The small signal model of the modular multilevel converter (MMC) is an effective method to analyze the resonance problem of the MMC-HVDC system. When MMC is in steady-state operation, the arm current and capacitor voltage contain harmonic components of various frequencies, which have typical multi-frequency response characteristics, and traditional modeling methods can hardly reflect its complex multi-harmonic coupling characteristics. In this paper, the frequency domain small signal impedance model of MMC is established by the harmonic state space (HSS) method. The modeling process takes into account the influence of the main circuit parameters, controller parameters, circulating current suppressor and time delay, and can describe the internal multi-harmonic coupling characteristic of MMC, and the model order is low, it is easy to program and realize. Then, on this basis, using the equivalent impedance of the AC grid and the equivalent impedance of the converter, the wide-frequency oscillation mechanism and system stability of the AC and DC system are analyzed, and measures to improve the stability of the AC and DC system are proposed. Finally, based on the electromagnetic transient simulation model in MATLAB/Simulink, the established small-signal impedance model and the stability analysis of AC and DC system are verified.

In order to improve the reliability of voltage source converter based high voltage direct current transmission (VSC-HVDC), each arm of modular multilevel converter (MMC) is equipped with a certain number of redundant submodules (SMs). Under the existing control strategy, once the redundant SMs are exhausted, the VSC-HVDC system should be shuted down immediately. This paper proposes that the capacitor voltage of SMs can be maintained at the rated value by keeping the sum of the number of inserted submodules in the upper and lower arms at the rated number. Therefore, it is not necessary to immediately stop the MMC when the redundant SMs are exhausted, and the VSC-HVDC system still has the ability to continue operation. The influence of the number of working SMs on the AC output voltage capability is analyzed, and the law of the maximum modulation index changing with the redundancy index is obtained. Furthermore, the limitation to the active and reactive power of MMC varied with redundancy index is analyzed. It is found that the redundancy exhaustion first leads to the decrease of capacitive reactive power limitation. If the redundancy exhaustion is within a certain extent, the MMC can still output the rated active power, which provides space for MMC redundancy optimization and operation capacity improvement, and thereby improve the reliability and availability of VSC-HVDC system.

Modular multilevel converter based high voltage direct current (MMC-HVDC) power transmission is an important transmission mode for building smart grids and a strategic choice to change the development pattern of large power grids. It has been widely used at home and abroad. MMC-HVDC voltage source converter valve is a kind of modular multi-level converter valve, in which each bridge arm uses a power module cascade mode. Many power modules will have partial fault bypass during the long-term operation of the power grid system. The power module fault types are divided into device faults, secondary board faults and optical communication faults. Traditional power module testing devices cannot effectively distinguish some of the fault types, and the accuracy of fault location is low. In order to effectively distinguish the fault types and improve the accuracy of fault location, this paper proposes an intelligent diagnosis device. The device uses the least square method to calculate the power consumption curve of the equalizing resistor, and based on the principle of energy conservation, accurately deduces the power consumption of the secondary board, and then diagnoses the fault types of the power module through the power attenuation of the secondary board, so the fault point diagnosis is accurate, Therefore, it has a great application prospect in the field of power module fault diagnosis.

DC transmission technology is developing rapidly, but its line protection still has some shortcomings. As a backup protection, current differential protection is susceptible to the influence of line distributed capacitance and has a long delay. In order to solve the long delay problem of DC line backup protection, this paper proposes a new principle of pilot protection based on low-frequency fault current F-test. The principle is based on the difference in the change trend of the fault current at both ends of the line at the time of the fault. The S-transform is used to extract the low-frequency fault current, and the F-test is used to perform the trend test to identify the internal and external faults. Finally, the double-terminal DC transmission system model experiment proves the correctness of the protection principle and verifies that the principle has strong resistance to transition resistance, distributed capacitance and noise interference.

Among different types of power supplies for transformer calibration, the common electric power supply does not have precise adjustment, the switching type inverter power supply is easy to produce harmonic interference, and the efficiency of linear power supply is relatively low. To solve these problems, this paper develops a high efficiency linear power amplifier scheme based on piecewise linearization, which can effectively improve the output waveform quality of power supply, improve the overall efficiency of power supply, greatly reduce the heat loss, and realize the fine regulation of output voltage. The proposed circuit topology has the characteristics of multi-level hierarchical structure. The sinusoidal signal is linearized by segments, which greatly reduces the voltage sag of the power transistor in operation. Therefore, the system efficiency is greatly improved, while the advantages of the traditional linear power amplifier, such as high output waveform quality and low electromagnetic interference, are retained. An eight-level power amplifier is built to verify the correctness of the scheme, and it turns out that the highest efficiency reaches 94%, showing good performance.

Aiming at the accident of open flames and smoke caused by turn-to-turn short-circuit fault of dry-type air-core reactor which can’t be identified quickly, by combining theory, simulation and test, this paper studies the scheme of online current monitoring system of dry-type air-core reactor based on Faraday magneto-optic effect. Results of verifiation test indicates three points. Firstly, optical signal is used in data transmission, which can effectively avoid the influence of strong magnetic field generated by air-core reactor on signal transmission. Secondly, the monitoring system could determine the turn-to-turn insulation fault within three cycles (0.06 s) by monitoring the current in spider arm and avoid the further expansion of accidents. Thirdly, the fault position of dry-type air core reactor can be located through data analysis and post-processing. The research results show that this monitoring system could effectively improve the online monitoring level of power transmission and transformation equipment status and provide an effective technical means for turn-to-turn insulation fault monitoring of dry-type air-core reactor.

By analyzing the modular multilevel converter (MMC) double star structure topology of cascaded half-bridge sub-modules, a static var generator topology of improved MMC containing gate turn-off thyristor (GTO) is proposed. Compared with topologies such as cascaded H-bridge static synchronous compensator (STATCOM), it only uses the lower half bridge-arm of the double-star MMC structure, and an H-bridge module is added to the upper layer of the lower half bridge-arm of the MMC structure. However, the H-bridge module requires to withstand higher voltage level in high-voltage scenarios. and has relatively low frequency requirements, so GTO is choosed to form the H-bridge circuit. The corresponding modulation and control strategy are designed for the circuit topology. Then, the improved MMC-STATCOM and the traditional cascaded H-bridge structure STATCOM are compared, the compensation current THD of this solution and the cascaded H-bridge are both less than 2% when the same cascaded modules are used, and the number of switching devices used in improved MMC-STATCOM is less. Finally, by comparing the reactive current command value with the reactive component of compensating current, the dynamic performance of the improved MMC-STATCOM in reactive power compensation and the effectiveness of the control scheme are verified.

With the increasingly prominent problem of fossil energy, wind power as a representative of clean energy has developed rapidly, but the uncertainty of wind power output poses a huge challenge to the safe and stable operation of the power grid. Firstly, this paper analyzes the uncertainty characteristics of wind power, and it is found that the forecast error of wind power has two characteristics of time sequence and interval. On this basis, according to the characteristics of forecast errors, this paper improves the Markov chain, and proposes a day-ahead scenario generation model of wind power based on the improved Markov chain, which can effectively describe the sequence and interval characteristics of forecast errors, build an effective and accurate day-ahead scenario set. The results show that the day-ahead wind power scenario generation method proposed in this paper is more reasonable than the wind power scenario generation by traditional methods, and it can incorporate more uncertain features. Finally, this paper takes the minimum probability distance as the objective to reduce the initial scenario set to obtain the typical scenario set, so as to improve the efficiency of stochastic optimization.

The general short-circuit current calculation method for distribution networks connected with distributed generations (DGs) uses nodal impedance matrix and iteration procedure, which needs more calculation steps and takes longer calculation time during high penetration level when penetration level of distributed generators is high. A new short-circuit calculation method is proposed based on the fault components superposition and equivalent compensation theorem. The new method uses the improved backward/forward sweep (BFS) method to estimate the normal and fault voltage components. The multiport equivalent compensation method is utilized in the looped systems. The proposed method can calculate short-circuit currents in radial networks as well as weakly meshed networks with DGs. Different cases are studied using the proposed and general methods, and results are examined by the simulations in MATLAB/Simulink. The comparison verifies the validity of the proposed method. The new method can obtain results without using the node impedance matrix and the calculation amount and time are less than the conventional methods, which can be used in radial and weakly meshed distribution networks with DG connection.

The massive access of renewable energy, electric vehicles makes it more difficult to estimate the real-time state of distribution network. The existing distribution network measurement system is difficult to meet the situational awareness requirements of distribution systems that contain a lot of uncertainty of distributed power and electric vehicle. For this reason, an optimal configuration method of phasor measurement unit in distribution systems for interval state estimation is proposed. First, a mathematical model of the minimum interval state estimation average interval width target is constructed based on the minimum path between each node and other nodes. Then, this paper takes the minimum number of PMU configurations, maximum redundancy and minimum interval state estimation average interval width as the goal and the configuration quantity as the constraint to establish a multi-objective optimization configuration model, and adopts the non-dominant ranking genetic algorithm based on non-dominated sorting genetic algorithm (NSGA-Ⅱ) to obtain the optimal solution set of the model. Finally, through NSGA-Ⅱ, the optimal solution obtained multiple times is solved, and a suitable configuration scheme is selected. A variety of scenario simulation verifications are performed on IEEE 30 power distribution system. The results show that the optimal configuration scheme obtained by the model can effectively reduce the interval width of the interval state estimation results, which proves the practicability, correctness and superiority of the configuration method.

With the popularization and application of synchronous phasor measurement unit (PMU) in the distribution network, the challenges in communication and timing are gradually emerging. In the aspect of time service, the cost of Beidou / GPS is high, and it is difficult to install in some occasions. In the aspect of communication, the stability of 4G is relatively poor and the investment cost of optical fiber is high. In this paper, PMU based on 5G is tested in laboratory environment, field test environment and demonstration project. In terms of time synchronization, pulse per second of 5G and standard clock source are compared and tested. The test results show that the errors are all within 300 ns, which meets the requirements of PMU device for time accuracy of one microsecond. In the aspect of communication delay, the test results show that the maximum bidirectional delay from PMU to main station is less than 14 ms, which can meet the higher requirements of PMU for communication delay. In terms of bandwidth, 5G can still meet the requirements of PMU for reliability and bandwidth under the limit condition that the program does not make any restrictions. 5G’s superior communication performances includes short delay, high reliability, high precision timing, high bandwidth, etc.are verified through the tests, which will provide communication and timing basis for the application of PMU in new power grid scenarios.

With the further development of digital grid construction, the scale of electrical energy data continues to expand, and the demand of data application is also increasing. Considering the limitations of the original application architecture of the electrical energy data platform of China Southern Power Grid, this paper designs a full stack and distributed architecture for large-scale electrical energy data, and applies it to the electrical energy data platform of China Southern Power Grid. The architecture integrates the distributed technology of computing, storage, message queue and application, which can achieve real-time data collection, unified management and unified application of electrical energy data. It supports the development of electricity consumption analysis, line loss management, power quality monitoring and other businesses of China Southern Power Grid, and improves the ability of analysis, application and management of electrical energy data.

Aiming at the problem of corona occurrence of power transmission and transformation fittings in high humidity areas, the influence of humidity on the corona characteristics of the rod-plate electrode is studied by simulating the electric field of the rod-plate electrode and the corona of the rod-plate electrode under different humidity law. The results show that when there are particles in the gap between the rod and the plate, the electric field distribution will be significantly distorted. When the humidity is high, depending on whether the connection direction of the particles is the same as the ray direction of the tip of the rod electrode, the presence of particles has a different effect on the electric field. When in the same direction, the presence of particles will promote the development of the flow column; when it is perpendicular, the presence of particles will hinder the development of the flow column; In the range of relative humidity of 30%~60%, the corona initiation voltage decreases with the increase of relative humidity. In the range of relative humidity of 60% to 85%, the voltage increases with the relative humidity, but the amplitde is smaller. For different electrodes, the relationship between corona initiation voltage and relative humidity shows the same trend.

During the spring festival, the distribution transformer load has obvious fluctuation characteristics due to the population flow. In the short term, the distribution transformer will be overloaded due to the load growth. The power supply bureau usually uses the method of emergency replacement of distribution transformer before the Spring Festival. Combined with the load forecasting model, this paper establishes a set of distribution transformer switching strategy algorithm to adapt to the Spring Festival period. Through the data analysis of material inventory distribution and distribution transformer in operation, the optimal switching strategy is obtained, which makes the distribution transformer safe operation and distribution transformer utilization rate achieve optimization. It can solve the problems of manual calculation data deviation, lack of reliable algorithm to formulate the exchange scheme, the implementation effect is not obvious, and can not realize the optimal allocation of resources. Through the actual example, the algorithm can optimize the goal of switching distribution transformer during the Spring Festival, improve the effective utilization of resources, and solve the risk of overload operation of distribution transformer.

In order to solve the problem of unified settlement of the electricity retail company, this paper conducts an in-depth analysis of the underlying logic of electricity retail unified settlement, sorts out the key factors of unified settlement in the electricity retail market, designs a mathematical model of unified settlement, and proposes the application of the settlement model in the power market technical support system. Finally, the feasibility and superiority of the model are verified by an example. The outcomes solve the unified settlement connection between the electricity wholesale market and the electricity retail market of the electricity retail company, and fully embodies the freedom, flexibility, and complexity of retail settlement in a fully competitive electricity market. The unified settlement model supports the efficient and accurate execution of electricity retail settlement work, and guarantees the realization of the goal of the electricity retail-side reform.

Considering multiple end-users taking part in the electric power market transaction of the smart grid, an energy trading market model of multiple end-users is constructed by introducing the double auction mechanism in this paper firstly. Then the model is transformed into a non-cooperative game model with incomplete information. To make energy trading among multiple users tend to be stable, an adaptive learning algorithm is proposed for the end-users playing the game to find the optimal mixed-strategy based on the Q learning, and the overall game comes up to Nash equilibrium, so that energy trading between the multiple end-users can run stably. Finally, the probability distribution of mixed-strategy for multiple end-users to obtain Nash equilibrium under numerical simulation, the validity of the proposed energy trading algorithm is verified.