Telegraph pole full life cycle intelligent operation and maintenance detection system based on wireless sensor network

Abstract

The invention relates to the technical field of power grid operation and maintenance, and discloses a telegraph pole full-life-cycle intelligent operation and maintenance detection system based on a wireless sensor network, and the system comprises an acquisition module which is used for monitoring the health state of a telegraph pole in real time; the data fusion module is used for obtaining health state estimation of the telegraph pole based on a Kalman filtering algorithm; the fault prediction module is used for predicting the fault occurrence probability of the telegraph pole in combination with a Bayesian inference algorithm; the early warning processing module is used for sending out an early warning signal and carrying out emergency response; the wireless transmission module is used for wireless transmission; and the modeling module is used for displaying information in real time. According to the telegraph pole health state monitoring system, continuous online monitoring is carried out on various health states of a telegraph pole through the acquisition module, so that the modeling and early warning module can carry out accurate operation based on a real state, real-time monitoring can be carried out on the telegraph pole in a large range at the same time, the emergency response speed is improved, and safe operation of a power system is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid operation and maintenance, and specifically to an intelligent operation and maintenance detection system for the whole life cycle of utility poles based on a wireless sensor network. Background Art

[0002] As an important supporting component of transmission lines, utility poles are widely used in various scenarios such as cities, rural areas, and mountainous areas, undertaking the tasks of power line erection and operation support. Especially in ultra-high voltage AC transmission systems at 750 kV and above, the security guarantee and defense system of large-scale power grids, and intelligent dispatching systems for multi-source collaboration, the stability and reliability of utility poles have a crucial impact on the continuity and security of the overall power supply system.

[0003] However, due to the long-term outdoor environment of utility poles, their service life and health status are easily affected by various factors, including climate change (such as storms, high humidity, strong sunlight), soil corrosion, mechanical fatigue, temperature stress, etc. The combined effect of these complex environmental factors on the structure will accelerate its aging and degradation process, affecting its life cycle and operation safety. Therefore, how to scientifically and effectively monitor the health and evaluate the life of utility poles is an important factor in ensuring the efficient operation of the power grid.

[0004] Currently, the number of utility poles in the power system is huge, covering a wide area, and some areas are even located in inaccessible areas such as mountains, forests, and deserts. The traditional maintenance method mainly based on manual inspection has been difficult to meet the requirements of large-scale, full-coverage, and real-time condition monitoring. On the one hand, manual inspection has a long cycle, high cost, and is affected by subjective factors such as missed inspection and misjudgment. On the other hand, some monitoring technologies still rely on indirect indicators (such as line current fluctuations, line loss rates, etc.) to infer the health status, lacking the direct perception ability of the utility pole body, and it is difficult to detect potential hidden dangers in time, which is likely to cause equipment damage and power grid failures, and even endanger the regional power supply stability in severe cases. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides an intelligent operation and maintenance detection system for the whole life cycle of utility poles based on a wireless sensor network, which solves the problems in the prior art that the number of utility poles is huge, the distribution is wide, the efficiency of manual inspection is low, the monitoring method is lagging, and it is difficult to timely grasp the health status of the pole body.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent operation and maintenance detection system for the whole life cycle of utility poles based on a wireless sensor network, including:

[0007] A collection module, used for real-time monitoring of the health status of utility poles;

[0008] A data fusion module, configured to receive data from a wireless sensor network and fuse the data based on the Kalman filtering algorithm to obtain an estimation of the health status of the utility pole;

[0009] A fault prediction module, based on the health status estimation output by the Kalman filter, combines the Bayesian inference algorithm to predict the probability of fault occurrence of the utility pole;

[0010] An early warning processing module, configured to issue early warning signals and perform emergency responses according to the health status estimation results and the probability of fault occurrence;

[0011] A wireless transmission module, configured to wirelessly transmit sensor data, health status estimation, probability of fault occurrence, and early warning signals;

[0012] A modeling module, based on the wireless transmission signals, virtually displays in real time the location of the utility pole, wireless sensor network data, health status estimation, probability of fault occurrence, and early warning signals.

[0013] Preferably, the acquisition module includes:

[0014] A strain sensor: configured to acquire strain data of the utility pole under external forces;

[0015] A corrosion sensor: configured to acquire data on the corrosion degree of the surface of the utility pole;

[0016] A temperature sensor: configured to acquire temperature data at the location of the utility pole;

[0017] A wind speed sensor: configured to acquire the wind speed at the location of the utility pole;

[0018] A humidity sensor: also configured to acquire the humidity at the location of the utility pole.

[0019] Preferably, the Kalman filtering algorithm is used to dynamically estimate the health status of the utility pole according to the historical health status of the utility pole and real-time sensor data, and the health status includes the strain force, corrosion degree, temperature at the location, wind speed at the location, and humidity at the location of the utility pole.

[0020] Preferably, the state equation of the Kalman filtering algorithm is:

[0021]

[0022] Wherein, represents the posterior estimation vector of the true state of the system at time t, that is, the estimated vector of the health status of the utility pole updated after x(t) is given; represents the prior estimation vector derived based on the state information at time t-1 and the state transition model at time t, which is the predicted value; K tdenotes the Kalman gain matrix, which is used to weight and correct the degree of prediction error. Its magnitude determines the degree of trust in the current observed data. H represents the observation matrix, which is used to map the state vector into the observation space. x(t) is the health state vector of the utility pole at time t.

[0023] Preferably, the fault prediction module constructs a fault probability model based on the Bayesian inference algorithm. By comparing the statistical correlation between the current health state and historical fault data, it calculates the probability of the utility pole failing within a set time window.

[0024] Preferably, the Bayesian inference algorithm includes:

[0025]

[0026] where P(F|x(t)) is the posterior probability of the utility pole failing F given the health state vector x(t) of the utility pole at time t; P(x(t)|F) is the conditional probability of the health state vector x(t) occurring given that the utility pole has failed F; P(F) is the prior probability of the fault F occurring; P(x(t)) is the total probability of the health state vector x(t) of the utility pole at time t.

[0027] Preferably, the emergency response includes sending out a warning signal, adjusting the power grid load, and dispatching maintenance personnel to carry out repairs.

[0028] Preferably, the modeling module establishes a digital twin model based on the location information of the utility pole and real-time displays the health state, health state estimation, fault occurrence probability, and emergency response of the utility pole based on the digital twin model.

[0029] Preferably, the wireless transmission module further includes a data security unit, which is used to encrypt the health state, health state estimation, fault occurrence probability, and warning signal, and adopts a symmetric encryption and authentication mechanism to prevent data leakage and tampering during wireless transmission.

[0030] Preferably, the symmetric encryption and authentication mechanism includes:

[0031] Encryption process:

[0032] C = E(K, P);

[0033] where C is the ciphertext; E is the encryption function; K is the encryption key; P is the plaintext data (data collected by the sensor);

[0034] Decryption process:

[0035] P = D(K, C);

[0036] Where D is the decryption function; K is the decryption key, which is the same as the encryption key; C is the ciphertext; P is the decrypted plaintext data;

[0037] Authentication Mechanisms:

[0038] MAC = H(K,M);

[0039] Among them, MAC is the message authentication code, which verifies the integrity of the message; H is the hash function; K is the key; and M is the message.

[0040] The present invention provides a full life cycle intelligent operation and maintenance detection system for electric poles based on a wireless sensor network.

[0041] It has the following beneficial effects:

[0042] 1. The present invention continuously monitors the health status of electric poles online through the acquisition module, collects and uploads parameters in real time, and enables subsequent modeling and early warning modules to accurately calculate based on the real status. This strategy not only improves the system's ability to respond to external disturbances, such as storms, thermal expansion and other extreme climates, but also reduces human inspection errors, providing great convenience for remote area operation and maintenance. It can also simultaneously monitor electric poles in a large area in real time, while improving the emergency response speed and ensuring the safe operation of the power system.

[0043] 2. The present invention adopts Kalman filtering to obtain the health status estimation of the utility pole, then uses the Bayesian inference algorithm to calculate the health status estimation, and finally predicts the failure probability of the utility pole, thereby achieving the technical effect of taking into account both the accuracy of state estimation and the sensitivity of risk prediction. Compared with the problem that a single statistical model or static limit judgment method in the prior art cannot dynamically reflect the structural degradation trend, the present invention effectively solves the technical problems of "non-linearity, noise interference, and high uncertainty" in structural health monitoring, and significantly improves the stability and credibility of the evaluation results.

[0044] 3. The present invention constructs a digital twin model by utilizing a modeling module, and combines real-time data to graphically visualize and dynamically update the location, status, and risk level of utility poles, thereby forming a three-dimensional virtual operation and maintenance interface that integrates monitoring, early warning, and dispatch instructions. Compared with the prior art that can only provide tabular data or offline charts, this solution completely changes the embarrassing situation of operation and maintenance personnel who "cannot see, cannot manage, and respond slowly", and provides new support for the unified monitoring and fine dispatching of large-scale power grid facilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Schematic diagram of the system architecture of the present invention. DETAILED DESCRIPTION

[0046] Next, in combination with the accompanying drawings in the specification of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts fall within the protection scope of the present invention.

[0047] To better understand the present invention, the above content will be described in detail below in combination with specific embodiments.

[0048] Please refer to the attached Figure 1 , the embodiment of the present invention provides an intelligent operation and maintenance detection system for the whole life cycle of electric poles based on a wireless sensor network, including:

[0049] An acquisition module for real-time monitoring of the health status of electric poles;

[0050] In this embodiment, to achieve comprehensive monitoring of the life cycle state of electric poles and subsequent intelligent evaluation, the system first collects raw data on the key operating parameters of electric poles through the acquisition module. The acquisition module, as the data input source of the system, is the basis for constructing a health state estimation model and a fault probability prediction logic. The collected data will be fused through the Kalman filter algorithm in the subsequent process and finally used by the modeling module for three-dimensional display and state rendering.

[0051] Therefore, a high-frequency, low-latency, and packet-loss-free data path should be formed between the acquisition module and the subsequent data fusion module to ensure the real-time and timeliness of the obtained health state data.

[0052] Specifically, the acquisition module includes multiple environmental and structural response sensing units, specifically including the following components:

[0053] A strain sensor: used to collect the axial strain or shear strain of the electric pole under the stress state in real time, reflecting the structural integrity of the electric pole. Generally, a resistance strain gauge or a fiber Bragg grating strain sensor can be used to convert the strain signal into an electrical signal.

[0054] A corrosion sensor: used to detect the degree of electrochemical corrosion on the surface of the electric pole, characterizing the degree of material degradation. In a possible implementation, the sensor is of the electrochemical impedance measurement type and can output the corrosion current density.

[0055] A temperature sensor: used to collect the temperature data of the environment where the electric pole is located. Specifically, a thermocouple or a thermistor can be selected as the sensor. Temperature changes not only affect material properties but also interfere with other sensor signals and need to be input into the Kalman filter model as an auxiliary correction parameter.

[0056] Wind speed sensor: The wind speed sensor is installed on the top or side of the utility pole to collect wind speed data in the horizontal or vertical direction. Generally, an ultrasonic anemometer or a cup anemometer is selected, with a measurement range of 0 - 30 m / s and an accuracy of ±0.1 m / s.

[0057] Humidity sensor: Used to measure the relative humidity in the air, usually a capacitive or dew point sensor is adopted. In this system, the relative humidity is not only positively correlated with the corrosion rate but also affects the elastic modulus correction of the strain response.

[0058] In some embodiments, the signals of each sensor are converted from analog to digital through the ADC module and uploaded to the edge computing gateway or the master control node via a wireless communication node (such as LoRa or NB-IoT). To maintain the time consistency of the data, each sensor's sampled data needs to be marked with a timestamp to form a unified data structure.

[0059] Specifically:

[0060] Strain sensor, model HBM-LY11-3 / 350 (resistance strain gauge) is used. It has high precision and small size, and is suitable for pasting on steel bars or metal brackets of utility poles;

[0061] Corrosion sensor, model Cosasco-8010-Electrical-Resistance-Probe is used. It is an industrial-grade corrosion probe, suitable for corrosion monitoring in soil, concrete, and air environments;

[0062] Temperature sensor, model Texas-Instruments-TMP117 is used. It is suitable for remote data acquisition, has low power consumption, and is internally calibrated.

[0063] Wind speed sensor, model NRG-Systems-40C (cup anemometer) is used. It is a classic wind speed meter for wind power plants, mechanical, with high cost performance and good accuracy, suitable for relatively mild environments

[0064] Humidity sensor, model Honeywell-HIH-5031 is used. It has the advantages of low power consumption, analog voltage output type, suitable for long-term operation, and has a dew condensation prevention function.

[0065] As an option, the acquisition module can adopt a self-powered design, providing power for the sensors through a thermal energy or vibration energy acquisition device (such as a piezoelectric generator), enhancing the long-term deployment ability of the system in remote areas.

[0066] In specific applications, to reduce signal drift and interference, in some embodiments, a signal preprocessing module is introduced, and a sliding average filter or a median filter is used to perform primary filtering on the original signal. This operation can be regarded as a primary data denoising step before the input of the Kalman filter, which helps to improve the convergence stability of the health state estimation model.

[0067] Furthermore, in the case where the geographical distribution of utility poles is extensive, multiple acquisition modules form a low-power wide-area sensing layer through a wireless transmission network, providing a high-coverage and high-redundancy data basis for the subsequent fusion and modeling layers.

[0068] Therefore, through the acquisition module, multi-source, high-frequency, and low-power acquisition of the physical state and environmental parameters of utility poles is achieved, providing sufficient and structured input conditions for the subsequent data fusion and fault probability modeling of the system, and ensuring the feasibility and technical completeness of the entire intelligent operation and maintenance system.

[0069] A data fusion module, configured to receive data from the wireless sensor network and fuse the data based on the Kalman filtering algorithm to obtain an estimated health state of the utility pole;

[0070] In this embodiment, to ensure that the multi-source sensor data obtained from the acquisition module can still be reliable and available under the interference of noise and errors, a data fusion module is specifically set up to complete the dynamic fusion of sensing signals and health state estimation. The data fusion module is located between the acquisition module and the modeling module, playing a central role in connecting the upper and lower levels. Specifically, this module receives various types of original perception data uploaded by the wireless sensor network and processes them through the Kalman filtering algorithm, thereby obtaining an estimated value of the health state of the utility pole at the current moment, providing a high-confidence data basis for the subsequent modeling module to construct a state evolution curve and three-dimensional visualization.

[0071] Specifically, the data fusion module is designed based on the Kalman filtering algorithm. By introducing a dual iterative mechanism of state prediction and observation correction, the algorithm can effectively mitigate the influence of sensor errors and signal drift on the final estimation result.

[0072] In a possible implementation, the health state vector of the utility pole at time t is defined as:

[0073]

[0074] where ∈(t) is the strain of the utility pole; c(t) is the corrosion degree of the utility pole; T(t) is the temperature at the location of the utility pole; v(t) is the wind speed at the location of the utility pole; h(t) is the humidity at the location of the utility pole; and x(t) is the health state vector of the utility pole at time t.

[0075] Generally, the state transition process can be described by the following state space model:

[0076] x(t + 1) = Ax(t) + Bu(t) + w(t);

[0077] Where, A is the state transition matrix, which is used to reflect the dynamic relationship between state variables; u(t) is the external input vector, which is used to represent the influence of environmental disturbances or manual interventions, etc.; B is the input influence matrix, which is used to describe the influence of the input on the state change; w(t) is the process noise vector, which reflects the sensor error; x(t + 1) is the health state vector of the utility pole at time t + 1;

[0078] As an option, to improve the system's adaptability under emergencies (such as lightning strikes, extreme storms), the Kalman filter algorithm is introduced in the prediction step to determine whether there are abnormal samples that deviate from the prediction ability range of the state space model, thereby triggering the abnormal state reporting mechanism.

[0079] Specifically, the Kalman filter formula includes:

[0080]

[0081] Where, represents the posterior estimation vector of the true state of the system at time t, that is, the estimated health state vector of the utility pole updated after x(t) is given; represents the prior estimation vector derived based on the state information at time t - 1 and the state transition model at time t, which is the predicted value; K t represents the Kalman gain matrix, which is used to weight and correct the degree of prediction error, and its size determines the degree of trust in the current observed data. H represents the observation matrix, which is used to map the state vector to the observation space.

[0082] In some embodiments, for the system architecture of multi-node collaborative monitoring, an extended Kalman filter (EKF) or an unscented Kalman filter (UKF) can be introduced to adapt to the non-linear forms of the state transition matrix A and the observation matrix H.

[0083] Furthermore, a health state trend prediction logic can also be embedded in the data fusion module, which is used to construct a short-term trend line based on the estimated x(t) series, providing basic data for the abnormal evolution detection and early warning model.

[0084] Therefore, through the above algorithm construction and matrix parameter definition, the data fusion module realizes the continuous dynamic integration of multi-source heterogeneous perception data, ensuring the accuracy and stability of the health state estimation results, and providing technical support for the subsequent risk modeling and maintenance decision-making of the system.

[0085] The fault prediction module predicts the probability of pole failure based on the health state estimation output by the Kalman filter and combines the Bayesian inference algorithm.

[0086] In this embodiment, to achieve a forward-looking assessment of the operation risk of the pole, after the acquisition and fusion processing of the sensing data are completed, the system further sets up a fault prediction module. As the direct subsequent logic unit of the data fusion module, this module constructs a fault probability model of the pole through the Bayesian inference algorithm based on the health state estimation result output by the Kalman filter, and is used to evaluate the potential failure risk of the target pole within a specific future time window. The role of this module is to convert the continuous state estimation into the probability inference of discrete events, so as to provide a decision basis for intelligent early warning and operation and maintenance decision-making.

[0087] Specifically, the fault prediction module takes the health state vector as the input, and combines the historical fault sample data and the statistical learning modeling method to calculate the failure probability of the pole under the given operating state.

[0088] Based on the foregoing health state vector x(t), and then based on the Bayesian inference algorithm, the module calculates the posterior probability of pole failure under the given state. The specific calculation formula is as follows:

[0089]

[0090] Among them, P(F|x(t)) is the posterior probability of pole failure F under the condition of the health state vector x(t) of the given pole at time t; P(x(t)|F) is the conditional probability of the health state vector x(t) appearing under the condition that the given pole has a failure F; P(F) is the prior probability of the occurrence of failure F; P(x(t)) is the total probability of the health state vector x(t) of the pole at time t.

[0091] Generally, P(x(t)|F)P(F) can be estimated by constructing a multi-dimensional Gaussian mixture model (GMM) or based on historical fault data for kernel density. To improve the modeling accuracy, in some embodiments, the state vector can be processed by principal component dimensionality reduction, and the high-dimensional variables can be compressed into the subspace with the strongest fault sensitivity for modeling.

[0092] In a possible implementation manner, the system continuously samples the health state data in a sliding time window manner and updates the Bayesian posterior probability in real time. If in any sliding window, the following conditions are met:

[0093] P(x(t)|F)≥θ;

[0094] Among them, θ is the set risk threshold, which can be taken as 0.75, for example, or optimized according to the ROC curve.

[0095] As an option, a dynamic prior update mechanism can be introduced, that is, during the operation of the system, P(F) is corrected in real time according to the observed data, so that the model can adapt to the risk change trend in different environments, thus avoiding the lag phenomenon of the static model.

[0096] In some embodiments, to further improve the generalization ability and fault tolerance of the model, the Bayesian inference result can also be jointly fused with other risk indicators (such as the state residual anomaly value and the long-term strain change rate) to form a multi-index risk assessment model.

[0097] In terms of the implementation architecture, the fault prediction module can be implemented through an edge computing node or a remote cloud server. If deployed in the cloud, it can be linked with the historical database to achieve horizontal multi-pole comparison and group risk ranking.

[0098] Through the above structural design and probability derivation method, the fault prediction module can, based on multi-source state inputs, utilize the Bayesian inference mechanism to complete the dynamic quantification of future failure trends, providing a theoretical basis and data support for subsequent intelligent scheduling, early warning management, and maintenance optimization.

[0099] An early warning processing module, which is used to issue an early warning signal and perform an emergency response according to the health state estimation result and the probability of fault occurrence;

[0100] In this embodiment, in the overall system structure of the present invention, the early warning processing module is a direct downstream logic unit of the fault prediction module. Its main function is to comprehensively judge the estimated health state result and the probability of fault occurrence, and timely trigger an early warning behavior based on the established response strategy and threshold logic. The early warning processing module not only plays the role of event judgment and response execution, but also undertakes the interface function between the system and the actual operation strategy of the power grid, and conducts judgment and notification response.

[0101] Specifically, the triggering condition of this module depends on the health state estimation result output by the data fusion module and the probability value of fault occurrence provided by the fault prediction module. The two cooperate as input signals for comprehensively judging the risk level and response level of the utility pole. Generally, the system sets multiple risk thresholds to implement a differential early warning and emergency response mechanism to improve the response efficiency and avoid resource waste.

[0102] Specifically, the early warning processing module receives the health state estimation vector from the data fusion module. This vector includes indicators such as the strain data, corrosion degree data, temperature, wind speed, and humidity of the utility pole at the current moment, and simultaneously receives the failure probability value provided by the fault prediction module. Under the condition of meeting the set conditions, the module performs early warning triggering and emergency response operations through the judgment logic.

[0103] Specifically, the early warning processing module conducts the judgment process based on the following core rules:

[0104] As an option, the system sets a fault probability threshold θ. When the predicted posterior probability P(F|x(t)) exceeds this threshold, it is determined that the current electric pole is in a potential risk state. In another setting, the system can further combine historical trend factors such as the state residual index and the corrosion growth rate to form a multi-parameter joint trigger logic for enhancing the sensitivity to hidden faults.

[0105] In a possible implementation, when the system detects that any of the following trigger conditions is satisfied:

[0106] In the current health state, the strain ∈(t) continuously exceeds the structural design limit;

[0107] The corrosion degree c(t) and the temperature-humidity combination characteristics indicate an accelerating trend of material degradation;

[0108] The predicted posterior probability P(F|x(t)) > 0.85 and remains not less than 0.75 within the sliding window for more than 12 hours;

[0109] The system immediately initiates the first-level early warning response process and takes different measures according to the response level.

[0110] In this embodiment, the emergency response mechanism includes but is not limited to the following operation contents:

[0111] First, the system issues an early warning signal, which can be achieved in various ways, such as a pop-up prompt on the visualization screen of the control center, pushing to the mobile terminal of the operation and maintenance personnel through the wireless communication channel, or giving a on-site prompt through the supporting sound and light alarm device;

[0112] Second, as an option, the system can automatically access the power grid dispatching interface to implement temporary load reduction, reallocation or load switching in the power supply area where the faulty electric pole is located, preventing the expansion of the fault point from triggering system-level risks. This load adjustment process can be linked with the SCADA system to ensure the closed-loop execution of the instructions;

[0113] Finally, the system will automatically generate an early warning report and push it to the operation and maintenance dispatching center. At the same time, it will call the maintenance resource management module, retrieve the nearest maintenance team and vehicle resources through GIS positioning, generate an inspection work order, and dispatch maintenance personnel to the location of the target electric pole for on-site detection and necessary handling.

[0114] In some embodiments, to avoid unnecessary responses caused by false alarms or signal jitters, the system can set an early warning confirmation mechanism to re-verify the trigger signal, such as re-determination and confirmation through a secondary model, an expert rule library or a historical comparison strategy.

[0115] Generally, the early warning processing module can also record all early warning triggering conditions, judgment logic processes, and response execution paths to form traceable event logs. These logs can be used for subsequent operation and maintenance decision-making analysis, model optimization training, and responsibility tracking.

[0116] Furthermore, in the distributed deployment solution of the system, the early warning processing module can also be deployed in edge computing nodes, enabling some response logics to be completed near the device end, reducing response latency, and improving system robustness.

[0117] Therefore, through the above structure and processing flow, the early warning processing module constructs a complete closed-loop mechanism starting from the health state estimation results and fault prediction results and facing on-site response operations, realizing the effective conversion from "data" to "action", and ensuring that the system of the present invention has the response ability and safety redundancy design for practical engineering applications.

[0118] The wireless transmission module is used for wirelessly transmitting sensor data, health state estimation, probability of fault occurrence, and early warning signals;

[0119] In this embodiment, to achieve efficient data interaction between multiple modules and remote information management, the system is further provided with a wireless transmission module. This module is used to achieve high-speed wireless communication between each functional unit, especially between the edge acquisition end and the remote center, ensuring that multi-source information collected from the original sensors, as well as subsequent health state estimation results, fault probability assessment information, and early warning response instructions, can be stably and securely transmitted and synchronously updated between devices at different levels.

[0120] The wireless transmission module is located in the underlying data communication path of the system, directly connecting to each functional unit including sensor nodes, data fusion modules, fault prediction modules, and early warning processing modules, and is the infrastructure to ensure the system's linkage response. Its core task is not only to achieve the wireless communication function, but more importantly to ensure the integrity, confidentiality, and anti-interference ability of data during the transmission process, especially putting forward higher security requirements for the transmission of key state information and risk signals.

[0121] In this embodiment, the wireless transmission module supports wireless communication protocols including but not limited to LoRa, NB-IoT, 4G, 5G, Wi-Fi, etc., and can adapt to various deployment environments and communication range requirements.

[0122] Specifically, the wireless transmission module has multi-channel scheduling capabilities and can achieve dynamic bandwidth allocation based on the priority of the transmission content. For example, the system can give priority to ensuring the immediacy of the health state estimation results and early warning signals, and delay the transmission of non-urgent original sensing data through queue caching to optimize communication efficiency.

[0123] In some embodiments, the module further includes a relay routing unit, which is used to form a grid-shaped self-organizing communication network in the scenario of pole networking deployment, improving the signal coverage ability and fault tolerance in complex environments.

[0124] To ensure the security and tamper-proof ability of the transmitted data, the wireless transmission module is further provided with a data security unit, which performs encryption and authentication operations on all key data types, including sensor raw data, health status vectors, state estimation results, probability of fault occurrence, and system warning signals, etc.

[0125] As an option, the data security unit adopts a symmetric encryption mechanism to encrypt the data to be transmitted through a pre-set or negotiated key. The system uses the same key to complete the encryption and decryption processes, thus enhancing data confidentiality while ensuring transmission efficiency. In some embodiments, lightweight encryption algorithms (such as AES-128) can be used to balance encryption strength and resource consumption, adapting to the limited computing power of edge nodes.

[0126] Generally, the encryption process is performed before the data is sent out from each module. The data to be encrypted (plaintext) is converted into ciphertext through an encryption function before entering the communication channel. The key is uniformly managed by the system and updated regularly to prevent security risks caused by long-term exposure of the key.

[0127] At the data receiving end, the system uses the same key to decrypt and restore the ciphertext to ensure that the confidentiality of the data during transmission is not compromised.

[0128] Furthermore, to prevent man-in-the-middle attacks and data tampering behaviors, the system is also equipped with an identity authentication mechanism, which adopts the Message Authentication Code (MAC) technology to generate a check code for the original message using a pre-shared key and a hash function. The receiving end recalculates the authentication code and compares it with the received value to confirm whether the message has been tampered with during transmission.

[0129] In a possible implementation, the system embeds the encryption process and identity authentication into the data frame structure at the same time, so that each data transmission unit contains a complete encryption header, authentication information, and integrity check identifier, thereby improving the security level.

[0130] Specifically, the symmetric encryption and identity authentication mechanism includes:

[0131] Encryption process:

[0132] C = E(K, P);

[0133] Where C is the ciphertext; E is the encryption function; K is the encryption key; P is the plaintext data (data collected by the sensor);

[0134] Decryption process:

[0135] P = D(K, C);

[0136] Wherein, D is a decryption function; K is a decryption key, which is the same as the encryption key; C is the ciphertext; P is the plaintext data after decryption;

[0137] Authentication mechanism:

[0138] MAC = H(K, M);

[0139] Wherein, MAC is a message authentication code for verifying the integrity of the message; H is a hash function; K is a key; M is a message (such as the health data of the utility pole).

[0140] In a possible implementation, the system embeds encryption processing and identity authentication into the data frame structure at the same time, so that each data transmission unit contains a complete encryption header, authentication information and integrity check identifier, thereby improving the security level.

[0141] In some embodiments, to avoid data loss or retransmission errors caused by network interruption, the wireless transmission module is also provided with a two-way handshake confirmation mechanism and a retransmission strategy to ensure the reliable delivery of critical data.

[0142] Furthermore, to adapt to the requirements of different operators and environments, the wireless module also supports a remote activation mechanism based on SIM card identity recognition, which can complete regional channel initialization and network registration during deployment.

[0143] In summary, through the above technical composition, the wireless transmission module not only realizes data interconnection and remote interaction between various functional units within the system of the present invention, but also constructs a flexible communication and encryption processing framework adaptable to various application scenarios on the basis of ensuring transmission security, integrity and real-time performance, meeting the actual requirements of high security level and high stability in the monitoring of power infrastructure.

[0144] The modeling module, based on the wireless transmission signal, virtually displays the position of the utility pole, the wireless sensor network data, the health status estimation, the probability of failure occurrence, and the warning signal in real time.

[0145] In this embodiment, to achieve an intuitive expression and dynamic monitoring of the structural state and operation information of the utility pole, a modeling module is further provided. This module is used to receive the data stream sent by the wireless transmission module and build a digital twin model of the utility pole on this basis.

[0146] This model can synchronously display key operating parameters such as the spatial location, structural health status, estimation and analysis results, failure risk probability, and warning and response status of utility poles. The modeling module serves as the terminal functional unit for information visualization and decision assistance in the system architecture, forming a complete closed loop of data and logic with the aforementioned modules such as sensing and acquisition, state estimation, risk prediction, and warning response.

[0147] Generally, the modeling module takes the basic spatial information of the utility pole as an anchor point, constructs a virtual entity of the utility pole through 3D modeling or graphical symbolization means, and binds the virtual entity to its corresponding dynamic operating state to achieve a state-space integrated visual output. The location information of the utility pole can be sourced from the Global Positioning System (GPS), Beidou system, or cellular network-based triangulation algorithms at the deployment site. In some embodiments, it can also be combined with a GIS geographic information system for coordinate correction, environmental overlay, and scene rendering, thereby achieving precise positioning and layout display of the utility pole distribution in a 2D or 3D map environment.

[0148] Specifically, the core data received by the modeling module includes: the data acquisition results from the wireless sensor network, the health status estimation values processed by the data fusion module, the risk occurrence probability output by the failure prediction module, and the warning information and response instructions generated by the warning processing module. After being received, various types of data are dynamically mapped to different dimensions of the digital twin model. For example, in some embodiments, the indicators of the health status (such as strain changes, corrosion levels, environmental factors) are visually embedded in the utility pole model body through color coding, intensity levels, graphical identifiers, etc., to achieve graphical expression of information.

[0149] Specifically, the modeling module dynamically adjusts the visual features of the model surface or interface by docking with the state estimation results. For example, when a certain structural indicator reaches the set risk threshold, the system can generate a red highlight prompt in the model interface, or produce a flashing special effect for the utility pole identifier to attract the attention of the operator. In another possible implementation, the modeling module can provide a state trend comparison chart, showing the fluctuation trajectory of the state indicators of the utility pole over a period of time in the past, so as to assist in judging its deterioration trend or maintenance cycle.

[0150] At the same time, the system also supports map heat zone rendering for the failure occurrence probability. For example, the high-probability areas are centrally marked as orange or red, and floating prompt information is generated on the model, listing in detail the risk factors and prediction details. This function has high value for identifying key areas of concern in a large-scale power grid.

[0151] In some embodiments, the modeling module further integrates the visualization display capability of early warning information. The system can mark the corresponding status on the model in real time according to the current early warning level, such as label contents like "Early warning to be confirmed", "Under maintenance", "Restored", etc., and can overlay auxiliary data such as maintenance personnel scheduling information and response time. In addition, the model interface supports docking with the scheduling system, and operators can directly dispatch tasks, confirm responses, adjust the scheduling order, etc. based on the model status.

[0152] In a possible implementation, the modeling module also supports the time backtracking function. The system can retrieve the status data of historical periods and restore and display them in the model for accident analysis, operation and maintenance assessment, and training simulation. Through the visual reconstruction of historical data, this function can effectively improve the depth and traceability ability of operation and maintenance management.

[0153] Furthermore, the modeling module supports the networked visual layout of multiple pole groups. In the scenario of multiple pole deployments, the modeling module can cluster and present multiple electric pole models in dimensions such as regions, lines, or risk levels, and supports map-level zooming, filtering, and searching operations to achieve an overview and screening of the structural status under a wide-area power grid. In some embodiments, the system can also automatically generate a wireless sensor network map based on information such as pole layout density and communication topology structure to support network-level maintenance and node diagnosis.

[0154] To ensure operation convenience and system openness, the modeling module can adopt a multi-platform architecture of the Web end, desktop end, or mobile end, and cooperate with a graphics rendering engine and a map service interface to achieve cross-platform display. In scenarios with limited network, the modeling module can also combine edge rendering technology to ensure the consistency of data loading and interaction experience.

[0155] In summary, the modeling module in the present invention not only realizes the comprehensive virtual expression of the spatial distribution and structural status of electric poles, but also has powerful state perception, risk manifestation, and emergency response assistance capabilities. Through the combination of the digital twin concept and the multi-data fusion mechanism, this module provides an unprecedented intelligent visualization monitoring method for power infrastructure and can be used as an extended interface for future AI analysis, system simulation, and decision-making optimization modules.

[0156] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent operation and maintenance detection system for the whole life cycle of electric poles based on a wireless sensor network, characterized in that, It includes: A collection module for real-time monitoring of the health status of utility poles. A data fusion module for receiving data from a wireless sensor network and fusing the data based on the Kalman filtering algorithm to obtain an estimated health status of the utility pole. A fault prediction module that, based on the health status estimate output by the Kalman filter, combines the Bayesian inference algorithm to predict the probability of a fault occurring in the utility pole. An early warning processing module for sending out early warning signals and conducting emergency responses according to the health status estimate results and the probability of a fault occurring. A wireless transmission module for wirelessly transmitting sensor data, health status estimates, probabilities of fault occurrence, and early warning signals. A modeling module that, based on the wireless transmission signals, virtually displays the location of the utility pole, wireless sensor network data, health status estimates, probabilities of fault occurrence, and early warning signals in real time.

2. The intelligent operation and maintenance detection system for the whole life cycle of electric poles based on a wireless sensor network according to claim 1, wherein The collection module includes: A strain sensor for collecting strain data of the utility pole under external forces. A corrosion sensor for collecting data on the corrosion degree of the surface of the utility pole. A temperature sensor for collecting temperature data at the location of the utility pole. An anemometer for collecting the wind speed at the location of the utility pole. A humidity sensor for also collecting the humidity at the location of the utility pole.

3. The intelligent operation and maintenance detection system for the whole life cycle of electric poles based on a wireless sensor network according to claim 1, characterized in that, The Kalman filtering algorithm is used to dynamically estimate the health status of the utility pole based on the historical health status of the utility pole and real-time sensor data. The health status includes the strain force, corrosion degree, temperature at the location, wind speed at the location, and humidity at the location of the utility pole.

4. The intelligent operation and maintenance detection system for the whole life cycle of a utility pole based on a wireless sensor network according to claim 3, characterized in that, The state equation of the Kalman filtering algorithm is: Among them, represents the posterior estimation vector of the true state of the system at time t, that is, the estimated vector of the pole health state updated after x(t) is given; represents the prior estimation vector derived based on the state information at time t-1 and the state transition model at time t, which is the predicted value; K t represents the Kalman gain matrix, which is used to weight and correct the degree of prediction error. Its size determines the degree of trust in the current observation data. H represents the observation matrix, which is used to map the state vector to the observation space, and x(t) is the health state vector of the pole at time t.

5. The intelligent operation and maintenance detection system for the whole life cycle of a utility pole based on a wireless sensor network according to claim 1, wherein The fault prediction module constructs a fault probability model based on the Bayesian inference algorithm and calculates the probability of a fault occurring in the utility pole within a set time window by comparing the statistical correlation between the current health status and historical fault data.

6. The intelligent operation and maintenance detection system for the whole life cycle of electric poles based on wireless sensor network according to claim 5, characterized in that, The Bayesian inference algorithm includes: Among them, P(F|x(t)) is the posterior probability of the utility pole having a fault F under the condition of the health status vector x(t) of the utility pole at time t; P(x(t)|F) is the conditional probability of the health status vector x(t) occurring given that the utility pole has a fault F; P(F) is the prior probability of the fault F occurring; P(x(t)) is the total probability of the health status vector x(t) of the utility pole at time t.

7. The intelligent operation and maintenance detection system for the whole life cycle of a utility pole based on a wireless sensor network according to claim 1, characterized in that, The emergency response includes sending out early warning signals, adjusting the power grid load, and dispatching maintenance personnel to carry out repairs.

8. The intelligent operation and maintenance detection system for the whole life cycle of a telegraph pole based on a wireless sensor network according to claim 1, characterized in that, The modeling module establishes a digital twin model based on the location information of the utility pole and displays the health status, health status estimate, probability of fault occurrence, and emergency response of the utility pole in real time based on the digital twin model.

9. The intelligent operation and maintenance detection system for the whole life cycle of electric poles based on a wireless sensor network according to claim 1, characterized in that The wireless transmission module also includes a data security unit, which is used to encrypt the health status, health status estimate, probability of fault occurrence, and early warning signals, and adopts a symmetric encryption and authentication mechanism to prevent data leakage and tampering during wireless transmission.

10. The intelligent operation and maintenance detection system for the whole life cycle of power poles based on a wireless sensor network according to claim 9, characterized in that, The symmetric encryption and authentication mechanism includes: Encryption process: C = E(K, P); Among them, C is the ciphertext; E is the encryption function; K is the encryption key; P is the plaintext data (data collected by the sensor); Decryption process: P = D(K, C); Among them, D is the decryption function; K is the decryption key, which is the same as the encryption key; C is the ciphertext; P is the decrypted plaintext data; Authentication mechanism: MAC = H(K, M); Among them, MAC is the message authentication code, which verifies the integrity of the message; H is the hash function; K is the key; M is the message.

Images