Variable cross-section pile whole cycle monitoring management system based on cloud computing

Abstract

This invention relates to the field of pile foundation monitoring technology, specifically a cloud-based full-cycle monitoring and management system for variable cross-section piles, including a data tagging module, a path mapping module, a module loading module, a monitoring and processing module, and a status archiving module. In this invention, by periodically classifying and mapping the monitoring data of variable cross-section piles at different stages, precise matching between engineering cycle types and data processing channels is achieved. Combined with channel identification, corresponding task content is automatically extracted and dynamically loaded and unloaded, forming a processing task sequence that can automatically switch with changes in the engineering cycle. The monitoring and processing stage distinguishes data characteristics such as stress, vibration, and temperature based on task content and performs dedicated analysis, effectively improving the relevance and accuracy of monitoring results at each stage of variable cross-section piles. Combined with multi-dimensional information archiving to the cloud monitoring area, a continuous and unified data closed-loop system is constructed, solving problems such as data interruption, task rigidity, and status assessment deviations in traditional models.

Description

Technical Field

[0001] This invention relates to the field of pile foundation monitoring technology, and in particular to a cloud computing-based full-cycle monitoring and management system for variable cross-section piles. Background Technology

[0002] The field of pile foundation monitoring technology involves the real-time collection, processing, and evaluation of information such as the state, stress, deformation, and construction quality of piles in building foundation structures. This includes monitoring the full-cycle state of piles during construction, operation, and maintenance to ensure structural safety and stability. It encompasses systematic aspects such as sensor deployment and installation, data acquisition and transmission, state analysis methods, and monitoring information management platforms, achieving digital monitoring and intelligent management of pile foundation operation. Among these, the variable cross-section pile full-cycle monitoring and management system refers to a system that collects and manages monitoring information throughout the entire service life of pile foundation structures with different diameters, cross-sectional shapes, or stress characteristic distributions. It primarily focuses on the characteristics of pile cross-section changes. In the early stages of construction, basic instruments such as strain gauges, settlement meters, or tilt sensors are manually installed. During construction, timed recorders are used to collect changes in physical quantities. In the operation phase, data sampling and breakpoint data are used to infer the structural state of the variable cross-section piles.

[0003] Existing technologies for monitoring pile foundation structures mainly rely on manual deployment of sensing equipment and phased sampling records in the early stages of construction. Data collection is intermittent and discontinuous, resulting in a lack of complete coverage of monitoring information over time. The method of inferring structural status through breakpoint data during the operation phase is difficult to ensure assessment accuracy in practical applications. Furthermore, monitoring data from different stages lacks a unified collection standard and a cross-stage data classification and management mechanism. In addition, sensor deployment and monitoring task scheduling rely on manually preset processes, making it impossible to achieve dynamic switching and updating of tasks according to stages. This causes the processing flow to lag behind the evolution of the structural status, ultimately affecting the accurate identification and response efficiency of the safety status of the pile throughout its entire life cycle. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a cloud computing-based full-cycle monitoring and management system for variable cross-section piles.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a cloud computing-based full-cycle monitoring and management system for variable cross-section piles includes: The data tagging module acquires continuous monitoring data of variable cross-section pile structures, matches the corresponding engineering cycle type identifier, and performs batch partitioning and reorganization of the mapping data of different cycle types to obtain cycle monitoring classified and organized data. The path mapping module performs channel mapping based on the engineering cycle type identifier in the cycle monitoring and classification data, synchronously records the corresponding channel name, and generates a set of channels corresponding to the cycle stage. The module loading module extracts the phase processing task list bound to the corresponding channel based on the channel name identified in the channel set corresponding to the cycle phase, loads the phase processing task list entries that have not been loaded by the cloud platform, and cancels the task content in the cloud platform that does not belong to the corresponding task list, thus obtaining the phase processing task sequence. The monitoring and processing module processes the data according to the task content in the phase processing task sequence, binds it to the project number of the original data source, and obtains the phase task structure monitoring results. Based on the monitoring results of the stage task structure, the status archiving module combines and collects the channel type, pile segment location, acquisition time and project number under the current task sequence, and archives them to the cloud full-cycle monitoring storage area to obtain the full-cycle monitoring management record of variable cross-section piles.

[0006] As a further aspect of the present invention, the project cycle type identifier includes the construction phase, operation phase, and maintenance phase.

[0007] As a further aspect of the present invention, the channel mapping specifically refers to mapping the construction phase to a continuous flow processing channel, the operation phase to a fixed-point extraction processing channel, and the maintenance phase to an anomaly screening processing channel. Specifically, the corresponding data processing refers to performing stress peak interval analysis in the continuous flow processing channel, performing vibration period trend clustering in the fixed-point extraction processing channel, and performing temperature drift distribution extraction in the anomaly screening processing channel.

[0008] As a further embodiment of the present invention, the periodic monitoring and classification data includes data label identification, engineering period classification results, and monitoring data reorganization structure; the periodic stage corresponding channel set includes stage type channel name mapping relationship, engineering period processing path set, and channel record index; the stage processing task sequence includes task name list, channel task mapping relationship, and activated task filter set; the structural monitoring results include stress peak interval results, vibration trend clustering results, and temperature drift distribution characteristics; and the variable cross-section pile full-cycle monitoring management record includes channel type identification, pile segment location index, acquisition timestamp, and project number label.

[0009] As a further aspect of the present invention, the data tag module includes: The data receiving submodule acquires continuous monitoring data generated by stress sensors, temperature sensors, current transmitters and vibration recorders on the variable cross-section pile structure. Based on the communication formats of various sensors, it performs channel identification and data standardization processing, integrates sampling timestamps and measurement numerical information, constructs a unified data recording unit, and generates a multi-source continuous monitoring data sequence by arranging the recording units in chronological order. The identifier matching submodule obtains the progress nodes and engineering cycle configuration list set in the project management platform based on the multi-source continuous monitoring data sequence, and performs interval discrimination on the monitoring data and the start and end times of the construction stage, operation stage and maintenance stage according to the timestamp, marks the engineering cycle type identifier to which it belongs, and generates a cycle attribution identifier data sequence. The classification and sorting submodule divides the record units into three subsequences—construction period, operation period, and maintenance period—based on the cycle attribution identifier data sequence and the engineering cycle type identifier. It then performs a reordering operation based on the internal channel and time field order of each subsequence to generate cycle monitoring classification and sorting data.

[0010] As a further aspect of the present invention, the path mapping module includes: The type identification submodule extracts the cycle label from each data record based on the engineering cycle type identifier in the cycle monitoring and classification data, establishes a label index in chronological order, and classifies and labels all records according to engineering cycle to generate an engineering cycle label index table. The channel retrieval submodule obtains the fixed matching standards for the three stages of construction, operation and maintenance set in the channel mapping relationship according to the project cycle tag index table, performs corresponding relationship filtering, matches the corresponding channel category for each cycle tag, and generates a list of channels corresponding to the cycle tag. The channel mapping submodule maps construction phase labels to continuous flow processing channels, operation phase labels to fixed-point extraction processing channels, and maintenance phase labels to anomaly screening processing channels based on the channel list corresponding to the cycle labels. It records the channel number and the corresponding structure of the label and establishes a set of channels corresponding to the cycle phase.

[0011] As a further aspect of the present invention, the module loading module includes: The channel task extraction submodule retrieves the stage processing task entries bound to each channel from the cloud platform configuration directory based on the channel names identified in the channel set corresponding to the periodic stage, deduplicates and integrates duplicate task entries, and generates a channel-bound task list table. The task activation comparison submodule obtains the task entries that are currently active in the cloud platform, filters the entries that do not appear in the active status table in the channel-bound task list table, establishes a new loading task list, and generates a sequence of task identifiers to be loaded. The task sequence adjustment submodule combines the task identifier sequence to be loaded with the channel-bound task list table, performs a loading operation based on the task identifier sequence to be loaded, loads missing task entries according to the binding order, and simultaneously cancels the task content that has been activated in the cloud platform but is not in the channel-bound task list table, and obtains the phase processing task sequence.

[0012] As a further aspect of the present invention, the monitoring and processing module includes: The stress interval extraction submodule identifies the corresponding task item of the continuous flow processing channel based on the task content in the stage processing task sequence, obtains the stress data corresponding to the construction stage in the periodic monitoring classification and sorting data, retrieves the stress peak position based on the timestamp sequence, calculates the time difference between adjacent peaks, and constructs an index structure consistent with the channel structure to obtain the stress peak interval distribution sequence. The vibration trend clustering submodule identifies the task content in the fixed-point extraction processing channel based on the stress peak interval distribution sequence and the stage processing task sequence, obtains the vibration data corresponding to the operation stage, and performs clustering according to the vibration amplitude sequence within the fixed sampling point aggregation period based on the amplitude fluctuation gradient value to obtain the vibration period clustering structure vector. The temperature drift detection submodule identifies the anomaly screening and processing channel task entries based on the vibration period clustering structure vector and the stage processing task sequence, obtains the maintenance stage temperature data, calculates the maximum drift amplitude in each sampling period and binds it to the project number in the original data, and establishes the stage task structure monitoring results.

[0013] As a further aspect of the present invention, the status archiving module includes: Based on the monitoring results of the stage task structure, the task field integration submodule extracts the channel type, pile location, collection time and project number from each data record, combines and aggregates them according to the task number, and sorts them to obtain the task attribution field aggregation table. The structure index construction submodule reads the task attribution field aggregation table, performs partitioning according to channel type, constructs a unique identifier based on pile location and collection time, binds it to each attribution data unit, and establishes an archive index coding matrix. The archive record generation submodule writes the belonging data units sequentially into the cloud full-cycle monitoring storage area according to the project number based on the archive index coding matrix, confirms the archive of the written entries and binds them with the channel identifier to obtain the full-cycle monitoring management record of the variable cross-section pile.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by periodically classifying and mapping the monitoring data of variable cross-section piles at different stages, the precise matching of engineering cycle types and data processing channels is achieved. Combined with channel identification, the corresponding task content is automatically extracted and dynamically loaded and unloaded, forming a processing task sequence that can be automatically switched according to the engineering cycle. The monitoring and processing link distinguishes data characteristics such as stress, vibration, and temperature according to the task content and performs dedicated analysis, effectively improving the pertinence and accuracy of monitoring results at each stage of variable cross-section piles. Combined with multi-dimensional information such as time, location, and number, the data is archived to the cloud monitoring area, constructing a continuous and unified data closed-loop system, solving problems such as data interruption, task rigidity, and state assessment deviation in the traditional mode. Attached Figure Description

[0015] Figure 1 This is a system flowchart of the present invention; Figure 2 This is a flowchart of the data tag module of the present invention; Figure 3 This is a flowchart of the path mapping module of the present invention; Figure 4 This is a flowchart of the module loading module of the present invention; Figure 5 This is a flowchart of the monitoring and processing module of the present invention; Figure 6 This is a flowchart of the status archiving module of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0018] Please see Figure 1 The cloud-based full-cycle monitoring and management system for variable cross-section piles includes: The data tagging module acquires continuous monitoring data generated by stress sensors, temperature sensors, current transmitters and vibration recorders on the variable cross-section pile structure. Based on the progress stages set in the project management platform, it matches the corresponding engineering cycle type identifiers, including construction stage, operation stage and maintenance stage. The data mapped to different cycle types are batch partitioned and reorganized to obtain cycle monitoring classification and sorting data. The path mapping module, based on the engineering cycle type identifier in the cycle monitoring classification data, maps the construction stage to the continuous flow processing channel, the operation stage to the fixed-point extraction processing channel, and the maintenance stage to the anomaly screening processing channel according to the fixed matching standard set in the channel mapping relationship, and records the corresponding channel name to generate the corresponding channel set for the cycle stage. The module loading module extracts the list of phase processing tasks bound to the mapped channel from the cloud platform configuration directory based on the channel name identified in the channel set corresponding to the period phase. It matches the list of tasks that are already in the active state, loads the phase processing task list entries that have not been loaded by the cloud platform, and cancels the task content in the cloud platform that does not belong to the corresponding task list, thus obtaining the phase processing task sequence. The monitoring and processing module processes the received classified monitoring data according to the task content in the phase processing task sequence. It performs stress peak interval analysis in the continuous flow processing channel, vibration period trend clustering in the fixed point extraction processing channel, and temperature drift distribution extraction in the anomaly screening processing channel. The data is then bound to the project number of the original data source to obtain the phase task structure monitoring results. Based on the phased task structure monitoring results, the status archiving module combines and collects the channel type, pile segment location, acquisition time and project number under the current task sequence, and archives them to the cloud full-cycle monitoring storage area to obtain the full-cycle monitoring management record of variable cross-section piles.

[0019] The periodic monitoring and classification data includes data labels, engineering cycle classification results, and restructured monitoring data. The corresponding channel set for each cycle stage includes the mapping relationship between stage type and channel name, the engineering cycle processing path set, and the channel record index. The stage processing task sequence includes the task name list, the channel task mapping relationship, and the activated task filter set. The structural monitoring results include the stress peak interval results, vibration trend clustering results, and temperature drift distribution characteristics. The full-cycle monitoring and management records for variable cross-section piles include channel type identifiers, pile segment location indexes, acquisition timestamps, and project number annotations.

[0020] Please see Figure 2 The data tagging module includes: The data receiving submodule acquires continuous monitoring data generated by stress sensors, temperature sensors, current transmitters and vibration recorders on the variable cross-section pile structure. Based on the communication formats of various sensors, it performs channel identification and data standardization processing, integrates sampling timestamps and measurement numerical information, constructs a unified data recording unit, and generates a multi-source continuous monitoring data sequence by arranging the recording units in chronological order. The data receiving submodule first initializes the communication interfaces with various sensors on key nodes of the variable cross-section pile structure (especially the interface between the expanded and contracted diameter sections of the pile). This interface is configured with an industrial-grade RS485 serial communication protocol, a baud rate of 9600 bits per second, 8 data bits, 1 stop bit, and no parity bit. Through this interface, the data receiving submodule captures the raw hexadecimal byte stream data sent by the sensors in real time. This data stream includes stress concentration values ​​at abrupt changes in cross-section sensed by stress sensors deployed at the variable cross-section junctions, deep hydration heat and ambient temperature values ​​collected by temperature sensors, analog current signals output by current transmitters, and weak vibration response signals of the pile at different cross-sections recorded by vibration recorders. The data receiving submodule incorporates multi-protocol parsing logic and pre-sets specific frame header identifiers and frame tail verification rules for each type of sensor. For example, when a data packet with a header of 0xAA and a length of 32 bytes is received, the module identifies it as stress data reflecting the variable cross-section bearing condition; when a data packet with a header of 0xBB and a length of 16 bytes is received, it is identified as temperature data. The data receiving submodule reads the payload segment from the data packet and converts the hexadecimal value to a decimal physical quantity value strictly according to the big-endian format, completing the data standardization process. During this process, the data receiving submodule calls a high-precision clock to obtain the precise time of arrival of the data packet. The time format adopts a UNIX timestamp, accurate to the millisecond level, and binds the timestamp with the parsed physical quantity value to construct a unified structured data record unit containing four core fields: "unique sensor number," "physical quantity type," "measured value," and "acquisition timestamp." The data receiving submodule sets up a temporary buffer queue with a capacity of 1024 records. Whenever the buffer queue is full or reaches a preset 1-second time threshold (whichever arrives first), the module triggers the sorting logic. It uses a fast sorting algorithm, with the acquisition timestamp as the first key and the sensor's unique number as the second key, to sort the record units in ascending order and generate a multi-source continuous monitoring data sequence.

[0021] The identifier matching submodule is based on multi-source continuous monitoring data sequence, obtains the progress node and engineering cycle configuration list set in the project management platform, and performs interval discrimination between the monitoring data and the start and end time of the construction phase, operation phase and maintenance phase according to the timestamp, marks the engineering cycle type identifier to which it belongs, and generates cycle attribution identifier data sequence; The identifier matching submodule receives the aforementioned multi-source continuous monitoring data sequence and simultaneously initiates a query request to the project management platform via an encrypted HTTPS interface to obtain the full lifecycle schedule configuration list for the variable cross-section pile project. This list details the specific start and end times for pile foundation drilling and grouting (construction phase), load-bearing operation (operation phase), and structural reinforcement (maintenance phase). The identifier matching submodule iterates through each record in the multi-source continuous monitoring data sequence, extracts its collection timestamp field, and uses a binary search algorithm to search and compare within the time interval of the schedule configuration list. Specifically, this module compares the recorded timestamp with the start and end times of each phase. If the timestamp of a record is greater than or equal to the start time of the construction phase and less than the end time of the construction phase, the module writes a "Construction Phase" type identifier code into the metadata field of that record, indicating that this phase focuses on the concrete forming quality at the variable cross-section. If the timestamp falls within the time interval of the operation phase, it writes an "Operation Phase" identifier code, indicating that this phase focuses on the fatigue characteristics at the changes in pile cross-section. If the timestamp belongs to the maintenance phase, it writes a "Maintenance Phase" identifier code. For abnormal timestamp data that fails to match any known phase, this module marks it as "pending" and stores it separately in the anomaly log. Through this full traversal and comparison operation, the identification matching submodule completes the period attribution definition of all raw data, generating a period attribution identification data sequence with clear periodic attributes.

[0022] The classification and sorting submodule divides the record units into three subsequences—construction period, operation period, and maintenance period—based on the cycle-attribution identifier data sequence and the engineering cycle type identifier. It then performs a re-sorting operation based on the internal channel and time field order of each subsequence to generate cycle monitoring classification and sorting data. The classification and sorting submodule reads the cycle-assigned identifier data sequence and initializes three independent dynamic array containers, named the construction period sequence container, operation period sequence container, and maintenance period sequence container, respectively. It iterates through the input sequence and distributes each record to the corresponding container based on the engineering cycle type identifier. For example, records identified as "construction period" are moved to the construction period sequence container, and records identified as "operation period" are moved to the operation period sequence container. After distribution, the classification and sorting submodule performs a secondary reordering operation on the data within each container. This operation first pre-sorts the data according to the sensor channel number of the strain section location from smallest to largest, ensuring that the data for the same pile section is stored continuously in physical storage; then, within the same channel, it again sorts the data in ascending order according to the acquisition timestamp, ensuring the temporal continuity of the data. After this series of distribution and reordering operations, the classification and sorting submodule finally outputs structured and ordered cycle monitoring classification and sorting data, as shown in Table 1, which displays some examples of the processed monitoring data structure.

[0023] Table 1. Classification and Organization Structure of Monitoring Data 1001 1704067200 stress sensor 25.5 MPa Construction period Variable cross-section expansion section 1002 1704067205 Vibration sensor 0.02 mm Operational period Reduced diameter section of the upper part of the pile 1003 1704067210 Temperature sensor 18.2 degrees Celsius Maintenance period Pile bottom bearing surface As shown in Table 1, different types of raw data have been successfully transformed into a unified format and given clear period identifiers, providing a standardized data foundation for subsequent path mapping for different life cycles of variable cross-section piles.

[0024] Please see Figure 3 The path mapping module includes: The type identification submodule extracts the cycle label from each data record based on the engineering cycle type identifier in the cycle monitoring and classification data, establishes a label index in chronological order, and classifies and labels all records according to engineering cycle to generate an engineering cycle label index table. The type identification submodule reads and categorizes the periodic monitoring data and initiates the hash index construction program. This module traverses each record unit in the data, extracts the engineering period type identifier as the hash key, uses the physical address of the record in memory as the hash value, and inserts it into a pre-built hash table. To improve retrieval efficiency, chaining is used to handle hash collisions, ensuring that all record addresses with the same period identifier are completely linked in the same hash bucket. After traversal, the type identification submodule traverses the hash table, extracts all unique period tags (i.e., construction period, operation period, maintenance period), and counts the total number of data records and the time span under each tag. Based on the time series order, this module organizes these tags and their metadata into a B+ tree index structure, generating an engineering period tag index table. In this index table, the root node stores the period type classification key value, and the leaf nodes point to the specific data record block address, thus enabling rapid location and access to massive amounts of variable cross-section pile monitoring data.

[0025] The channel retrieval submodule obtains the fixed matching standards for the three stages of construction, operation and maintenance set in the channel mapping relationship based on the project cycle tag index table, performs corresponding relationship filtering, matches the corresponding channel category for each cycle tag, and generates a list of channels corresponding to the cycle tag. The channel retrieval submodule parses the engineering cycle label index table based on a pre-defined mapping rule base. This rule base is fixed during the initialization phase, explicitly defining the channel processing strategies for different physical characteristic stages of variable cross-section piles. The channel retrieval submodule reads configuration items from the rule base, such as "construction stage (requiring real-time monitoring of concrete pouring pressure at variable cross-sections) matches the continuous flow processing channel," "operation stage (requiring long-term structural fatigue analysis) matches the fixed-point extraction processing channel," and "maintenance stage (requiring capture of crack propagation anomalies) matches the anomaly screening processing channel." A string matching algorithm is used to compare the cycle labels in the index table with the stage names in the rule base. When a match is successful, the module extracts the corresponding channel category code from the rule base and establishes a temporary association object with the cycle label. For example, for the "construction period" label, the module finds that its corresponding standard processing mode is high-frequency continuous monitoring, therefore matching it to the "continuous flow processing channel" category. Through the sequential retrieval and matching of all cycle labels, the module generates a detailed list of channels corresponding to each cycle label, clearly indicating which type of processing logic the data for each time stage should flow to.

[0026] The channel mapping submodule maps construction phase labels to continuous flow processing channels, operation phase labels to fixed-point extraction processing channels, and maintenance phase labels to anomaly screening processing channels based on the channel list corresponding to the cycle labels. It records the channel number and the corresponding structure of the label and establishes a set of channels corresponding to the cycle phase. The channel mapping submodule performs specific logical address mapping operations based on the channel list corresponding to the periodic labels. It maintains a global channel resource pool containing the unique IDs and status information of all currently available data processing channels. The submodule iterates through each item in the list, assigning specific continuous flow processing channel IDs (e.g., channel ID101) to the "Construction Phase" label to ensure no data loss during the variable cross-section forming process; assigning specific fixed-point extraction processing channel IDs (e.g., channel ID202) to the "Operation Phase" label to accommodate long-term structural health monitoring; and assigning specific anomaly screening processing channel IDs (e.g., channel ID303) to the "Maintenance Phase" label to sensitively detect potential damage at abrupt changes in pile cross-section. During the allocation process, the channel load status is checked to ensure that the allocated channels are idle or reusable. After allocation, a key-value mapping structure is constructed, where the key is the periodic label and the value is the channel ID sequence. This structure is serialized and stored, establishing the channel set corresponding to the periodic phase. This set establishes a data flow routing table, ensuring that subsequent modules can accurately load the corresponding processing tasks based on this routing table.

[0027] Please see Figure 4 The module loading module includes: The channel task extraction submodule retrieves the stage processing task entries bound to each channel from the cloud platform configuration directory based on the channel name identified in the channel set corresponding to the period stage, removes duplicate task entries, and integrates them to generate a channel-bound task list. The channel task extraction submodule parses the channel set corresponding to the cycle stage and extracts a list of all activated channel names. For each channel name in the list, it sends an HTTP GET request to the cloud platform configuration center to read the JSON configuration file stored in the cloud. This configuration file details the specific processing task entries bound to each channel, which are specifically designed for the characteristics of variable cross-section piles. For example, the "continuous flow processing channel" may be bound to "variable cross-section pressure wave filtering task," "peak extraction task," and "storage task." The channel task extraction submodule parses the JSON response packet and stores the parsed task entry names into a temporary HashSet collection. Utilizing the data structure characteristics of HashSet, duplicate task entries caused by configuration redundancy or multiple requests are automatically removed, ensuring that each task appears only once in the list. After traversal and deduplication, a channel-bound task list table is generated, which lists all calculation and processing tasks required to be executed in the current project cycle without duplicates.

[0028] The task activation comparison submodule obtains the task entries that are currently active in the cloud platform, filters the entries in the channel binding task list that do not appear in the active status table, establishes a list of newly added loading tasks, and generates a sequence of task identifiers to be loaded. The task activation comparison submodule queries the kernel interface to retrieve a list of all processes and threads in the current runtime environment that are in a "running" or "suspended" state, obtains the IDs and names of currently activated task instances, and generates an activation status table. Subsequently, it executes the set difference operation logic. Specifically, it uses the channel-bound task list table as the minuend and the activation status table as the subtrahend, calculating the difference between the two. The elements in this difference set are the task entries that need to be executed but have not yet been loaded. For example, if the list table contains tasks A (pile integrity analysis), B (stress wave velocity calculation), and C (section impedance analysis), while the activation status table only contains tasks A and B, then the difference operation result is task C. The task activation comparison submodule marks these difference set elements as "to be loaded" and sorts them by priority, generating a sequence of task identifiers to be loaded.

[0029] The task sequence adjustment submodule combines the task identifier sequence to be loaded with the channel-bound task list table, performs the loading operation according to the task identifier sequence to be loaded, loads the missing task entries according to the binding order, and cancels the task content that has been activated in the cloud platform but is not in the channel-bound task list table, and obtains the phase processing task sequence. The task sequence adjustment submodule reads the sequence of task identifiers to be loaded, calls the dynamic link library loader, and loads the corresponding algorithm components or code blocks from disk or network storage into memory according to the task identifiers in the sequence. It allocates independent memory space and execution threads for each newly loaded task, completing the task loading operation. Simultaneously, it performs a set inverse difference operation again to identify task entries that exist in the active state table but are not in the channel-bound task list table (i.e., tasks no longer needed in the current stage). For these tasks, a termination signal is sent to safely release their occupied memory resources and handles, completing the unregistration operation. Finally, all task entries confirmed to be active and matching the current stage are chained together according to the logical order of data flow processing (e.g., filtering first, then feature extraction) to obtain the stage processing task sequence.

[0030] Table 2 Example of Comparison between Task Loading and Activation Status Pressure wave filtering Continuous flow channel Activated Variable cross-section reflected wave processing Keep high Peak Extraction Continuous flow channel Not activated Cross-section stress concentration point monitoring load middle Spectrum Analysis Operation Channel Activated Pile body natural frequency analysis Cancellation Low As shown in Table 2, the lifecycle of the task was dynamically adjusted based on the comparison results to ensure that resources are used only for the currently necessary processing steps, especially prioritizing critical tasks such as variable cross-section reflected wave processing.

[0031] Please see Figure 5 The monitoring and processing module includes: The stress interval extraction submodule identifies the corresponding task items of the continuous flow processing channel based on the task content in the stage processing task sequence, obtains the stress data corresponding to the construction stage in the periodic monitoring classification and sorting data, retrieves the stress peak position based on the timestamp sequence, calculates the time difference between adjacent peaks, and constructs an index structure consistent with the channel structure to obtain the stress peak interval distribution sequence. The stress interval extraction submodule first identifies the task node marked "continuous stream processing" in the phase processing task sequence and subscribes to the stress monitoring data stream of the construction phase from the data bus. This data stream mainly reflects the stress state at the locations of abrupt changes in cross-section during the grouting or driving of variable cross-section piles. A sliding window algorithm is used to scan the stress data in real time, with the window size set to 50 sampling points. Within each window, the local maximum value of the stress value is found and marked as a potential peak. To avoid interference from construction noise and wave impedance reflection at the variable cross-section, a peak discrimination threshold is set. The threshold is obtained as follows: the first 1000 stress sampling points from the initial historical phase are selected, and their arithmetic mean and standard deviation are calculated. For example, if the average stress value of the first 1000 sampling points is 10 MPa and the standard deviation is 2 MPa, a weighted summation operation is performed, adding the average stress value to three times the standard deviation, resulting in a peak discrimination threshold of 16 MPa. This 16 MPa threshold setting aims to filter out normal fluctuations and only capture abnormally high stresses that may lead to concrete cracking at the variable cross-section. If the current local maximum value exceeds 16 MPa, it is confirmed as an effective stress peak. The stress interval extraction submodule records the precise timestamp corresponding to each effective peak and calculates the time difference between two adjacent peaks. For example, if the first peak occurs at 10 seconds and the second peak occurs at 25 seconds, the difference calculation result is 15 seconds. These differences are stored sequentially in a doubly linked list structure, with the index key being the channel ID and the value being the difference sequence, thus obtaining the stress peak interval distribution sequence.

[0032] The vibration trend clustering submodule identifies the task content in the fixed-point extraction processing channel based on the stress peak interval distribution sequence and the stage processing task sequence, obtains the vibration data corresponding to the operation stage, and clusters the vibration amplitude sequence within the aggregation period according to the amplitude fluctuation gradient value based on the fixed sampling point to obtain the vibration period clustering structure vector. The vibration trend clustering submodule activates the task logic of the fixed-point extraction processing channel to acquire vibration monitoring data during the operation phase, aiming to analyze the dynamic response characteristics of variable cross-section piles under long-term loads. The sampling aggregation period is set to 60 seconds, with 3000 vibration amplitude data points collected in each period. The fluctuation gradient value of the amplitude sequence within this period is calculated as follows: First, the arithmetic mean of the amplitudes of the last 100 sampling points and the arithmetic mean of the amplitudes of the first 100 sampling points are obtained; second, the average of the last 100 points is subtracted from the average of the first 100 points to obtain the amplitude change; finally, this change is divided by the sampling time interval. For example, if the average amplitude of the last 100 points is 0.8 mm, the average amplitude of the first 100 points is 0.2 mm, and the time interval is 60 seconds, then 0.8 minus 0.2 equals 0.6 mm, and 0.6 is divided by 60 to calculate a fluctuation gradient value of 0.01 mm / s. This gradient value reflects the dissipation or accumulation trend of pile vibration energy. Subsequently, the calculated gradient values ​​were classified using the K-Means clustering algorithm. The preset K value was 3, representing three states: "stable," "slight fluctuations (minor fatigue at the cross-section)," and "severe fluctuations (potential structural loosening at the cross-section)." The algorithm initialized three cluster center vectors, calculated the Euclidean distance between each gradient value and the cluster center, assigned it to the nearest category, and iteratively updated the cluster centers until convergence. Finally, the algorithm outputs the category identifier vector for each sampling period, obtaining the vibration period clustering structure vector.

[0033] The temperature drift detection submodule identifies the anomaly screening and processing channel task items based on the vibration period clustering structure vector and the stage processing task sequence, obtains the temperature data of the maintenance stage, calculates the maximum drift amplitude in each sampling period and binds it to the project number in the original data, and establishes the stage task structure monitoring results. The temperature drift detection submodule targets the anomaly screening and processing channel, acquiring temperature data sequences during the maintenance phase. It focuses on monitoring the risk of crack propagation within the variable cross-section pile body due to residual heat of hydration or environmental temperature differences. A 24-hour drift detection window is used. Within each window, the temperature data is traversed, identifying the maximum and minimum temperature values, and the difference between them is calculated as the maximum drift amplitude. Simultaneously, to exclude normal temperature rise caused by solar radiation, a baseline drift parameter is introduced. This parameter is set by selecting environmental meteorological data from the same period over the past year for the project site and calculating the average daily temperature difference. For example, if historical data shows an average daily temperature difference of 10 degrees Celsius for the season, then 10 degrees Celsius is set as the baseline drift parameter. The temperature drift detection submodule compares the measured maximum drift amplitude with the baseline drift parameter. If the measured drift amplitude exceeds 1.2 times the baseline parameter (i.e., 10 multiplied by 1.2 equals 12 degrees Celsius), it is determined to be an abnormal drift, which may indicate abnormal thermal expansion and contraction effects or crack leakage at the variable cross-section joint. The calculated drift amplitude value and anomaly judgment result are strongly bound to the project number in the original data to construct a stage task structure monitoring result containing a quadruple of "project ID-time window-drift value-state judgment".

[0034] Please see Figure 6 The status archiving module includes: The task field integration submodule extracts the channel type, pile location, collection time and project number from each data record based on the phase task structure monitoring results, combines and aggregates them according to the task number, and sorts them to obtain the task attribution field aggregation table. The task field integration submodule receives the monitoring results of the phased task structure and initiates the field extraction parser. This parser traverses each data object in the result set, using reflection or key-value indexing to precisely extract four key metadata fields: "channel type," "pile segment location (clearly identified as expanded diameter segment, reduced diameter segment, or straight pile segment)," "collection time," and "project number." After extraction, the data is grouped and aggregated based on the task number. A hash mapping table with the task number as the key is maintained, grouping all records belonging to the same task number into the same list structure. Subsequently, within each list, a multi-level sorting algorithm is executed. First, the data is sorted in ascending order according to the ASCII code of the project number. If the project numbers are the same, a second-level sorting is performed according to the chronological order of the collection time. Through this process, the disorganized monitoring results are reorganized into a logically clear and hierarchically structured task attribution field aggregation table, ensuring the systematic nature of the variable cross-section pile's full lifecycle data.

[0035] The structure index construction submodule reads the task attribution field aggregation table, performs partitioning based on channel type, constructs a unique identifier based on pile location and collection time, binds it to each attribution data unit, and establishes an archive index coding matrix. The structure index construction submodule reads the task attribution field aggregation table. First, it divides the data table into different logical partitions based on the "channel type" field (e.g., continuous flow, fixed-point extraction, anomaly screening). For each record within each partition, a unique identifier generation logic is executed. This generation logic combines string concatenation and hash operations: the "pile location" string (e.g., "Pile-A-VariableSection-1") is concatenated with the "collection time" string (e.g., "20240101120000") to obtain the basic string "Pile-A-VariableSection-120240101120000". Then, the MD5 digest algorithm is called to operate on this basic string, generating a 128-bit unique hash value, which is then converted into a 32-bit hexadecimal string as the unique identifier. This unique identifier is added as a new field and directly bound to the corresponding attribution data unit. A two-dimensional matrix structure is constructed, with the row index being the project number, the column index being the channel type, and the matrix elements being the unique identifier, thus establishing the archive index encoding matrix.

[0036] The archive record generation submodule writes the data units belonging to the archive index coding matrix into the cloud full-cycle monitoring storage area in sequence according to the project number, confirms the archive of the written entries and binds them with the channel identifier to obtain the full-cycle monitoring management record of the variable cross-section pile. The archive record generation submodule establishes an encrypted transmission channel with the cloud-based full-cycle monitoring storage area based on the archive index encoding matrix. Following the project number order recorded in the matrix, it serializes the assigned data units into JSON or Binary Large Object (BLOB) format and writes them to the specified directory of the cloud storage bucket via HTTPPUT or FTP upload commands. For each successfully written data entry, the cloud storage returns an ACK signal and a storage address pointer. The archive record generation submodule receives this ACK signal, binds it to the local channel identifier, and updates the archive status field of the record to "archived" in the local database. If the write fails, an exponential backoff retry mechanism is executed, retrying a maximum of 3 times. After all data is written and confirmed to be correct, a comprehensive full-cycle monitoring management record for variable cross-section piles, including storage path, total data volume, and verification hash value, is generated, as shown in Table 3.

[0037] Table 3. Example of Full-Cycle Monitoring Archive Records a1b2c3d4... PROJ-001 Continuous flow 2024-01-0810:00 / cloud / data / s / VS-node1 e5f6g7h8... PROJ-001 Targeted sampling 2024-01-0810:05 / cloud / data / v / VS-node2 i9j0k1l2... PROJ-002 Mutation screening 2024-01-0810:10 / cloud / data / t / VS-node3 As shown in Table 3, the final generated records provide traceable index information, marking the completion of the closed-loop monitoring data processing flow for each key node of the variable cross-section pile.

[0038] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A variable cross-section pile full-cycle monitoring management system based on cloud computing, characterized in that, include: The data tagging module acquires continuous monitoring data of variable cross-section pile structures, matches the corresponding engineering cycle type identifier, and performs batch partitioning and reorganization of the mapping data of different cycle types to obtain cycle monitoring classified and organized data. The path mapping module performs channel mapping based on the engineering cycle type identifier in the cycle monitoring and classification data, synchronously records the corresponding channel name, and generates a set of channels corresponding to the cycle stage. The engineering cycle type identifier includes construction stage, operation stage, and maintenance stage. The channel mapping specifically refers to mapping the construction stage to the continuous flow processing channel, the operation stage to the fixed-point extraction processing channel, and the maintenance stage to the anomaly screening processing channel. The module loading module extracts the phase processing task list bound to the corresponding channel based on the channel name identified in the channel set corresponding to the cycle phase, loads the phase processing task list entries that have not been loaded by the cloud platform, and cancels the task content in the cloud platform that does not belong to the corresponding task list, thus obtaining the phase processing task sequence. The monitoring and processing module processes the data according to the task content in the phase processing task sequence, binds it to the project number of the original data source, and obtains the phase task structure monitoring results. The monitoring and processing module includes: The stress interval extraction submodule identifies the corresponding task item of the continuous flow processing channel based on the task content in the stage processing task sequence, obtains the stress data corresponding to the construction stage in the periodic monitoring classification and sorting data, retrieves the stress peak position based on the timestamp sequence, calculates the time difference between adjacent peaks, and constructs an index structure consistent with the channel structure to obtain the stress peak interval distribution sequence. The vibration trend clustering submodule identifies the task content in the fixed-point extraction processing channel based on the stress peak interval distribution sequence and the stage processing task sequence, obtains the vibration data corresponding to the operation stage, and performs clustering according to the vibration amplitude sequence within the fixed sampling point aggregation period based on the amplitude fluctuation gradient value to obtain the vibration period clustering structure vector. The temperature drift detection submodule identifies the anomaly screening and processing channel task entries based on the vibration period clustering structure vector and the stage processing task sequence, obtains the temperature data of the maintenance stage, calculates the maximum drift amplitude in each sampling period and binds the project number in the original data, and establishes the stage task structure monitoring results. Based on the monitoring results of the stage task structure, the status archiving module combines and collects the channel type, pile segment location, acquisition time and project number under the current task sequence, and archives them to the cloud full-cycle monitoring storage area to obtain the full-cycle monitoring management record of variable cross-section piles.

2. The cloud computing-based full-cycle monitoring and management system for variable cross-section piles according to claim 1, characterized in that: Specifically, the corresponding data processing refers to performing stress peak interval analysis in the continuous flow processing channel, performing vibration period trend clustering in the fixed-point extraction processing channel, and performing temperature drift distribution extraction in the anomaly screening processing channel.

3. The cloud computing-based full-cycle monitoring and management system for variable cross-section piles according to claim 1, characterized in that: The periodic monitoring and classification data includes data tags, engineering cycle classification results, and monitoring data reorganization structure. The channel set corresponding to each cycle stage includes a stage type channel name mapping relationship, an engineering cycle processing path set, and a channel record index. The stage processing task sequence includes a task name list, a channel task mapping relationship, and an activation task filter set. The stage task structure monitoring results include stress peak interval results, vibration trend clustering results, and temperature drift distribution characteristics. The full-cycle monitoring and management record of the variable cross-section pile includes channel type identifiers, pile segment location indexes, acquisition timestamps, and project number annotations.

4. The cloud computing-based full-cycle monitoring and management system for variable cross-section piles according to claim 1, characterized in that, The data tag module includes: The data receiving submodule acquires continuous monitoring data generated by stress sensors, temperature sensors, current transmitters and vibration recorders on the variable cross-section pile structure. Based on the communication formats of various sensors, it performs channel identification and data standardization processing, integrates sampling timestamps and measurement numerical information, constructs a unified data recording unit, and generates a multi-source continuous monitoring data sequence by arranging the recording units in chronological order. The identifier matching submodule obtains the progress nodes and engineering cycle configuration list set in the project management platform based on the multi-source continuous monitoring data sequence, and performs interval discrimination on the monitoring data and the start and end times of the construction stage, operation stage and maintenance stage according to the timestamp, marks the engineering cycle type identifier to which it belongs, and generates a cycle attribution identifier data sequence. The classification and sorting submodule divides the record units into three subsequences—construction period, operation period, and maintenance period—based on the cycle attribution identifier data sequence and the engineering cycle type identifier. It then performs a reordering operation based on the internal channel and time field order of each subsequence to generate cycle monitoring classification and sorting data.

5. The cloud computing-based full-cycle monitoring and management system for variable cross-section piles according to claim 1, characterized in that, The path mapping module includes: The type identification submodule extracts the cycle label from each data record based on the engineering cycle type identifier in the cycle monitoring and classification data, establishes a label index in chronological order, and classifies and labels all records according to engineering cycle to generate an engineering cycle label index table. The channel retrieval submodule obtains the fixed matching standards for the three stages of construction, operation and maintenance set in the channel mapping relationship according to the project cycle tag index table, performs corresponding relationship filtering, matches the corresponding channel category for each cycle tag, and generates a list of channels corresponding to the cycle tag. The channel mapping submodule maps construction phase labels to continuous flow processing channels, operation phase labels to fixed-point extraction processing channels, and maintenance phase labels to anomaly screening processing channels based on the channel list corresponding to the cycle labels. It records the channel number and the corresponding structure of the label and establishes a set of channels corresponding to the cycle phase.

6. The cloud computing-based full-cycle monitoring and management system for variable cross-section piles according to claim 1, characterized in that, The module loading module includes: The channel task extraction submodule retrieves the stage processing task entries bound to each channel from the cloud platform configuration directory based on the channel names identified in the channel set corresponding to the periodic stage, deduplicates and integrates duplicate task entries, and generates a channel-bound task list table. The task activation comparison submodule obtains the task entries that are currently active in the cloud platform, filters the entries that do not appear in the active status table in the channel-bound task list table, establishes a new loading task list, and generates a sequence of task identifiers to be loaded. The task sequence adjustment submodule combines the task identifier sequence to be loaded with the channel-bound task list table, performs a loading operation based on the task identifier sequence to be loaded, loads missing task entries according to the binding order, and simultaneously cancels the task content that has been activated in the cloud platform but is not in the channel-bound task list table, and obtains the phase processing task sequence.

7. The cloud computing-based full-cycle monitoring and management system for variable cross-section piles according to claim 1, characterized in that, The status archiving module includes: Based on the monitoring results of the stage task structure, the task field integration submodule extracts the channel type, pile location, collection time and project number from each data record, combines and aggregates them according to the task number, and sorts them to obtain the task attribution field aggregation table. The structure index construction submodule reads the task attribution field aggregation table, performs partitioning according to channel type, constructs a unique identifier based on pile location and collection time, binds it to each attribution data unit, and establishes an archive index coding matrix. The archive record generation submodule writes the belonging data units sequentially into the cloud full-cycle monitoring storage area according to the project number based on the archive index coding matrix, confirms the archive of the written entries and binds them with the channel identifier to obtain the full-cycle monitoring management record of the variable cross-section pile.

Images