First，the working principle of SIEMENS phase-locked loop （PLL） is analyzed，then the equivalent conversion of FIR filter and inverse tangent segment is conducted，and the analysis of the frequency tracker is carried out，and the discrete domain mathematical model of the phase-locked loop is established. Next，the transfer function of the PLL is analyzed from three aspects of stability，dynamic response and steady-state error，and the restricted range of each parameter of the PLL system is obtained. Then，the new and old phase locked loops are tested under the interference conditions of frequency jump，frequency slope change，phase jump and harmonic intrusion，and the new parameters are found to have better phase locked performance. Finally，phase-locked loops of the old and new parameters are simulated and compared in the CIGRE HVDC benchmark model and Guizhou—Guangdong Ⅱ project model. The results show that the new parameter PLL can suppress the subsequent commutation failure more effectively under the AC faults，and improve the fault recovery performance of HVDC transmission system.

Small resistance grounding is an effective means to improve the protection sensitivity of AC and DC distribution network and reduce the overvoltage level. Arcing grounding overvoltage threatens equipment insulation，so it is of great value to analyze this problem for the safe operation of AC and DC distribution network. In this paper，an AC/DC hybrid distribution network is built to simulate the arc overvoltage scenario of the distribution network on the 10 kV AC side，and the AC/DC power grid overvoltage levels and fault ride-through characteristics are studied. Firstly，based on PSCAD/EMTDC simulation software，a flexible DC distribution network simulation model and arc grounding overvoltage mathematical model are built. Then，the overvoltage fault response curves of the non-fault phase after the AC/DC hybrid distribution network system fails are simulated，and the arc overvoltage ride-through characteristics caused by zero-sequence and negative-sequence components when a single-phase ground fault occurs on the AC side are analyzed. By comparing and analyzing the overvoltage data of AC and DC power grids under different ground resistance conditions，the change rule is obtained. The results show that the effect of suppressing overvoltage is the best when the ground resistance is 6 Ω.

In order to study the problems that the measured value of the intensity of the power field in high humidity area is larger than the actual value，this paper starts from the influence of environmental humidity on the electrical characteristic parameters of wood insulating brackets in measurement system，describes the electrical characteristics of wood insulating brackets using complex permittivity，and builds quasi-static field solution model. Finite element method （FEM） is applied to solve the spatial electric field distribution under the transmission line，and the influencing law of environmental humidity on the measurement results of space electric field is obtained by numerical simulation. The research shows that under the actual working conditions，the environmental humidity affects the electrical conductivity of the insulating brackets through changing the moisture content of wood，which distort the spatial electric field near insulating brackets，which is the main factor leading to the increase of electric field strength. Based on the results of calculation and experiment，a method for measuring the intensity of emergency power frequency electric field using metal brackets in high humidity area and correct the results is proposed.

The sensitivity and effectiveness of the commonly used partial discharge detection technology such as transient earth voltage （TEV） and acoustic emission （AE） have been greatly affected due to the good sealing performance of switchgear. A new partial discharge detection technology is proposed in the paper，an ultra-high frequency current transformer （UHFCT） is installed on the live display device of switchgear，to couple high-frequency transient signals in the electrical circuit. A test platform based on live display device is set up and typical partial discharge defects are set. UHFCT，high frequency current transformer （HFCT），detection impe-dance and ultra-high frequency sensor （UHFS） are used for PD detection. It’s found that UHFCT detection is effective and sensitive，its results are better than HFCT and UHFS. Pilot application is also carried out in actual operation switch cabinet，and typical partial discharge signals are detected by UHFCT.

To improve the effect of transformer fault diagnosis，a transformers fault diagnosis method based on weighted comprehensive loss optimization deep learning and dissolved gas-in-oil analysis （DGA） is proposed. The presented method takes DGA characteristic variables as input and probability distribution of fault states in Softmax layer as output，and builds a transformer fault diagnosis model based on stack sparse auto-encoder （SSAE） deep learning. In order to solve the problem that the transformer fault diagnosis effect is low under the normal cross entropy loss function，and the unbalanced distribution of training samples affects the fault diagnosis effect，the weighted comprehensive loss function is used to optimize the deep learning model. The application results show that compared with the traditional methods，the presented method can reduce the adverse effect of training sample asymmetry on transformer fault diagnosis and improve the level of transformer fault diagnosis. The accuracy of the method in this paper can be maintained above 90% for each training set.

Aiming at the drawbacks of manual judgement reliance and low efficiency in power cable fault locating technologies，a novel algorithm using fault soundwave auto-recognition and initial point auto-marking is proposed. Four general features are defined by analyzing the fault soundwave characteristics，by performing feature extractions on fault and non-fault soundwaves，training set and test set are forged. An AdaBoost-support vector machine （SVM） based algorithm for fault soundwave recognition is proposed，which is used to train the four proposed features in spatial distributing difference of discharge and noise signals. A fault soundwave initial point marking algorithm is proposed by combining discrete wavelet transformation and Gaussian distribution regularities. Experiment verifies that the proposed algorithms increase its efficiency in fault pinpointing while maintaining the accuracy.

The load density of the Guangdong power grid keeps rising，and the structure of the power grid is becoming more and more complex. The original double-loop network structure can no longer meet the development needs，which increases the risk of large-scale power outages. Meanwhile，the AC-DC interaction problems are prominent，the short-circuit current is becoming difficult to control，and the power flow congestion problem of the Pearl River Delta region is outstanding which further limits the ability of accident support of power gird. Based on Guangdong-Hong Kong-Macao Greater Bay Area，Guangdong’s socioeconomic development and resource endowment，the key development target of the Guangdong transmission grid is analyzed. Meanwhile，considering the requirements of steady operation of the power grid and engineering construction conditions in practice，quantitative index and criteria for comparison are formed. Based on the technical route of the West-East electricity transmission projects as well as the developed scheme of the main grid of China Southern Power Grid，this paper contrasts different technical routes that include flexible DC back-to-back asynchronous grid interconnection，constructing a 500 kV outer ring of the Guangdong-Hong Kong-Macao Greater Bay Area based on flexible DC back-to-back grid interconnection，and constructing a 1 000 kV Ultra-high voltage AC outer ring of the Guangdong-Hong Kong-Macao Greater Bay Area. The technical route of constructing a new 500 kV outer ring of the Guangdong-Hong Kong-Macao Greater Bay Area based on flexible DC back-to-back grid interconnection is recommended as the target network planning scheme of the Guangdong Power Grid because it could meet the Guangdong development needs and control the short-circuit current with good economic efficiency and construction feasibility.

Electromagnetic ring network is the product of the transition stage of power grid development. There are some stability problems such as transient stability and thermal stability after some electromagnetic ring network faults in China southern power grid，which make the operation of the main grid difficult control. In extreme cases，the operation risk is extremely high. In this paper，the typical electromagnetic ring network with problems of CSG is studied. According to the analysis of security and stability，the relevant operation of the electromagnetic ring network is formulated. The principle and operation requirements of the electromagnetic ring network are put forward，which can provide the reference for electromagnetic ring network operation.

Aiming at the problem of equilibrium and dynamic development of electricity market，a bi-level model of non-cooperative game is firstly established in this paper，which can directly reflect the changes of market equilibrium interests and bidding behavior. Then，an improved firefly algorithm is used to solve this model and compared with the classical firefly algorithm to verify its superiority. Finally，the bi-level equilibrium model is simulated and analyzed from the perspectives of load change，bidding strategy adjustment，price difference contract and cooperative behavior，and the internal relationship between different factors and dynamic development of market equilibrium is obtained，which will provide theoretical reference for guiding market equilibrium.

The transmission pricing mechanism，as a crucial part in electricity market，is one of the most important missions in Chinese electricity market reform. A reasonable transmission pricing mechanism can provide effective price signals for transmission users and promote the optimization of power resources. This paper conducts a deep research on Norwegian transmission pricing mechanism，including the basic ideas of Norwegian transmission pricing，the components and the impact factors of the transmission grid tariffs. Since the transimission grid tariffs in Norway is divided into variable tariffs and fixed tariffs，this paper goes into details of the calculation method of variable tariff based on marginal loss rate and the influence of adjustiment of marginal loss rate on variable tariff. Meanwhile，this paper aslo introduces the reduction standard and the reduction degree of fixed tariff，with futher research on the difference of fixed cost which load side bears with or without reduction measures. Morever，a 7-node model is used to analyze the effectiveness of Norwegian transmission pricing mechanism in guiding behaviors of generators/consumers and investments of power resources. The conclusion of this paper is promoted in the end that Norwegian transmission pricing mechanism can effectively guide the reasonable layout of power supply/load and it can be used as a reference for Chinese transmission and distribution pricing.

The fixed electricity price mechanism has been adopted in the medium-long trans-provincial electricity market in China，but it is unable to play the regulation role of the electricity market. To solve the shortcomings of the fixed price mechanism，the principle of electricity-dividing according to the trading agreement curve and the execution curve are proposed，and the mechanism of time-sharing price settlement is analyzed. According to the occurrence period of the external power of the curve and the power scale ，three different pricing mechanisms for the external power of the peak and valley curves are proposed. Finally，the calculation cases of the extra-curve price and transaction cost of the mechanism are carried out，which quantitatively explain the influence of the deviation power scale on the settlement price under different pricing modes，and achieve the purpose of guiding the market entity to consciously perform the contract according to the curve of the transaction agreement.

Aiming at building a world-class enterprise vision with global competitiveness of China Southern Power Grid （CSG），the paper constructs a world-class electric power enterprise benchmarking model which combines indicator benchmarking and management capability benchmarking by using various big data quantitative analysis methods，such as content analysis，Delphi method，analytic hierarchy process，coefficient of variation method，performance excellence model and Likert scale. The practice has proved that the model is groundbreaking，objective and quantitative，convenient to popularize，which has been applied well in CSG and has explored new methods for benchmarking work in electric power industry.