Intelligent storage and retrieval device for construction data of yard engineering pipe network

Abstract

This invention relates to the field of construction monitoring technology for stockyard engineering, and discloses an intelligent storage and retrieval device for construction data of stockyard engineering pipeline networks. The device includes a data access module, a differential sampling module, a hardware protection module, a topology verification module, a differential storage module, and a coordinate reconstruction module. The invention utilizes the differential sampling module to adaptively control the sampling frequency based on the pore water pressure change rate, reducing the accumulation of redundant invalid data at the source. The hardware protection module blocks writing to the main storage medium when vibration exceeds limits and uses non-volatile memory to temporarily store data, avoiding underlying physical damage and data read / write errors caused by strong earthquakes. The topology verification module, combined with the spatial connection relationship of the pipeline network, intercepts abnormal false displacement data. In conjunction with the differential storage and coordinate reconstruction modules, incremental storage and a full snapshot mechanism are used to restore the spatial position at a specified time, ensuring the authenticity of the reconstructed coordinates while improving the retrieval and output efficiency of historical monitoring data.

Description

Technical Field

[0001] This invention relates to the field of construction monitoring technology for stockyard engineering, specifically to an intelligent storage and retrieval device for construction data of pipeline networks in stockyard engineering. Background Technology

[0002] In the monitoring of pipeline construction in large-scale stockyard projects, real-time acquisition of displacement and deformation data of the pipeline network through automated sensor networks is a key means of assessing soil stability and construction safety. Current monitoring equipment and storage systems typically use preset fixed sampling frequencies for data recording. However, this approach exhibits significant limitations in scenarios like stockyard construction, which spans long periods. Because the soil consolidation process is characterized by intense deformation in the early stages and gradual stabilization in the later stages, fixed sampling frequencies are prone to missing critical strain peaks due to insufficient sampling density during the active deformation phase. Furthermore, once the secondary consolidation and stabilization phase begins, a large amount of static redundant data with no physical meaning is generated, placing a heavy burden on storage media and data transmission bandwidth.

[0003] Meanwhile, the complex working environment at the construction site also poses a severe challenge to the reliability of storage hardware. The strong and continuous vibrations generated by the frequent operation of large machinery in the storage yard can easily interfere with the magnetic or semiconductor physical read and write processes of the main storage medium. This can lead to errors in writing current monitoring data or damage to file indexes, or even permanent physical damage to storage sectors. Existing devices often lack a real-time redirection protection mechanism that is linked to the on-site vibration conditions.

[0004] Furthermore, during the data retrieval and coordinate reconstruction phases, the scale of records in the time-series database grows exponentially with the extension of the monitoring cycle. Traditional retrieval methods, when processing large-scale historical displacement data, often suffer from delayed query responses due to a lack of scientific data organization, making it difficult to meet the real-time requirements of construction early warning. More critically, existing coordinate reconstruction logics mostly only superimpose values ​​from a single node, ignoring the constraints of the physical topology relationships between nodes in the pipeline network. This results in false displacements caused by long-term sensor drift not being effectively intercepted, directly affecting the accuracy of spatial coordinate reconstruction at the construction site. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent storage and retrieval device for construction data of storage yard engineering pipelines. This device solves the problems faced in long-term construction monitoring of storage yard engineering pipelines, such as the excessive storage space consumption of massive redundant data, the easy damage to storage hardware and data loss caused by strong vibration environments, and the low efficiency of retrieving massive historical monitoring data and reconstructing real spatial coordinates.

[0006] To achieve the above objectives, the present invention provides an intelligent storage and retrieval device for construction data of a stockpile engineering pipeline network, comprising: The data access module is used to receive time-series monitoring data and load the initial spatial reference coordinates of each node in the pipeline network. The differential sampling module is used to evaluate the soil consolidation state based on time-series monitoring data, control the sampling frequency, and calculate the differential displacement vector. The hardware protection module is used to receive differential displacement vectors and monitor vibration parameters based on time-series monitoring data. When the parameters are within limits, the differential displacement vectors are directly output. When the parameters are exceeded, the differential displacement vectors are temporarily stored and the main memory is blocked from being written. The output is written back after the vibration subsides. The topology verification module is used to receive the differential displacement vector output by the hardware protection module, compare it with the spatial topology of the pipeline network and the physical limit strain constant, intercept abnormal data and output the differential displacement vector that has passed the verification. The differential storage module is used to write the verified differential displacement vector and its corresponding timestamp into the time series database; The coordinate reconstruction module is used to parse data retrieval requests, combine the initial spatial reference coordinates with the historical differential displacement vectors extracted from the time series database, reconstruct and output the spatial coordinates at the requested time.

[0007] Furthermore, the time-series monitoring data includes pore water pressure, vibration acceleration, and axial strain along the pipeline network collected by the front-end sensor network. After receiving the time-series monitoring data, the data access module uses the sampling time of the highest-frequency acceleration as the reference time axis, and performs time node completion on the pore water pressure and axial strain along the pipeline network using zero-order hold or linear interpolation methods to generate a time-aligned multi-dimensional time-series data set. This configuration eliminates the time phase deviation caused by the difference in sampling frequencies of different sensors by aligning the clock periods of multi-source heterogeneous data to a high-frequency reference, thus unifying the time reference for subsequent data processing in the system.

[0008] Furthermore, the differential sampling module's assessment of the soil consolidation state and control of the sampling frequency specifically includes: The design extracts pore water pressure from time-series monitoring data and calculates its time derivative to obtain the pore water pressure change rate. This rate is then compared to a preset consolidation state determination threshold. If the pore water pressure change rate is greater than or equal to the threshold, the underlying hardware timer switches to high-frequency sampling. If the rate is less than the threshold, the soil is determined to have entered a sub-consolidation stable phase, and the underlying hardware timer switches to low-frequency heartbeat sampling. This design uses the pore water pressure change rate as a physical parameter representing the soil consolidation phase, directly intervening in the interruption cycle of the underlying hardware. This ensures that sampling resources match the actual foundation deformation rate in real time, reducing the generation of invalid redundant data during the consolidation plateau period from the data source.

[0009] Furthermore, the differential sampling module calculates the differential displacement vector specifically including: An initial spatial geometric path along the pipeline network is generated by fitting the initial spatial reference coordinates, and the unit tangent vector of the initial spatial geometric path is obtained. The corresponding physical lengths of the pipeline nodes on the optical fiber laying path are obtained. The axial strain along the pipeline network in the time-series monitoring data is numerically integrated along the initial spatial geometric path and projected onto the unit tangent vector to derive a differential displacement vector containing displacement components in three coordinate directions. This step integrates the scalar axial strain acquired by distributed optical fibers along the three-dimensional spatial topology path of the pipeline network, realizing the dimensional transformation of deformation data from a one-dimensional scalar to a three-dimensional spatial vector.

[0010] Furthermore, the hardware protection module monitors vibration parameters specifically including: The vibration acceleration is extracted from the time-series monitoring data, the summation magnitude of the vibration acceleration in three dimensions is calculated, and the summation magnitude is compared with a preset physical interface tolerance safety threshold. When the summation magnitude is greater than or equal to the physical interface tolerance safety threshold, the vibration is deemed to have exceeded the limit. The commit queue facing the main storage medium is blocked, a bus suspension command is issued, and the differential displacement vector is redirected to a non-volatile random access memory for temporary storage. When the summation magnitude falls back to less than the physical interface tolerance safety threshold, and does not exceed the limit again within the preset seismic observation window, the commit queue is released, and the temporarily stored differential displacement vectors are written back and output in batches. If the summation magnitude is greater than or equal to the physical interface tolerance safety threshold again within the preset seismic observation window, the preset seismic observation window is reset, and the bus suspension command and the temporary storage of the redirected differential displacement vectors are maintained. This mechanism utilizes the sum of vibration acceleration modulus to trigger hardware-level protection. Under strong impact conditions, it cuts off the physical write path of the main memory at the bus level, directs the data flow to the shock-resistant cache medium, and uses the observation window to isolate aftershock interference, thus avoiding the momentary interruption of physical contact of the main memory and damage to the file system caused by severe vibration.

[0011] Furthermore, the topology verification module performs comparisons based on the spatial topology relationships of the pipeline network, specifically including: The received differential displacement vector is added to the corresponding initial spatial reference coordinates to obtain the current absolute spatial coordinates of the target pipeline node. A pre-constructed and stored topological adjacency matrix reflecting the actual pipeline layout is retrieved. Based on this matrix, adjacent nodes of the target pipeline node are identified. A spatial distance algorithm is used to calculate the real-time spatial distance between the current absolute spatial coordinates of the target pipeline node and its adjacent nodes. The reference spatial distance between the target pipeline node and its adjacent nodes in the initial static state is retrieved. The calculated real-time spatial distance is subtracted from the reference spatial distance to obtain the relative deformation. The topology verification module intercepts abnormal data by comparing the absolute value of the calculated relative deformation with a preset physical tolerance threshold, which serves as the physical limit strain constant. When the absolute value is greater than or equal to the preset physical tolerance threshold, it is determined that the pipe segment between the corresponding adjacent nodes has a structural failure risk. The corresponding abnormal data is intercepted, and the pipe segment is marked as abnormal in the underlying topology status table. When the absolute value is less than the preset physical tolerance threshold, the differential displacement vector that has passed the verification is output. This module introduces a topological adjacency matrix that reflects the actual connection relationship of the pipeline network. By comparing the relative spatial distance changes of adjacent physical nodes with the theoretical mechanical tolerance of the pipe material, it eliminates non-real spatial displacements caused by long-term drift of the sensor body or communication errors.

[0012] Furthermore, the differential storage module writes to the time-series database including: The system retrieves the historical reference spatial coordinates from the last persistent storage and uses a spatial distance algorithm to calculate the scalar drift of the current absolute spatial coordinates relative to the historical reference spatial coordinates, corresponding to the received, verified differential displacement vector. When the scalar drift is greater than or equal to a preset differential storage trigger threshold, a valid displacement is determined, and the current absolute spatial coordinates and corresponding timestamp are encapsulated into an incremental data packet and persistently written to the time-series database. The corresponding historical reference spatial coordinates in memory are then overwritten with the current absolute spatial coordinates. When the scalar drift is less than the preset differential storage trigger threshold, no valid displacement is determined, the persistent write operation is not performed, and the corresponding historical reference spatial coordinates in memory remain unchanged. This logic uses the absolute change in spatial distance as the criterion for triggering persistent writes, ensuring that the database only records the node states that produce substantial spatial positional shifts, directly discarding invalid minor fluctuations, effectively controlling the storage scale of massive time-series data.

[0013] Furthermore, the method of writing to the time-series database also includes maintaining an independent hardware clock counter. Every preset snapshot period, ignoring the comparison results of scalar drift, it forces the current absolute spatial coordinates of all nodes in the pipeline network and the underlying hardware operating status to be packaged into a full state snapshot file and persistently stored in the time-series database. The periodic full state snapshot mechanism establishes a global reference anchor point on the time axis, which can block the small truncation errors accumulated by long-term incremental calculations and ensure the spatial authenticity of the overall pipeline network topology model.

[0014] Furthermore, the coordinate reconstruction module reconstructs and outputs the spatial coordinates at the requested time, specifically including: The process involves retrieving the target full-state snapshot file (whose generation time is less than and closest to the requested time) in reverse order from the time-series index of the time-series database. The current absolute spatial coordinates from these snapshots are then extracted as the initial reference coordinates for the reconstruction calculation. All incremental data packets within the interval from the generation time of the target full-state snapshot file to the requested time are extracted to form an update sequence. If the update sequence is not empty, the last incremental data packet with the largest timestamp that is less than or equal to the requested time is extracted. Zero-order hold logic is used to assign the current absolute spatial coordinates of this incremental data packet to the requested time and output it. If the update sequence is empty, the initial reference coordinates are directly used as the spatial coordinates of the requested time and output. This reconstruction process combines periodic snapshots and discrete incremental packets to construct a complete logic for retrieving full and incremental data. Zero-order hold logic is used to deduce the spatial location of unrecorded times, reducing the computational overhead of traversing large-scale historical records.

[0015] This invention provides an intelligent storage and retrieval device for construction data of pipeline networks in stockyard engineering. It has the following beneficial effects: 1. This invention utilizes a differential sampling module to identify the soil consolidation phase using the time derivative of pore water pressure, achieving dynamic closed-loop control of the sampling frequency. This mechanism samples as needed based on the physical state of the construction site, ensuring high-density capture during periods of intense deformation and reducing the accumulation of invalid data through heartbeat sampling after entering the stable period. This solves the problem of monitoring data redundancy at its source, reduces hardware power consumption and transmission pressure, ensures the database records have extremely high value, and balances monitoring accuracy and storage efficiency.

[0016] 2. This invention establishes a data redirection mechanism based on vibration monitoring through a hardware protection module, enabling proactive protection of the storage medium under strong vibration conditions during construction. The system compares the physical interface tolerance threshold in real time, blocking main memory writes during periods of excessive vibration and temporarily storing differential displacement vectors in non-volatile memory. Batch write-back is then performed once the environment stabilizes. This isolates external environmental interference, eliminates the risk of storage damage caused by vibration, and ensures the integrity of monitoring data and the lifespan of the equipment.

[0017] 3. This invention combines a topology verification module and a coordinate reconstruction module, utilizing the pipeline network physical topology adjacency matrix and a full state snapshot mechanism to ensure the authenticity and retrieval speed of the reconstructed data. Topology verification eliminates false displacement data caused by sensor drift by calculating real-time spatial distances and comparing them with physical limit strains; while incremental storage combined with full snapshots eliminates the need for reconstruction to traverse the entire history, shortening response time and providing an efficient means for deformation monitoring and early warning. Attached Figure Description

[0018] Figure 1 This is a flowchart of a method for intelligent storage and retrieval of construction data for a stockpile engineering pipeline network according to an embodiment of the present invention; Figure 2 This is a time-series curve of pore water pressure dissipation in the foundation according to an embodiment of the present invention; Figure 3 This is a graph showing the adaptive sampling frequency control of the underlying hardware in an embodiment of the present invention. Figure 4 This is a graph comparing the cumulative data storage volume of embodiments of the present invention with that of traditional fixed-frequency devices. Detailed Implementation

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This invention provides an intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network, deployed on an edge computing node, comprising: a data access module; a differential sampling module; a hardware protection module; a topology verification module; a differential storage module; and a coordinate reconstruction module.

[0021] The device establishes a hardware connection with a sensor network embedded in the construction area via a field communication bus. The sensor network includes pore water pressure sensors to monitor the consolidation state of the foundation, triaxial accelerometers to sense the dynamic loads of construction machinery, and distributed fiber optic strain sensors installed alongside the pipeline network. The device's mainboard integrates an industrial-grade solid-state drive as the primary storage medium and incorporates non-volatile random access memory as an independent cache.

[0022] See attached document Figure 1 This invention provides an intelligent storage and retrieval method for construction data of pipeline networks in stockyard engineering, comprising the following steps: S10: Receive timing monitoring data transmitted from the sensor network and synchronously load the initial spatial reference coordinates of each node in the pipeline network. S20, calculate the time derivative of pore water pressure to assess the soil consolidation state, control the data sampling frequency based on the assessment results, and extract fiber optic strain data to calculate the differential displacement vector of the pipeline node; S30 monitors triaxial acceleration data. When the vibration parameters exceed the preset tolerance, a bus suspension command is issued to block the main storage medium write path and redirect the differential displacement vector to a non-volatile random access memory for temporary storage. After the vibration subsides, batch processing is performed to write back the temporarily stored data. S40 combines the spatial topology of the pipeline network with the physical limit strain constant of the pipe material to calculate the theoretical spatial distance after the node displacement, and intercepts abnormal displacement data that exceeds the physical limit. S50 writes the verified differential displacement vector and its corresponding timestamp into the time series database; S60, parse the target node and time parameters in the data retrieval request, extract the initial spatial reference coordinates of the target node and the historical differential displacement vector within the specified time interval, reconstruct the spatial coordinates of the pipeline node at the requested time through time-series accumulation operation and output them.

[0023] In this embodiment, the data access module is mainly responsible for physical quantity acquisition, initial spatial reference coordinate establishment, and time-series data synchronization and alignment. In the actual application scenario of soft soil foundation treatment engineering in stockpile areas, due to the coexistence of slow foundation consolidation settlement and instantaneous mechanical compaction impact, the system needs to simultaneously acquire pore water pressure reflecting the long-term stress state of the soil, vibration acceleration reflecting transient dynamic loads, and axial strain reflecting the deformation of the pipeline network body, in order to provide complete basic data input. Specifically, the processing flow of this module includes the following steps: S101 establishes the underlying communication link between the field sensor network and the edge computing nodes. In specific engineering field deployments, pore water pressure sensors are buried inside the foundation, piezoelectric triaxial accelerometers are deployed on the surface, and distributed fiber optic strain sensors are laid along the pipeline network. Edge computing nodes are connected to the pore water pressure sensors via RS485 communication buses, to the piezoelectric triaxial accelerometers via controller area network buses, and to the distributed fiber optic strain sensors via fiber optic demodulators. The specific selection of multi-source sensing devices and the layout of peripheral adapter circuits can be configured by those skilled in the art according to the actual accuracy requirements of the project. The hardware infrastructure wiring is well-known in the field and will not be elaborated upon here.

[0024] S102, Input the initial spatial reference coordinates of the pipeline network nodes and establish spatial mapping associations. In this embodiment, when the pipeline network is laid and the foundation is in an initial static state, the three-dimensional position information of each node in the pipeline network is obtained as a reference. A basic mapping table containing node numbers and three-dimensional coordinate values ​​is established internally by the system. The node number in the pipeline network node set is set to... ,in This is a positive integer, ranging from 1 to the total number of network nodes. The initial spatial reference coordinates are: The three-dimensional coordinate values ​​are represented using an independent engineering coordinate system, including the X-axis coordinate. Y-axis coordinate and Z-axis coordinates The basic spatial relationships are represented as follows: ; The aforementioned initial spatial reference coordinates serve as the spatial reference for subsequent calculations of node differential displacements and the execution of topology constraint verification. Simultaneously, the system records the laying path of the distributed optical fibers on the pipeline network and calibrates the distance measurement point from the fiber optic cable to the corresponding pipeline node. Actual physical length This forms the spatial location of the optical fiber. discrete nodes of the pipeline network The corresponding association table is used to provide a basis for accurate node positioning of subsequent strain data.

[0025] S103 acquires real-time physical quantities and performs time-series data alignment. In practice, the edge computing node motherboard integrates a hardware clock source to assign a unified timestamp to asynchronous data received from different communication buses. The pore water pressure sensor continuously outputs a pore water pressure value that varies over time, denoted as... The piezoelectric triaxial accelerometer outputs a three-dimensional vibration acceleration vector, denoted as [vector]. Distributed fiber optic strain sensors acquire axial strain along the pipeline network based on the Brillouin frequency shift effect. The fiber optic demodulator emits probe pulses into the fiber and receives backscattered light, analyzing the Brillouin frequency shift in the scattered light. Since the Brillouin frequency shift has a linear correlation with the axial deformation of the fiber, edge computing nodes can calculate the spatial location along the pipeline network accordingly. In time The continuous axial strain distribution is denoted as The calculation principle for converting Brillouin frequency shifts into specific strain values ​​can be achieved by those skilled in the art using conventional photoelectric demodulation algorithms. Its physical layer analysis is a well-known technique in the field and will not be elaborated here.

[0026] It should be noted that the frequencies of change of various physical quantities differ. For example, pore water pressure reflects a slowly changing process, and its sampling frequency is typically set on the order of minutes; while vibration acceleration reflects a transient process, and its sampling frequency is typically set on the order of kilohertz. To achieve asynchronous data synchronization, the edge computing nodes will extract the pore water pressure values... Vibration acceleration vector and continuous axial strain distribution Unified mapping to discrete time points under the same clock period ,in The sequence number is a positive integer that increases sequentially in time. In this embodiment, the alignment method uses the highest frequency acceleration sampling time as the reference time axis. The time nodes of the low-frequency pore water pressure and strain data are filled in using zero-order hold or linear interpolation methods, thereby generating an aligned multidimensional time series data set, which is then transmitted to the subsequent differential sampling module and hardware protection module for further analysis and processing.

[0027] In this embodiment, the differential sampling module mainly utilizes the pore water pressure dissipation characteristics during the foundation consolidation process as its driving logic. It achieves adaptive adjustment of the data calculation frequency through state machine switching and derives the displacement changes of discrete nodes by combining continuous strain data. Specifically, the processing flow of this module includes the following steps: S201, calculate the time derivative of pore water pressure and determine the consolidation phase. In specific implementation, the system extracts the time-aligned pore water pressure value. The rate of change of pore water pressure at the current moment is calculated using the discrete difference algorithm. The formula for calculating the rate of change of pore water pressure is: ; in, For the previous sampling time, This is the current sampling time. After calculating the rate of change, the system compares it with a preset consolidation state determination threshold. A comparison is performed. In this embodiment, a threshold is determined. The specific value is usually derived from the indoor standard consolidation test data of the soil at the engineering site, and its magnitude is closely related to the permeability coefficient and compression modulus of the soil. In conventional soft soil foundation treatment projects, this threshold is usually set between 0.5 kPa and 2.0 kPa, with the specific value depending on the engineering survey report.

[0028] S202 performs adaptive switching of the sampling frequency based on the consolidation phase evaluation results. At the underlying hardware implementation level, the edge computing node has a built-in programmable hardware timer used to send interrupt requests for data extraction and computation to the system bus. When the comparison result is... When the system determines that the soil beneath the current pipeline is in the primary consolidation phase with rapid pore water discharge, or has been subjected to severe disturbance during dynamic compaction, the underlying controller of the edge computing node writes high-frequency period parameters to the hardware timer, for example, setting the sampling calculation period to the order of 100ms to 1 second, thereby switching the system to high-frequency sampling mode. Conversely, when the comparison result is... At this point, the system determines that the pore water pressure difference within the soil has largely dissipated, entering a subconsolidation stable phase with extremely slow deformation. The underlying controller then adjusts the hardware timer register, for example, setting the low-frequency heartbeat sampling period to the order of 10 minutes to 1 hour, smoothly switching the system to low-frequency heartbeat sampling mode. This control mechanism, which directly intervenes in the underlying hardware timer period based on the soil's mechanical state, effectively reduces the amount of redundant data processed by the system.

[0029] S203, deriving the differential displacement vector of discrete nodes based on the continuous strain distribution of the pipeline network. At each valid calculation moment triggered by a hardware timer. The system extracts the fiber optic axial strain distribution data at the corresponding time. The underlying physical principle is that distributed fiber optic measurements only obtain the axial tensile or compressive scalar deformation of a local area of ​​the pipeline network. To obtain the actual three-dimensional spatial displacement, this scalar deformation needs to be projected onto the actual direction of the pipeline in three-dimensional space. Therefore, the system uses the initial spatial reference coordinates of the pipeline nodes to fit and generate the initial spatial geometric path along the pipeline network, and calculates the distance of this path from the starting point... unit tangent vector at point In this embodiment, the unit tangent vector is specifically composed of its three directional cosine components in the X, Y, and Z axes.

[0030] S204, Perform spatial path integral calculation. Further, the system utilizes the previously established fiber optic spatial location... With pipeline nodes Obtain the corresponding association table to retrieve the pipeline nodes. Corresponding physical length on the fiber optic path The system numerically integrates the axial strain along the spatial path to derive the nodal parameters. exist The differential displacement vector at time relative to the initial state of the system The integral equation for calculating the differential displacement vector is: ; In practical digital signal processing, the above continuous integral equation needs to be implemented using a discrete accumulation algorithm. The system divides the integration interval into multiple infinitesimal lengths. The system will determine the length of the infinitesimal element. The average axial strain within the segment is multiplied by the unit tangent vector at the corresponding location to obtain the three-dimensional deformation vector of that infinitesimal element. This vector is then successively accumulated along the integration path to the length. The differential displacement vector calculated from this is... This includes displacement components along the X, Y, and Z axes. For a specific infinitesimal element length... The partitioning criteria and the specific coding implementation of the discrete accumulation algorithm can be readily configured and selected by those skilled in the art based on the microprocessor's computing performance. The fundamental numerical analysis methods are well-known in the field and will not be elaborated upon here. The final generated differential displacement vector is then transmitted to the buffer queues of the hardware protection module and the topology verification module, awaiting processing in subsequent steps.

[0031] In this embodiment, the hardware protection module is mainly used to cope with extreme impact conditions such as dynamic compaction construction in the stockpile. Through underlying bus intervention and storage routing switching, it ensures the physical safety of the main storage medium of the edge computing node and the integrity of the data. Its core physical principle is that severe vibration environments may cause microscopic contact interruptions at the gold fingers of the storage medium interface, or cause logical errors in the flash memory conversion layer inside the solid-state drive during the programming cycle. Specifically, the module's processing flow includes the following steps: S301 calculates the real-time vibration acceleration modulus and performs a safety threshold comparison. In specific implementation, the system receives the time-aligned three-dimensional vibration acceleration vector in real time. This vector contains acceleration components along three mutually perpendicular physical axes. The system uses its magnitude calculation logic to derive the conjunction magnitude of the acceleration vector at the current moment. The formula for calculating the sum modulus is: ; in, , and Representing the current sampling time respectively The acceleration components are calculated in three physical dimensions. After determining the modulus, the system compares it with a preset physical interface tolerance safety threshold for the main storage medium. Real-time comparison is performed. In one embodiment of the present invention, this security threshold... The specific value depends on the vibration resistance specifications of the industrial-grade solid-state drives and their physical slots used in the edge computing nodes. To prevent reflection interference or voltage drops in the slot's electrical signals, this safety threshold is typically set at the 3G to 5G gravitational acceleration level in conventional heavy machinery construction sites. The specific value can be calibrated according to the vibration tolerance manual provided by the hardware manufacturer.

[0032] S302 triggers a suspend interrupt on the underlying memory bus based on the threshold determination result. When the system comparison finds that the condition is met... At this point, it is determined that the shock wave has reached the physical limits of the underlying hardware. To prevent storage write operations during this period from causing the logical ring block or file system to crash, the underlying controller of the edge computing node performs intervention control. Specifically, the underlying controller modifies the command scheduling algorithm of the storage driver layer, blocks the commit queue facing the main storage medium, and intercepts incomplete write requests.

[0033] Furthermore, the underlying controller sends an I / O suspend interrupt command to the main storage medium via the bus, forcing the solid-state drive controller chip into an instruction hold state, thereby physically blocking the normal data write path. The specific encoding, encapsulation, and trigger timing configuration of the underlying bus interrupt command can be implemented using conventional embedded low-level driver development techniques by those skilled in the art. Its hardware bus communication protocol is well-known in the field and will not be elaborated upon here.

[0034] S303, synchronously execute hardware-level routing redirection of differential displacement vector data. In this embodiment, the system triggers data routing switching logic when the main storage medium write path is blocked. The edge computing node motherboard integrates a non-volatile magnetoresistive random access memory as an independent temporary storage area. In specific implementation, the underlying controller dynamically modifies the destination address register of the direct memory access controller to redirect the differential displacement vector data stream originally pointing to the main storage medium buffer to the physical address space of the non-volatile magnetoresistive random access memory. This switching mechanism based on hardware address mapping ensures that critical pipeline deformation data can be completely and with low latency temporarily stored during the main storage suspension period.

[0035] S304 monitors the vibration decay and performs batch write-back of cached data or state reset. As the impact load decays, the system continuously monitors and detects that the combined vibration modulus has fallen below the safe threshold, thus satisfying the condition. In this embodiment, to avoid the aftershocks caused by soil rebound, the system initiates a preset seismic observation window of 500ms to 1000ms after detecting a rebound. If the vibration does not exceed the limit again within this window, the underlying controller sends an I / O recovery command to the communication bus, releasing the blocked submission queue and reopening the write path to the main storage medium. Subsequently, the system starts a batch write-back program, traversing the temporary data blocks in the non-volatile magnetoresistive random access memory and moving them in batches to the preset storage address of the main storage medium according to their timestamp numbers via the direct memory access channel. After the data write-back is completed and verified, the system releases the address space of the temporary storage area to prepare for the next impact event.

[0036] Conversely, if the system detects that the combined vibration modulus is again greater than or equal to the physical interface tolerance safety threshold within the preset seismic observation window, the system determines that the current hardware is still in the period of active aftershock interference. At this time, the underlying controller immediately clears and resets the seismic observation window timer, continues to keep the bus suspended and main memory locked, and continuously appends the newly generated differential displacement vector to the non-volatile magnetoresistive random access memory until the system passes through a complete observation window without exceeding the limit, and then performs the aforementioned release and write-back operations.

[0037] In this embodiment, the topology verification module is mainly used to assess the relative deformation between adjacent nodes based on real-time displacement data of pipeline nodes and the physical connection relationship of the pipeline network, in order to monitor whether structural failure risks such as pipeline joint detachment or excessive compression have occurred. Specifically, the processing flow of this module includes the following steps: S401 updates the real-time spatial coordinates of the pipeline nodes. In practice, the system retrieves the pipeline node coordinates from the main storage medium or cache queue. At the current sampling time Three-dimensional differential displacement vector The system compares the differential displacement vector with the initial spatial reference coordinates of the node established by the data access module. Perform vector addition to calculate the current absolute spatial coordinates of the node. The formula for calculating coordinate updates is: ; in, Includes initial coordinate components The differential displacement vector contains displacement change components. To meet the requirements for subsequent spatial distance calculation, in this embodiment, the system uses absolute spatial coordinates. It can be directly analyzed into three mutually perpendicular coordinate axis components, denoted as follows: , and Specifically, the real-time value of each coordinate axis component is equal to the scalar algebraic sum of the corresponding initial coordinate component and the displacement change component. The spatial coordinate components updated by the above calculation objectively represent the actual physical position of the pipeline node after experiencing foundation consolidation settlement or construction disturbance.

[0038] S402, extract the pipeline network topology connections and initial physical spacing. Further, the system pre-builds and stores a topological adjacency matrix reflecting the actual pipeline network layout in memory. In this embodiment, based on this adjacency matrix, the system identifies nodes connected to the pipeline network. There are directly physically connected adjacent nodes Based on this, the system retrieves pipeline network nodes. With neighboring nodes Reference space distance in the initial static state In practice, the reference spatial distance can be directly calculated using the spatial distance formula from the initial spatial reference coordinates of the two nodes, or it can be provided by the actual pipe section length calibration data during the project laying.

[0039] S403 calculates the real-time spatial distance between adjacent nodes and determines the relative deformation. The underlying physical and mechanical principle is that due to the heterogeneity of the foundation soil properties at the engineering site or the uneven distribution of disturbance energy during dynamic compaction, different uneven settlement or horizontal displacement will occur at each node along the pipeline network, resulting in relative displacement between adjacent physical nodes. (Obtain pipeline network node information.) With neighboring nodes After determining the current absolute spatial coordinate components, the system uses the Euclidean distance algorithm to calculate the distance between the two nodes. Real-time spatial distance at any moment In fact, the formula for calculating spatial distance is: ; in, , and Adjacent nodes Adopted with pipeline nodes The same analytical logic is used to calculate the real-time three-dimensional coordinate components. Subsequently, the system calculates the relative deformation between adjacent nodes. In this embodiment, the relative deformation is equivalent to the algebraic difference between the real-time spatial distance and the reference spatial distance, expressed as: ; S404, Perform joint tolerance threshold verification and status marking. Specifically, relative deformation amount. The sign and magnitude of the value directly reflect the local stress and deformation state at the pipe network joint. A positive value indicates that the joint is under axial tension; a negative value indicates that the joint is under axial compression. The system will calculate the absolute value of the relative deformation. With the preset physical tolerance threshold of the pipeline joint A comparison is performed. In this embodiment, the physical tolerance threshold... The specific value is determined based on the characteristics of the pipe material and the joint connection process. For example, for high-density polyethylene pipes using heat fusion connection, the allowable elastic expansion and contraction deformation is relatively large; while for reinforced concrete pipes using socket connection, the allowable displacement is extremely small. In conventional municipal or storage yard pipeline projects, this threshold is usually set to one to three percent of the nominal length of the joint. The specific safety threshold needs to be strictly set based on the mechanical performance test report of the corresponding pipe material.

[0040] When the comparison results meet When the system determines that the pipe segment between adjacent nodes is at risk of detachment and leakage or structural failure due to compression, it immediately marks the pipe segment corresponding to that node as abnormal in the underlying topology status table and triggers the push of location alarm data containing the node number and deformation amount to the engineering monitoring terminal. The encapsulation and transmission protocol of the monitoring terminal alarm data and the visualization rendering logic of the user interface can be fully implemented by those skilled in the art using conventional communication handshake protocols and front-end development technologies. The data presentation and interactive parsing are well-known technologies in the field and will not be elaborated upon here.

[0041] In this embodiment, the differential storage module is mainly used to address the conflict between continuous monitoring data and the storage space of edge computing nodes. The physical mechanism is that the foundation consolidation process typically takes a long time, and the displacement changes of pipeline nodes are minimal during most regular sampling cycles. If all high-frequency sampling data were to be permanently written, it would not only accelerate the depletion of solid-state drive capacity but also shorten the lifespan of the storage medium due to frequent write / erase operations. Specifically, the processing flow of this module includes the following steps: S501, extract the current monitoring frame data and read historical benchmark snapshots. In specific implementation, the system retrieves data from the topology verification module's cache queue, containing all network nodes at the current sampling time. The absolute spatial coordinate set. Simultaneously, the underlying controller retrieves the historical baseline spatial coordinate set saved during the last persistent storage execution from the main storage medium. In this embodiment, for a specific node... The absolute spatial coordinate components at the current moment are denoted as follows: , and The corresponding historical reference spatial coordinate components are denoted as follows: , and .

[0042] S502 calculates the scalar drift of the node's spatial position. Further, to quantify the degree of change of the current coordinates relative to the historical reference coordinates, the system employs a spatial distance algorithm to calculate the scalar drift of each network node. This scalar drift represents the linear spatial distance between the node's current physical location and its last recorded location, and its calculation formula is as follows: ; Through this calculation process, the system converts the changes in the three-dimensional coordinate vector into intuitive absolute distance offset values.

[0043] S503 executes differential storage trigger determination and incremental write. In practice, after calculating the scalar drift amount of each node, the system compares it with a preset differential storage trigger threshold. A comparison is performed. In this embodiment, the trigger threshold... The specific value is determined by the strain measurement accuracy and spatial resolution parameters of the front-end fiber optic demodulation equipment. To shield against interference from thermal expansion and contraction caused by ambient temperature fluctuations, this trigger threshold is typically set to 0.5mm to 2.0mm in conventional pipeline monitoring projects, and is configured according to the factory calibration parameters of the fiber optic instrument.

[0044] When the comparison results meet When the system determines that a valid displacement has occurred at a certain point, it activates the write channel of the main storage medium, encapsulates only the node number of the valid displacement, the current absolute spatial coordinates, and the corresponding timestamp into an incremental data packet, and persistently writes it to the solid-state drive of the edge computing node. After completing the partial write operation, the system only overwrites the historical reference spatial coordinates of the aforementioned out-of-limit nodes with the current absolute spatial coordinates in memory, using this as the physical reference for the next comparison of that node. Conversely, if a specific node meets the following conditions... If the node is in a stable state, the system will discard its redundant coordinate data for that moment and not trigger any disk writes to that node. This fine-grained node-level intervention mechanism achieves differential storage in the strictest sense.

[0045] S504, perform periodic full state snapshot persistence. Furthermore, although node-level incremental writes significantly reduce storage overhead, long-term reliance on local difference comparisons can easily lead to overall topology model distortion due to the accumulation of small computational truncation errors. In this embodiment, the system maintains an independent hardware clock counter at the system's underlying layer. Every preset snapshot period, the system ignores the aforementioned scalar drift determination results and forcibly packages the absolute spatial coordinates of all nodes in the network at the current moment along with the underlying hardware operating status, generating a full state snapshot file and storing it in the main storage medium. In specific implementations, this snapshot period is typically set to 12 to 24 hours.

[0046] By periodically saving full state snapshots, the system can quickly reconstruct the complete physical model of the pipeline network using the most recent snapshot file after a power outage or hardware restart. The specific implementation of the underlying file system's data packaging protocol and the solid-state drive wear leveling algorithm can be readily adapted by those skilled in the art based on the selected embedded operating system; the underlying storage driver principles are well-known technologies in the field and will not be elaborated upon here.

[0047] In this embodiment, the coordinate reconstruction module is mainly used to restore the physical spatial location of the pipeline network node at any specified time when the edge computing node undergoes a power outage and restart, or when the monitoring center initiates a historical deformation trajectory query. This is based on discrete differential storage data and a full state snapshot. Its underlying technical logic lies in using the full snapshot as a physical reference anchor point, combined with the sequential extrapolation mechanism of the differential data, to recover the compressed temporal spatial data. Specifically, the processing flow of this module includes the following steps: S601: Receives a coordinate reconstruction request and locates the initial reference snapshot. In specific implementation, the system receives a network node containing the target. and the target timestamp The coordinate reconstruction instruction. The underlying file system uses this timestamp. A reverse search is performed in the time-series index of the main storage medium to find the closest one whose generation time is less than or equal to that of the previous generation. A full state snapshot file. Extract the network nodes recorded in this snapshot file. The absolute spatial coordinate components are used as the initial reference coordinates for reconstruction calculation, and are denoted as follows: , and The corresponding snapshot generation time is denoted as In this embodiment, this step establishes a reliable physical starting point for subsequent coordinate derivation.

[0048] S602, extract the valid state update sequence within the time interval. Further, the system retrieves the time interval from the underlying database. All differential data frames within the time interval. Since the aforementioned differential storage module only triggers writing when the node's physical displacement exceeds a safety threshold, the extracted update sequence exhibits a non-continuous, discrete state on the time axis. Specifically, assuming a total of [number missing] data frames were retrieved within this time interval... A node The system arranges the differential data frames according to their timestamp order.

[0049] S603, perform the sequential evolution of the incremental state and extract the latest anchor point. In this embodiment, the system determines the order based on the number of retrieved differential data frames. Determine the timestamp of the distance from the target The most recent valid physical state. If This indicates that within this specific time interval, the pipeline nodes... If no valid physical displacement occurs that reaches the differential storage trigger threshold, the system directly updates the latest valid timestamp. The equivalent assignment is the initial snapshot time. The corresponding latest absolute space coordinate components are denoted as follows: , and And the values ​​are respectively equal to , and Conversely, if the judgment is... The system directly extracts the timestamp with the largest timestamp from the arranged update sequence, that is, the timestamp that is closest to and less than or equal to the target timestamp. The last differential data frame is used to generate a timestamp, which is then recorded as the latest valid timestamp. The system reads and parses the absolute spatial coordinate components contained in the data frame, and directly assigns them to... , and .

[0050] S604, zero-order hold logic is applied to determine the final reconstructed coordinates and output them. In this embodiment, for the latest valid timestamp... To the target time During the time period in question, based on the physical mechanism of the aforementioned differential storage module, the system determines that any physical drift occurring in a node during that period is strictly less than the trigger threshold. The physical and engineering basis for this is that as long as no new differential data is generated, it indicates that the current position of the pipeline network within the engineering tolerance range has not fundamentally changed. Therefore, the system uses zero-order hold logic to process the coordinate evolution for this period, that is, directly using the latest valid time. The absolute spatial coordinates of each component are directly assigned to the target time. This serves as the final reconstructed coordinate. Unnecessary expansion equations have been removed; equivalent assignment can be achieved simply by overwriting memory values.

[0051] Through the above steps, the system achieves the restoration of the historical physical displacement of pipeline nodes while avoiding the persistence of all high-frequency data. Although there may be slight deviations between the reconstructed coordinates and the actual microscopic locations, this spatial error is strictly limited within the boundary range of the differential storage trigger threshold, fully meeting the accuracy requirements of actual engineering monitoring. Regarding the database timestamp index construction method and multi-threaded parallel scheduling algorithm involved in the edge computing nodes, those skilled in the art can fully implement them using conventional time-series database query optimization techniques. The data retrieval engine and concurrency control are well-known technologies in this field and will not be elaborated upon here.

[0052] Specific application examples: This embodiment takes the construction monitoring of a high-density polyethylene (HDPE) drainage pipe network in a coastal port storage yard as an example. The project uses dynamic compaction for foundation reinforcement, and pore water pressure sensors, piezoelectric triaxial accelerometers, and distributed optical fibers are deployed on-site. The edge computing node motherboard is equipped with an industrial-grade solid-state drive and 256MB of non-volatile magnetoresistive random access memory (MRAM).

[0053] In the initial state, the system acquires pipeline nodes. and adjacent nodes The initial spatial reference coordinates are denoted as and And calculate the initial reference space distance. The thickness is 6000.0 mm. During the dynamic compaction stage, the rammer drops its heavy hammer, generating a violent shock wave on the ground surface. Simultaneously, the pore water pressure inside the soil increases sharply and then dissipates rapidly. The system performs the following dynamic processing within this timeframe: First, the data access module aligns the low-frequency sensor data with a 1000Hz acceleration sampling rate as the reference time axis. The differential sampling module extracts data from two consecutive sampling moments. and The pore water pressure value, according to the formula The current pore water pressure change rate was calculated. The value is 2.8 kPa / h. The system compares this value with the preset consolidation state determination threshold. (In this embodiment, the pressure is set to 1.5 kPa / h) for comparison. Since 2.8 is greater than 1.5, the system determines that the soil is in a period of severe deformation, and the bottom controller immediately switches the hardware timer to a high-frequency sampling state of 100ms. Simultaneously, the system extracts the fiber optic strain distribution and applies it according to the integral equation. Derive the node differential displacement vector The value is [3.2, -1.5, -12.8] mm.

[0054] Subsequently, the hardware protection module extracts the three-dimensional acceleration components, according to the formula... Calculate the sum modulus Due to the impact of the heavy hammer, the current mold length has reached 4.2G. The system compares this with the physical interface tolerance safety threshold. (Set to 3.0G) Comparison. Because 4.2 is greater than or equal to 3.0, the system triggers a low-level bus suspension interrupt, blocking solid-state drive writes and redirecting the aforementioned differential displacement vector to MRAM for temporary storage. After an 800ms observation window, the vibration modulus drops back to 1.2G, the system releases the bus and performs batch writeback.

[0055] In the topology verification phase, the system calculates the updated absolute spatial coordinates and applies the formula... The calculated real-time spatial distance is 6008.5 mm. (Based on the formula...) The relative deformation was determined to be 8.5 mm. This value is less than the physical tolerance threshold of HDPE pipe joints. (Set to 30.0mm), the data passed the verification and no pipe section abnormality alarm was triggered.

[0056] Finally, the differential storage module extracts the historical reference space coordinates, according to the formula... Calculate the scalar drift. Assume the result of this calculation is... The value is 13.3mm, which is greater than the set differential storage trigger threshold. (1.0mm), the system determines it as a valid displacement, encapsulates the current coordinates and timestamp into an incremental data packet and stores it in the time series database, and updates the reference coordinates in memory.

[0057] Experimental verification and effect comparison: To verify the actual technical effect of the present invention, a 30-day continuous monitoring experiment was conducted in the aforementioned stockpile project. A control group was set up in the experiment. The control group used a traditional fixed-frequency (fixed 1 second / time) full-data storage monitoring device, which did not have differential sampling, hardware protection, or incremental storage mechanisms.

[0058] See attached document Figure 2 -Appendix Figure 3 The smooth solid line at the top of the figure represents the dissipation and attenuation of pore water pressure as construction progresses, while the dotted-dashed step curve at the bottom represents the system's actual sampling frequency. (See attached...) Figure 2 -Appendix Figure 3 It can be seen that during the intense compaction period from day 1 to day 5, the pore water pressure derivative is relatively large, and the lower step curve remains in the high-frequency sampling range of 10Hz. After day 15, the soil enters the secondary consolidation phase, the upper pressure curve tends to flatten, and the system automatically reduces the sampling frequency to a low-frequency heartbeat sampling every 10 minutes. This sampling control driven by a physical mechanism effectively conforms to the objective laws of foundation consolidation.

[0059] See attached document Figure 4 The horizontal axis represents the number of monitoring days, and the vertical axis represents the amount of data persistently written to the solid-state drive. The dashed line in the graph represents the data volume trend of traditional devices, which, due to their fixed frequency and full-volume writing, exhibits a steep linear increase, accumulating approximately 45.6GB of data over 30 days. The solid line represents the data volume trend of the device of this invention. Due to the adoption of differential sampling and incremental storage strategies based on scalar drift, its trajectory shows a non-linear evolution characteristic of slight initial increase followed by rapid flattening. After the 30-day experimental period, the actual storage space occupied by this invention was only 2.1GB, achieving a data compression ratio of approximately 21.7:1.

[0060] Furthermore, regarding hardware reliability, 412 instances of transient strong vibrations exceeding 3.0G were recorded at the construction site during the experiment. Reading the system's underlying logs revealed that the control group's solid-state drive experienced 27 I / O write timeouts and generated two logical bad blocks in the file system; while the device of this invention, due to the intervention of a non-volatile storage redirection mechanism, intercepted 100% of high-risk write commands, achieving zero errors and zero bad blocks on the main storage medium. In terms of data query and retrieval, for a historical coordinate reconstruction request spanning 15 days, the control group system needed to traverse over 1.2 million full records, with an average response time of 4.2 seconds; this invention, relying on full state snapshots and zero-order persistence extrapolation of local incremental data packets, reduced the average response time for a single query to less than 0.15 seconds, fully meeting the real-time topology early warning requirements of the engineering site.

Claims

1. An intelligent storage and retrieval device for construction data of a stockpile engineering pipeline network, characterized in that, include: The data access module is used to receive time-series monitoring data and load the initial spatial reference coordinates of each node in the pipeline network. The differential sampling module is used to evaluate the soil consolidation state based on the time-series monitoring data, control the sampling frequency, and calculate the differential displacement vector. The hardware protection module is used to receive the differential displacement vector and monitor the vibration parameters based on the timing monitoring data. When the parameters are within the limits, the differential displacement vector is directly output. When the parameters are exceeded, the differential displacement vector is temporarily stored and the main memory is blocked from being written. After the vibration subsides, the output is written back. The topology verification module is used to receive the differential displacement vector output by the hardware protection module, compare it with the spatial topology of the pipeline network and the physical limit strain constant, intercept abnormal data and output the differential displacement vector that has passed the verification. A differential storage module is used to write the verified differential displacement vector and its corresponding timestamp into a time-series database; The coordinate reconstruction module is used to parse the data retrieval request, combine the initial spatial reference coordinates with the historical differential displacement vector extracted from the time series database, reconstruct and output the spatial coordinates at the requested time.

2. The intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network according to claim 1, characterized in that, The time-series monitoring data includes pore water pressure, vibration acceleration, and axial strain along the pipeline network, all collected by the front-end sensor network. After receiving the time-series monitoring data, the data access module is further used to use the highest frequency acceleration sampling time as the reference time axis, and to fill in the time nodes of the pore water pressure and the axial strain along the pipeline using zero-order hold or linear interpolation methods to generate a time-aligned multidimensional time-series data set.

3. The intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network according to claim 2, characterized in that, The differential sampling module assesses the soil consolidation state and controls the sampling frequency, specifically including: Extract the pore water pressure from the time-series monitoring data, and calculate the time derivative of the pore water pressure to obtain the rate of change of pore water pressure; The pore water pressure change rate is compared with a preset consolidation state determination threshold. If the rate of change of pore water pressure is greater than or equal to the consolidation state determination threshold, then the underlying hardware timer is controlled to switch to high-frequency sampling state. If the rate of change of pore water pressure is less than the consolidation state determination threshold, the soil is determined to have entered a sub-consolidation stable phase, and the underlying hardware timer is controlled to switch to low-frequency heartbeat sampling state.

4. The intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network according to claim 2, characterized in that, The differential sampling module calculates the differential displacement vector specifically by: The initial spatial geometric path along the pipeline is generated by fitting the initial spatial reference coordinates, and the unit tangent vector of the initial spatial geometric path is obtained. Obtain the corresponding physical length of the pipeline node on the optical fiber laying path, numerically integrate the axial strain along the pipeline in the time-series monitoring data along the initial spatial geometric path, and project it onto the unit tangent vector to derive the differential displacement vector containing displacement change components in three coordinate directions.

5. The intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network according to claim 2, characterized in that, The hardware protection module monitors vibration parameters specifically including: Extract the vibration acceleration from the time-series monitoring data, calculate the conjunction modulus of the vibration acceleration in three dimensions, and compare the conjunction modulus with a preset physical interface tolerance safety threshold. When the summation modulus is greater than or equal to the physical interface tolerance safety threshold, it is determined that the vibration exceeds the limit, the submission queue facing the main storage medium is blocked and a bus suspension command is issued, and the differential displacement vector is redirected to a non-volatile random access memory for temporary storage. When the combined modulus falls back to less than the physical interface tolerance safety threshold, and does not exceed the limit again within the preset seismic observation window, the submission queue is released, and the temporarily stored differential displacement vectors are written back and output in batches. If the combined modulus length is greater than or equal to the physical interface tolerance safety threshold again within the preset seismic observation window, the preset seismic observation window is reset, and the bus suspension command and differential displacement vector redirection temporary storage state are maintained.

6. The intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network according to claim 1, characterized in that, The topology verification module performs a comparison based on the spatial topology relationship of the pipeline network, specifically including: The received differential displacement vector is added to the corresponding initial spatial reference coordinates to obtain the current absolute spatial coordinates of the target pipeline node. Retrieve a pre-constructed and stored topological adjacency matrix reflecting the actual pipeline network layout, identify adjacent nodes of the target pipeline node based on the topological adjacency matrix, and use a spatial distance algorithm to calculate the real-time spatial distance between the current absolute spatial coordinates of the target pipeline node and the current absolute spatial coordinates of the adjacent nodes. The reference spatial distance between the target pipeline node and the adjacent node in the initial static state is retrieved, and the relative deformation is obtained by subtracting the calculated real-time spatial distance from the reference spatial distance.

7. The intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network according to claim 6, characterized in that, The topology verification module intercepts abnormal data specifically including: The absolute value of the obtained relative deformation is compared with the preset physical tolerance threshold, which is the physical limit strain constant. When the absolute value is greater than or equal to the preset physical tolerance threshold, it is determined that there is a risk of structural failure in the pipe segment between the adjacent nodes, the corresponding abnormal data is intercepted, and the pipe segment is marked as abnormal in the underlying topology status table. When the absolute value is less than the preset physical tolerance threshold, the differential displacement vector that has passed the verification is output.

8. The intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network according to claim 6, characterized in that, The differential storage module writes to the time-series database including: Retrieve the historical reference spatial coordinates from the last persistent storage, and use a spatial distance algorithm to calculate the scalar drift of the current absolute spatial coordinates relative to the historical reference spatial coordinates corresponding to the received verified differential displacement vector; When the scalar drift is greater than or equal to the preset differential storage trigger threshold, it is determined that a valid displacement has occurred. The current absolute spatial coordinates and the corresponding timestamp are encapsulated into an incremental data packet and persistently written into the time series database. The corresponding historical reference spatial coordinates in memory are overwritten with the current absolute spatial coordinates. When the scalar drift is less than the preset differential storage trigger threshold, it is determined that no effective displacement has occurred, the persistent write operation is not performed, and the corresponding historical reference space coordinates in memory remain unchanged.

9. The intelligent storage and retrieval device for construction data of a stockyard engineering pipeline network according to claim 8, characterized in that, The method of writing to the time-series database also includes maintaining an independent hardware clock counter, and at each preset snapshot period, ignoring the comparison results of the scalar drift, forcibly packaging the absolute spatial coordinates of all nodes in the network at the current moment and the underlying hardware operating status into a full state snapshot file, and persistently storing it in the time-series database.

10. The intelligent storage and retrieval device for construction data of a stockpile engineering pipeline network according to claim 9, characterized in that, The coordinate reconstruction module reconstructs and outputs the spatial coordinates at the requested time, specifically including: The target full state snapshot file with a generation time less than and closest to the requested time is retrieved in reverse order from the time series index of the time series database, and the absolute spatial coordinate components in the target full state snapshot file are extracted as the initial reference coordinates for reconstruction calculation. Extract all incremental data packets within the interval from the generation time of the target full state snapshot file to the requested time to form an update sequence; If the update sequence is not empty, then extract the last incremental data packet in the update sequence that has the largest timestamp and is less than or equal to the requested time, use zero-order hold logic to assign the absolute spatial coordinates corresponding to the incremental data packet to the requested time and output it; If the update sequence is empty, the initial reference coordinates are directly used as the spatial coordinates of the requested time and output.

Images