Power system-oriented power optical fiber communication and inductance integrated transmission network architecture

Abstract

The invention discloses a power system-oriented power optical fiber communication and inductance integrated transmission network architecture, which is characterized in that the new architecture is of a five-layer model structure, namely an intelligent physical layer, a data layer, a cooperative computing layer, an intelligent model layer and an application layer. Wherein the intelligent physical layer is used for power system state sensing and communication signal transmission, the data layer processes and analyzes global data obtained by the intelligent physical layer and stores the global data in a cloud space, and the cooperative computing layer further optimizes the data stored in the cloud space by using computing power resources. The intelligent model layer carries out power grid fault assessment and risk prediction according to optimized data in combination with artificial intelligence and other modes, and the application layer monitors the state of a power system in real time and solves problems in time. Through mutual cooperation of the five layers of structures, not only is communication and sensing integration based on the electric power system realized, but also the problems of mutual interference of communication and sensing and high error rate of communication signals in a communication and sensing integrated system are solved, and the state monitoring precision and the instruction transmission reliability of the electric power system are improved while the accuracy and the stability of data transmission are improved; the operation and maintenance cost of the power grid is reduced, resource waste is avoided, effective implementation of intelligent management of the power grid is ensured, and support is provided for active defense and lean operation and maintenance of the intelligent power grid.

Description

Technical Field

[0001] The present invention relates to an electric power optical fiber communication integrated transmission network architecture for an electric power system, and belongs to the field of electric power communications. Background Art

[0002] With the continuous improvement of intelligent and automated power systems, power companies are increasingly demanding monitoring, protection, and dispatching of power network operations. Traditional electrical fiber-optic communication methods are subject to signal attenuation and transmission delays, especially in harsh environments such as high voltage, strong electromagnetic interference, and long-distance transmission. At the same time, the amount of sensor information in power equipment and systems in traditional power systems is increasing year by year. Monitoring methods based on electrical sensors and current and voltage measurement devices cannot meet the requirements of real-time performance, accuracy, and remote transmission. Furthermore, they cannot simultaneously communicate and sense data on the same transmission fiber. Therefore, how to efficiently and in real time acquire and transmit large amounts of monitoring data has become a critical issue facing power systems.

[0003] In recent years, optical fiber has become an ideal data transmission medium due to its high bandwidth, low loss, and strong anti-interference properties. Furthermore, the development of optical fiber sensing technology has further expanded its application in monitoring physical quantities such as temperature and strain. Optical fiber not only enables high-speed transmission of power data but also enables a variety of sensing functions, such as monitoring the status of power equipment and collecting environmental parameters. It is particularly suitable for power system applications that require high monitoring accuracy, fast response time, and reliable transmission.

[0004] However, traditional power fiber-optic communication networks primarily focus on data transmission, lacking comprehensive optimization solutions for the multi-dimensional sensing requirements of power systems. Existing fiber-optic sensing systems often operate independently, resulting in a disconnect between the fiber-optic sensing network and the power fiber-optic communication network. This prevents resource sharing and collaborative operation, and also presents challenges in transmission distance, signal quality, and equipment complexity. Therefore, a network architecture integrating power fiber-optic communication and sensing capabilities is urgently needed. This architecture can provide high-speed data transmission while effectively integrating sensor data from various power systems, thereby enhancing the intelligent operation and maintenance of power systems. Summary of the Invention

[0005] To solve the problem that the current power system architecture cannot solve the problem of communication and sensing integration and the mutual interference between sensing signals and communication signals, the present invention provides a power fiber optic communication and sensing integrated transmission network architecture for power systems.

[0006] The technical solution adopted in the present invention is:

[0007] An integrated power optical fiber transmission network architecture for power systems, including:

[0008] The intelligent physical layer ensures efficient transmission and real-time dynamic optimization of communication and sensor signals. As the basic transmission medium of the architecture, it guarantees signal quality and network stability. It has three module functions, including signal generation and modulation module, wavelength division multiplexing module, and optical power dynamic adjustment and monitoring module.

[0009] The data layer collects, stores and manages data from communication and sensor signals, extracts key features, and provides comprehensive data information support for system operation and analysis. It has three module functions, including data storage module, data acquisition and organization module, and basic data analysis and processing module.

[0010] The collaborative computing layer collaboratively utilizes edge computing and cloud computing resources to achieve signal processing, bit error rate optimization, nonlinear compensation and sensor performance improvement. It has three module functions, including signal processing module, computing resource scheduling module and nonlinear compensation module.

[0011] The intelligent model layer builds models through machine learning and big data analysis to achieve signal optimization, anomaly prediction and dynamic decision-making, providing intelligent support for system optimization. It has three module functions, including prediction module, optimization decision module and fault detection module.

[0012] The application layer provides intuitive visualization and management functions for specific businesses such as power monitoring, fault diagnosis, and resource scheduling, ultimately serving the actual power system operation needs. It has three module functions, including a monitoring and management module, a fault diagnosis module, and a visualization display module.

[0013] The present invention designs an integrated power optical fiber communication and sensing transmission network architecture for power systems. Compared with existing architectures, the present invention has the following advantages: the architecture of the present invention can realize the integration of communication signals and sensing signals on a single optical fiber, reduce the mutual interference between communication signals and sensing signals, reduce the bit error rate of communication signals, and use artificial intelligence for dynamic adjustment and fault prediction, timely detect and handle related problems, improve the security and processing efficiency of the power grid, reduce costs and minimize losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a block diagram of the integrated power fiber optic transmission network architecture designed for the power system by the present invention.

[0015] Figure 2 This is the signal transmission process described by the intelligent physical layer introduced in this invention.

[0016] Figure 3 This is the signal receiving process described for the data layer introduced in the present invention. DETAILED DESCRIPTION

[0017] The present invention will be described in further detail below with reference to the accompanying drawings.

[0018] See also Figure 1 , an integrated power optical fiber transmission network architecture for power systems, including an intelligent physical layer, a data layer, a collaborative computing layer, an intelligent model layer, and an application layer.

[0019] like Figure 2 As shown, the intelligent physical layer first generates a communication signal using a 1550nm signal source. A 1450nm narrow-linewidth laser generates a Brillouin Optical Time Domain Reflectometer (BOTDR) sensor signal, and a 1510nm narrow-linewidth laser generates an OTDR (Optical Time Domain Reflectometer) sensor signal. After generating the 1550nm communication signal, the electrical signal is processed using forward error correction coding (FEC). An electro-optical modulator then modulates the communication signal into optical light for propagation. A power modulator then controls the optical power of the communication signal within a certain range to prevent interference with the high-power sensor signal. The sensor signal is then modulated into a high-power narrow pulse using an acousto-optic modulator (AOM), ensuring high-precision measurement over long distances.

[0020] A wavelength division multiplexer (WDM) multiplexes the three signals into one signal for transmission. Variable optical attenuators (VAs) are incorporated into the transmitter and link. By interacting with the upper layer, they dynamically adjust the power of each signal to control the total power of the sensing signal and ensure the fiber's nonlinear threshold. The transmission line uses G.654 fiber and incorporates Raman amplifiers and erbium-doped optical fiber amplifiers (EDFAs). This allows for dynamic adjustment of the pump position, minimizing interference between the sensing and communication signals.

[0021] After the signal of the smart physical layer is transmitted to the data layer, Figure 3As shown, the data layer first uses a wavelength division multiplexer (WDM) and narrowband optical filters to separate the transmission signal into three channels: 1550nm, 1450nm, and 1510nm. These channels also eliminate spurious noise and improve signal quality. The data layer then processes the communication and sensor signals separately. The communication signal is equalized, denoised, and decoded using digital signal processing (DSP) modules. The high-order modulated signal is recovered and its frequency domain characteristics are analyzed using a fast Fourier transform (FFT). The sensor signal is fitted using BOTDR sensing technology to obtain monitoring data. The fitted data is analyzed and Brillouin frequency shifts are extracted to calculate changes in temperature, strain, and bending. The Brillouin spectrum is then used to identify the location of the problem, and the OTDR analyzes the time reflection curve to locate fractures and aging. The processed data is uniformly encoded and stored in a distributed database or cloud space, providing real-time data support for artificial intelligence (AI).

[0022] The implementation methods of the collaborative computing layer include: the design of the collaborative computing layer adopts a layered architecture to achieve efficient collaboration. Heterogeneous computing units integrating FPGA (Field Programmable Gate Array) and GPU (Graphics Processing Unit) are deployed on the edge side. The FPGA is responsible for processing real-time signal streams with strict timing, such as industrial equipment control instructions and sensor data frame parsing, and the GPU accelerator card uses the CUDA (Compute Unified Device Architecture) parallel computing architecture to perform local image preprocessing and spectrum analysis; a high-performance computing cluster based on a distributed architecture is built on the cloud side, and a transmission channel is established through a low-latency fiber optic dedicated line to dynamically migrate computing-intensive tasks such as deep learning model training and multi-node joint simulation to the cloud GPU node group; the intelligent scheduling center has a built-in multi-dimensional evaluation model to continuously monitor the task computing volume, network throughput and equipment load status. When a machine vision data stream is detected, the diversion strategy is automatically triggered, and the edge GPU is preferentially called to perform feature extraction. At the same time, the 3D point cloud reconstruction task is dispatched to the cloud computing card equipped with Tensor Core, forming a computing resource pool that complements the end and the cloud.

[0023] The intelligent model layer acquires processed fiber-optic sensing and communication signals from the data layer and uses deep learning models to perform pattern recognition and prediction on these signals, detecting cable breaks, aging, and environmental anomalies. The analysis results are then fed back to the collaborative computing layer for optimization of signal processing strategies (such as dynamically adjusting communication signal power). Simultaneously, the intelligent model layer provides real-time status monitoring and fault location support to the application layer through device health assessment and fault diagnosis, helping to achieve self-healing capabilities for the power grid.

[0024] Based on the analysis results of the intelligent model layer, the application layer dynamically monitors fiber status and rapidly responds to faults. Combined with the resource scheduling capabilities of the collaborative computing layer, it optimizes fiber resource utilization and communication bandwidth allocation. Through this layered collaboration, the data layer provides high-quality signals, the collaborative computing layer optimizes processing efficiency, and the intelligent model layer implements intelligent analysis. Ultimately, the application layer enables real-time monitoring, fault repair, and resource management, forming a highly efficient and intelligent integrated power fiber optic communication system.

[0025] These five layers work closely together to form a closed-loop system: the intelligent physical layer generates and modulates communication and sensor signals to ensure high-quality transmission; the data layer separates, filters, extracts features from signals, and stores structured data; the collaborative computing layer uses hardware accelerators to optimize signal processing and dynamically allocate computing resources; the intelligent model layer uses deep learning and pattern recognition to perform fault diagnosis and prediction, and dynamically optimizes system parameters; the application layer monitors the power grid status in real time, combines AI to achieve self-healing and resource management, and feeds optimization suggestions back to the physical layer, forming a closed-loop control from signal generation to intelligent decision-making, ensuring efficient and reliable operation of the power grid.

Claims

1. A power optical fiber integrated transmission network architecture for power systems, characterized by: Includes the following: The intelligent physical layer provides efficient transmission and real-time dynamic optimization of communication and sensor signals. As the basic transmission medium of the architecture, it ensures signal quality and network stability. The data layer collects, stores, and manages communication and sensor signal data, extracts key features, and provides comprehensive data information support for system operation and analysis; Collaborative computing layer, which collaboratively utilizes edge computing and cloud computing resources to achieve signal processing, bit error rate optimization, nonlinear compensation, and sensor performance improvement; The intelligent model layer builds models through machine learning and big data analysis to achieve signal optimization, anomaly prediction, and dynamic decision-making, providing intelligent support for system optimization; The application layer provides intuitive visualization and management functions for specific businesses such as power monitoring, fault diagnosis, and resource scheduling, ultimately serving the actual power system operation needs.

2. The smart physical layer according to claim 1, characterized in that: It has three module functions, including signal generation and modulation module, wavelength division multiplexing module, and optical power dynamic adjustment and monitoring module.

3. The data layer according to claim 1, characterized in that: It has three module functions, including data storage module, data collection and organization module, and basic data analysis and processing module.

4. The collaborative computing layer according to claim 1, characterized in that: It has three module functions, including signal processing module, computing resource scheduling module, and nonlinear compensation module.

5. The intelligent model layer according to claim 1, characterized in that: It has three module functions, including prediction module, optimization decision module and fault detection module.

6. The application layer according to claim 1, characterized in that: It has three module functions, including monitoring and management module, fault diagnosis module, and visualization display module.

Images