whether it is a false rate include:

[0036] Obtain the historical maximum rate extreme value that represents the actual settlement peak value in the past period, selected from the historical first-order difference rate sequence set;

[0037] The scaling factor of the current mutation rate relative to the historical peak limit is calculated.

[0038] Compare the scaling factor exceeding the multiple with the preset allowable threshold for deviation of physical extreme inertia;

[0039] If the ratio factor exceeds the allowable threshold, the external environmental load module will be activated to investigate indirect compliance evidence.

[0040] Send a joint query message to an external traffic database;

[0041] Obtain actual traffic flow statistics and vehicle load monitoring records for road sections that completely overlap with historical time periods;

[0042] Determine whether the actual traffic flow statistics record exceeds the preset daily average load ratio limit and retrieve the overweight vehicle passage records;

[0043] If the actual traffic flow does not exceed the upper limit of the load ratio and there are no overweight vehicles, it is determined that there are no external conditions to cause severe subsidence, and it is classified as a false rate.

[0044] Furthermore, if the rate is determined to be false, the settling motion state machine is forcibly set to transient state, and the steps for generating replacement values ​​include:

[0045] The current window average rate value that is determined to be a physically untrusted false rate is intercepted from the transmission path to the security warning execution module.

[0046] Retrieve the average rate cache record value output by the previous complete sliding verification window of the settlement monitoring point that has passed a complete physical consistency verification cycle;

[0047] The average rate cache records the values ​​by overwriting them with memory overwrite operations, thus generating a smooth-transition rate replacement value parameter.

[0048] The generated rate replacement value parameter is only granted read and write permissions to the backend database archive storage module and the trend curve rendering module, and warning and control actions are prohibited from being triggered based on this value.

[0049] Send a status rewrite command to the global status registry to forcibly change the settlement motion state machine attribute corresponding to the settlement monitoring point from a transitional state to a transient identifier representing short-term data isolation and observation period to block external traffic intervention commands.

[0050] Furthermore, the steps for generating discontinuity correction flags and performing downgrading processing on the data from multiple subsequent sliding verification windows of the settlement monitoring point include:

[0051] Once the underlying layer successfully executes the rate replacement overwrite operation designed to smooth out false data, an exception event response logging program is automatically triggered.

[0052] Based on the current time and the unified encoding of the sensor that triggered the action, the discontinuity correction flag with a unique traceability identifier is generated;

[0053] The discontinuous correction flag is deeply bound to the current sliding time period that caused the anomaly and permanently stored in an independently allocated system audit log library;

[0054] For the settlement monitoring point carrying the transient marker, a comprehensive data degradation observation process is initiated, with the duration covering multiple subsequent preset independent sliding cycles.

[0055] During the downgrade process coverage period, the flow of newly calculated window average rate data for each subsequently generated sliding verification window is forcibly restricted to any warning threshold comparison algorithm container.

[0056] Furthermore, the steps for outputting the arithmetic average rate of the independent sliding verification window to the early warning control logic to remove the downgrade restriction include:

[0057] Continuously monitor the remaining execution cycle of the data degradation observation process until it is confirmed that the set degradation penalty time has been completely exhausted and the project is closed;

[0058] Strict requirements were initially imposed that the settlement monitoring point must continuously and successfully construct a preset number of independent sliding verification windows that do not overlap with each other on the time axis;

[0059] The start and end time span of each independent sliding verification window must be an integer multiple of the original data acquisition time interval;

[0060] The aforementioned inertial rejection logic, which includes historical extreme value comparison, is re-run for each of the pre-defined number of independent sliding verification windows that are continuously constructed.

[0061] If and only if the average rate of the new windows generated by each of the independent sliding verification windows in the batch does not trigger the rejection condition of being judged as a physically unreliable false rate;

[0062] Extract the numerical sequence of the average rate of this batch of continuous safe new windows, and obtain the arithmetic mean rate of this batch of safe rates by summing and dividing by the number of statistical items;

[0063] The confirmed arithmetic average rate result value will be re-enabled and output to the early warning and control logic.

[0064] This application provides an IoT-based dynamic settlement control system for soft soil culvert bridges, used to implement a method for dynamic settlement control of soft soil culvert bridges based on IoT monitoring, including:

[0065] The module includes a state machine construction module, a verification window filtering module, an average rate calculation module, a change factor comparison module, a false rate determination module, a degradation processing module, and a restriction removal module.

[0066] The state machine construction module is used to construct the settlement motion state machine for each settlement monitoring point and set the initial state of all settlement monitoring points to a steady state.

[0067] The verification window filtering module is used to construct a sliding verification window for settlement monitoring points, extract the local timestamps of the current data point and the previous data point already in the database, and filter continuous valid sampling points to fill the sliding verification window.

[0068] The average rate calculation module is used to calculate the average rate of the window when the sliding verification window is filled with a preset number of valid sampling points.

[0069] The change factor comparison module is used to compare the average rate of the window with the average rate of the previous window corresponding to the steady state in the settling motion state machine. If the change factor exceeds the set change factor, the settling motion state machine is switched to the transition state and the inertial rejection logic is triggered.

[0070] The false rate determination module is used to extract a first-order difference sequence composed of historical valid sampling points from the historical database and calculate the historical maximum rate to determine whether it is a false rate.

[0071] The degradation processing module is used to, if determined to be a false rate, force the settlement motion state machine to be set to transient state, generate replacement values ​​and discontinuity correction flags to perform degradation processing on the data of multiple consecutive sliding verification windows of the settlement monitoring point.

[0072] The restriction removal module is used to output the arithmetic average rate of the independent sliding verification window to the early warning control logic to remove the downgrade restriction.

[0073] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0074] By constructing a settlement motion state machine and sliding verification window for each monitoring point, the average rate of the window is calculated by selecting continuous effective sampling points, and the inertial rejection logic is triggered to retrieve historical differential sequences and external traffic load data to judge false rates. In this way, false high rates caused by retransmission can be accurately identified when the data is discontinuous, and the cumulative settlement can be misjudged as instantaneous and severe settlement.

[0075] Furthermore, after determining that the rate is false, the state machine is forcibly switched to transient state, a replacement value and a discontinuity correction flag are generated, and a degradation process is performed on multiple subsequent sliding windows to restrict their data from entering the early warning comparison. This ensures that the rate spike after communication is restored will not trigger the load limit or drop button command, and the degradation period isolation prevents the continuous interference of a single discontinuous event.

[0076] Furthermore, this method relies solely on state transitions, window filtering, extreme value comparison, and multi-window continuous verification, requiring no additional hardware. It can track the actual settlement trend under adverse conditions such as power fluctuations and communication interruptions, reducing unnecessary traffic disruptions and operational losses, and improving the reliability of the control system. Attached Figure Description

[0077] Figure 1 Flowchart of a method for dynamic settlement control of soft soil culvert bridges based on Internet of Things monitoring, provided in an embodiment of this application;

[0078] Figure 2 This is a schematic diagram of the settlement dynamic control system for soft soil culvert bridges based on Internet of Things monitoring, provided in an embodiment of this application. Detailed Implementation

[0079] This application provides a method and system for dynamic control of settlement of soft soil culvert bridges based on Internet of Things monitoring. It solves the problem in the prior art where retransmitted data is misjudged as instantaneous high rate due to power fluctuations and communication interruptions, thus triggering unnecessary load limits or pole drops. By constructing a settlement motion state machine, a sliding verification window, and inertial veto logic to identify false rates and perform degradation processing, it achieves the technical effect of accurately suppressing false settlement rate peaks and avoiding ineffective traffic control actions under data interruption conditions.

[0080] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0081] like Figure 1 The diagram shows a flowchart of a method for dynamic settlement control of soft soil culvert bridges based on Internet of Things (IoT) monitoring, provided in an embodiment of this application. This method is applied to a system for dynamic settlement control of soft soil culvert bridges based on IoT monitoring. The method includes the following steps: constructing a settlement motion state machine for each settlement monitoring point and setting the initial state of all settlement monitoring points to a steady state.

[0082] A sliding verification window for settlement monitoring points is constructed. The local timestamps of the current data point and the previous data point in the database are extracted, and the time difference between the two is calculated to filter out continuous valid sampling points to fill the sliding verification window.

[0083] When the sliding verification window is filled with a preset number of valid sampling points, the cumulative settlement difference and local timestamp difference between the first and last valid sampling points in the sliding verification window are extracted, and the average window rate is calculated.

[0084] The average rate of the window is compared with the average rate of the previous window corresponding to the steady state in the settling motion state machine by a factor of change. If the factor of change exceeds the set factor of change, the settling motion state machine is switched to the transition state and the inertial rejection logic is triggered.

[0085] The inertial rejection logic is executed, the first-order difference sequence composed of historical valid sampling points is extracted from the historical database and the historical maximum rate is calculated. The average rate of the window is compared with the historical maximum rate, and traffic flow data is used to determine whether it is a false rate.

[0086] If the rate is determined to be false, the settlement motion state machine is forcibly set to transient, and the average rate of the previous window is used to replace the average rate of the window to generate a replacement value. The replacement value is used for historical record display, and the discontinuity correction flag is used to downgrade the data of the subsequent multiple sliding verification windows of the settlement monitoring point.

[0087] When the downgrade processing period expires and the average rate of the new window calculated by multiple independent sliding verification windows is not judged as a false rate, the arithmetic average rate of the independent sliding verification windows is output to the early warning control logic to remove the downgrade restriction, and long-term trend update and state machine reset operations are periodically performed to maintain the data model.

[0088] In this embodiment, during the system initialization phase, a corresponding "settlement motion state machine" object is constructed in the dynamic allocation area of ​​the system memory for each settlement monitoring point on the culvert bridge. This state machine object is composed of a structure (Struct) in the underlying data structure, which includes a state enumeration variable (Enum_State), a timestamp register, an accumulator, and a flag pointer.

[0089] During initialization, the control program iterates through the identifiers (IDs) of all settlement monitoring points and forces the enumeration variables of all settlement motion state machines to be set to steady state (STEADY). Subsequently, a continuous, fixed-length ring buffer is allocated in memory for each monitoring point to construct a sliding verification window. The length of this buffer is set according to the preset number of valid sampling points (e.g., 3).

[0090] The system receives data packets from IoT nodes via polling or interruption mechanisms, extracts the cumulative settling value and local timestamp from the data packets, and then proceeds to subsequent verification, calculation, and rejection logic. Through state machine switching and the intervention of inertial rejection logic, it can automatically isolate the pollution of the control logic by false high-rate data when encountering intermittent data transmission, ultimately achieving closed-loop degradation and recovery control.

[0091] Furthermore, the steps for selecting consecutive valid sampling points to fill the sliding verification window include:

[0092] Receive current settlement monitoring data points collected by sensors and extract the latest local timestamp information;

[0093] Access the historical database to query the last local timestamp information of the previous data point that has been stored in the database for the same settlement monitoring point;

[0094] The actual data reception time interval is obtained by calculating the difference between the latest local timestamp information and the previous local timestamp information.

[0095] Extract the preset maximum normal sampling time interval parameter and multiply it by the tolerance factor to obtain the maximum allowable discontinuity time threshold;

[0096] If the actual data reception time interval is determined to be no greater than the maximum discontinuity time threshold, the current data point is marked as a valid sampling point and sequentially filled into the internal storage queue of the sliding verification window;

[0097] If the actual data reception time interval is determined to be greater than the maximum discontinuity time threshold, the current data point is marked as an invalid sampling point, preventing it from entering the storage queue, storing it in the historical database with an additional discontinuity tag, and clearing the storage queue counter.

[0098] In this embodiment, when the communication front-end receives a current settlement monitoring data packet uploaded by an IoT sensor via an NB-IoT or LoRa network, the parsing module extracts the local timestamp information of the current data point from the packet header. The timestamp uses the UNIX standard time format (accurate to milliseconds).

[0099] Next, an SQL query command is sent to the Time-Series Database to retrieve the local timestamp information of the last successfully entered data point with a valid tag, based on the sensor's unique MAC address. .

[0100] The central processing unit (CPU) performs the difference calculation to obtain the actual data reception time interval. :

[0101] ;

[0102] The maximum normal sampling time interval parameter is set in advance in the configuration file. (For example, setting the normal sampling period to 10 minutes). To accommodate normal network handshake delays, a tolerance factor is introduced. (Default value is 2.0), calculate the maximum allowable discontinuous time threshold. :

[0103] ;

[0104] The judgment logic is as follows: If This indicates that the data has not experienced severe network congestion or network outage retransmission. The current data point is instantiated as a "valid sampling point" object, and a pointer of the object is pushed to the end of the circular buffer (enqueue operation). At the same time, the storage queue counter is incremented by one.

[0105] like This indicates that the data point is a centralized retransmission point after the network outage was restored, and it has a large amount of time compression distortion.

[0106] At this point, mark it as an "invalid sampling point", discard the enqueue request (prevent it from entering the circular buffer), and directly store it into the history database through an asynchronous write thread. Append the string label "Interrupted" to the "status bar" of the data table, and trigger a reset command to clear the current circular buffer and reset the counter to zero to prevent contamination of subsequent windows.

[0107] Furthermore, when the sliding verification window is filled with a preset number of valid sampling points, the step of calculating the window average rate includes:

[0108] Continuously monitor the cumulative number of valid sampling points in the internal storage queue of the sliding verification window;

[0109] When the cumulative amount is identified to have precisely reached the preset full load threshold, the initial cumulative settlement value and the initial local timestamp record contained in the first valid sampling point of the time series in the storage queue are located and extracted.

[0110] Extract the cumulative settlement value and local timestamp record at the end of the last valid sampling point in the time series.

[0111] Subtract the initial cumulative settlement value from the cumulative settlement value at the end to obtain the absolute cumulative settlement difference within the span;

[0112] Subtract the initial local timestamp record from the end local timestamp record to obtain the absolute time elapsed difference within the span;

[0113] The window average rate is generated by dividing the absolute cumulative settlement difference by the absolute time elapsed difference; it is then determined whether the current settlement motion state machine is in a steady state.

[0114] If the system is in a steady state, extract the average rate of the previous window corresponding to the previous sampling period, and calculate the absolute change factor of the average rate of the window compared to the average rate of the previous window as the basis for control.

[0115] In this embodiment, the cumulative number of valid sampling points inside the circular buffer is continuously monitored in a background thread. When the value of the storage queue counter is exactly equal to a preset full-load threshold (e.g., 3), a "window full-load interrupt" is triggered.

[0116] In the interrupt service routine, the initial cumulative settlement value of the first valid sampling point (index 0) in the sequence is extracted based on the position of the head pointer of the circular buffer. (Unit: mm) and initial local timestamp Simultaneously, based on the position of the tail pointer, the cumulative settlement value at the end of the last valid sampling point (subscript 2) is extracted. and local timestamp at the end .

[0117] Since soft soil settlement is physically a slow, macroscopic movement, its average settlement rate The calculation is based on the classical kinematic mean speed formula:

[0118] ;

[0119] In acquiring Then, read the current enumeration value of the settlement motion state machine at the monitoring point. If the current state is steady (STEADY), retrieve the window average rate of the previous complete cycle for that monitoring point from the cache. Then, the absolute change factor of the current rate compared to the previous rate is calculated. :

[0120] ;

[0121] This parameter It is a dimensionless ratio used to quantify the severity of sudden changes in settlement rate and serves as the core basis for whether to trigger veto control measures.

[0122] Furthermore, the steps of extracting the first-order difference sequence composed of historical valid sampling points from the historical database and calculating the historical maximum rate include:

[0123] When the comparison reveals that the absolute change factor parameter exceeds the system's set safe floating factor threshold, the settlement motion state machine is updated from steady state to a transitional state flag awaiting verification, and the inertial veto logic for abnormal settlement rate mutations is activated simultaneously.

[0124] Extract a historical time period of a preset length based on the start time of the current sliding verification window;

[0125] Scan and extract all historical valid sampling point datasets with valid labels stored within the historical time period and count the total number of data entries;

[0126] If the total number of data entries is less than the minimum required number, it is determined that there are insufficient historical samples, the inertial rejection logic is terminated, the state machine is brought back to a steady state, and the current window average rate is adopted.

[0127] If the total number of data entries reaches the minimum required number, sort all extracted historical valid sampling points according to the time series.

[0128] Calculate the rate of settlement change between adjacent historical valid sampling points, construct a set of first-order difference sequences that reflect the historical settlement fluctuation pattern, and screen out the historical maximum rate extreme value in the set of sequences.

[0129] In this embodiment, the step of extracting historical data for self-verification includes:

[0130] By setting a safety float multiple threshold (Typically set between 2.0 and 3.0 to avoid normal measurement noise). The calculated absolute change factor parameter... The system suspects a non-physical jump in the current rate. At this point, the state machine properties are rewritten, changing the steady state (STEADY) to a transitional state (TRANSITION). This is equivalent to adding a "software lock," pausing the output of warning commands, and simultaneously calling the "inertial veto subroutine."

[0131] The subroutine first uses Using time as a reference, the starting time of a historical query is calculated by extrapolating backwards. Using the database interface, query all historical data with valid tags within this 6-hour timeframe.

[0132] If the total number of valid entries returned This indicates that the historical sample is sparse and insufficient to support a statistically significant comparison. To ensure safety and adhere to the principle of "quality over quantity," the rejection logic is terminated, the state machine is reverted to a steady state, and the current data is directly updated. The valid value is output to the early warning module.

[0133] If the total number of valid entries , will this The data points are sorted in chronological order to form a set. Next, the set is traversed, and the settling rate between every two adjacent data points is calculated to construct a set of first-order difference sequences. Among them, the first The calculation method for each differential rate is as follows:

[0134] ;

[0135] After the entire sequence is constructed, a sorting algorithm (such as quicksort) is called to select the element with the largest absolute value from the set of first-order difference sequences, and this element is assigned to the historical maximum rate extreme value variable. This serves as a reference benchmark for the physical limit of motion of soft soil in recent history.

[0136] Furthermore, the steps of extracting the first-order difference sequence composed of historical valid sampling points from the historical database and calculating the historical maximum rate to determine whether it is a false rate include:

[0137] Obtain the historical maximum rate extreme value that represents the actual settlement peak value in the past period, selected from the historical first-order difference rate sequence set;

[0138] Divide the calculated average rate value of the window by the historical maximum rate extreme value to obtain the excess multiple scaling factor of the current mutation rate relative to the historical limit peak.

[0139] Compare the scaling factor exceeding the multiple with the preset allowable threshold for deviation of physical extreme inertia;

[0140] If the ratio factor exceeds the allowable threshold, the external environmental load module will be activated to investigate indirect compliance evidence.

[0141] Send a joint query message to an external traffic database;

[0142] Obtain actual traffic flow statistics and vehicle load monitoring records for road sections that completely overlap with historical time periods;

[0143] Determine whether the actual traffic flow statistics record exceeds the preset daily average load ratio limit and retrieve the overweight vehicle passage records;

[0144] If the actual traffic flow does not exceed the upper limit of the load ratio and there are no overweight vehicles, it is determined that there are no external conditions to cause severe subsidence, and it is classified as a false rate.

[0145] In this embodiment, the average rate of the current window is extracted. and compared with the extracted historical maximum rate extreme value Divide to calculate the scaling factor that exceeds the multiple. :

[0146] ;

[0147] A preset threshold for allowable deviation of physical extreme inertia has been established. (For example, according to the theory of soft soil consolidation, this value is usually set to 5.0 if there is no major damage in a short period of time). If the discriminator outputs... If the data itself is highly likely to be corrupted due to timestamp backlog caused by network retransmission, a cross-module call to the "External Environmental Load Verification" API is made to prevent unforeseen circumstances (such as extreme heavy vehicle crushing leading to actual collapse). A joint query JSON message containing the bridge section number and historical time period is sent to the traffic management department's traffic counting database and Dynamic Weighing Information (WIM). After parsing the returned message, the actual total traffic volume and heavy vehicle axle load records are extracted. If the actual traffic volume does not exceed the corresponding proportion of the section's daily average load design (e.g., 25%), and WIM does not record any overweight vehicles exceeding the design axle load limit, then a closed-loop logic is formed: the internal data shows a sudden change of more than tenfold, while the external physical cause (traffic load) is completely absent. Based on this, the expert logic determiner outputs a "TRUE" boolean value, setting the current... It is precisely characterized as a "pseudo-rate".

[0148] Furthermore, if the rate is determined to be false, the settling motion state machine is forcibly set to transient state, and the steps for generating replacement values ​​include:

[0149] At the logical level, the transmission path of the current window average rate value, which is judged to be a false rate of physical untrustworthiness, to the security warning execution module is intercepted.

[0150] Retrieve the average rate cache record value output by the previous complete sliding verification window of the settlement monitoring point that has passed a complete physical consistency verification cycle;

[0151] The average rate cache records the values ​​by overwriting them with memory overwrite operations, thus generating a smooth-transition rate replacement value parameter.

[0152] The generated rate replacement value parameter is only granted read and write permissions to the backend database archive storage module and the trend curve rendering module, and warning and control actions are prohibited from being triggered based on this value.

[0153] Send a status rewrite command to the global status registry to forcibly change the settlement motion state machine attribute corresponding to the settlement monitoring point from a transitional state to a transient identifier representing short-term data isolation and observation period to block external traffic intervention commands.

[0154] In this embodiment, once the qualitative result is a false rate, the core routing module immediately locks the current rate. The data flows to the "Safety Early Warning Execution Module" (which controls the audible and visual alarms and the raising and lowering of the barrier gate). Subsequently, the memory manager initiates a pointer redirection operation. The uncontaminated average rate of the previous window, generated from the previous complete verification cycle, is retrieved from the cache (Redis or a local memory hash table). .

[0155] Through a memory overwrite operation, the original memory structure stored in the output structure allocated to the current window is overwritten. The memory address value was forcibly replaced with The value of generates a rate replacement value parameter for a smooth transition. The replacement value It is marked as "Display_Only" in the data dictionary.

[0156] The database write process and the front-end web trend curve rendering engine can still read it to ensure that the curve is unbroken and the data is not abrupt; however, the code logic of the early warning and control module will automatically skip execution after reading the flag.

[0157] Finally, the underlying instruction sends a write command to the global state registry, forcibly rewriting the state machine variable of the settlement monitoring point from the transition state to the transient state. The monitoring point in the transient state is equivalent to being sent to a "data isolation ward" and will no longer issue any traffic restrictions or alarm instructions based on the data it currently generates.

[0158] Furthermore, the steps for generating discontinuity correction flags and performing downgrading processing on the data from multiple subsequent sliding verification windows of the settlement monitoring point include:

[0159] Once the underlying layer successfully executes the rate replacement overwrite operation designed to smooth out false data, an exception event response logging program is automatically triggered.

[0160] Based on the current time and the unified encoding of the sensor that triggered the action, the discontinuity correction flag with a unique traceability identifier is generated;

[0161] The discontinuous correction flag is deeply bound to the current sliding time period that caused the anomaly and permanently stored in an independently allocated system audit log library;

[0162] For the settlement monitoring point carrying the transient marker, a comprehensive data degradation observation process is initiated, with the duration covering multiple subsequent preset independent sliding cycles.

[0163] During the downgrade process coverage period, the flow of newly calculated window average rate data for each subsequently generated sliding verification window is forcibly restricted to any warning threshold comparison algorithm container.

[0164] The average rate of the new window mentioned above is strictly limited to internal benchmark reference data during the degradation period; at the same time, a formatted diagnostic report containing only plain text information is pushed to the operation and maintenance mobile terminal, stating that the network latency and intermittent connection are suspected and that the error correction mechanism has been initiated.

[0165] In this embodiment, while performing memory overwrite, the background audit microservice generates a globally unique identifier (UUID) as a "discontinuity correction flag" for this discontinuity event. This UUID, along with the sensor device number and the timestamp of the occurrence, is packaged into a JSON format and written to the audit log library mounted on a separate disk volume for future traceability.

[0166] Simultaneously, a degradation processing timer and counter are activated for this "transient" measurement point. The degradation process specifically involves setting a degradation window counter variable. (The threshold is set to 3). In the following acquisition cycles, even if the measurement point network recovers and new sliding windows are continuously generated to fill 3 valid sampling points, the average rate of these newly calculated windows is all marked as internal reference level (Level_Internal).

[0167] exist Before three such instances, the interceptor remains in effect, preventing any data from penetrating to the warning algorithm container. During this period, a message queue (such as RabbitMQ) generates a formatted diagnostic message: "[Device ID: XXX] is suspected of experiencing network latency and triggering a large backlog of historical data. Inertial self-rejection logic has been executed, and the device is currently in a protective degradation observation period; warnings are suspended. Maintenance personnel should monitor network power supply status." This message is pushed to the maintenance personnel's WeChat mini-program or dedicated monitoring app client.

[0168] Furthermore, the steps for outputting the arithmetic average rate of the independent sliding verification window to the early warning control logic to remove the downgrade restriction include:

[0169] Continuously monitor the remaining execution cycle of the data degradation observation process until it is confirmed that the set degradation penalty time has been completely exhausted and the project is closed;

[0170] Strict requirements were initially imposed that the settlement monitoring point must continuously and successfully construct a preset number of independent sliding verification windows that do not overlap with each other on the time axis;

[0171] The start and end time span of each independent sliding verification window must be an integer multiple of the original data acquisition time interval;

[0172] The aforementioned inertial rejection logic, which includes historical extreme value comparison, is re-run for each of the pre-defined number of independent sliding verification windows that are continuously constructed.

[0173] If and only if the average rate of the new windows generated by each of the independent sliding verification windows in the batch does not trigger the rejection condition of being judged as a physically unreliable false rate;

[0174] Extract the numerical sequence of the average rate of this batch of continuous safe new windows, and obtain the arithmetic mean rate of this batch of safe rates by summing and dividing by the number of statistical items;

[0175] The confirmed arithmetic average rate result value will be re-enabled and output to the early warning and control logic.

[0176] In this embodiment, the background degradation monitoring process continuously polls over time. The value of is strictly checked. The subsequent generated independent sliding verification windows are strictly checked (here, "independent" means that each window does not overlap at the beginning and end, and the time span is strictly equal to an integer multiple of the original sampling interval, proving that the network transmission has truly recovered to a steady-state rhythm).

[0177] For the three independent sliding verification windows generated during the downgrade period, each one is still tested using the "extreme value comparison and traffic volume query" rejection logic described in Examples 4 and 5. If any window is again determined to be a "false rate," the downgrade counter... The number of cases was forcibly cleared, and the quarantine period started again.

[0178] If and only if the average rate of the new windows of these three consecutive independent sliding windows (denoted as ) If all safety checks pass, meaning no rejection conditions are triggered, then the arithmetic mean calculation library is invoked.

[0179] ;

[0180] This As a result value confirming safety, the "Display_Only" and "Level_Internal" interception tags in the data dictionary are removed, and the memory read permission of this result value to the core algorithm module of the early warning and control logic is restored, so that the device is completely unbound and resumes its responsibility for protecting the actual settlement safety of the bridge deck.

[0181] like Figure 2 The diagram shown is a structural schematic of the settlement dynamic control system for soft soil culvert bridges based on Internet of Things monitoring provided in this application embodiment. The settlement dynamic control system for soft soil culvert bridges based on Internet of Things monitoring provided in this application embodiment includes: a state machine construction module, a verification window filtering module, an average rate calculation module, a change factor comparison module, a false rate judgment module, a degradation processing module, and a restriction removal module.

[0182] The state machine construction module is used to construct the settlement motion state machine for each settlement monitoring point and set the initial state of all settlement monitoring points to a steady state.

[0183] The verification window filtering module is used to construct a sliding verification window for settlement monitoring points, extract the local timestamps of the current data point and the previous data point already in the database, and filter continuous valid sampling points to fill the sliding verification window.

[0184] The average rate calculation module is used to calculate the average rate of the window when the sliding verification window is filled with a preset number of valid sampling points.

[0185] The change factor comparison module is used to compare the average rate of the window with the average rate of the previous window corresponding to the steady state in the settling motion state machine. If the change factor exceeds the set change factor, the settling motion state machine is switched to the transition state and the inertial rejection logic is triggered.

[0186] The false rate determination module is used to extract a first-order difference sequence composed of historical valid sampling points from the historical database and calculate the historical maximum rate to determine whether it is a false rate.

[0187] The degradation processing module is used to, if determined to be a false rate, force the settlement motion state machine to be set to transient state, generate replacement values ​​and discontinuity correction flags to perform degradation processing on the data of multiple consecutive sliding verification windows of the settlement monitoring point.

[0188] The restriction removal module is used to output the arithmetic average rate of the independent sliding verification window to the early warning control logic to remove the downgrade restriction.

[0189] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0190] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0191] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0192] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0193] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0194] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for dynamic settlement control of soft soil culvert bridges based on Internet of Things (IoT) monitoring, characterized in that, Includes the following steps: Construct a settlement motion state machine for each settlement monitoring point, and set the initial state of all settlement monitoring points to a steady state; A sliding verification window for settlement monitoring points is constructed. The local timestamps of the current data point and the previous data point in the database are extracted, and consecutive valid sampling points are selected and filled into the sliding verification window. When the sliding verification window is filled with a preset number of valid sampling points, the average window rate is calculated. The average rate of the window is compared with the average rate of the previous window corresponding to the steady state in the settling motion state machine by a factor of change. If the factor of change exceeds the set factor of change, the settling motion state machine is switched to the transition state and the inertial rejection logic is triggered. Extract the first-order difference sequence of historical valid sampling points from the historical database and calculate the historical maximum rate to determine whether it is a false rate; If the rate is determined to be false, the settlement motion state machine is forcibly set to transient, and replacement values ​​and discontinuity correction flags are generated to perform downgrade processing on the data of multiple consecutive sliding verification windows of the settlement monitoring point. Output the arithmetic average rate of the independent sliding verification window to the early warning and control logic to remove the downgrade restriction.

2. The method for dynamic settlement control of soft soil culvert bridges based on IoT monitoring as described in claim 1, characterized in that, The steps for selecting consecutive valid sampling points to fill the sliding verification window include: Receive current settlement monitoring data points collected by sensors and extract the latest local timestamp information; Access the historical database to query the last local timestamp information of the previous data point that has been stored in the database for the same settlement monitoring point; The actual data reception time interval is obtained by calculating the difference between the latest local timestamp information and the previous local timestamp information. Extract the preset maximum normal sampling time interval parameter and multiply it by the tolerance factor to obtain the maximum allowable discontinuity time threshold; If the actual data reception time interval is determined to be no greater than the maximum discontinuity time threshold, the current data point is marked as a valid sampling point and sequentially filled into the internal storage queue of the sliding verification window; If the actual data reception time interval is determined to be greater than the maximum discontinuity time threshold, the current data point is marked as an invalid sampling point.

3. The method for dynamic settlement control of soft soil culvert bridges based on IoT monitoring as described in claim 1, characterized in that, When the sliding verification window is filled with a preset number of valid sampling points, the steps for calculating the average window rate include: Continuously monitor the cumulative number of valid sampling points in the internal storage queue of the sliding verification window; When the cumulative amount is identified to have precisely reached the preset full load threshold, the initial cumulative settlement value and the initial local timestamp record contained in the first valid sampling point of the time series in the storage queue are located and extracted. Extract the cumulative settlement value and local timestamp record at the end of the last valid sampling point in the time series. Subtract the initial cumulative settlement value from the cumulative settlement value at the end to obtain the absolute cumulative settlement difference within the span; Subtract the initial local timestamp record from the end local timestamp record to obtain the absolute time elapsed difference within the span; The window average rate is generated by dividing the absolute cumulative settlement difference by the absolute time elapsed difference; it is then determined whether the current settlement motion state machine is in a steady state. If the system is in a steady state, extract the average rate of the previous window corresponding to the previous sampling period, and calculate the absolute change factor of the average rate of the window compared to the average rate of the previous window as the basis for control.

4. The method for dynamic settlement control of soft soil culvert bridges based on IoT monitoring as described in claim 1, characterized in that, The steps for extracting a first-order difference sequence of historical valid sampling points from a historical database and calculating the historical maximum rate include: When the comparison reveals that the absolute change factor parameter exceeds the system's set safe floating factor threshold, the settlement motion state machine is updated from steady state to a transitional state flag awaiting verification, and the inertial veto logic for abnormal settlement rate mutations is activated simultaneously. Extract a historical time period of a preset length based on the start time of the current sliding verification window; Scan and extract all historical valid sampling point datasets with valid labels stored within the historical time period and count the total number of data entries; If the total number of data entries is less than the minimum required number, it is determined that there are insufficient historical samples, the inertial rejection logic is terminated, the state machine is brought back to a steady state, and the current window average rate is adopted. If the total number of data entries reaches the minimum required number, sort all extracted historical valid sampling points according to the time series. Calculate the rate of settlement change between adjacent historical valid sampling points, construct a set of first-order difference sequences that reflect the historical settlement fluctuation pattern, and screen out the historical maximum rate extreme value in the set of sequences.

5. The method for dynamic settlement control of soft soil culvert bridges based on Internet of Things monitoring as described in claim 1, characterized in that, The steps for extracting a first-order difference sequence from historical valid sampling points from the historical database, calculating the historical maximum rate, and determining whether it is a false rate include: Obtain the historical maximum rate extreme value that represents the actual settlement peak value in the past period, selected from the historical first-order difference rate sequence set; The scaling factor of the current mutation rate relative to the historical peak limit is calculated. Compare the scaling factor exceeding the multiple with the preset allowable threshold for deviation of physical extreme inertia; If the ratio factor exceeds the allowable threshold, the external environmental load module will be activated to investigate indirect compliance evidence. Send a joint query message to an external traffic database; Obtain actual traffic flow statistics and vehicle load monitoring records for road sections that completely overlap with historical time periods; Determine whether the actual traffic flow statistics record exceeds the preset daily average load ratio limit and retrieve the overweight vehicle passage records; If the actual traffic flow does not exceed the upper limit of the load ratio and there are no overweight vehicles, it is determined that there are no external conditions to cause severe subsidence, and it is classified as a false rate.

6. The method for dynamic settlement control of soft soil culvert bridges based on IoT monitoring as described in claim 1, characterized in that, If the rate is determined to be false, the settling motion state machine is forcibly set to transient state, and the steps for generating replacement values ​​include: The current window average rate value that is determined to be a physically untrusted false rate is intercepted from the transmission path to the security warning execution module. Retrieve the average rate cache record value output by the previous complete sliding verification window of the settlement monitoring point that has passed a complete physical consistency verification cycle; The average rate cache records the values ​​by overwriting them with memory overwrite operations, thus generating a smooth-transition rate replacement value parameter. The generated rate replacement value parameter is only granted read and write permissions to the backend database archive storage module and the trend curve rendering module, and warning and control actions are prohibited from being triggered based on this value. Send a status rewrite command to the global status registry to forcibly change the settlement motion state machine attribute corresponding to the settlement monitoring point from a transitional state to a transient identifier representing short-term data isolation and observation period to block external traffic intervention commands.

7. The method for dynamic settlement control of soft soil culvert bridges based on IoT monitoring as described in claim 1, characterized in that, The steps for generating discontinuity correction flags and performing downgrading processing on data from multiple consecutive sliding verification windows of the settlement monitoring point include: Once the underlying layer successfully executes the rate replacement overwrite operation designed to smooth out false data, an exception event response logging program is automatically triggered. Based on the current time and the unified encoding of the sensor that triggered the action, the discontinuity correction flag with a unique traceability identifier is generated; The discontinuous correction flag is deeply bound to the current sliding time period that caused the anomaly and permanently stored in an independently allocated system audit log library; For the settlement monitoring point carrying the transient marker, a comprehensive data degradation observation process is initiated, with the duration covering multiple subsequent preset independent sliding cycles. During the downgrade process coverage period, the flow of newly calculated window average rate data for each subsequently generated sliding verification window is forcibly restricted to any warning threshold comparison algorithm container.

8. The method for dynamic settlement control of soft soil culvert bridges based on Internet of Things monitoring as described in claim 1, characterized in that, The steps to output the arithmetic average rate of the independent sliding verification window to the early warning control logic to remove the degradation restriction include: Continuously monitor the remaining execution cycle of the data degradation observation process until it is confirmed that the set degradation penalty time has been completely exhausted and the project is closed; Strict requirements were initially imposed that the settlement monitoring point must continuously and successfully construct a preset number of independent sliding verification windows that do not overlap with each other on the time axis; The start and end time span of each independent sliding verification window must be an integer multiple of the original data acquisition time interval; The aforementioned inertial rejection logic, which includes historical extreme value comparison, is re-run for each of the pre-defined number of independent sliding verification windows that are continuously constructed. If and only if the average rate of the new windows generated by each of the independent sliding verification windows in the batch does not trigger the rejection condition of being judged as a physically unreliable false rate; Extract the numerical sequence of the average rate of this batch of continuous safe new windows, and obtain the arithmetic mean rate of this batch of safe rates by summing and dividing by the number of statistical items; The confirmed arithmetic average rate result value will be re-enabled and output to the early warning and control logic.

9. A dynamic settlement control system for soft soil culvert bridges based on Internet of Things (IoT) monitoring, used to implement the dynamic settlement control method for soft soil culvert bridges based on IoT monitoring as described in any one of claims 1-8, characterized in that, include: The module includes a state machine construction module, a verification window filtering module, an average rate calculation module, a change factor comparison module, a false rate determination module, a degradation processing module, and a restriction removal module. The state machine construction module is used to construct the settlement motion state machine for each settlement monitoring point and set the initial state of all settlement monitoring points to a steady state. The verification window filtering module is used to construct a sliding verification window for settlement monitoring points, extract the local timestamps of the current data point and the previous data point already in the database, and filter continuous valid sampling points to fill the sliding verification window. The average rate calculation module is used to calculate the average rate of the window when the sliding verification window is filled with a preset number of valid sampling points. The change factor comparison module is used to compare the average rate of the window with the average rate of the previous window corresponding to the steady state in the settling motion state machine. If the change factor exceeds the set change factor, the settling motion state machine is switched to the transition state and the inertial rejection logic is triggered. The false rate determination module is used to extract a first-order difference sequence composed of historical valid sampling points from the historical database and calculate the historical maximum rate to determine whether it is a false rate. The degradation processing module is used to, if determined to be a false rate, force the settlement motion state machine to be set to transient state, generate replacement values ​​and discontinuity correction flags to perform degradation processing on the data of multiple consecutive sliding verification windows of the settlement monitoring point. The restriction removal module is used to output the arithmetic average rate of the independent sliding verification window to the early warning control logic to remove the downgrade restriction.

Images