国网辽宁省电力有限公司

In the novel power distribution system,photovoltaic(PV)charging stations have attracted much atten-tion as a typical distributed resource aggregation form.Since both distributed PV generation and charging loads are characterized by randomness and volatility,the load forecasting task of PV charging stations is particularly complex.In this paper,considering the dynamic impact of PV charging station access on the regional load profile,a PV char-ging station load forecasting method based on multi-layer limit learning machine and quantile regression theory is proposed.Firstly,the factors affecting the load of PV charging station are feature extracted and key feature quanti-ties are extracted.Secondly,combined with the quantile regression algorithm,a multi-layer kernel limit learning machine deep neural network model is constructed to realize the load interval prediction of PV charging station un-der different confidence levels,and the improved sparrow optimization algorithm is used for the parameter optimiza-tion and the optimal model is selected for the load prediction.Finally,the load data of a photovoltaic charging sta-tion in a northern region of China is selected for example analysis.The results demonstrate that the proposed PV charging station load prediction method oriented to the novel distribution system achieves satisfactory forecasting performance and enables more accurate acquisition of load forecasting information.

In the process of suppressing backward stray light in the satellite ground laser communication system terminal,traditional stray light suppression methods ignore the local detail information of the target image during the suppression process,resulting in color distortion and edge discontinuity,which affects the imaging quality.The impact of different polarization states(0°,45°,90°)on the suppression effect of stray light varies significantly,and it is necessary to achieve stability and consistency of the suppression results under multi band polarization conditions.In response to this issue,researches were conducted on stray light suppression methods considering multi band polarization.By analyzing the optical structure of backward stray light involved in satellite ground laser communication terminals,a stray light diffusion function was constructed.The multi-band polarization degree was calculated and the image pixels were constrained under the influence of stray light.A binarization mask approach was employed to generate a mask image,thereby enabling the separation of stray light and diffuse reflected light.By fully accounting for the local detail information of the target,the light reflectance of stray light was calculated.A weighted regularization constraint was introduced to mitigate issues such as color distortion and edge discontinuity that might arise after stray light suppression.By combining the weighted regularization information completion method,the problem of suppressing stray light was transformed into an energy function minimization problem,thereby suppressing stray light.The experimental results show that this method is not affected by complex polarization environments.The suppressed light intensities at 0°,45°,and 90° polarization states are 3.4×104 W/m2,3.6×104 W/m2,and 4.0×104 W/m2,respectively,which are consistent with the ideal light intensity.It shows that this method has good suppression reliability.

Graph data,as a data type with strong expressive power,is difficult to model efficiently due to its complex structure.How to effectively capture its intrinsic information has become a challenging problem.Graph embedding methods have received increasing attention by mapping high-dimensional sparse graphs into low-dimensional dense feature vectors,while preserving the structural information of graphs.However,the existing reviews do not summarize the graph embedding methods comprehensively enough,especially paying less attention to complex graph embedding,which leads to the failure to systematically sort out the current status of research on graph embedding in dealing with diverse graph data.Therefore,this paper presents a systematic review of graph embedding learning methods from simple to complex graphs.Firstly,it gives the common definitions of various types of graphs and graph embedding.Secondly,it systematically summarizes the embedding methods on simple graphs,including shallow and deep embedding methods.Then,it summarizes the embedding methods on complex graphs according to the types of graphs,focusing on the application of deep embedding techniques in complex graph structures such as dynamic graphs,heterogeneous graphs,multiplex graphs,and hypergraphs,to fill the gaps in the existing literature that is insufficiently researched on complex graph structures.Finally,it discusses the practical application scenarios of graph embedding techniques,and looks forward to the future development directions.

Under the background of"dual carbon",a low-carbon economic scheduling strategy considering carbon trading mechanism and flexible load is proposed for the demand of multi-energy utilization and low carbon operation in the park.The framework of multi-energy coupling system is constructed,and the electric vehicle load is involved in system dispatching as a flexible load to increase the flexibility of the load side,while carbon trading cost is intro-duced to internalize this part of external environmental cost.An economic dispatching model of multi-energy cou-pling system in the park is constructed with the objective of minimizing the comprehensive cost,including the ener-gy purchase cost,carbon transaction cost and surplus electricity income.The whale optimization algorithm(WOA)is adopted,and the Levy flight strategy and adaptive inertia weight are introduced to improve the global search and local optimization ability of the algorithm,so as to realize the efficient solution of the low-carbon economic schedu-ling model.The simulation results show that the convergence efficiency of the improved algorithm has increased by 50％.The introduction of carbon trading,flexible load and electric and thermal coupling equipment can reduce the overall carbon emissions of the park,promote the consumption of renewable energy,and improve the economy and environmental protection of the system.

The modular multilevel converter based high voltage direct current(MMC-HVDC)is a dynamic power balancing system.The control system of MMC generally adopts a dual closed-loop vector control strategy based on the traditional instantaneous power model under an asymmetric grid state,which has a complex control structure and low control accuracy.This paper introduces a flexible instantaneous power model and establishes a general power equation with active power and new reactive power as control objects. In order to eliminate the twice grid-frequency ripples in both active and reactive power under asymmetric grid states,an enhanced sliding-mode MMC-HVDC direct power control strategy based on the new instantaneous power model is proposed.This approach combines the improved sliding-mode control method with the flexible instantaneous power model.Furthermore,it omits the inner-loop controller and power compensation terms while optimizing the control structure.According to simulation results,the suggested approach can more effectively take advantage of the benefits of the flexible instantaneous power model because it has superior dynamic responsiveness,control precision,and robustness under operating conditions including asymmetric grid state and parameter perturbation.