A refrigerated foundation temperature and humidity remote monitoring method and system based on the Internet of Things

Abstract

This invention belongs to the field of Internet of Things (IoT) monitoring and wireless communication transmission technology, specifically relating to an IoT-based remote monitoring method and system for the temperature and humidity of frozen foundations. The method includes: acquiring raw temperature and humidity data of the frozen foundation; obtaining preprocessed temperature and humidity sequences after filtering and detrending; calculating transient volatility and spatial consistency to obtain final correction weights; correcting the temperature and humidity using these weights; establishing node edges; calculating topological weight coefficients based on the corrected temperature and constructing a Laplace matrix; extracting the second smallest eigenvalue; determining the closure state of the frozen soil curtain based on the second smallest eigenvalue; and calculating the target output power ratio of the chiller unit by combining the energy deviation of the corrected temperature and humidity with the dynamic load evolution trend. This invention reduces construction and operation energy consumption and improves the monitoring accuracy of artificial ground freezing projects.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) monitoring and wireless communication transmission technology. More specifically, this invention relates to a method and system for remote monitoring of temperature and humidity in frozen foundations based on the Internet of Things. Background Technology

[0002] In artificial ground freezing projects, real-time acquisition of temperature evolution data and moisture distribution within the frozen foundation via an Internet of Things (IoT) system is a core method for assessing the overall strength of frozen soil walls, determining the state of frozen pipe connections, and preventing frost heave disasters. Since the formation process of frozen soil walls involves complex phase change heat conduction and moisture migration, monitoring data can provide construction teams with direct feedback on the development progress of the frozen soil curtain, thereby ensuring the safety of underground excavation operations.

[0003] Currently, the industry primarily uses buried sensor arrays for monitoring, transmitting sensor signals to a ground monitoring center via wired or wireless communication networks. Some systems have introduced basic automated early warning modules, triggering alarm signals when monitored data exceeds safe limits by setting instantaneous thresholds for temperature and humidity. With the development of IoT technology, low-power wide-area network technologies such as narrowband IoT are beginning to be applied to deep underground monitoring to solve the problems of signal penetration and long-term power supply in complex underground environments.

[0004] However, the existing technologies still have room for improvement in practical applications. The existing monitoring logic treats each measuring point as an isolated observation unit and lacks the ability to determine the integrity of the frozen soil curtain closure from a global topological perspective. When local sensors fail or geological conditions are heterogeneous, it is difficult to accurately assess whether a closed waterproof curtain has been formed between frozen soil columns by relying solely on single-point thresholds, resulting in a blind spot in the judgment. In addition, the power regulation of the refrigeration unit relies heavily on manual experience for step-by-step adjustments and lacks a mathematical control model based on the real-time evolution logic of temperature and humidity, resulting in a low degree of matching between cooling output and actual demand. Summary of the Invention

[0005] To address the aforementioned technical problem of inaccurate intersection determination and evaluation, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a method for remote monitoring of temperature and humidity in frozen foundations based on the Internet of Things, comprising: Raw temperature and humidity data from multiple temperature and humidity sensors within the frozen foundation are acquired and filtered and detrended to obtain preprocessed temperature and humidity sequences. The transient volatility and spatial consistency of the temperature and humidity sensors are calculated based on the local dispersion and neighborhood conduction bias of the preprocessed temperature sequences. A final correction weight is obtained by combining the transient volatility and spatial consistency. The preprocessed temperature and humidity sequences are corrected using the final correction weight to obtain corrected temperature and corrected humidity. Any temperature and humidity sensor is used as a node, and node edges are established between its neighboring sensors. The topological weight coefficients of each node edge are calculated based on the spatial distribution mean of the corrected temperature, and a Laplace matrix is ​​constructed using these coefficients. The second smallest eigenvalue of the Laplace matrix is ​​extracted. The closure state of the frozen soil curtain is determined based on the relationship between the second smallest eigenvalue and a preset interconnection threshold. The target output power ratio of the chiller unit is calculated based on the energy deviation of the corrected temperature and humidity and the dynamic load evolution trend.

[0007] This invention utilizes narrowband IoT technology to acquire raw temperature and humidity data within frozen foundations. Kalman filtering and detrending processing improve the quality of the basic data, and a final correction weight based on transient volatility and spatial consistency is introduced to reduce characteristic scale distortion caused by environmental interference, ensuring the logical consistency of temperature and humidity monitoring results. Furthermore, by establishing node edges and topological weight coefficients, the evolution logic of the frozen foundation is transformed into an algebraic connectivity assessment of the Laplace matrix. The second smallest eigenvalue is used to objectively determine the closure state of the permafrost curtain, reducing the risk of misjudgment from manual experience. By combining energy deviation and evolution trend to adjust the target output power ratio of the refrigeration unit, a closed-loop linkage between monitoring data and equipment control is achieved, which helps improve the construction safety and energy consumption control level of artificial ground freezing projects.

[0008] Preferably, the acquisition of the preprocessing temperature sequence and the preprocessing humidity sequence includes: Temperature and humidity sensors with integrated narrowband IoT communication modules are arranged at multiple depths inside the frozen foundation. The temperature and humidity sensors are connected to an industrial gateway with narrowband IoT transparent transmission function via RS485 bus. The industrial gateway collects raw temperature data and raw humidity data from all temperature and humidity sensors at a preset sampling frequency at high speed and uploads them to the IoT cloud platform. By adaptively filtering the raw temperature and humidity data of any temperature and humidity sensor using Kalman filtering, a preliminary purified temperature and humidity sequence for the corresponding temperature and humidity sensor is obtained. The temperature and humidity sequences obtained from the initial purification are detrended to obtain the preprocessed temperature and humidity sequences corresponding to the temperature and humidity sensors.

[0009] This invention deploys temperature and humidity sensors with integrated narrowband IoT modules at different depths within a frozen foundation, and utilizes an industrial gateway for high-speed data acquisition and cloud synchronization, ensuring stable transmission and large-scale connectivity coverage of deep stratum monitoring data. Kalman filtering is used to perform adaptive recursive correction on the raw temperature and humidity data, combined with first-order difference methods for detrending processing, thereby reducing systematic zero-point shifts caused by the high water pressure environment or stratum creep in deep foundations, which helps improve the baseline stability of the pre-processed temperature and humidity sequences.

[0010] Preferably, the transient volatility satisfies the following relationship: ; In the formula, Indicates the first A temperature and humidity sensor at all times Transient volatility; Indicates the window length; Indicates the first A temperature and humidity sensor is constantly inside the window. Temperature data; Indicates the first The average value of the preprocessed temperature sequence of a temperature and humidity sensor within a window; Indicates a preset micro value; Represents the absolute value symbol.

[0011] Preferably, the spatial consistency satisfies the following relation: ; In the formula, Indicates the first A temperature and humidity sensor at all times Spatial consistency; Indicates the first The total number of neighboring sensors of a temperature and humidity sensor; Indicates the first The initial purification temperature sequence of the temperature and humidity sensor at time [time] Temperature data; Indicates the first Within the neighborhood of the temperature and humidity sensor, the first... The initial purification temperature sequence of the temperature and humidity sensor at time [time] Temperature data; Represents the spatial temperature gradient; Indicates the first The temperature and humidity sensor and the first The straight-line distance between the temperature and humidity sensors; Indicates the spatial distance attenuation coefficient; This represents an exponential function with the natural constant as its base.

[0012] Preferably, the topological weight coefficients of each node edge satisfy the following relationship: ; In the formula, Indicates the first The temperature and humidity sensor, and the sensor in its neighborhood. The node edges between the temperature and humidity sensors at time... Topological weight coefficients; Indicates the sensitivity coefficient; Indicates the first A temperature and humidity sensor at all times The corrected temperature; Indicates the first Within the neighborhood of the temperature and humidity sensor, the first... A temperature and humidity sensor at all times The corrected temperature; Indicates the critical temperature of phase transition; This represents an exponential function with the natural constant as its base.

[0013] Preferably, the step of constructing a Laplacian matrix using the topological weight coefficients and extracting the second smallest eigenvalue of the Laplacian matrix includes: A Laplace matrix is ​​constructed using the topological weight coefficients of all node edges. The negative value of the topological weight coefficient corresponding to each node edge is filled into the corresponding off-diagonal element position. At the same time, the sum of the topological weight coefficients of the corresponding temperature and humidity sensor and all its neighboring sensors is filled into each diagonal position of the Laplace matrix. For temperature and humidity sensors without neighboring sensors, the corresponding Laplace matrix element value is uniformly recorded as zero. The Laplacian matrix is ​​decomposed using a power-law algorithm to extract the second smallest eigenvalue.

[0014] This invention utilizes the topological weight coefficients of all node edges to establish a Laplace matrix, transforming the local temperature distribution mean into the algebraic structural features of the matrix. The sum of the topological weight coefficients is introduced at the diagonal position of the Laplace matrix to assess the correlation strength between nodes. By extracting the second smallest eigenvalue, the core index for evaluating the global connectivity of frozen foundations is obtained. This simplifies the complex spatial temperature and humidity evolution process into an algebraic connectivity determination using a graph theory model, enabling the capture of subtle interlocking patterns within the permafrost curtain and reducing inaccurate identification of weak points in the curtain due to geological heterogeneity.

[0015] Preferably, the closed state of the permafrost curtain includes: When the second smallest feature value is greater than or equal to the preset connection threshold, it is determined that the frozen soil curtain has formed a continuous closed body with waterproof function, and the connection completion determination signal is output to the Internet of Things cloud platform; when the second smallest feature value is less than the preset connection threshold, it is determined that the frozen soil curtain has not yet formed a connection, triggering the calculation of the target output power ratio of the chiller unit, and sending it to the chiller unit in real time to perform enhanced freezing control.

[0016] Preferably, the calculation of the target output power ratio of the refrigeration unit includes: The target output power ratio of the refrigeration unit is obtained by summing the difference between the current corrected temperature and the target maintained temperature of all temperature and humidity sensors, the product of the local influence volume of the temperature and humidity sensors, multiplying by the temperature contribution weighting factor, and adding the product of the ratio of the current corrected humidity average value of all temperature and humidity sensors to the previous corrected humidity average value, the inverse of the sampling frequency, and the humidity disturbance compensation factor.

[0017] This invention assesses the volumetric effect of foundation heat exchange by accumulating the product of corrected temperature deviation and local influence volume, and uses the rate of change of the corrected mean humidity over time to perform feedback compensation for the latent heat disturbance caused by moisture migration. This allows the cooling output to adaptively match the actual needs of frozen soil curtain development, thereby reducing energy waste caused by excessive cooling and ensuring sufficient cooling support during critical transition stages. This improves the energy efficiency of the construction process and ensures the quality and safety of frozen soil curtain construction under complex geological conditions.

[0018] Preferably, the acquisition of the corrected temperature and corrected humidity includes: The corrected temperature at the current moment is obtained by multiplying the final correction weight of any temperature and humidity sensor with the temperature data at the current moment, and adding the complement of the final correction weight with the average temperature of the corresponding neighborhood sensors at the current moment. The corrected humidity at the current moment is obtained by multiplying the final correction weight of any temperature and humidity sensor by the humidity data at the current moment, and adding the complement of the final correction weight by the product of the average humidity of the corresponding neighboring sensors at the current moment.

[0019] This invention obtains corrected temperature and corrected humidity by fusing monitoring data from a single measuring point with the average temperature and humidity values ​​of corresponding neighboring sensors. By utilizing the spatiotemporal coupling characteristics of the temperature and humidity field within the frozen foundation to perform complementary calibration of neighboring data, it reduces the misleading effect on the overall monitoring results caused by physical damage or signal drift of a single temperature and humidity sensor, enhances the system's fault tolerance and robustness under harsh underground conditions, and helps maintain the continuity and reliability of the frozen soil curtain closure determination process.

[0020] Secondly, the present invention provides an Internet of Things-based remote monitoring system for temperature and humidity of frozen foundation, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned Internet of Things-based remote monitoring method for temperature and humidity of frozen foundation is implemented.

[0021] By adopting the above technical solution, a computer program for remote monitoring of temperature and humidity of frozen foundation based on the Internet of Things is generated and stored in a memory so that it can be loaded and executed by a processor. A terminal device can then be made based on the memory and the processor for convenient use.

[0022] The beneficial effects of this invention are as follows: (1) This invention uses narrowband Internet of Things technology to acquire temperature evolution data and moisture distribution inside frozen foundations in real time. It uses Kalman filtering and first-order difference method to improve the smoothness of monitoring data, thereby reducing the systematic zero-point offset caused by the high water pressure environment of deep foundations or stratum creep. It also introduces the final correction weight based on transient volatility and spatial consistency to reduce the characteristic scale distortion caused by single-point failure or local electromagnetic interference of temperature and humidity sensors, and ensure the continuity and authenticity of monitoring data in physical evolution. Furthermore, it transforms discrete measurement point data into the algebraic connectivity evaluation of the Laplace matrix, and uses the second smallest eigenvalue to objectively determine the closure state of the frozen soil curtain, thereby reducing the judgment blind spot and misjudgment risk brought by the traditional empirical judgment method. (2) The present invention calculates the target output power ratio of the refrigeration unit through a continuous control function, so that the output of the refrigeration capacity can adaptively match the actual physical needs of the development of the permafrost curtain. While ensuring sufficient cooling capacity support in the key stage of the transition, it reduces energy waste caused by excessive refrigeration and improves the energy efficiency and construction safety level of the artificial stratum freezing project. Attached Figure Description

[0023] Figure 1 The flowchart illustrates a remote monitoring method for temperature and humidity of frozen foundations based on the Internet of Things in this invention. Figure 2 A schematic diagram illustrating the final corrected weight distribution of the temperature and humidity sensor is shown. Figure 3 This schematic diagram illustrates the connectivity constructed based on the topological weight coefficients of each node's edges. Detailed Implementation

[0024] This invention discloses a method for remote monitoring of temperature and humidity in frozen foundations based on the Internet of Things (IoT), referring to... Figure 1 This includes steps S100-S400: S100: Acquire raw temperature and humidity data from multiple temperature and humidity sensors within the frozen foundation, and perform filtering and detrending processing to obtain preprocessed temperature and humidity sequences.

[0025] It should be noted that, since the original signal acquisition process is susceptible to random interference from deep water pressure fluctuations and electromagnetic environment, and considering that high-frequency discrete sampling can capture the state switching during the evolution of formation temperature and humidity, this invention uses narrowband Internet of Things technology to achieve synchronous acquisition of original data from multiple points in the formation, thereby reducing the loss of characteristic signals due to excessively long sampling periods and improving the real-time perception of the underlying physical state by the monitoring terminal.

[0026] Specifically, temperature and humidity sensors, each integrated with a narrowband IoT communication module, are deployed at multiple depths within the frozen foundation. It should be noted that these temperature and humidity sensors employ a PT100 high-precision temperature probe and a capacitive polymer thin-film humidity-sensitive element.

[0027] The temperature and humidity sensors are connected to an industrial gateway with narrowband IoT pass-through functionality via an RS485 bus. The industrial gateway rapidly acquires raw temperature and humidity data from all the sensors at a preset sampling frequency and uploads it to an IoT cloud platform. For example, the sampling frequency is set to 100Hz.

[0028] Adaptive filtering is applied to the raw temperature and humidity data of any temperature and humidity sensor to obtain a preliminary purified temperature and humidity sequence for the corresponding sensor. For example, the adaptive filtering is a Kalman filter. Kalman filtering recursively corrects noisy measurements to extract the closest approximate temperature and humidity estimates from the interference-laden raw temperature and humidity data. Kalman filtering is existing technology and will not be elaborated upon here.

[0029] The initially purified temperature and humidity sequences are detrended to obtain the pre-processed temperature and humidity sequences for the corresponding temperature and humidity sensors. For example, the detrending process is a first-order difference method. The first-order difference method can eliminate the systematic zero-point shift of the temperature and humidity composite sensor caused by deep water pressure or formation creep. The first-order difference method is existing technology and will not be elaborated upon here.

[0030] Thus, the preprocessed temperature sequence and the preprocessed humidity sequence were obtained.

[0031] S200. Calculate the transient fluctuation rate and spatial consistency of the temperature and humidity sensor based on the local dispersion of the preprocessed temperature sequence and the neighborhood conduction deviation.

[0032] It should be noted that, due to the impact of abnormal leakage from the freezing pipe on local strata, the signals fed back by the temperature and humidity sensors may exhibit non-stationary jumps. Considering that the degree of discrete deviation within the time-domain window can reflect the fluctuation intensity of the data relative to the static evolution trend, this invention introduces a transient volatility rate based on window mean normalization to assess the severity of environmental random noise interference on single-point data, thereby suppressing the impact of abnormal disturbance data on subsequent judgment conclusions and improving the signal-to-noise ratio of the monitoring sequence.

[0033] Specifically, the transient fluctuation rate of all temperature and humidity sensors is calculated based on the intensity of temperature amplitude disturbances, including: In the In the preprocessed temperature sequence of a temperature and humidity sensor, a window of length L is set with the current time as the end point, and the start time of the window is determined to be... For example, L is set to 10, which can be determined based on the thermal inertia characteristics of the frozen foundation and the sampling frequency of the narrowband IoT communication module.

[0034] The transient fluctuation rate of any temperature and humidity sensor satisfies the following relationship: ; In the formula, Indicates the first A temperature and humidity sensor at all times Transient volatility; Indicates the window length; Indicates the first A temperature and humidity sensor is constantly inside the window. Temperature data; Indicates the first The average value of the preprocessed temperature sequence of a temperature and humidity sensor within a window; This represents a preset microvalue to prevent the denominator from being 0; it can be set to 0.001. Represents the absolute value symbol.

[0035] In this relation, It represents the absolute deviation of the temperature value at a single moment from the average temperature within the window. This represents the sum of the dispersion of all times within the window relative to the average temperature within the window. The larger the sum of dispersion, the more random noise is mixed into the temperature data of the temperature and humidity sensor, or the sensor is experiencing signal drift and physical environmental interference, such as severe seepage of local groundwater carrying away cold energy, resulting in abnormal temperature fluctuations. In this case, the reliability of the temperature data from the temperature and humidity sensor decreases. Conversely, it indicates that the temperature data from the temperature and humidity sensor does not have outlier jumps, reflecting the stability of the cryowall development and the high reliability of data acquisition.

[0036] Thus, the transient fluctuation rates of all temperature and humidity sensors were obtained.

[0037] It should be noted that, as the temperature field evolution of adjacent temperature and humidity sensors follows the physical laws of spatial heat conduction during the development of the permafrost curtain, a logically consistent gradient relationship should be maintained between temperature and humidity sensors. Considering that the deviation between the measured spatial temperature difference and the steady-state heat conduction expectation can identify isolated anomalies, this invention establishes a neighborhood temperature difference expectation deviation evaluation mechanism based on the spatial arrangement position, thereby reducing logical conflicts caused by soft faults of single-point temperature and humidity sensors and improving the reliability of the monitoring network in the spatial distribution dimension.

[0038] Preferably, the spatial consistency of all temperature and humidity sensors is calculated based on the expected deviation of temperature difference within the neighborhood of each temperature and humidity sensor, including: Calculate the first using the Euclidean distance formula The linear distance between a temperature and humidity sensor and any other temperature and humidity sensor is defined by a preset distance threshold. If the linear distance is less than or equal to the distance threshold, the corresponding temperature and humidity sensor is considered the first one. The neighboring sensors of a temperature and humidity sensor; if the straight-line distance is greater than a distance threshold, it indicates that there is no direct thermal conduction relationship between the two temperature and humidity sensors, and therefore the temperature and humidity sensor is not included in the count. A neighboring sensor of a temperature and humidity sensor.

[0039] The spatial consistency of any temperature and humidity sensor satisfies the following relationship: ; In the formula, Indicates the first A temperature and humidity sensor at all times Spatial consistency; Indicates the first The total number of neighboring sensors of a temperature and humidity sensor; Indicates the first The initial purification temperature sequence of the temperature and humidity sensor at time [time] Temperature data; Indicates the first Within the neighborhood of the temperature and humidity sensor, the first... The initial purification temperature sequence of the temperature and humidity sensor at time [time] Temperature data; The spatial temperature gradient is a theoretical value determined by the steady-state heat conduction formula based on the distribution location of the freezing pipes. Indicates the first The temperature and humidity sensor and the first The straight-line distance between the temperature and humidity sensors; This represents the spatial distance attenuation coefficient, set to 2.5; This represents an exponential function with the natural constant as its base.

[0040] In this relation, This indicates the deviation between the measured temperature difference and the theoretically expected temperature difference. The larger the value, the more likely it is to be the first. The measured temperature evolution of the temperature and humidity sensor deviated from the preset physical laws of heat conduction. Indicates the first The cumulative spatial logic distortion of a temperature and humidity sensor relative to all its neighboring sensors; the larger the cumulative spatial logic distortion, the more significant the increase in the number of sensors. The temperature and humidity sensor has extremely low physical correlation with multiple neighboring sensors, presenting itself as an isolated point of abrupt change in spatial dimension; under the influence of the exponential function, spatial consistency will decrease, determining the... The first temperature and humidity sensor is severely affected by environmental interference and has low reliability; conversely, it indicates that the second... The temperature and humidity sensor shows a high degree of matching with the evolution trend of the neighborhood temperature field, and the spatial consistency tends to be 1, indicating that the first temperature and humidity sensor... The temperature and humidity sensor is highly reliable.

[0041] This achieves spatial consistency across all temperature and humidity sensors.

[0042] S300. Combining the transient volatility and spatial consistency, the final correction weight is obtained; the preprocessed temperature sequence and preprocessed humidity sequence are corrected according to the final correction weight to obtain the corrected temperature and corrected humidity; taking any temperature and humidity sensor as a node, and establishing node edges between its neighboring sensors; calculating the topological weight coefficient of each node edge according to the spatial distribution mean of the corrected temperature, and using the topological weight coefficient to construct a Laplace matrix, and extracting the second smallest eigenvalue of the Laplace matrix.

[0043] It should be noted that the temperature drop of local temperature and humidity sensors does not equate to the formation of ice crystal connections with physical strength between nodes. Considering that the deviation of the average temperature between adjacent temperature and humidity sensors from the freezing point can reflect the probability of phase change in the soil in that area, this invention improves the analytical accuracy of the monitoring model for the spatial evolution process of the frozen soil curtain by converting the temperature evolution state into a confidence ratio describing the connectivity strength between temperature and humidity sensors, and suppresses misjudgments of interconnection caused by ignoring the connectivity quality between nodes based solely on single-point compliance.

[0044] Specifically, based on the freezing phase transition connectivity probability between adjacent temperature and humidity sensors, the topological weight coefficients of all node edges are calculated, including: The spatial consistency of any temperature and humidity sensor is calculated and multiplied by a power function with the natural constant as the base and the negative of the transient volatility as the exponent, to obtain the final corrected weight of the corresponding temperature and humidity sensor.

[0045] For example, Figure 2 This is a schematic diagram of the final correction weight distribution for the temperature and humidity sensor. The horizontal axis represents the sensor number, and the vertical axis represents the final correction weight value. The diagram reflects the real-time correction and interference identification results for each measurement point based on transient fluctuation rate and spatial consistency; higher values ​​indicate higher reliability of the temperature and humidity sensor.

[0046] The corrected temperature at the current moment is obtained by multiplying the final correction weight of any temperature and humidity sensor by the temperature data at the current moment, and adding the complement of the final correction weight by the product of the average temperature of the corresponding neighboring sensors at the current moment.

[0047] The corrected humidity at the current moment is obtained by multiplying the final correction weight of any temperature and humidity sensor by the humidity data at the current moment, and adding the complement of the final correction weight by the product of the average humidity of the corresponding neighboring sensors at the current moment.

[0048] Use any temperature and humidity sensor as a node, and establish node edges between its neighboring sensors.

[0049] The topological weight coefficients of edges at any node satisfy the following relationship: ; In the formula, Indicates the first The temperature and humidity sensor, and the sensor in its neighborhood. The node edges between the temperature and humidity sensors at time... Topological weight coefficients; This represents the sensitivity coefficient, set to 1.2, based on the nonlinear evolution law of soil phase transformation characteristics. Indicates the first A temperature and humidity sensor at all times The corrected temperature; Indicates the first Within the neighborhood of the temperature and humidity sensor, the first... A temperature and humidity sensor at all times The corrected temperature; This represents the critical temperature for phase transition, set to 0℃. This represents an exponential function with the natural constant as its base.

[0050] In this relation, This indicates that the two temperature and humidity sensors are at time... The higher the total temperature, the higher the local heat reserve, and the warmer the soil as a whole, which is not conducive to the formation of physical bonds in the frozen soil; conversely, the lower the total temperature, the lower the overall temperature of the region, and the more suitable it is for the accumulation of cold energy. This indicates the degree of deviation of the average temperature of the medium between the two temperature and humidity sensors from the freezing point of the phase transition. The greater the deviation, the more the average temperature between the two temperature and humidity sensors is much higher than the freezing point, and the soil medium is still in a liquid or molten state, lacking the physical conditions for topological connectivity. Conversely, it indicates that the soil between the two temperature and humidity sensors has undergone a phase transition or has already frozen.

[0051] At this point, the topological weight coefficients of all node edges have been obtained.

[0052] It should be noted that since the frozen connectivity between local temperature and humidity sensors only represents the physical evolution of the geological microstructure, and considering that the Laplace matrix in algebraic graph theory can integrate the local connectivity properties of temperature and humidity sensors across the entire field, and that its spectral characteristics can describe the global algebraic connectivity of the network structure, this invention improves the quantitative assessment of the overall closure state of the permafrost curtain by extracting the second smallest eigenvalue reflecting the weakest connectivity state across the entire field as the criterion, and suppresses misjudgments of the interlocking state caused by uneven spatial distribution of measuring points.

[0053] Preferably, a Laplace matrix is ​​constructed using the topological weight coefficients of all node edges. The negative value of the topological weight coefficient corresponding to each node edge is filled into the corresponding off-diagonal element position. At the same time, the sum of the topological weight coefficients of the corresponding temperature and humidity sensor and all its neighboring sensors is filled into each diagonal position of the Laplace matrix. For temperature and humidity sensors without neighboring sensors, the corresponding Laplace matrix element value is uniformly recorded as zero.

[0054] A power-law iteration algorithm is used to perform eigenvalue decomposition on the Laplacian matrix, and the second smallest eigenvalue in the Laplacian matrix is ​​extracted as the core indicator for evaluating the global connectivity of the permafrost foundation. It should be noted that this second smallest eigenvalue dynamically increases with the radial expansion and fusion of the permafrost columns between the underground temperature and humidity sensor nodes, and is used to characterize the overall closure of the topological structure of all node edges within the monitoring area. It should be noted that the power-law iteration algorithm is existing technology and will not be elaborated upon here.

[0055] For example, Figure 3 This is a connectivity diagram constructed based on the topological weight coefficients of each node edge. The diagram uses temperature and humidity sensors as nodes, and establishes node edges between neighboring sensors with thermal conduction correlations. The density of the node edge connections reflects the topological connectivity and global closure state of the permafrost curtain.

[0056] S400. Based on the relationship between the second smallest characteristic value and the preset interconnection threshold, determine the closure state of the frozen soil curtain, and calculate the target output power ratio of the chiller unit based on the energy deviation of the corrected temperature and humidity and the dynamic load evolution trend.

[0057] It should be noted that, since the heat load requirements of the refrigeration system vary at different development stages, and the dynamic convection heat transfer caused by underground seepage is the core reason for the loss of cold energy, considering that the deviation of the measured temperature from the target can reflect the amount of cold storage that the current stratum still needs to replenish, and the rate of change of humidity can reflect the rate at which groundwater carries away the cold energy, this invention adopts a comprehensive correction strategy based on the total energy deviation of the stratum and the dynamic migration load of moisture when the non-circular state is determined, thereby suppressing the energy redundancy caused by blind constant power operation and improving the response matching effect between the output power of the refrigeration unit and the actual load of the stratum.

[0058] Specifically, when the second smallest feature value is greater than or equal to the preset connection threshold, it is determined that the frozen soil curtain has formed a continuous closed body with waterproof function, and a connection completion determination signal is output to the IoT cloud platform; when the second smallest feature value is less than the preset connection threshold, it is determined that the frozen soil curtain has not yet formed a connection, triggering the calculation of the target output power ratio of the chiller unit, and sending it to the chiller unit in real time to execute enhanced freezing control. For example, the connection threshold is set to 0.1, which is comprehensively set based on the design perimeter of the frozen soil curtain and the arrangement density of temperature and humidity sensors.

[0059] Based on the combined value of the total heat deviation and the dynamic moisture migration load, the target output power ratio of the chiller unit is calculated, including: Divide the total design volume of the frozen soil curtain by the total number of temperature and humidity sensors to obtain the local influence volume of the temperature and humidity sensors.

[0060] The target output power ratio satisfies the following relationship: ; In the formula, Indicates the time of the refrigeration unit The target output power ratio; This represents the preset temperature contribution weighting factor, used to eliminate dimensions; Indicates the total number of deployed temperature and humidity sensors; Indicates the first A temperature and humidity sensor at all times The corrected temperature; Indicates the target temperature to be maintained, set to ℃; This indicates the localized volume of influence of the temperature and humidity sensor. This represents the preset humidity disturbance compensation factor, used to eliminate dimensions; This indicates that all temperature and humidity sensors are at time [time missing]. The corrected average humidity; This indicates that all temperature and humidity sensors were at the previous moment. The corrected average humidity; It represents the time interval between two consecutive sensor data acquisitions, and is equal to the reciprocal of the sampling frequency.

[0061] In this relation, This indicates the total heat deviation of the permafrost medium relative to the target freezing state. The larger the total heat deviation, the higher the overall temperature of the current permafrost curtain is than the target maintenance temperature, and the more heat is accumulated in the stratum. The refrigeration unit needs to output a large amount of cooling capacity to fill the energy gap. Conversely, it indicates that the overall temperature field of the stratum has approached or reached the target maintenance temperature, and the energy state tends to be stable. At this time, the refrigeration unit only needs a small amount of power to maintain the status quo. This indicates the intensity of convective heat transfer caused by groundwater migration. The greater the intensity of convective heat transfer, the more intense the movement of water within the stratum, and the faster the groundwater carries away the cold, generating a dynamic convective heat load. The system must increase its power to offset this heat loss and prevent the ice wall from being eroded into weak points. Conversely, a lower intensity indicates that the humidity is stable, the water within the stratum is basically still or has been frozen and locked, there is no obvious convective heat transfer, and no additional compensation power is required.

[0062] Thus, the adaptive power adjustment of the refrigeration unit based on real-time monitoring data was completed.

[0063] This invention also discloses an Internet of Things (IoT)-based remote monitoring system for the temperature and humidity of frozen foundations, comprising a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement an IoT-based remote monitoring method for the temperature and humidity of frozen foundations according to the present invention.

[0064] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0065] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.

Claims

1. A method for remote monitoring of temperature and humidity in frozen foundations based on the Internet of Things, characterized in that, include: The raw temperature and humidity data from multiple temperature and humidity sensors within the frozen foundation were acquired and then filtered and detrended to obtain preprocessed temperature and humidity sequences. Based on the local dispersion and neighborhood conduction bias of the preprocessed temperature sequence, the transient volatility and spatial consistency of the temperature and humidity sensor are calculated. The final corrected weights are obtained by combining the transient volatility and spatial consistency. The preprocessed temperature and humidity sequences are corrected according to the final correction weights to obtain the corrected temperature and corrected humidity; arbitrary temperature and humidity sensors are used as nodes, and node edges are established between their neighboring sensors; the topological weight coefficients of each node edge are calculated according to the spatial distribution mean of the corrected temperature, and the Laplace matrix is ​​constructed using the topological weight coefficients, and the second smallest eigenvalue of the Laplace matrix is ​​extracted. Based on the relationship between the second smallest eigenvalue and the preset interconnection threshold, the closure state of the frozen soil curtain is determined, and the target output power ratio of the chiller unit is calculated based on the energy deviation of the corrected temperature and humidity and the dynamic load evolution trend.

2. The method for remote monitoring of temperature and humidity of frozen foundation based on the Internet of Things according to claim 1, characterized in that, The acquisition of the preprocessing temperature sequence and the preprocessing humidity sequence includes: Temperature and humidity sensors with integrated narrowband IoT communication modules are arranged at multiple depths inside the frozen foundation. The temperature and humidity sensors are connected to an industrial gateway with narrowband IoT transparent transmission function via RS485 bus. The industrial gateway collects raw temperature data and raw humidity data from all temperature and humidity sensors at a preset sampling frequency at high speed and uploads them to the IoT cloud platform. By adaptively filtering the raw temperature and humidity data of any temperature and humidity sensor using Kalman filtering, a preliminary purified temperature and humidity sequence for the corresponding temperature and humidity sensor is obtained. The temperature and humidity sequences obtained from the initial purification are detrended to obtain the preprocessed temperature and humidity sequences corresponding to the temperature and humidity sensors.

3. The method for remote monitoring of temperature and humidity of frozen foundation based on the Internet of Things according to claim 1, characterized in that, The transient volatility satisfies the following relationship: ； In the formula, Indicates the first A temperature and humidity sensor at all times Transient volatility; Indicates the window length; Indicates the first A temperature and humidity sensor is constantly inside the window. Temperature data; Indicates the first The average value of the preprocessed temperature sequence of a temperature and humidity sensor within a window; Indicates a preset micro value; Represents the absolute value symbol.

4. The method for remote monitoring of temperature and humidity of frozen foundation based on the Internet of Things according to claim 1, characterized in that, The spatial consistency satisfies the following relation: ； In the formula, Indicates the first A temperature and humidity sensor at all times Spatial consistency; Indicates the first The total number of neighboring sensors of a temperature and humidity sensor; Indicates the first The initial purification temperature sequence of the temperature and humidity sensor at time [time] Temperature data; Indicates the first Within the neighborhood of the temperature and humidity sensor, the first... The initial purification temperature sequence of the temperature and humidity sensor at time [time] Temperature data; Represents the spatial temperature gradient; Indicates the first The temperature and humidity sensor and the first The straight-line distance between the temperature and humidity sensors; Indicates the spatial distance attenuation coefficient; This represents an exponential function with the natural constant as its base.

5. The method for remote monitoring of temperature and humidity of frozen foundation based on the Internet of Things according to claim 1, characterized in that, The topological weight coefficients of each node edge satisfy the following relationships: ； In the formula, Indicates the first The temperature and humidity sensor, and the sensor in its neighborhood. The node edges between the temperature and humidity sensors at time... Topological weight coefficients; Indicates the sensitivity coefficient; Indicates the first A temperature and humidity sensor at all times The corrected temperature; Indicates the first Within the neighborhood of the temperature and humidity sensor, the first... A temperature and humidity sensor at all times The corrected temperature; Indicates the critical temperature of phase transition; This represents an exponential function with the natural constant as its base.

6. The method for remote monitoring of temperature and humidity of frozen foundation based on the Internet of Things according to claim 1, characterized in that, The step of constructing a Laplacian matrix using the topological weight coefficients and extracting the second smallest eigenvalue of the Laplacian matrix includes: A Laplace matrix is ​​constructed using the topological weight coefficients of all node edges. The negative value of the topological weight coefficient corresponding to each node edge is filled into the corresponding off-diagonal element position. At the same time, the sum of the topological weight coefficients of the corresponding temperature and humidity sensor and all its neighboring sensors is filled into each diagonal position of the Laplace matrix. For temperature and humidity sensors without neighboring sensors, the corresponding Laplace matrix element value is uniformly recorded as zero. The Laplacian matrix is ​​decomposed using a power-law algorithm to extract the second smallest eigenvalue.

7. The method for remote monitoring of temperature and humidity of frozen foundation based on the Internet of Things according to claim 1, characterized in that, The closed state of the permafrost curtain includes: When the second smallest feature value is greater than or equal to the preset connection threshold, it is determined that the frozen soil curtain has formed a continuous closed body with waterproof function, and the connection completion determination signal is output to the Internet of Things cloud platform; when the second smallest feature value is less than the preset connection threshold, it is determined that the frozen soil curtain has not yet formed a connection, triggering the calculation of the target output power ratio of the chiller unit, and sending it to the chiller unit in real time to perform enhanced freezing control.

8. The method for remote monitoring of temperature and humidity of frozen foundation based on the Internet of Things according to claim 1, characterized in that, The calculation of the target output power ratio of the chiller unit includes: The target output power ratio of the refrigeration unit is obtained by summing the difference between the current corrected temperature and the target maintained temperature of all temperature and humidity sensors, the product of the local influence volume of the temperature and humidity sensors, multiplying by the temperature contribution weighting factor, and adding the product of the ratio of the current corrected humidity average value of all temperature and humidity sensors to the previous corrected humidity average value, the inverse of the sampling frequency, and the humidity disturbance compensation factor.

9. A method for remote monitoring of temperature and humidity in frozen foundations based on the Internet of Things, as described in claim 1, is characterized in that... The acquisition of the corrected temperature and corrected humidity includes: The corrected temperature at the current moment is obtained by multiplying the final correction weight of any temperature and humidity sensor with the temperature data at the current moment, and adding the complement of the final correction weight with the average temperature of the corresponding neighborhood sensors at the current moment. The corrected humidity at the current moment is obtained by multiplying the final correction weight of any temperature and humidity sensor by the humidity data at the current moment, and adding the complement of the final correction weight by the product of the average humidity of the corresponding neighboring sensors at the current moment.

10. A remote monitoring system for temperature and humidity of frozen foundation based on the Internet of Things, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a remote monitoring method for temperature and humidity of frozen foundations based on the Internet of Things, as described in any one of claims 1-9.

Images