广东电网有限责任公司

Accurate detection of internal partial discharge(PD)in gas-insulated switchgear(GIS),a core compo-nent of modern power systems,is crucial for assessing insulation condition,ensuring equipment reliability,and maintaining power grid stability.However,the practical application of traditional ultra-high frequency(UHF)elec-tromagnetic sensors in GIS condition-based maintenance faces significant limitations,including severe signal attenu-ation(20～40 dB/m)caused by the metal enclosure,the risk of SF6 leakage due to compromised airtightness with window-based installation,and a limited dynamic range(60～80 dB)insufficient for covering the full range of dis-charge magnitudes.To overcome these technical bottlenecks,this paper innovatively proposes a quantum-based UHF electric field sensing method utilizing the Rydberg atomic electromagnetically induced transparency(EIT)effect.The core advantages of this method lie in:leveraging the extremely high electric polarizability of highly ex-cited cesium atoms(n=72D5/2)to achieve intrinsically safe measurement through an optical readout mechanism;establishing a stable EIT window using a dual-frequency laser system(852 nm probe light and 509 nm pump light)combined with precise frequency-locking technology;and designing a non-metallic atomic vapor cell antenna to re-alize the first near-field sensing of transient pulse electric fields in the 0.5～3 GHz band excited by particle dis-charge within GIS.This research demonstrates the capability of Rydberg atomic sensing technology to effectively de-tect discharges from a typical GIS defect(metal particles)for the first time.By employing quantum precision meas-urement,it provides a novel optical fiber sensing solution characterized by high sensitivity,broadband capability,and intrinsic safety for the online insulation monitoring of power equipment.This approach holds significant promise for advancing condition-based maintenance towards intelligent and precise evolution,offering key supporting tech-nologies for the development of smart grids.

After the distributed photovoltaic power source is connected to the distribution network,its output is af-fected by environmental factors such as sudden changes in irradiance and cloud cover,showing significant asymmet-ric fluctuations and thick tailed distribution characteristics,resulting in a multi-modal probability distribution of node voltage(i.e.multiple peaks in the probability distribution),which makes it difficult to accurately calculate the risk index of voltage over-limit of the distribution network nodes and increases the risk of voltage over-limit.On this basis,a new method for regulating the output of distributed photovoltaic power generation in novel power system considering the risk of voltage over-limit is proposed.The multi-head attention neural network-factorization machine(MANN-FM)model is utilized to predict future photovoltaic output,capturing the randomness and intermittency characteristics of photovoltaic output more accurately.Based on these prediction results,the Cornish-Fisher method is used to calculate the risk index of voltage over-limit at distribution network nodes connected to distributed photo-voltaic power,and to evaluate the risk of voltage over-limit.When it is predicted that the output of photovoltaic power generation will exceed the risk value of grid voltage over-limit,the output of distributed photovoltaic power generation is dynamically adjusted by implementing a comprehensive control strategy of active power and reactive power.When the risk of voltage over-limit is relatively small,only reactive power regulation is implemented.If the risk indicators continue to rise and exceed the preset threshold,the active and reactive power cooperative control mode will be activated.Through the linked control of limiting the maximum output power and enhancing reactive power compensation,the voltage of the distribution network is ensured to remain stable within the safe operation range.The experimental results show that in the typical scenario based on the 10 kV medium-voltage distribution network,the proposed method has been verified for effectiveness in the IEEE 33-node system.The voltage limit is exceeded only once,and the power fluctuation is controlled around 10 kW.It can ensure that the distribution net-work voltage operates within a safe range and is suitable for practical applications.

At present,due to the complex structure of multi-segment and multi-branch in distribution network,there are problems of long time consumption and low efficiency in line grounding fault location.Taking a typical complex distribution network area as an example,the grounding fault rate reached 10 times/100 kilometers in the past year,and the traditional location method takes about 25 minutes on average.When the traditional binary parti-cle swarm optimization algorithm is applied to this scene,it is easy to fall into local optimization and global optimi-zation is limited because of the lack of constraint of speed update mechanism.Moreover,the single fitness function of other methods cannot give consideration to both accuracy and efficiency,which leads to insufficient positioning accuracy and is difficult to meet the actual needs.Therefore,this paper proposes a method of cross-domain line grounding fault location in distribution network based on improved binary particle swarm optimization.Firstly,the fault criterion is established by constructing the distribution network operation topology based on directed graph,and the regional location of cross-domain line grounding fault is realized.Then,an improved binary particle swarm optimization algorithm is proposed on the basis of regional positioning.By introducing a dynamic adaptive Sigmoid function to dynamically limit the particle motion speed,the negative influence of particle local saturation on the global optimization result is effectively avoided.Meanwhile,a dual fitness function is constructed based on the fault area information,in which one fitness function focuses on improving the positioning accuracy and the other focuses on the convergence speed of the algorithm,thus significantly improving the search efficiency while ensuring the po-sitioning accuracy.Finally,the effectiveness of the method is verified by experiments.The experimental results show that the average error of the improved binary particle swarm optimization algorithm is about 2.5 m,which is 34.2%lower than that of the traditional particle swarm optimization(PSO),and it can be effectively applied to the location of grounding faults in distribution network,providing support for the operation and maintenance as well as safe and stable operation of power grid.

Power big data covers all links such as power generation,transmission,transformation,distribution and consumption,featured with the characteristics of wide data sources,complex data structure,high real-time require-ments and containing a large amount of sensitive information.During the sharing process,once the data is leaked,it will not only infringe upon the privacy of users,but also threaten the safe and stable operation of the power sys-tem.It has significantly increased the difficulty of sharing power big data.Traditional methods usually complete en-cryption directly before data sharing,but they cannot ensure the security of the encryption environment for power big data,resulting in the problem of theft still occurring during the sharing process.Therefore,a secure sharing technology for power big data based on consortium chain and attribute encryption is proposed.A power big data sharing platform is established based on alliance chain,on the consortium chain,attribute encryption methods are used to set encryption access strategies for power big data.The consortium chain can provide a trusted environment where all parties can record the sharing process and usage of data through blockchain,ensuring that only users who meet specific attribute conditions can access the corresponding power data;based on Lagrange interpolation and Shamir secret sharing scheme,power data is segmented and distributed to ensure the security and privacy of power data during the sharing process,while improving the efficiency and flexibility of power data access.The experimen-tal results indicate that the proposed data security sharing method is more suitable for practical applications and has higher security.

Aiming at the problems of gas pressure fluctuation and background noise in the insulation state detection of gas insulated switchgear(GIS)equipment,a high sensitivity detection method for the insulation state of eco-friendly gases GIS equipment based on photoacoustic spectroscopy technology was proposed.The method used photoacoustic spectroscopy technology to suppress background noise and accurately detect the components of eco-friendly gases.Combined with the maximum correlation minimum redundancy criterion,the insulation state characteristics were optimized to improve the detection pertinence.High sensitivity detection was achieved by building a model with support vector machine.The detection process included photoacoustic signal detection,noise suppression,component concentration determination,feature selection and classification.Experiment results on C4/CO2 gas mixtures demonstrated a fitting accuracy of 0.978 for C4 compoment,and the dynamic response entropy index consistently remaining below 0.1,indicating that the proposed method has excellent detection sensitivity for monitoring the insulation state of eco-friendly gas GIS equipment,capable of accurately capturing the micro molecular movement changes.This provides a reliable basis for the early diagnosis of equipment insulation deterioration.