Power transmission and transformation intelligent monitoring system based on wireless local area network

Abstract

The invention relates to the technical field of power system monitoring and wireless local area networks, and particularly discloses a power transmission and transformation intelligent monitoring system based on a wireless local area network, which comprises a self-energized monitoring node, a wireless access gateway and a central processing platform. The self-energized monitoring node obtains energy from the environment through the multi-source composite micro-energy acquisition module and cooperates with the self-adaptive power consumption management module to realize intelligent energy allocation. The wireless access gateway adopts a routing algorithm taking node residual energy and link stability as a core to realize efficient and reliable data relay and network energy consumption balance; and the central processing platform performs data analysis and early warning through the intelligent diagnosis engine. According to the invention, energy self-sufficiency and permanent operation of the monitoring nodes are realized, the operation and maintenance cost is reduced, the reliability of the whole monitoring network is improved, and the service life of the whole monitoring network is prolonged.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of power system monitoring and wireless local area network, and particularly relates to a power transmission and transformation intelligent monitoring system based on a wireless local area network. BACKGROUND

[0002] In the field of power system automation and intelligence, real-time and reliable monitoring of the operation state of power transmission and transformation equipment and lines is a key link to ensure the safe and stable operation of the power grid and improve operation and maintenance efficiency. The power transmission and transformation intelligent monitoring system based on a wireless local area network aims to use wireless communication technology to realize data collection and remote transmission of widely distributed and complex environment monitoring nodes, so as to overcome the limitations of traditional wired monitoring in deployment flexibility and cost.

[0003] The monitoring nodes equipped with various sensors and wireless communication modules gather the collected temperature, current, voltage and other data to the monitoring center through the wireless local area network. However, such systems face serious challenges in long-term unattended power transmission scenarios. In the prior art, a large number of monitoring nodes rely on battery power supply, and in remote, high-altitude or difficult-to-wire monitoring points, frequent manual battery replacement not only brings high operation and maintenance cost, but also has safety risks, which seriously restricts the long-term, continuous and reliable operation of the monitoring system. At the same time, the limited battery capacity also limits the sampling frequency of the sensor and the communication frequency, which may cause missed reports or delays of critical state information, and cannot meet the high requirements of rapid early warning and accurate positioning of power grid faults. SUMMARY

[0004] The purpose of the present application is to provide a power transmission and transformation intelligent monitoring system based on a wireless local area network to solve the technical contradiction that the monitoring nodes relying on battery power supply have high operation and maintenance cost, limited reliability and are difficult to guarantee long-term continuous operation in the prior art.

[0005] The present application provides a power transmission and transformation intelligent monitoring system based on a wireless local area network, which includes self-powered monitoring nodes deployed at key monitoring points of power transmission and transformation equipment and lines, wireless access gateways deployed at fixed facilities such as substations or power towers, and a central processing platform located in a monitoring center. The system realizes sustainable and highly reliable monitoring of power transmission and transformation state by building a wireless sensor network that integrates energy self-management, data intelligent relay and collaborative diagnosis.

[0006] Further, the self-powered monitoring node is the core sensing unit of the system, which integrates a micro-energy collection module, a super capacitor energy storage module, a self-adaptive power consumption management module and a sensing and communication integrated module. The node completely abandons the traditional chemical battery, and its energy supply completely relies on the collection and conversion of various physical field energies around the power transmission and transformation equipment and lines.

[0007] As an embodiment of the present application, the micro energy harvesting module adopts a multi-source composite design, specifically including an electromagnetic induction power unit, a vibration energy harvesting unit, and a thermoelectric power generation unit. The electromagnetic induction power unit converts the alternating magnetic field energy around the wire into electrical energy through the micro high-permeability magnetic core and coil around the current-carrying wire or busbar. The vibration energy harvesting unit adopts a piezoelectric cantilever beam structure, whose natural frequency is tuned to match the main frequency of the mechanical vibration of the target power transmission and transformation equipment, such as a transformer or a circuit breaker, to maximize the energy conversion efficiency. The thermoelectric power generation unit uses semiconductor thermoelectric materials, whose hot end is tightly attached to the heat generating part of the equipment through heat-conducting silicone, and the cold end is in contact with the air through the heat dissipation fins, thereby converting the waste heat generated by the equipment operation into electrical energy.

[0008] Further, the supercapacitor energy storage module serves as an energy buffer pool of the node, with its input end connected to the output end of each unit of the micro energy harvesting module through rectification and maximum power point tracking circuit, and its output end connected to the adaptive power consumption management module. The capacity of the module is configured according to the worst energy harvesting conditions of the target monitoring point and the minimum maintenance power consumption of the node, ensuring that the node can maintain standby and minimum frequency heartbeat signal sending capability for at least 72 hours in the case of continuous no energy input.

[0009] As an embodiment of the present application, the adaptive power consumption management module is built-in a multi-threshold energy state machine. The module monitors the terminal voltage of the supercapacitor energy storage module in real time and maps it to discrete energy levels. When the energy level is at the highest level, the module instructs the sensor and communication integrated module to enter the full-function working mode to perform high-frequency data acquisition and real-time data reporting. When the energy level drops to the intermediate level, the module automatically switches to the power saving working mode, reduces the sensor sampling rate and uses data compression algorithm to reduce the data volume of single communication. When the energy level reaches the lowest level, the module forces the node to enter a deep sleep state, only retaining the minimum monitoring function required to maintain the wireless network connection, until the energy level rises above the intermediate level threshold.

[0010] Further, the sensor and communication integrated module integrates at least one sensor for monitoring electrical or non-electrical quantities, such as a non-contact temperature sensor, a Hall current sensor, or a partial discharge ultrasonic sensor, and integrates a radio frequency transceiver chip conforming to the wireless local area network communication protocol. The working mode and parameters of the module are completely dynamically regulated by the adaptive power consumption management module.

[0011] In one embodiment of the present invention, the wireless access gateway not only undertakes data aggregation and forwarding functions but also serves as a regional energy and data collaborative management unit. Each gateway is responsible for managing a cluster of self-powered monitoring nodes within its wireless signal coverage area. The gateway has a built-in node topology maintenance unit and a dynamic routing calculation unit. The node topology maintenance unit periodically receives heartbeat signals and link quality reports sent by each node in the cluster to construct and update the network topology map in real time. The dynamic routing calculation unit calculates the optimal or suboptimal multi-hop relay path for data packets to travel from the source node to the gateway based on the topology map and global task instructions issued from the central processing platform.

[0012] Furthermore, the dynamic routing calculation unit employs a routing algorithm with node remaining energy and link stability as core metrics. This algorithm defines a comprehensive cost function for each potential relay path, which is a linear combination of the weighted sum of the reciprocals of the remaining energy of all relay nodes on the path and the weighted sum of the reciprocals of the path's average link quality index. The goal of the dynamic routing calculation unit is to select the path with the minimum comprehensive cost function value for each data packet to be transmitted. This allows data traffic to automatically avoid nodes with insufficient energy or unstable communication, guiding it to nodes with sufficient energy and reliable links for relaying, thereby achieving balanced network energy consumption and improved communication reliability at the global level.

[0013] In one embodiment of the present invention, the central processing platform includes a data warehouse, an intelligent diagnostic engine, and a system configuration manager. The data warehouse receives and stores time-series monitoring data uploaded from all wireless access gateways. The intelligent diagnostic engine performs in-depth analysis of the aggregated data based on a preset expert rule base and a machine learning model. The expert rule base contains pre-set multi-level alarm thresholds and logical criteria for typical equipment faults, such as transformer overheating, insulator flashover, and excessive line sag. The machine learning model employs a time-series prediction algorithm based on long short-term memory networks. This algorithm uses historical monitoring data as training samples to learn the normal evolution patterns of equipment state parameters, such as winding temperature and leakage current, and predicts their future short-term trends in real time. When real-time data continuously deviates from the prediction range or triggers existing expert rules, the intelligent diagnostic engine generates a corresponding level of warning or fault event.

[0014] Furthermore, the system configuration manager provides remote parameter configuration capabilities for all self-powered monitoring nodes across the network. Based on the output of the intelligent diagnostic engine or instructions from maintenance personnel, the configuration manager generates configuration commands for specific nodes or node groups, which are then distributed via the wireless access gateway. Configurable parameters include, but are not limited to, sensor sampling periods, data reporting thresholds, energy level thresholds for the adaptive power management module, and operating mode switching strategies. Through this remote flexible configuration capability, the system can dynamically adjust the behavior patterns of nodes according to actual monitoring needs and grid operating conditions, further optimizing overall energy utilization efficiency while ensuring key monitoring objectives are met.

[0015] In one embodiment of the present invention, the communication protocol between the wireless access gateway and the self-powered monitoring node adopts a hybrid media access control mechanism combining time division multiple access (TDMA) and carrier sense multiple access (CSM). During system initialization or significant changes in network topology, the gateway coordinates a centralized TDMA scheduling phase, allocating conflict-free fixed time slots to nodes within the cluster for transmitting critical control signaling and high-priority data. During stable operation, the system switches to a random contention access mode based on carrier sense multiple access to improve channel utilization and meet the needs of sudden data transmission. The gateway provides a time reference for the entire network through periodically broadcast synchronization beacon frames, ensuring clock synchronization among all nodes.

[0016] Furthermore, the sensing and communication integrated module of the self-powered monitoring node supports a fast wake-up mechanism with in-band signaling. When the node is in deep sleep mode, its radio frequency receiving circuit periodically listens to the channel with an extremely low duty cycle. If a wireless access gateway or adjacent relay node needs to send data or instructions to the sleep node, it will first send a preamble sequence of a specific pattern. Once the sleep node detects this specific preamble within the listening interval, it immediately starts the complete receiving circuit to receive subsequent data frames. This mechanism minimizes the node's standby power consumption while ensuring network reachability and timely response to any node.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention, by designing a multi-source composite micro-energy harvesting module and a supercapacitor energy storage module, enables monitoring nodes to continuously obtain energy from the electromagnetic field, mechanical vibration, and temperature difference of power transmission and transformation equipment, achieving complete self-sufficiency and permanence of energy supply. This fundamentally eliminates the high maintenance costs and safety risks caused by battery replacement, laying a solid foundation for the large-scale and permanent deployment of intelligent monitoring systems for power transmission and transformation.

[0018] 2. This invention introduces a dynamic routing algorithm with node remaining energy and link stability as core metrics, enabling data transmission paths in the network to adaptively avoid nodes with low energy or poor communication quality. This collaborative optimization of data flow and energy status not only significantly improves the data transmission reliability of the entire wireless sensor network, but more importantly, it achieves automatic energy balance among network nodes, preventing some nodes from prematurely exhausting their energy due to excessive relaying, thereby greatly extending the overall service life and stability of the entire network.

[0019] 3. This invention, through the intelligent diagnostic engine in the central processing platform, integrates expert rules and machine learning prediction models, achieving a leap from passive threshold-based alarms to proactive trend-based early warnings for equipment status. The system can identify potential faults earlier and more accurately. Furthermore, combined with the remote flexible configuration capabilities of the system configuration manager, the monitoring intensity of relevant monitoring nodes can be dynamically adjusted, achieving an intelligent balance between monitoring accuracy and system energy consumption, comprehensively improving the ability to perceive and prevent risks to power grid operation. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the overall technical solution architecture of the present invention; Figure 2 This is a schematic diagram of the core principle framework of the self-powered monitoring node in this invention; Figure 3 This is a logic framework diagram of the energy state management and working mode switching of the adaptive power consumption management module in this invention; Figure 4 This is a schematic diagram of the core algorithm principle of the dynamic routing calculation unit within the wireless access gateway in this invention; Figure 5 This is a logical framework diagram of the intelligent diagnosis and collaborative configuration of the central processing platform in this invention. Detailed Implementation

[0021] Example 1: The overall technical architecture of the intelligent power transmission and transformation monitoring system based on wireless local area network described in this invention is shown in the attached figure. Figure 1 To be continued Figure 5 As shown, the system consists of three main parts: self-powered monitoring nodes deployed at key monitoring points of power transmission and transformation equipment and lines; wireless access gateways deployed at fixed facilities such as substations or power towers; and a central processing platform located in the monitoring center. These three components form a closed-loop monitoring system with autonomous energy, collaborative data, and intelligent diagnostics through a wireless local area network communication protocol, enabling long-term, continuous, and highly reliable perception and early warning of the operating status of power transmission and transformation equipment.

[0022] Please refer to the attached document. Figure 2The self-powered monitoring node, as the core sensing unit of the system, comprises a micro-energy harvesting module, a supercapacitor energy storage module, an adaptive power management module, and an integrated sensing and communication module. This node completely abandons traditional chemical batteries; all its operating energy comes from the real-time energy capture and conversion of the physical field surrounding the power transmission and transformation equipment and lines. The micro-energy harvesting module adopts a multi-source composite design, specifically including three sub-units: an electromagnetic induction energy harvesting unit, a vibration energy harvesting unit, and a thermoelectric power generation unit. These three units work in parallel and complement each other to cope with energy supply fluctuations under different operating conditions.

[0023] The electromagnetic induction power harvesting unit forms a closed loop by wrapping a miniature, high-permeability magnetic core around a current-carrying conductor or busbar, and then winding multiple turns of copper coil around it. When an alternating current flows through the conductor, an alternating magnetic field is generated in the surrounding space. This magnetic field passes through the magnetic core and induces an electromotive force (EMF) in the winding. The induced EMF is converted into a DC voltage by a rectifier bridge and then input to the subsequent energy management circuitry. This unit can stably output power when the conductor current is not less than 5 amps, with a typical output voltage range of 3.3 volts to 12 volts and an output power of over 50 milliwatts.

[0024] The vibration energy harvesting unit employs a piezoelectric cantilever beam structure, with one end fixed inside the node housing and the other end freely suspended with a mass block attached. The natural frequency of this structure is precisely tuned to the 45 Hz to 65 Hz range to match the dominant mechanical vibration frequencies of typical power transmission and transformation equipment, such as transformer core vibration and circuit breaker operating shocks. When the equipment vibrates during operation, the cantilever beam deforms, causing charge separation within the piezoelectric material, thereby outputting AC power. This unit can output an average power of approximately 10 milliwatts when the vibration acceleration reaches 0.5g. The thermoelectric power generation unit utilizes the Seebeck effect of semiconductor thermoelectric materials. Its hot end is tightly attached to heat-generating components such as the transformer tank and circuit breaker housing using high thermal conductivity silicone grease, while the cold end is connected to aluminum heat sink fins exposed to the air. Under normal operating temperature rise conditions (hot end temperature 60 degrees Celsius, ambient temperature 25 degrees Celsius), a stable temperature difference of 35 degrees Celsius can be maintained, thus continuously outputting 5 to 15 milliwatts of DC power.

[0025] The outputs of the three energy harvesting subunits are all connected to an integrated energy conditioning circuit. This circuit includes a synchronous rectifier, a maximum power point tracking (MPPT) controller, and a multiplexer. The MPPT controller uses a perturbation-observation method to dynamically adjust the working load impedance of each harvesting unit with a period of 100 milliseconds, ensuring that it can extract maximum usable power from the environment under any lighting, current, vibration, or temperature difference conditions. The multiplexer prioritizes the optimal energy source to charge the supercapacitor storage module based on the real-time output power and voltage stability of each unit, while isolating inefficient or failed units to prevent reverse current loss.

[0026] The supercapacitor energy storage module serves as the node's energy buffer. Its core consists of a series of double-layer capacitors with a total nominal voltage of 5.5 volts and a total capacity of 10 farads. The module's input is connected to the energy conditioning circuit via a reverse-connection protection diode, while its output is connected to the power input of the adaptive power management module. Its capacity configuration is strictly calculated based on the historical lowest energy acquisition conditions of the target monitoring point and the node's minimum sustaining power consumption. Specifically, in an extreme scenario simulating 72 consecutive hours without any external energy input (i.e., conductor current below 1 ampere, equipment shutdown and no vibration, ambient temperature difference less than 5 degrees Celsius), the energy stored in the supercapacitor energy storage module is still sufficient to support the node in performing the following operations: sending a heartbeat signal containing node identification and basic link quality every 30 minutes, with each communication lasting 80 milliseconds and an RF transmission power of 0 dB / mW, while maintaining the operation of the real-time clock and low-power monitoring circuitry. Verified by actual testing, this configuration ensures that the node maintains network reachability for at least 72 hours under harsh conditions of complete darkness, no current, and no vibration.

[0027] Please refer to the appendix. Figure 3 The adaptive power management module incorporates a multi-threshold energy state machine, the core of which consists of a 12-bit analog-to-digital converter (ADC), a state decision logic unit, and multiple power domain control switches. The ADC samples the terminal voltage of the supercapacitor energy storage module once per second and converts this analog voltage value into a digital quantity. The state decision logic unit maps this digital quantity to three discrete energy levels: high energy level (terminal voltage ≥ 4.8 V), medium energy level (3.6 V ≤ terminal voltage < 4.8 V), and low energy level (terminal voltage < 3.6 V). Each energy level corresponds to a preset operating mode strategy.

[0028] When the energy level is high, the state decision logic unit outputs a full-function enable signal, instructing the integrated sensing and communication module to activate all sensors. This enables the acquisition of non-contact infrared temperature, Hall current, and ultrasonic partial discharge signals at a sampling frequency of 10 Hz. The raw data is then encapsulated directly through the wireless LAN protocol stack without compression and reported to the corresponding wireless access gateway every 5 seconds. At this time, the node's total power consumption is approximately 25 milliwatts.

[0029] When the energy level drops to the medium level, the state decision logic unit automatically switches to a power-saving operating mode. In this mode, the sensor sampling frequency drops to 1 Hz, and a lightweight data compression algorithm is activated. This algorithm uses differential encoding for temperature and current data, and wavelet packet decomposition for ultrasonic signals, retaining only the first three principal coefficients. The compressed data packet size is reduced by approximately 60%, and the single communication time is shortened to 30 milliseconds. The data reporting cycle is extended to once every 30 seconds. At this time, the node's overall power consumption drops to 8 milliwatts.

[0030] When the energy level reaches a low level, the state decision logic unit forces the node into a deep sleep state. In this state, power to all functional modules except the real-time clock, low-power RF monitoring circuit, and the state decision logic unit itself is cut off. The node listens to the wireless channel only once every 10 seconds with an extremely low duty cycle (0.1%), waiting for a specific wake-up preamble. At the same time, the heartbeat signal transmission cycle is extended to once every 30 minutes, and only contains the most basic node identification and voltage level information. At this time, the standby power consumption of the entire node is only 150 microwatts.

[0031] The integrated sensing and communication module integrates three sensing units: a non-contact infrared temperature sensor, an open-loop Hall current sensor, and a piezoelectric ultrasonic sensor, along with a Sub-1GHz RF transceiver chip compliant with the IEEE 802.11ah standard. The module's power supply, sampling triggering, data format, and communication parameters are all dynamically controlled by an adaptive power management module via an internal bus. For example, when switching to power-saving mode, the module disables the ultrasonic sensor's excitation circuit, only temporarily activating it upon receiving a specific command; the temperature sensor switches to a low-resolution mode (accuracy ±2 degrees Celsius) to reduce the power consumption of the analog-to-digital converter.

[0032] Wireless access gateways are deployed in locations with a stable mains power supply, such as substation control rooms or crossarms of power transmission towers. Their role is not only as data aggregation points but also as the core of regional energy and data collaborative management. Each gateway manages a cluster of self-powered monitoring nodes within its wireless signal coverage radius (typically 300 meters), with a cluster size of up to 128 nodes. The gateway internally includes a node topology maintenance unit and a dynamic routing calculation unit.

[0033] The node topology maintenance unit constructs a dynamically updated adjacency matrix by parsing the link quality indicator field and received signal strength indicator from the heartbeat signals periodically sent by each self-powered monitoring node. This matrix records the communication quality score between any two nodes, ranging from 0 to 100, with higher values ​​indicating more stable links. Simultaneously, each node embeds its current energy level (high, medium, low) into its heartbeat signal; this information is also recorded by the topology maintenance unit and synchronized to the dynamic routing calculation unit.

[0034] Please refer to the attached document. Figure 4 The dynamic routing calculation unit employs a routing algorithm that uses node remaining energy and link stability as core metrics. This algorithm calculates the routing path for each node from the source node. To the gateway The potential multi-hop path P is defined by the comprehensive cost function. Its mathematical expression is as follows:

[0035] in Representing a path Upper Normalized residual energy of each relay node (1.0 for high energy level, 0.6 for medium energy level, and 0.2 for low energy level). This represents the quality score of the link between nodes i and j; and These are configurable weighting coefficients used to balance the priority between energy balance and communication reliability; the default value for both is 1. After receiving a data packet from the source node, the dynamic routing calculation unit traverses all feasible paths (with a maximum of 4 hops) and calculates the weighting coefficient for each path. Value, and select The shortest path is used as the relay route for this transmission. For example, if path A consists of two high-energy nodes but has average link quality (Q=70), while path B consists of one medium-energy node and one high-energy node but has excellent link quality (Q=95), the algorithm may choose path B to avoid excessive consumption of the high-energy node's energy, thereby extending the overall network lifetime.

[0036] The routing decision is sent to the relevant relay nodes in the form of routing instructions. Upon receiving the instructions, the relay nodes cache them in their local routing tables and forward the data packets within the specified time slots. If a link interruption or a sudden drop in the energy level of the next-hop node is detected during transmission, a route failure report is immediately sent to the gateway, triggering a new round of path recalculation.

[0037] The communication protocol between the wireless access gateway and the self-powered monitoring node employs a hybrid media access control mechanism combining Time Division Multiple Access (TDMA) and Carrier Sense Multiple Access (CSMA). During system initialization or when significant changes occur in the network topology (such as the addition of new nodes or widespread node outages), the gateway broadcasts a "Scheduling Mode Activation" command, entering the centralized TDMA phase. In this phase, the gateway assigns unique time slot numbers and lengths to each active node based on the current list of active nodes. Critical control signaling (such as configuration updates and emergency alarms) and high-priority data (such as partial discharge surge events) are confined to their respective assigned time slots, completely avoiding channel conflicts. The TDMA phase lasts for 5 minutes, after which the system automatically switches to CSMA mode.

[0038] In CSMA mode, nodes listen for channel idle time before transmitting data. If the idle time exceeds the distributed coordination function (TDMA) inter-frame interval, a random backoff is initiated, and transmission is attempted. To improve efficiency, the system employs an exponential backoff algorithm with an initial contention window size of 16 and a maximum retransmission count of 3. The gateway broadcasts a synchronization beacon frame every 100 milliseconds, containing a global timestamp, current network load status, and a preview of the next TDMA scheduling window. All self-powered monitoring nodes achieve microsecond-level clock synchronization by receiving this beacon frame, ensuring the accuracy of TDMA time slot alignment and sleep / wake-up timing.

[0039] Specifically, the sensing and communication integrated module of the self-powered monitoring node supports a fast wake-up mechanism with in-band signaling. When the node is in deep sleep, its RF receiving front-end turns on every 10 seconds, each turn lasting 1 millisecond, for channel listening. If a wireless access gateway or adjacent relay node needs to send instructions to the sleep node (such as emergency sampling commands or parameter updates), it will first send a specific 32-bit preamble sequence. This sequence is generated by a pseudo-random code generator according to a preset key and has low autocorrelation and high detection robustness. The sleep node's baseband processor performs correlation calculations on the received signal within the listening window. Once the correlation peak exceeds a preset threshold (typically 0.85), it is determined to be a valid wake-up signal, and the main processor, sensors, and complete RF link are immediately activated to prepare for receiving subsequent data frames. The entire wake-up process has a delay of less than 5 milliseconds, while the average listening power consumption during standby is only 10 microwatts, achieving a balance between extremely low power consumption and high responsiveness.

[0040] Please refer to the attached document. Figure 5 The central processing platform is deployed in provincial or municipal power dispatch centers, and its core components include a data warehouse, an intelligent diagnostic engine, and a system configuration manager. The data warehouse adopts a distributed time-series database architecture, receiving and storing structured monitoring data uploaded from all wireless access gateways across the network. Each data record includes a timestamp, unique node identifier, device type, monitoring point location, sensor type, raw data value, data quality flags, and routing path information, with a write throughput of up to 100,000 records per second.

[0041] The intelligent diagnostic engine consists of two parts: an expert rule inference engine and a machine learning prediction model. The expert rule base contains multi-level criteria for over 20 typical faults, including transformer winding overheating, insulator pollution flashover, abnormal conductor sag, and circuit breaker mechanical jamming. For example, for transformer overheating faults, the rule is defined as follows: when the top oil temperature continuously exceeds 85 degrees Celsius for 10 minutes, and the load current is greater than 90% of the rated value, while the ambient temperature is above 35 degrees Celsius, a level two warning is triggered; if the oil temperature further rises to 95 degrees Celsius, it is upgraded to a level one alarm. All rules support dynamic loading and version management.

[0042] The machine learning prediction model employs a time-series prediction algorithm based on a long short-term memory network. The model uses multi-dimensional time series data, including winding temperature, oil temperature, load current, and ambient temperature, collected every 5 minutes over the past 7 days, as input features to train a deep neural network, learning the state evolution patterns of the equipment under normal operating conditions. After training, the model can predict the trends of key parameters in real time for the next 4 hours and output a 95% confidence interval. When the actual monitored values ​​exceed this prediction interval for three consecutive sampling points, and the deviation exceeds 5%, it is judged as an early sign of an anomaly, generating a level-three warning. The model is automatically incrementally trained weekly to adapt to pattern drift caused by equipment aging or seasonal changes.

[0043] The system configuration manager provides a graphical user interface and application programming interface (API), supporting remote and flexible configuration of all network nodes by maintenance personnel or upper-level scheduling systems. Configuration commands include: shortening the temperature sampling period of a specific transformer monitoring node from 30 seconds to 5 seconds; increasing the partial discharge alarm threshold of a heavily loaded line node to suppress false alarms; and lowering the energy level switching threshold in the adaptive power management module of nodes in remote areas to extend their standby time. These commands, after being encrypted and signed, are sent to the corresponding wireless access gateway through the central processing platform, and then reliably delivered to the target node via the aforementioned hybrid MAC mechanism. Upon receiving the command, the node verifies its integrity and permissions; if valid, it updates its local configuration parameters and returns a confirmation response.

[0044] Through the above mechanisms, this system achieves four core capabilities: energy self-sufficiency, network self-healing, intelligent diagnosis, and flexible configuration. In a pilot deployment at a 500 kV substation, a total of 128 self-powered monitoring nodes were installed, covering key equipment such as main transformers, busbars, circuit breakers, and disconnect switches.

[0045] Example 2: Based on Example 1, this example optimizes the structure of the micro-energy harvesting module of the self-powered monitoring node to adapt to the special electromagnetic environment of UHVDC transmission lines. In UHVDC scenarios, there is no alternating magnetic field around the conductor, therefore the electromagnetic induction energy harvesting unit cannot operate. Therefore, this example replaces the electromagnetic induction energy harvesting unit with an electric field coupling energy harvesting unit.

[0046] The electric field coupling energy harvesting unit consists of a pair of parallel metal plates, each with an area of ​​100 square centimeters and a spacing of 5 centimeters, installed near the insulator string. The plates are connected to a rectifier and voltage regulator circuit via a high-resistance current-limiting resistor (100 megohms). During ±800 kV DC line operation, a potential difference of several kilovolts can be induced between the plates. After multi-stage voltage multiplier rectification and regulation, a stable 5-volt DC voltage can be output, with an average power of approximately 8 milliwatts. This unit, together with the vibration energy harvesting unit and the thermoelectric power generation unit, constitutes a new three-source composite acquisition architecture, ensuring continuous power supply capability for the node in DC transmission scenarios.

[0047] Meanwhile, considering the low vibration frequency of DC lines (typically below 10 Hz), this embodiment replaces the piezoelectric cantilever beam structure of the vibration energy harvesting unit with a low-frequency resonator driven by a magnetostrictive material. This resonator adjusts its resonant frequency to the range of 5 Hz to 15 Hz by a bias magnetic field, and can stably output 6 milliwatts of power under wind-induced vibration excitation of the conductor.

[0048] Furthermore, this embodiment enhances the algorithm of the dynamic routing calculation unit of the wireless access gateway. Considering that the monitoring nodes of the UHV line are distributed in a linear topology, the selection of relay paths is limited, and the original comprehensive cost function is prone to getting trapped in local optima. Therefore, a path diversity factor is introduced into the formula. The modified cost function is:

[0049] in, Representing a path With recent The Hamming distance of the used path is used to encourage traffic dispersion. The default value is 0.2, which can be dynamically adjusted in the system configuration manager. This improvement significantly enhances network load balancing capabilities in linear topologies, preventing a few relay nodes from becoming energy bottlenecks.

[0050] On the central processing platform side, the intelligent diagnostic engine has added expert rules for faults specific to DC lines, such as fitting overheating and uneven insulator contamination. It also introduces a graph neural network model to enhance the ability to identify local anomalies by utilizing the spatial topological relationships between nodes. The system configuration manager also supports batch distribution of differentiated configuration strategies by line segment. For example, for high-risk segments crossing rivers or mountains, the sampling frequency and reporting priority are automatically increased.

Claims

1. A power transmission and transformation intelligent monitoring system based on wireless local area network, characterized in that, include: Self-powered monitoring nodes deployed at key monitoring points of power transmission and transformation equipment and lines, wireless access gateways deployed at fixed facilities in substations or power towers, and a central processing platform located in the monitoring center. The self-powered monitoring node includes a micro-energy acquisition module, a supercapacitor energy storage module, an adaptive power management module, and an integrated sensing and communication module. The micro-energy acquisition module is used to collect the physical field energy around the power transmission and transformation equipment and lines and convert it into electrical energy. Its output terminal is connected to the supercapacitor energy storage module. The supercapacitor energy storage module serves as an energy buffer, with its input end connected to the micro-energy acquisition module and its output end connected to the adaptive power management module. The adaptive power management module is used to monitor the terminal voltage of the supercapacitor energy storage module in real time, and dynamically adjust the working mode of the integrated sensing and communication module according to the discrete energy level mapped by the terminal voltage. The sensing and communication integrated module integrates at least one sensor and a wireless local area network radio frequency transceiver chip, and its working mode is controlled by the adaptive power management module. The wireless access gateway is used to manage the cluster of self-powered monitoring nodes within its wireless signal coverage area and to act as a data aggregation and forwarding node. The central processing platform is used to receive and store monitoring data from the wireless access gateway, and to analyze and process the data.

2. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 1, characterized in that, The micro-energy harvesting module adopts a multi-source composite design, including an electromagnetic induction energy harvesting unit, a vibration energy harvesting unit, and a thermoelectric power generation unit. The electromagnetic induction energy harvesting unit converts the energy of the alternating magnetic field around the conductor into electrical energy through a miniature high-permeability magnetic core and coil surrounding the current-carrying conductor or busbar. The vibration energy harvesting unit adopts a piezoelectric cantilever beam structure, and its natural frequency is tuned to match the main mechanical vibration frequency of the target power transmission and transformation equipment. The thermoelectric power generation unit utilizes semiconductor thermoelectric materials. Its hot end is tightly attached to the heat-generating part of the device through thermally conductive silicone grease, while its cold end is in contact with the air through heat dissipation fins.

3. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 1, characterized in that, The adaptive power management module has a built-in multi-threshold energy state machine; When the energy level is at the highest level, the adaptive power management module commands the integrated sensing and communication module to enter the full-function working mode and perform high-frequency data acquisition and real-time data reporting. When the energy level drops to the intermediate level, the adaptive power management module automatically switches to a power-saving mode, reduces the sensor sampling rate, and uses a data compression algorithm to reduce the amount of data in a single communication. When the energy level reaches the lowest level, the adaptive power management module forces the node into a deep sleep state, retaining only the minimum listening function required to maintain the wireless network connection.

4. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 1, characterized in that, The wireless access gateway has a built-in node topology maintenance unit and a dynamic routing calculation unit. The node topology maintenance unit periodically receives heartbeat signals and link quality reports sent by each node in the cluster, and constructs and updates the network topology map in real time. The dynamic routing calculation unit calculates multi-hop relay paths for the transmission of data packets from the source node to the gateway based on the network topology diagram and global task instructions issued from the central processing platform.

5. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 4, characterized in that, The dynamic routing calculation unit adopts a routing algorithm with node remaining energy and link stability as the core metrics; The routing algorithm defines a comprehensive cost function for each potential relay path, which is a linear combination of the weighted sum of the reciprocals of the remaining energy of all relay nodes on the path and the weighted sum of the reciprocals of the average link quality index of the path. The goal of the dynamic routing calculation unit is to select the path with the minimum comprehensive cost function value for each data packet to be transmitted.

6. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 1, characterized in that, The central processing platform includes a data warehouse, an intelligent diagnostic engine, and a system configuration manager; The data warehouse is used to receive and store time-series monitoring data uploaded from all wireless access gateways; The intelligent diagnostic engine performs in-depth analysis of the aggregated data based on a preset expert rule base and machine learning model, and generates early warnings or fault events. The system configuration manager is used to generate remote configuration instructions for specific self-powered monitoring nodes or node groups based on the output of the intelligent diagnostic engine or the instructions of maintenance personnel, and to distribute them through the wireless access gateway.

7. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 6, characterized in that, The machine learning model in the intelligent diagnostic engine adopts a time-series prediction algorithm based on long short-term memory networks; The time-series prediction algorithm uses historical monitoring data as training samples to learn the normal evolution pattern of equipment state parameters and predict their future short-term trends in real time. When real-time data continues to deviate from the prediction range, the intelligent diagnostic engine generates an early warning event.

8. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 1, characterized in that, The communication protocol between the wireless access gateway and the self-powered monitoring node adopts a hybrid media access control mechanism that combines time division multiple access and carrier sense multiple access. When the system is initialized or the network topology undergoes significant changes, the gateway coordinates the entry into a centralized time-division multiple access scheduling phase to allocate conflict-free fixed time slots to nodes within the cluster. During stable operation, the system switches to a random contention access mode based on carrier sense multiple access; The gateway provides a time reference for the entire network by periodically broadcasting synchronization beacon frames.

9. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 1, characterized in that, The integrated sensing and communication module supports a fast wake-up mechanism with in-band signaling; When a node is in a deep sleep state, its radio frequency receiving circuit periodically listens to the channel with an extremely low duty cycle. If a wireless access gateway or adjacent relay node needs to send data or instructions to the dormant node, it will first send a preamble sequence of a specific pattern. Once the dormant node detects the specific preamble during the listening interval, it immediately activates the full receiving circuit to receive subsequent data frames.

10. The intelligent power transmission and transformation monitoring system based on wireless local area network according to claim 2, characterized in that, The capacity configuration of the supercapacitor energy storage module is calculated and determined based on the worst energy acquisition conditions of the target monitoring point and the minimum maintenance power consumption of the node. The capacity configuration ensures that the node can maintain standby for at least 72 hours and transmit the lowest frequency heartbeat signal even without continuous power input.

Images