In view of the ice melting of the ground wire of the deactivated line of a ±500 kV double-circuit DC transmission line with the same tower, when the ground wire of the deactivated line is melted under the condition of "one shutdown and the other operating", the failure of the ice melting device and the breakdown of the automatic wiring device and the lightning arrester caused by excessive induced voltage are studied. According to the calculation and analysis of 6 working conditions, the overvoltage level induced on the shutdown line when the corona initiation voltage of the live running circuit conductor is calculated, and the overvoltage suppression level and line energy are comprehensively considered, and the overvoltage suppression measures with 5 kΩ grounding high resistance are proposed, which can realize the induction voltage of more than 100 kV to less than 20 kV, and the effectiveness is verified by the actual line ice melting operation.

An improved modulation model predictive control (MMPC) method is proposed to enhance the control performance and efficiency of the three-level indirect matrix converter-permanent magnet synchronous motor (TLIMC-PMSM) system, which suffers from the non-fixed switching frequency issue in model predictive control (MPC) and the heavy computational burden in traditional modulation model predictive control (MMPC). Firstly, the topology structure and principles of the TLIMC-PMSM system are introduced in detail. Secondly, the issues of the control system based on MPC and traditional MMPC are identified. Then, the improved MMPC method is proposed, which simplifies the sector selection method of the inverter level, adopts the improved output voltage prediction MMPC method, and modifies the inverter level value function. Finally, the experimental verification on the MATLAB/Simulink demonstrates that the TLIMC-PMSM system based on the proposed improved MMPC method exhibits superior control performance compared to the two traditional control methods.

Using deionized water as the cooling medium, the cooling system of phase-change valve faces challenges in balancing insulation and heat dissipation performance,which affects the reliable operation of the system.Leveraging superior heat dissipation capabilities, phase-change cooling techniques are being gradually applied for cooling high-voltage high-power power electronic devices.However, bubbles in the liquid dielectric pose risks of insulation degradation.To better understand the migration characteristics and insulation characteristics of bubbles in the extremely inhomogeneous electric field associated with phase-change cooling, numerical simulation methods are applied to analyze the effects of three key factors on the trajectory of bubble motion,which are bubble diameter, velocity and electric field strength. Furthermore, after entering low field strength regions, the impact of bubbles with different contact angles, on the electric field distribution over insulator surfaces has been studied.The results reveal significant effects of various factors on bubble motion trajectories, provide an improved understanding of bubble migration and insulation characteristics in phase-change,and hold important reference value for the further application of phase-change cooling in DC equipment.

In order to ensure the safety of personnel and equipment during normal live working of distribution network, abnormal behavior recognition is an indispensable technical means. However, the existing live working behavior recognition methods have problems such as low accuracy, few identifiable types, and missed detection and false detection caused by background interference. An abnormal behavior recognition method is proposed for 10 kV live working video based on improved spatial temporal graph convolutional networks (ST-GCN). Firstly, the method of target detection and tracking is used to add a mask in the video human region to eliminate the influence of complex background. Then, the human skeleton is obtained by using the lightweight improved pose estimation model, and the spatio-temporal graph is constructed with multi-frame skeleton sequence. Finally, the spatial posture and timing information of the spatio-temporal map are extracted by ST-GCN, and the squeeze-and-excitation networks (SENet) module is introduced to strengthen the action features to complete the live working behavior recognition. The behavior dataset is constructed with some live working videos, and five typical behaviors such as insulation detection are selected for verification. The experimental results show that this method improves the speed of human pose estimation, constrains the skeleton detection area, and reduces the false detection rate and missed detection rate of limbs. It can effectively identify the live working behavior and reach 88% average accuracy rate. It has good generalization in complex environment and provides an effective reference for the intelligent safety monitoring of live working.

The dual-carbon target and the rapid development of new energy generation are the symbols of the new power system. High proportion of renewable energy and power electronic equipments bring new features and challenges to the new power system in dynamic characteristics, power balance and dispatching operation mode, and also promote the speed of digital grid technology and application. From the aspect of the connotation, characteristics and development requirements of the new power system, the evaluation index set of the construction of the new power system is put forward considering the differences in resource endowments and the mutual contribution of trans-regional electric energy. The system time series production simulation and the quantitative calculation of the new power system index are carried out for the planning year 2025 and 2030 of China Southern Power Grid. On this basis, the new power system planning and construction in China Southern Power Grid is evaluated and analyzed.

As a part of the new energy system, energy storage has been widely used in the distribution network due to its excellent regulation effect. However, the energy storage life will be affected by its operation characteristics, resulting in inaccurate planning results. In order to solve this problem, a distribution network energy storage planning model is proposed considering the cyclic discharge depth constraints. Firstly, the basic principle of rain flow counting method is introduced. Combining with the idea of piecewise linearization, the corresponding relationship between discharge depth and energy storage life is established. On this basis, an energy storage life model considering the constraint of equivalent cyclic discharge depth is proposed. Then, taking the minimum total cost as the objective function, a planning model considering the energy storage life is established. Secondly, the non-convex terms in the model are convex and transformed into a second-order cone programming model. Finally, simulation verification is carried out on the improved Portugal 54 example system to verify the effectiveness of the proposed model.

The continuous integration of high proportion new energy and diverse loads presents dynamic and AC/DC coexistence operating characteristics in the distribution network, which poses new challenges for the reconstruction of the distribution network. A multi-objective model for dynamic reconfiguration of AC/DC distribution networks was established, taking into account the consumption problems caused by the high proportion of new energy access, with the objectives of maximizing the daily consumption of new energy, minimizing the comprehensive cost of reconfiguration, and minimizing the mean square error of net load. A method combining candidate solution sets and improved fireworks algorithm (IFWA) was proposed to optimize and solve the multi-objective model based on the topology structure and operational technical requirements of the distribution network. The candidate solution set is generated based on network topology and operational technology constraints, with the aim of eliminating a large number of infeasible solutions during the optimization process and improving search efficiency. On this basis, the speed of multi-objective optimization was improved by improving the explosion and mutation operators of the fireworks algorithm. Adopting an improved IEEE 69 node system for simulation comparison and analysis. The results indicate that the established model can effectively coordinate the consumption of new energy and reconstruction costs, and the proposed optimization solution strategy has significantly improved operational efficiency compared to before the improvement.

The energy endowment and load characteristics of various region are different in integrated energy system due to geographical dispersion, which causes an unreasonable resource allocation inevitably. In addition, energy systems in various region are usually different energy operators, so the traditional centralized scheduling calculation cannot protect the data privacy of each region. Aiming at the economic dispatch problem of complex and high-dimensional integrated energy system, a distributed crisscross optimization with sine cosine algorithm(DCSO-SCA) is proposed. Firstly, sine cosine algorithm (SCA) is added to crisscross optimization algorithm (CSO), which improves the diversity of population and the ability of algorithm development. Secondly, DCSO-SCA algorithm realizes the parallel optimization of regional scheduling without centralized controller, the proposed algorithm employs CSO-SCA independently to optimize the area dispatch in parallel, reduces the solving dimension of economic dispatch problem, solves the integrated energy system economic dispatch problem in a completely decentralized way and protects the data privacy of each region. Finally, by comparing the case with other technologies, DCSO-SCA algorithm can obtain a multi-regional integrated energy system economic dispatch scheme with high efficiency and low energy consumption.

Against the backdrop of the increasingly expanding impact of energy transformation, the integration of high proportion of renewable energies and power electronic devices brings challenges to the research of the new power system. The traditional power system is beginning to shift towards a comprehensive dynamic balance of power grid, power source, load and energy storage. The model simulation of the new power system requires more accurate and rapid dynamic simulation to verify. However, there is currently a lack of efficient simulation solutions and effective simulation tools for large-scale electromagnetic transient simulation of the new power system. Based on the domestic real-time simulator (UREP) and taking the Simulink model of the medium voltage distribution network as an example, this paper proposes the concept of "full topology, full electromagnetic transient, and full configuration", and uses ideal transformer model(ITM) segmentation to reduce the order of the system model and multi-core parallel computing technology to kernel the model, the results show that the proposed modeling method not only improves the simulation accuracy of the new power system model, but also greatly shortens the simulation time of the model, which verifies the feasibility of large-scale new power system simulation.

Wind turbines connected to the power grid through series and parallel compensation can reduce line losses and effectively improve line transmission power and system stability. However, the interaction between the compensation capacitor and the wind turbine control device may cause a wide-band resonance stability problem including sub-synchronous resonance(SSR)and high frequency resonance(HFR). Considering two types of grid-connected lines, series compensation and parallel compensation, the impedance model of doubly-fed wind turbine grid-connected system is deduced and established. An impedance method is used to analyze the mechanism of wide-band resonance of wind turbine grid connected system, and an adaptive wide-band resonance suppressor is proposed and designed, which is connected in parallel to the point of common coupling (PCC). By extracting the resonant voltage of PCC and using the method of harmonic content limitation, a damping resistor is synthesized to suppress the wide-band resonance generated by the system. Finally, an equivalent system model of wind power grid integration system is built in MATLAB/Simulink, and the correctness and effectiveness of the adaptive wide-band resonance suppressor are verified by time domain simulation.

In order to solve the problem that single-phase grounding fault detection (especially high-resistance grounding fault) in traditional distribution network is affected by weak fault characteristics and difficult to detect, this paper proposes a new detection method for single-phase grounding fault detection in eutral point ungrounded distribution network based on two-terminal measurement of power factor angle from the perspective that the power of distribution network system will change significantly before and after the fault. The two voltage transformers connected with the neutral point inject controllable current signals into the neutral point and measure the return voltage signals, so as to accurately measure the active and reactive power of the distribution network at the current operating state, and then obtain the power factor angle of the system. Combined with the power factor angle of the distribution network under normal operation, the ratio coefficient of power factor angle before and after the system failure is measured in real time. Based on the comparison between the ratio coefficient and the ratio threshold coefficient, the sensitive detection of high resistance grounding fault is realized. The smart application of dual voltage transformer eliminates the measurement error caused by internal resistance and detuning resistance of the transformer in principle and greatly improves the effective detection range of high resistance grounding fault in distribution network. The proposed method is verified by setting up a 10 kV neutral point ungrounded distribution network in the simulation environment of PSCAD/EMTDC. Different types of high resistance grounding faults are simulated to verify the proposed method. The simulation results show that the proposed method can detect high resistance grounding faults reliably, and the detection range is up to 20 kΩ.

A decoupling linear auto disturbance rejection control strategy is proposed to address the phenomenon of voltage fluctuations in the DC bus of hybrid microgrids caused by factors such as sudden load changes, disturbance addition, and system grid connection. Firstly, a mathematical model of the hybrid microgrid converter is established based on the circuit structure of the hybrid microgrid. Then, in order to improve the system′s anti-interference ability, decoupling treatment is performed on the linear self disturbance rejection to form a decoupling type linear self disturbance rejection. Furthermore, in order to achieve real-time parameter optimization, a fuzzy adaptive optimization algorithm is introduced to form a fuzzy decoupling auto disturbance rejection control strategy. Finally, a physical and digital model of the hybrid microgrid is established and simulation and semi physical simulation experiments are conducted. The results show that the fuzzy decoupling active disturbance rejection control strategy has excellent ability to suppress DC bus voltage fluctuations.

Aiming at the problem that residual current detection in low-voltage distribution areas containing photovoltaic power supply is easily affected by multiple factors and difficult to achieve accurate detection of leakage faults, a leakage fault detection method for low-voltage distribution areas containing photovoltaic power supply is proposed based on the random forest algorithm, taking into account the residual current disturbance factors. By mining and analyzing residual current disturbance factors from multiple perspectives, the residual current deviation method is used to quantitatively analyze the impact of residual current disturbance factors on residual current. The frequency domain characteristics of leakage faults considering residual current disturbance factors are analyzed, and multidimensional fault feature vectors and feature datasets are constructed. A leakage fault detection model based on random forest algorithm is established. Through simulation analysis and verification using a simulation model, the results show that the proposed method can detect leakage faults with high accuracy. Compared with commonly used methods, the fault detection accuracy and stability of the proposed method are higher, and the anti-interference ability is stronger.

Under the background of insufficient development of user-side flexible resources, it is necessary to conduct optimization research on the production process of high energy-consuming machining industry. This paper takes the flexible load of the mechanical processing as the starting point, analyzes the corresponding relationship between different equipments and flexible loads by sorting out the electricity consumption characteristics of the mechanical processing, and establishes a calculation model for demand response transactions. On this basis, the photovoltaic-storage synergy model is added to establish a complete demand response oriented mechanical processing production process optimization and solar-storage synergy strategy. This model considers the work constraints of the production equipment corresponding to the flexible load and constraints of the photovoltaic-storage equipment, with the goal of maximizing the total revenue, optimizes the production process of mechanical processing enterprise, and obtains the optimized work flow of the production equipment corresponding to the flexible load, and the maximum benefit that can be obtained. Finally, a copper pipe processing enterprise is used for simulation verification. The calculation example shows that the optimization strategy can effectively improve the economy of the machining industry and increase the consumption rate of new energy.

At present, the researchs on electric vehicle load prediction are mostly divided by functional areas in space, and the influences of complex spatial forms are rarely considered. So a fast charging load forecast method is proposed for electric vehicles considering urban spatial structure. Firstly, according to the urban road and points of interest (POI) data, the kernel density analysis method is used to determine the urban spatial structure, and the actual electric vehicle travel chain law is simulated by combining the central land theory and the improved gravity model. Then, the flow-speed-power consumption model of real-time traffic flow is constructed in the urban road network. Subsequently, considering the influence of subjective factors of decision-makers, the adsorption model is used to give the user′s limited rational decision-making method. Finally, taking the actual spatial data of a city as an example, the Monte Carlo method is used to obtain the temporal and spatial distribution map of fast charging load in the region, which verifies the effectiveness of the proposed method.

Aiming at the charging/discharging scheduling problem for charging station aggregation of electric vehicles, an optimal energy scheduling strategy is proposed for a electric vehicle aggregator (EVA) that takes into account the demands of vehicle owners with the goal of minimizing the long-term power purchase cost of EVA. Firstly, adequate consideration of vehicle owners demands and the time-varying nature of external grid tariffs, an operational framework for EVA energy scheduling management is established. Secondly, the electric vehicles (EVs) are classified into three charging modes according to the difference of users' charging demands, that is, two-way-dispatch EVs, one-way-dispatch EVs and fast-dispatch EVs, and load models are established respectively. Then, based on reinforcement learning theory the real-time energy scheduling strategy is designed for EVA. Finally, the reasonableness and effectiveness of the proposed algorithm are verified by simulation examples of real data and comparing with other greedy algorithms. The results show that the first two scheduling modes based on the proposed strategy can save 54.1% and 47.5% of the cost of EVA in one month, compared with the scheduling mode under the greedy algorithms.