Iron tower foundation environment data fusion control method and system for low temperature environment

Abstract

The application provides a tower foundation environment data fusion control method and system for a low-temperature environment, relates to the technical field of data fusion, and comprises the following steps: configuring a tower foundation environment sensing network; performing preliminary abnormal sensing on a plurality of environment sensing data sets; inputting the plurality of environment abnormal sensing data sets into a low-temperature collection frequency adjustment module to perform low-temperature environment abnormal filtering, generating an adaptive collection frequency according to the filtered plurality of environment abnormal sensing data sets, and obtaining a plurality of updated environment sensing data sets; judging whether a mode switching instruction is triggered, and if the mode switching instruction is triggered, performing fusion processing on the plurality of updated environment sensing data sets through a central data fusion mode. The application can solve the technical problem of low tower foundation environment data fusion control precision in the prior art, and achieve the technical effect of improving the tower foundation environment data fusion control precision.

Description

Technical Field

[0001] This application relates to the field of data fusion technology, and in particular to a data fusion control method and system for tower foundations in low-temperature environments. Background Technology

[0002] With the continuous expansion of power transmission networks and the increasing complexity of tower operating environments, environmental monitoring and safety assessment technologies for tower structures are gradually becoming key means to ensure line stability. In high-altitude, cold, windy, or mountainous areas, the temperature, humidity, wind load, and vibration load of the environment where towers are located often exhibit significant spatiotemporal fluctuations. Sudden environmental events can occur within a very short time, thus requiring continuous and high-precision environmental sensing based on large-scale distributed sensors. However, due to the significant spatial and temporal differences in monitoring scenarios, and the dynamic changes in the effectiveness and importance of different monitoring points, traditional monitoring systems often employ fixed acquisition frequencies and fixed data processing modes, making it difficult to adapt to the monitoring needs of multiple scenarios, multiple loads, and strong fluctuations.

[0003] In summary, existing technologies suffer from technical problems such as insufficient accuracy in environmental anomaly detection, incomplete structural status assessment, and low data processing efficiency due to fixed acquisition frequency, inflexible anomaly identification, coarse damage analysis models, and lack of mode switching. These issues further affect the safety of tower operation, the real-time performance of monitoring, and the reliability of large-scale monitoring deployments. Summary of the Invention

[0004] The purpose of this application is to provide a data fusion control method and system for tower foundations in low-temperature environments, in order to solve the technical problems in the prior art that result in insufficient accuracy of environmental anomaly detection, incomplete structural status assessment, and low data processing efficiency due to fixed acquisition frequency, inflexible anomaly identification, coarse damage analysis model, and lack of mode switching. These problems further affect the safety of tower operation, real-time monitoring, and reliability of large-scale monitoring deployment.

[0005] In view of the above problems, this application provides a method and system for environmental data fusion control of iron tower foundations in low-temperature environments.

[0006] Firstly, this application provides a data fusion control method for tower foundation environmental conditions in low-temperature environments, implemented through a data fusion control system for tower foundation environmental conditions in low-temperature environments. The method includes: configuring a tower foundation environmental sensing network, wherein the network includes multiple environmental sensors, each independently connected via a low-temperature acquisition frequency adjustment module; acquiring multiple environmental perception datasets from the multiple environmental sensors based on the network, performing preliminary anomaly detection on these datasets to obtain multiple environmental anomaly perception datasets; inputting these datasets into the low-temperature acquisition frequency adjustment module for low-temperature environmental anomaly filtering, generating an adaptive acquisition frequency based on the filtered datasets, and acquiring multiple updated environmental perception datasets using the adaptive acquisition frequency; determining whether a mode switching command is triggered based on the updated datasets, and if so, fusing the updated datasets using a central data fusion mode.

[0007] Preferably, the method for controlling the environmental data fusion of tower foundations in low-temperature environments further includes: configuring an adaptive acquisition frequency threshold, comparing the multiple adaptive acquisition frequencies corresponding to the multiple updated environmental perception datasets with the adaptive acquisition frequency threshold, and obtaining the proportion of quantities greater than the adaptive acquisition frequency threshold; if the proportion of quantities is greater than a preset proportion threshold, triggering a mode switching instruction, wherein the mode switching instruction is used to switch the edge data fusion mode to the center data fusion mode.

[0008] Preferably, the data fusion control method for tower foundation environment in low-temperature environment further includes: acquiring an edge computing unit under edge data fusion mode, analyzing the configuration parameters of the edge computing unit to obtain edge computing resource indicators and edge computing power scaling indicators; and optimizing the adaptive acquisition frequency threshold according to the edge computing resource indicators and edge computing power scaling indicators to obtain an adaptive acquisition frequency threshold in which the probability of missed risk is within the expected probability.

[0009] Preferably, the method for controlling the environmental data fusion of tower foundations in low-temperature environments further includes: constructing historical environmental health perception data samples based on the multiple environmental sensors; training multiple environmental anomaly identification models with the historical environmental health perception data samples; performing preliminary anomaly perception on the multiple environmental perception datasets based on the multiple environmental anomaly identification models; and obtaining multiple environmental anomaly perception datasets.

[0010] Preferably, the method for data fusion control of tower foundation environment in low-temperature environments further includes: acquiring a low-temperature environment sensing data fingerprint database, wherein the low-temperature environment sensing data fingerprint database includes a first type of low-temperature event, a second type of low-temperature event, and a third type of low-temperature event, wherein the first type of low-temperature event is a sudden low-temperature anomaly event, the second type of low-temperature event is a continuous low-temperature anomaly event, and the third type of low-temperature event is a drifting low-temperature anomaly event; the low-temperature acquisition frequency adjustment module performs data anomaly fingerprint matching on the multiple environmental anomaly sensing datasets by calling the low-temperature environment sensing data fingerprint database to acquire a low-temperature environment anomaly sensing dataset; and obtaining multiple filtered environmental anomaly sensing datasets based on the low-temperature environment anomaly sensing datasets.

[0011] Preferably, the method for controlling the environmental data fusion of tower foundations in low-temperature environments further includes: optimizing the compensation acquisition frequency based on the low-temperature environment anomaly perception dataset to obtain a low-temperature compensation acquisition frequency where the low-temperature data restoration quality is greater than a preset quality threshold; and updating the adaptive acquisition frequency based on the low-temperature compensation acquisition frequency to obtain the updated adaptive acquisition frequency.

[0012] Preferably, the method for controlling the environmental data fusion of iron tower foundations in low-temperature environments further includes: identifying the abnormal temporal distribution characteristics of multiple filtered environmental anomaly perception datasets; calculating the anomaly perception confidence level of the current acquisition frequency according to the abnormal temporal distribution characteristics to obtain a confidence index; and responding with the current acquisition frequency and the corresponding confidence index as input variables, using the preset confidence index as the adaptive target, to obtain an adaptive acquisition frequency.

[0013] Preferably, the method for environmental data fusion control of iron tower foundations in low-temperature environments further includes: obtaining the monitoring item weights of the multiple environmental sensors, generating adjustment frequency weights based on the monitoring item weights, and optimizing the generated adaptive acquisition frequency with the adjustment frequency weights to obtain the optimized adaptive acquisition frequency.

[0014] Preferably, the method for controlling the environmental data fusion of iron tower foundations in low-temperature environments further includes: extracting features from the plurality of updated environmental sensing datasets to obtain a plurality of damage feature datasets; extracting the sensor locations of the plurality of updated environmental sensing datasets and the location of the iron tower to construct a damage weight network; and fusing the plurality of damage feature datasets according to the damage weight network to obtain a damage fusion index.

[0015] Preferably, the method for controlling the environmental data fusion of iron tower foundations in low-temperature environments further includes: if the multiple filtered environmental anomaly perception datasets return empty, a mode return instruction is triggered, wherein the mode return instruction is used to process the multiple updated environmental perception datasets through an edge data fusion mode.

[0016] Secondly, this application also provides a data fusion control system for the foundation environment of a steel tower in a low-temperature environment, used to execute the data fusion control method for the foundation environment of a steel tower in a low-temperature environment as described in the first aspect, comprising: a network configuration module for configuring a foundation environment sensing network of the steel tower, wherein the foundation environment sensing network includes multiple environmental sensors, each of which is independently connected through a low-temperature acquisition frequency adjustment module; a dataset acquisition module for acquiring multiple environmental perception datasets from the multiple environmental sensors based on the foundation environment sensing network, performing preliminary anomaly detection on the multiple environmental perception datasets, and acquiring multiple environmental anomaly perception datasets; a frequency generation module for inputting the multiple environmental anomaly perception datasets into the low-temperature acquisition frequency adjustment module for low-temperature environment anomaly filtering, generating an adaptive acquisition frequency based on the filtered multiple environmental anomaly perception datasets, and acquiring multiple updated environmental perception datasets through the adaptive acquisition frequency; and a fusion processing module for determining whether a mode switching command is triggered based on the multiple updated environmental perception datasets, and if a mode switching command is triggered, performing fusion processing on the multiple updated environmental perception datasets through a central data fusion mode.

[0017] Thirdly, this application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the method for environmental data fusion control of iron tower foundations in low-temperature environments as described in any one of the first aspects above.

[0018] Fourthly, a computer-readable storage medium storing a computer program that, when executed, implements the steps of the above-described method for environmental data fusion control of iron tower foundations in low-temperature environments, as described in the first aspect.

[0019] The technical solution provided in this application has at least the following technical effects or advantages: by realizing the adaptive, refined and scenario-based collection and control of tower foundation environmental data in low temperature environment, and constructing a dynamic data fusion system that integrates edge computing capabilities and central computing capabilities, the technical effect of improving the effectiveness of monitoring data, enhancing the structural damage perception capability of multi-source data fusion, reducing the probability of abnormal missed reports and improving the overall real-time response capability of monitoring in the context of abnormal and diverse environments.

[0020] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating the data fusion control method for tower foundations in low-temperature environments according to this application.

[0023] Figure 2 This is a schematic diagram of the structure of the tower foundation environmental data fusion control system for low-temperature environments used in this application.

[0024] Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application.

[0025] Explanation of reference numerals in the attached diagram: Network configuration module 1, Data set acquisition module 2, Frequency generation module 3, Fusion processing module 4, Bus 300, Receiver 301, Processor 302, Transmitter 303, Memory 304, Bus interface 305. Detailed Implementation

[0026] This application provides a data fusion control method and system for tower foundations in low-temperature environments. It addresses the technical problems in existing technologies, such as fixed acquisition frequencies, inflexible anomaly identification, coarse damage analysis models, and lack of mode switching. These problems lead to insufficient accuracy in environmental anomaly detection, incomplete structural condition assessment, and low data processing efficiency, further impacting tower operation safety, monitoring real-time performance, and the reliability of large-scale monitoring deployments. The method aims to achieve adaptive, refined, and scenario-based acquisition and control of tower foundation environmental data in low-temperature environments, and to construct a dynamic data fusion system integrating edge computing and central computing capabilities. This results in improved monitoring data effectiveness, enhanced structural damage perception capabilities through multi-source data fusion, reduced anomaly false alarm probability, and improved overall real-time monitoring response capabilities in the face of diverse environmental anomalies.

[0027] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0028] Example 1, please refer to the appendix. Figure 1 This application provides a data fusion control method for tower foundations in low-temperature environments, which is applied to a data fusion control system for tower foundations in low-temperature environments. The method specifically includes the following steps: S1: Configure the basic environmental sensing network of the tower. The basic environmental sensing network includes multiple environmental sensors, each of which is independently connected through a low-temperature acquisition frequency adjustment module.

[0029] Specifically, configuring the basic environmental sensing network for the iron tower refers to pre-deploying sensing units in the foundation area of ​​the iron tower to sense the environmental conditions. Through site planning, deployment density, and installation methods, the sensing units can cover the key stress areas and parts of the iron tower that are susceptible to low temperatures, and are used to continuously collect basic environmental parameters such as temperature, humidity, frost heave displacement, and foundation moisture content, thereby forming the underlying structure for environmental data acquisition.

[0030] The basic environmental sensing network comprises multiple environmental sensors. Different types of sensing devices with varying measurement ranges and response speeds are deployed within the same tower foundation area, including temperature sensors, pressure sensors, humidity sensors, and micro-vibration sensors, to acquire multi-dimensional sensing data reflecting changes in the foundation environment. These sensors are interconnected via wired or wireless data transmission methods and work collaboratively in a distributed manner, forming redundant detection links.

[0031] Each environmental sensor is independently connected through a low-temperature acquisition frequency adjustment module. The data acquisition rhythm of each sensor no longer operates at a fixed frequency, but is dynamically adjusted by the low-temperature acquisition frequency adjustment module according to the degree of environmental change, the risk level of low-temperature events, and the energy consumption constraints of the sensor itself, so that each sensor has a differentiated sampling strategy.

[0032] S2: Based on the basic environmental sensing network, obtain multiple environmental perception datasets from the multiple environmental sensors, perform preliminary anomaly detection on the multiple environmental perception datasets, and obtain multiple environmental anomaly perception datasets.

[0033] Furthermore, this application also includes: constructing historical environmental health perception data samples based on the multiple environmental sensors; training multiple environmental anomaly recognition models with the historical environmental health perception data samples; performing preliminary anomaly perception on the multiple environmental perception datasets based on the multiple environmental anomaly recognition models; and obtaining multiple environmental anomaly perception datasets.

[0034] Specifically, environmental sensors deployed within the tower foundation area continuously collect environmental parameters such as temperature, humidity, soil moisture content, frost heave displacement, and micro-vibrations through their respective acquisition mechanisms, forming multiple datasets according to time series or spatial distribution. Based on the topology of the basic environmental sensor network, real-time data from different sensor nodes are aggregated to edge nodes or a central server for subsequent analysis.

[0035] Historical environmental health perception data samples based on multiple environmental sensors are constructed. Stable, normal environmental data accumulated during long-term operation that are not affected by low temperature are screened, labeled and structured to form a multi-dimensional environmental characteristic sample set that can reflect the health status of the tower foundation, including typical data under different time conditions, meteorological conditions and foundation operation conditions.

[0036] Multiple environmental anomaly detection models are trained using historical environmental health perception data samples. These models learn the temporal patterns of environmental data under healthy conditions and the correlations between variables, thereby gaining the ability to identify abnormal deviations. The environmental anomaly detection models include time series prediction models based on long short-term memory networks, probabilistic models based on Gaussian distributions, and anomaly detection models based on isolated forests, using different mechanisms to determine whether environmental parameters have undergone abnormal changes.

[0037] Multiple environmental anomaly detection models are used to perform preliminary anomaly detection on multiple environmental sensing datasets. These models, already trained, are used to detect real-time environmental data, determining whether there are parameter values ​​or trends that significantly deviate from healthy samples, such as sudden temperature drops, abrupt displacements, or abnormal humidity increases. This results in multiple environmental anomaly sensing datasets. Anomaly results are then output based on the prediction bias, anomaly score, or confidence index of the multiple environmental anomaly detection models.

[0038] S3: Input the multiple environmental anomaly sensing datasets into the low-temperature acquisition frequency adjustment module for low-temperature environmental anomaly filtering, generate an adaptive acquisition frequency based on the filtered multiple environmental anomaly sensing datasets, and obtain multiple updated environmental sensing datasets through the adaptive acquisition frequency.

[0039] Furthermore, this application also includes: acquiring a low-temperature environment sensing data fingerprint database, wherein the low-temperature environment sensing data fingerprint database includes a first type of low-temperature event, a second type of low-temperature event, and a third type of low-temperature event, wherein the first type of low-temperature event is a sudden low-temperature anomaly event, the second type of low-temperature event is a continuous low-temperature anomaly event, and the third type of low-temperature event is a drifting low-temperature anomaly event; the low-temperature acquisition frequency adjustment module, by calling the low-temperature environment sensing data fingerprint database, performs data anomaly fingerprint matching on the multiple environmental anomaly sensing datasets to acquire a low-temperature environment anomaly sensing dataset; and, based on the low-temperature environment anomaly sensing dataset, obtains multiple filtered environmental anomaly sensing datasets.

[0040] Furthermore, this application also includes: optimizing the compensation acquisition frequency based on the low-temperature environment anomaly perception dataset to obtain a low-temperature compensation acquisition frequency where the low-temperature data restoration quality is greater than a preset quality threshold; and updating the adaptive acquisition frequency based on the low-temperature compensation acquisition frequency to obtain the updated adaptive acquisition frequency.

[0041] Furthermore, this application also includes: identifying the abnormal temporal distribution characteristics of multiple filtered environmental anomaly perception datasets; calculating the anomaly perception confidence level of the current acquisition frequency according to the abnormal temporal distribution characteristics to obtain a confidence index; and responding with the current acquisition frequency and the corresponding confidence index as input variables, using the preset confidence index as the adaptive target, to obtain an adaptive acquisition frequency.

[0042] Furthermore, this application also includes: obtaining the monitoring item weights of the plurality of environmental sensors, generating adjustment frequency weights based on the monitoring item weights; and optimizing the generated adaptive acquisition frequency with the adjustment frequency weights to obtain an optimized adaptive acquisition frequency.

[0043] Specifically, acquiring a low-temperature environment sensing data fingerprint database refers to extracting representative low-temperature event patterns from long-term monitoring and multi-regional collection of low-temperature environment data, and forming a standardized feature set based on event type, rate of change, duration, and characteristic parameters for subsequent anomaly identification and comparison. The low-temperature environment sensing data fingerprint database includes the following categories: The first category of low-temperature events refers to sudden temperature drops occurring within a very short period, characterized by a rapid temperature decrease exceeding a certain threshold within minutes, such as a drop of 3 degrees Celsius within 5 consecutive minutes; the second category refers to temperatures remaining below the normal range for a relatively long period, such as temperatures remaining below zero for 12 consecutive hours; and the third category refers to a slow deviation in temperature or related environmental parameters, with changes not significant in the short term but gradually deviating from the healthy baseline over several days.

[0044] The low-temperature acquisition frequency adjustment module performs data anomaly fingerprint matching on multiple environmental anomaly sensing datasets by calling the low-temperature environment sensing data fingerprint database. It compares the similarity between the data's change amplitude, change rate, and fluctuation pattern and existing event templates in the fingerprint database to determine which category of low-temperature event the abnormal data belongs to or whether it is a composite anomaly event, thus outputting a classified low-temperature environment anomaly sensing dataset. The matching process employs algorithms such as distance measurement, similarity scoring, and template alignment, and determines the anomaly level based on the degree of matching.

[0045] Multiple filtered environmental anomaly perception datasets are obtained from the low-temperature environment anomaly perception dataset. This involves removing non-low-temperature factors from the original anomaly data based on the identified low-temperature event category and event level, retaining only anomalies sensitive to or triggered by low temperatures. This process removes noisy events or false alarms related to other environmental factors. The filtering process reduces the complexity of subsequent analysis and improves the purity of the anomaly identification results.

[0046] Based on the low-temperature environment anomaly sensing dataset, the compensation sampling frequency is optimized. According to the type, intensity, and rate of change of the anomaly, the potential amount of missing data or data sparsity under the current environmental conditions is calculated. An optimization algorithm is then used to find the optimal sampling frequency that can recover the true change curve of the low-temperature event. Evaluation is performed based on a data restoration model or time series reconstruction model to compensate for the loss of time-series information caused by insufficient sampling. The optimization of the compensation sampling frequency aims to ensure that the low-temperature data restoration quality is greater than a preset quality threshold, for example, setting the quality threshold to 0.85. The final output is the low-temperature compensation sampling frequency that meets this condition, ensuring that key low-temperature events can be completely represented in the data.

[0047] The adaptive acquisition frequency is updated based on the low-temperature compensation acquisition frequency. After obtaining a compensation acquisition frequency that improves data reconstruction quality, the original adaptive acquisition frequency strategy is corrected so that it can take into account both general environmental changes and the special needs of low-temperature events in subsequent acquisition processes. The constraint update mechanism is implemented based on upper bound constraints, lower bound constraints, or weight correction methods, so that the adaptive acquisition frequency maintains energy efficiency under normal conditions and enhances agility under low-temperature conditions.

[0048] The study identifies the temporal distribution characteristics of anomalies in multiple filtered environmental anomaly perception datasets. After obtaining environmental anomaly data that has excluded non-low-temperature interference factors, the distribution patterns over time are analyzed to extract characteristic parameters such as the density, duration, rate of change, and periodicity of anomalies. These temporal distribution characteristics include the instantaneous jump characteristics of sudden anomalies, the long-term stable shift characteristics of persistent anomalies, and the slow trend-like change characteristics of drifting anomalies.

[0049] Anomaly perception confidence is calculated based on the anomaly time series distribution characteristics of the current sampling frequency to obtain a confidence index. The suitability of the current sampling frequency for capturing anomaly events is calculated based on the importance, frequency of occurrence, and trend complexity of the anomaly time series characteristics, and a quantified confidence index is output. The confidence index indicates whether the current sampling frequency is sufficient to accurately record the change process of anomaly events. As the complexity of anomaly time series characteristics increases from a single fluctuation pattern to a multi-segment composite curve, the feature dimensions involved in the confidence calculation also increase accordingly, making the confidence results more refined.

[0050] Using a pre-set confidence index as the adaptive target, and the current acquisition frequency and the corresponding confidence index as input variables, the adaptive acquisition frequency is obtained. That is, based on the minimum confidence requirement and the correspondence between the current acquisition frequency and the confidence, the frequency is adjusted by adjusting the model, and the acquisition frequency that can meet or exceed the target confidence is output. The response process uses function fitting, interpolation optimization or adaptive control algorithm to adjust the acquisition frequency up or down.

[0051] The weights of monitoring items from multiple environmental sensors are obtained. This involves calculating the importance of each monitoring item within the overall monitoring system based on the impact, sensitivity, and risk level of the parameters monitored by different types of sensors in the tower foundation environmental monitoring. This importance is then expressed as a weight value. Monitoring item weights can be obtained based on historical event contribution, expert rules, risk sensitivity curves, or data-driven models. Frequency adjustment weights are generated based on these weights, mapping the importance of monitoring items to reference factors for adjusting the acquisition frequency. This ensures that sensors with higher risk levels correspond to higher frequency weights, and sensors with lower risk levels correspond to lower frequency weights. This guides the adjustment of the acquisition frequency, ensuring that the acquisition strategy reflects the distribution of monitoring priorities.

[0052] The generated adaptive acquisition frequency is optimized by adjusting the frequency weights. After obtaining the initial adaptive acquisition frequency, the frequency weights of each monitoring item are incorporated into the frequency adjustment model, enabling globally differentiated configuration of the acquisition frequencies of different sensors, thereby obtaining an optimized acquisition frequency that better meets monitoring requirements. This optimization process can be based on weighted adjustment, linear combination, or multi-objective optimization models. By establishing a correlation between the acquisition frequency and the weights, the sensitivity of the acquisition frequency to key monitoring items is significantly improved.

[0053] Multiple updated environmental sensing datasets are acquired through adaptive acquisition frequency, enabling dynamic data acquisition from multiple environmental sensors deployed in the tower foundation environment. The adaptive acquisition frequency, a sampling rate that is dynamically adjusted based on the degree of environmental change, the importance of the monitored items, and historical data fluctuations, is used to control the number of times each environmental sensor acquires environmental parameters per unit time.

[0054] S4: Determine whether a mode switching command is triggered based on the multiple updated environmental perception datasets. If a mode switching command is triggered, fuse the multiple updated environmental perception datasets through the central data fusion mode.

[0055] Furthermore, this application also includes: configuring an adaptive acquisition frequency threshold, comparing the multiple adaptive acquisition frequencies corresponding to the multiple updated environmental perception datasets with the adaptive acquisition frequency threshold, and obtaining the proportion of quantities greater than the adaptive acquisition frequency threshold; if the proportion of quantities is greater than a preset proportion threshold, triggering a mode switching instruction, the mode switching instruction being used to switch the edge data fusion mode to the center data fusion mode.

[0056] Furthermore, this application also includes: collecting edge computing units under edge data fusion mode, analyzing the configuration parameters of the edge computing units to obtain edge computing resource indicators and edge computing power scaling indicators; and performing adaptive collection frequency threshold optimization according to the edge computing resource indicators and edge computing power scaling indicators to obtain an adaptive collection frequency threshold in which the probability of missed risk is within the expected probability.

[0057] Furthermore, this application also includes: extracting features from the plurality of updated environmental perception datasets to obtain a plurality of damage feature datasets; extracting the sensor locations and the location of the tower from the plurality of updated environmental perception datasets to construct a damage weight network; and fusing the plurality of damage feature datasets according to the damage weight network to obtain a damage fusion index.

[0058] Furthermore, this application also includes: if the filtered multiple environmental anomaly perception datasets return empty, a mode return instruction is triggered, wherein the mode return instruction is used to process the multiple updated environmental perception datasets through an edge data fusion mode.

[0059] Specifically, the process involves acquiring edge computing units in edge data fusion mode to obtain information about computing nodes within this mode. Edge computing units refer to local computing devices deployed near sensors, capable of performing data preprocessing, initial fusion, and rapid response computing tasks. The configuration parameters of these edge computing units are analyzed to obtain edge computing resource indicators and edge computing power scaling indicators. These parameters include edge processor type, available memory capacity, cache structure, data bandwidth, and real-time processing queue length. Edge computing resource indicators quantify the overall availability of hardware and operating resources, such as combining available processing power, available memory, and available communication bandwidth into a weighted numerical indicator. Edge computing power scaling indicators describe the performance improvement or degradation under different task loads. Subsequently, an adaptive acquisition frequency threshold is optimized based on the edge computing resource indicators and edge computing power scaling indicators. Using these two indicators as constraints, an optimization strategy is employed to find an acquisition frequency threshold that balances resource consumption and task performance. The adaptive acquisition frequency threshold represents the boundary value at which higher-level processing needs to be initiated after the sensor sampling frequency reaches a certain upper limit. Finally, the adaptive sampling frequency threshold obtained, which is within the expected probability of risk miss probability, indicates that the determined threshold must ensure the reliability of risk identification capability even when edge computing power is limited. The risk miss probability is used to characterize the probability of failing to identify or collect in time during abnormal environmental events, while the expected probability is used to define the maximum miss level allowed by the system.

[0060] Subsequently, multiple adaptive acquisition frequencies corresponding to various updated environmental perception datasets were compared with adaptive acquisition frequency thresholds to statistically analyze the range where sensor acquisition frequencies were higher than normal operating conditions within the current monitoring period. The percentage represents the proportion of all sensors or all time windows where the adaptive acquisition frequency exceeded the adaptive acquisition frequency threshold out of all acquisition frequencies.

[0061] Next, if the proportion of data exceeds a preset threshold, a mode switching command is triggered. The preset threshold indicates that the current environmental fluctuations have reached a level where data fusion accuracy must be improved. The mode switching command drives the data processing module to switch from edge data fusion mode to center data fusion mode. Edge data fusion mode is used for rapid, low-cost, and low-energy data preprocessing near sensor nodes, while center data fusion mode uploads large amounts of raw data to a central computing node for high-precision fusion, deep correlation analysis, and global situational awareness.

[0062] If a mode switching command is triggered, feature extraction is performed on multiple updated environmental sensing datasets, i.e., structured processing is carried out on updated data from multiple environmental sensors. Feature extraction refers to identifying a representative set of parameters from continuous or discrete environmental monitoring data, such as temperature gradient change characteristics, humidity abrupt change amplitude, wind load disturbance frequency, vibration energy distribution, or the slope of low-temperature response curves. The extracted features are quantified according to time, space, or physical dimensions, ultimately forming multiple damage feature datasets, which are used as input in the subsequent assessment of the tower's structural condition.

[0063] Subsequently, the sensor locations and tower locations from multiple updated environmental perception datasets are extracted to construct a damage weighting network. This network establishes a mapping relationship between each monitoring point and the actual stress area of ​​the tower based on the latitude and longitude coordinates of the sensor installation points, the relative height from the pile foundation to the tower body, and the overall spatial layout of the tower. The damage weighting network is used to assign weights to different monitoring points through spatial distance attenuation rules, node connectivity rules, or structural sensitivity rules, thereby reflecting the law that the perception contribution decreases with increasing distance.

[0064] Next, the damage feature datasets are fused according to the damage weight network to obtain the damage fusion index. The damage features contributed by each sensor are combined according to their weights. The fusion process includes steps such as weighted averaging, spatial correlation convolution, feature superposition, or multi-source consistency test. Finally, the damage fusion index is obtained to characterize the development trend of potential damage to the tower during the monitoring period.

[0065] If multiple filtered environmental anomaly perception datasets return empty, it indicates that after anomaly screening of the environmental perception data, no data samples meeting the anomaly judgment criteria were found within the current monitoring period. An empty return value signifies that the result set processed by various anomaly identification models, fingerprint matching modules, or confidence filtering mechanisms does not contain any abnormal event data, meaning the current environment is in a normal and stable state.

[0066] Subsequently, a mode return command is triggered. When the anomaly is insufficient to support the continued use of high-cost or high-resource-consumption data processing modes, a mode adjustment mechanism is initiated, sending a command to the upper-layer control module to switch the current data fusion and processing mode back to edge-side execution. The mode return command is used to process multiple updated environmental awareness datasets through edge data fusion mode, re-enabling the edge computing unit to perform near-source processing on the updated data, including preliminary denoising, edge feature extraction, and low-latency fusion steps, so that it can switch to a low-resource-consumption mode in the absence of anomalies.

[0067] In summary, the data fusion control method for tower foundation environmental conditions in low-temperature environments provided in this application has the following technical effects: by achieving adaptive, refined, and scenario-based acquisition and control of tower foundation environmental data in low-temperature environments, and constructing a dynamic data fusion system that integrates edge computing capabilities and central computing capabilities, the method achieves the technical effects of improving the effectiveness of monitoring data, enhancing the structural damage perception capability of multi-source data fusion, reducing the probability of missed anomalies, and improving the overall real-time response capability of monitoring in the context of diverse environmental anomalies.

[0068] Example 2: Based on the same inventive concept as the data fusion control method for tower foundations in low-temperature environments described in the foregoing examples, this application also provides a data fusion control system for tower foundations in low-temperature environments. Please refer to the appendix. Figure 2 The system includes: a network configuration module 1, used to configure the basic environmental sensing network of the tower, which includes multiple environmental sensors, each of which is independently connected through a low-temperature acquisition frequency adjustment module; a dataset acquisition module 2, used to acquire multiple environmental perception datasets from the multiple environmental sensors based on the basic environmental sensing network, perform preliminary anomaly detection on the multiple environmental perception datasets, and acquire multiple environmental anomaly perception datasets; a frequency generation module 3, used to input the multiple environmental anomaly perception datasets into the low-temperature acquisition frequency adjustment module for low-temperature environmental anomaly filtering, generate an adaptive acquisition frequency based on the filtered multiple environmental anomaly perception datasets, and acquire multiple updated environmental perception datasets through the adaptive acquisition frequency; and a fusion processing module 4, used to determine whether a mode switching command is triggered based on the multiple updated environmental perception datasets, and if a mode switching command is triggered, to perform fusion processing on the multiple updated environmental perception datasets through a central data fusion mode.

[0069] Furthermore, the tower foundation environmental data fusion control system for low-temperature environments is also used to: configure an adaptive acquisition frequency threshold, compare the multiple adaptive acquisition frequencies corresponding to the multiple updated environmental perception datasets with the adaptive acquisition frequency threshold, and obtain the proportion of quantities greater than the adaptive acquisition frequency threshold; if the proportion of quantities is greater than a preset proportion threshold, trigger a mode switching instruction, the mode switching instruction being used to switch the edge data fusion mode to the center data fusion mode.

[0070] Furthermore, the data fusion control system for tower foundation environment in low-temperature environment is also used for: collecting edge computing units in edge data fusion mode, analyzing the configuration parameters of the edge computing units to obtain edge computing resource indicators and edge computing power scaling indicators; and optimizing the adaptive acquisition frequency threshold according to the edge computing resource indicators and edge computing power scaling indicators to obtain an adaptive acquisition frequency threshold in which the probability of missed risk is within the expected probability.

[0071] Furthermore, the tower foundation environmental data fusion control system for low-temperature environments is also used to: construct historical environmental health perception data samples based on the multiple environmental sensors; train multiple environmental anomaly recognition models with the historical environmental health perception data samples; perform preliminary anomaly perception on the multiple environmental perception datasets based on the multiple environmental anomaly recognition models; and obtain multiple environmental anomaly perception datasets.

[0072] Furthermore, the tower foundation environmental data fusion control system for low-temperature environments is also used to: acquire a low-temperature environment sensing data fingerprint database, which includes a first type of low-temperature event, a second type of low-temperature event, and a third type of low-temperature event, wherein the first type of low-temperature event is a sudden low-temperature anomaly event, the second type of low-temperature event is a continuous low-temperature anomaly event, and the third type of low-temperature event is a drifting low-temperature anomaly event; the low-temperature acquisition frequency adjustment module, by calling the low-temperature environment sensing data fingerprint database, performs data anomaly fingerprint matching on the multiple environmental anomaly sensing datasets to acquire a low-temperature environment anomaly sensing dataset; and obtains multiple filtered environmental anomaly sensing datasets based on the low-temperature environment anomaly sensing datasets.

[0073] Furthermore, the data fusion control system for tower foundation environment in low-temperature environments is also used for: optimizing the compensation acquisition frequency based on the low-temperature environment anomaly perception dataset to obtain a low-temperature compensation acquisition frequency where the low-temperature data restoration quality is greater than a preset quality threshold; and updating the adaptive acquisition frequency based on the low-temperature compensation acquisition frequency to obtain the updated adaptive acquisition frequency.

[0074] Furthermore, the data fusion control system for tower foundations in low-temperature environments is also used to: identify the abnormal temporal distribution characteristics of multiple filtered environmental anomaly perception datasets; calculate the anomaly perception confidence level of the current acquisition frequency according to the abnormal temporal distribution characteristics to obtain a confidence index; and respond with the current acquisition frequency and the corresponding confidence index as input variables, using the preset confidence index as the adaptive target, to obtain an adaptive acquisition frequency.

[0075] Furthermore, the data fusion control system for tower foundations in low-temperature environments is also used to: obtain the monitoring item weights of the multiple environmental sensors, generate adjustment frequency weights based on the monitoring item weights, and optimize the generated adaptive acquisition frequency with the adjustment frequency weights to obtain the optimized adaptive acquisition frequency.

[0076] Furthermore, the tower foundation environmental data fusion control system for low-temperature environments is also used for: extracting features from the multiple updated environmental sensing datasets to obtain multiple damage feature datasets; extracting the sensor locations of the multiple updated environmental sensing datasets and the location of the tower to construct a damage weight network; and fusing the multiple damage feature datasets according to the damage weight network to obtain a damage fusion index.

[0077] Furthermore, the tower foundation environmental data fusion control system for low-temperature environments is also used to: if the filtered multiple environmental anomaly perception datasets return empty, trigger a mode return instruction, the mode return instruction being used to process the multiple updated environmental perception datasets through edge data fusion mode.

[0078] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The data fusion control method and specific examples for tower foundations in low-temperature environments described in Embodiment 1 are also applicable to the data fusion control system for tower foundations in low-temperature environments in this embodiment. Through the foregoing detailed description of the data fusion control method for tower foundations in low-temperature environments, those skilled in the art can clearly understand the data fusion control system for tower foundations in low-temperature environments in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0079] Example 3: Based on the inventive concept of the data fusion control method for tower foundations in low-temperature environments described in the foregoing embodiments, this application also provides an electronic device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the data fusion control method for tower foundations in low-temperature environments described in any one of the above Examples 1.

[0080] Appendix Figure 3 This is a schematic diagram of the structure of an exemplary electronic device of this application. Figure 3In this document, the bus architecture is represented by bus 300. Bus 300 may include any number of interconnected buses and bridges, and bus 300 connects various circuits including one or more processors represented by processor 302 and memory represented by memory 304. Bus 300 may also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 305 provides an interface between bus 300 and receiver 301 and transmitter 303. Receiver 301 and transmitter 303 may be the same element, i.e., a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 302 is responsible for managing bus 300 and general processing, while memory 304 can be used to store data used by processor 302 during operation.

[0081] Example 4: Based on the same inventive concept as the data fusion control method for tower foundations in low-temperature environments described in the foregoing examples, this application also provides a computer-readable storage medium storing a computer program. When executed, the computer program implements the steps of the data fusion control method for tower foundations in low-temperature environments described in any one of the above examples.

[0082] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0083] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A data fusion control method for tower foundations in low-temperature environments, characterized in that, The method includes: Configure a basic environmental sensing network for the tower, which includes multiple environmental sensors, each of which is independently connected through a low-temperature acquisition frequency adjustment module; Based on the basic environmental sensing network, multiple environmental perception datasets from the multiple environmental sensors are obtained, and preliminary anomaly detection is performed on the multiple environmental perception datasets to obtain multiple environmental anomaly perception datasets. The multiple environmental anomaly sensing datasets are input into the low-temperature acquisition frequency adjustment module for low-temperature environmental anomaly filtering. An adaptive acquisition frequency is generated based on the filtered multiple environmental anomaly sensing datasets, and multiple updated environmental sensing datasets are obtained through the adaptive acquisition frequency. Based on the multiple updated environmental awareness datasets, determine whether a mode switching command is triggered. If a mode switching command is triggered, fuse the multiple updated environmental awareness datasets using the central data fusion mode. The method for inputting the multiple environmental anomaly perception datasets into the low-temperature acquisition frequency adjustment module for low-temperature environmental anomaly filtering includes: A low-temperature environment sensing data fingerprint database is obtained, which includes a first type of low-temperature event, a second type of low-temperature event, and a third type of low-temperature event. The first type of low-temperature event is a sudden low-temperature anomaly event, the second type of low-temperature event is a continuous low-temperature anomaly event, and the third type of low-temperature event is a drifting low-temperature anomaly event. The low-temperature acquisition frequency adjustment module obtains the low-temperature environment abnormality sensing dataset by calling the low-temperature environment sensing data fingerprint library and performing data abnormality fingerprint matching on the multiple environmental abnormality sensing datasets. Based on the aforementioned low-temperature environment anomaly perception dataset, multiple filtered environmental anomaly perception datasets are obtained. After acquiring the low-temperature environment anomaly sensing dataset, the low-temperature acquisition frequency adjustment module further includes: Based on the low-temperature environment anomaly perception dataset, the compensation acquisition frequency is optimized to obtain the low-temperature compensation acquisition frequency where the low-temperature data restoration quality is greater than a preset quality threshold. The adaptive acquisition frequency is constrained and updated based on the low-temperature compensation acquisition frequency to obtain the updated adaptive acquisition frequency. The step of determining whether to trigger a mode switching command based on the multiple updated environment-aware datasets includes: Configure an adaptive acquisition frequency threshold, compare the multiple adaptive acquisition frequencies corresponding to the multiple updated environmental perception datasets with the adaptive acquisition frequency threshold, and obtain the proportion of the number greater than the adaptive acquisition frequency threshold; if the proportion of the number is greater than a preset proportion threshold, trigger a mode switching instruction, which is used to switch the edge data fusion mode to the center data fusion mode. The step of generating an adaptive sampling frequency based on multiple filtered environmental anomaly perception datasets includes: Identify the anomalous temporal distribution characteristics of multiple filtered environmental anomaly perception datasets; Based on the aforementioned abnormal time-series distribution characteristics, the anomaly perception confidence level of the current acquisition frequency is calculated to obtain a confidence index; Using a preset confidence index as the adaptive target, and the current acquisition frequency and the corresponding confidence index as input variables, the adaptive acquisition frequency is obtained.

2. The method for environmental data fusion control of iron tower foundations in low-temperature environments as described in claim 1, characterized in that, Configuring an adaptive sampling frequency threshold can be achieved through the following methods: The edge computing unit in the edge data fusion mode is collected, and the configuration parameters of the edge computing unit are analyzed to obtain edge computing resource indicators and edge computing power scaling indicators. Based on the edge computing resource indicators and edge computing power scaling indicators, an adaptive sampling frequency threshold is optimized to obtain an adaptive sampling frequency threshold where the probability of missed risk is within the expected probability.

3. The method for environmental data fusion control of iron tower foundations in low-temperature environments as described in claim 1, characterized in that, The method for performing preliminary anomaly detection on the multiple environmental perception datasets to obtain multiple environmental anomaly perception datasets includes: Construct a historical environmental health perception data sample based on the aforementioned multiple environmental sensors; Multiple environmental anomaly identification models are trained using the historical environmental health perception data samples. Based on the multiple environmental anomaly identification models, preliminary anomaly perception is performed on the multiple environmental perception datasets to obtain multiple environmental anomaly perception datasets.

4. The method for environmental data fusion control of iron tower foundations in low-temperature environments as described in claim 1, characterized in that, The method generates an adaptive sampling frequency based on multiple filtered environmental anomaly perception datasets. The method also includes: Obtain the monitoring item weights of the multiple environmental sensors, and generate adjustment frequency weights based on the monitoring item weights; The generated adaptive acquisition frequency is optimized using the adjusted frequency weights to obtain the optimized adaptive acquisition frequency.

5. The method for environmental data fusion control of iron tower foundations in low-temperature environments as described in claim 1, characterized in that, The method involves fusing the multiple updated environmental awareness datasets using a central data fusion model, including: Feature extraction is performed on the multiple updated environmental awareness datasets to obtain multiple damage feature datasets; The sensor locations and the location of the tower are extracted from the multiple updated environmental perception datasets to construct a damage weighting network; The multiple damage feature datasets are fused according to the damage weight network to obtain a damage fusion index.

6. The method for environmental data fusion control of iron tower foundations in low-temperature environments as described in claim 1, characterized in that, After fusing the multiple updated environmental awareness datasets using a central data fusion mode, the method further includes: If the filtered multiple environmental anomaly perception datasets return empty, a mode return instruction is triggered. The mode return instruction is used to process the multiple updated environmental perception datasets through edge data fusion mode.

7. A data fusion control system for tower foundations in low-temperature environments, characterized in that, The steps for implementing the data fusion control method for tower foundation environmental conditions in low-temperature environments as described in any one of claims 1 to 6 include: The network configuration module is used to configure the basic environmental sensing network of the tower. The basic environmental sensing network includes multiple environmental sensors, each of which is independently connected through a low-temperature acquisition frequency adjustment module. The dataset acquisition module is used to acquire multiple environmental perception datasets from the multiple environmental sensors based on the basic environmental sensing network, perform preliminary anomaly detection on the multiple environmental perception datasets, and acquire multiple environmental anomaly perception datasets. The frequency generation module is used to input the multiple environmental anomaly perception datasets into the low temperature acquisition frequency adjustment module for low temperature environmental anomaly filtering, generate an adaptive acquisition frequency based on the filtered multiple environmental anomaly perception datasets, and obtain multiple updated environmental perception datasets through the adaptive acquisition frequency. The fusion processing module is used to determine whether a mode switching command is triggered based on the multiple updated environmental perception datasets. If a mode switching command is triggered, the multiple updated environmental perception datasets are fused using the central data fusion mode.

8. An electronic device, characterized in that, include: At least one processor; A memory that is communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the steps of the data fusion control method for tower foundation environmental conditions in low-temperature environments as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed, implements the steps of the data fusion control method for the environmental conditions of iron tower foundations in low-temperature environments as described in any one of claims 1 to 6.

Images