Along with the commissioning of UHV AC/DC transmission projects in China, the "strong DC links while weak AC grid" problem is outstanding, the interaction influences among AC/DC lines, multiple DC ones and sending/receiving ends are increasingly important, and the operation complexity of power grid is growing rapidly, so higher requirements on the scale, accuracy and speed of power grid simulation are put forward. From three aspects as power system modeling, multi-time-scale simulation and digital-analog hybrid real-time simulation, the current situation of simulation technology was explored, and the challenges from the development of AC/DC hybrid large-scale grid were analyzed. To increase the capability for large-scale power grid simulation, researches will focused on improving modeling accuracy of complex models such as DC system, renewable energy, FACTS （flexible alternative current transmission systems） and load, enlarging scale of electromagnetic simulation, and increasing efficiency of large-scale power grid simulation.
With the development of power system, traditional power system gradually evolved into a large-scale complex hybrid AC/DC power system. To design an efficient and fast large-scale power system electromagnetic transient （EMT） parallel simulator, become a research hotspot. Firstly, this paper discussed the limitations of traditional coarse-grained parallel approach in accelerating large-scale power system EMT simulation；and then introduced an averaging strategy in converter modeling to improve the traditional EMT algorithm. In addition, the paper proposed a vectorization strategy to design a fine-grained parallel computing method which could realize the acceleration of EMT simulation of large scale power system on new type of computing device GPU with high acceleration rate. Finally, this paper emphasized the development direction of large-scale complex power system simulation, and proposed a real-time heterogeneous simulation platform architecture for large-scale power system.
With the wide application of power electronic devices in power system, such as high voltage direct current （HVDC）, flexible AC transmission systems （FACTS） and so on, the electromagnetic transients program （EMTP） gradually becomes an important tool for power system analysis. However, the simulation scale of EMTP is limited due to the small time step, occupying large memory and slow computing speed. Based on the rotating transformation of single phase coordinate, this paper proposed a fast simulation method with large time step, which could reduce the signal frequency through the rotation transform. Compared with the electromechanical transient simulation, this method can reflect the electromagnetic transient behavior of the system and improve the simulation accuracy. It also can improve the simulation speed compared with the traditional electromagnetic transient simulation.
With the formation of large-scale nationwide AC-DC interconnected power system in China, the dynamic characteristics of power system are increasingly complicated. Unified Dynamic Simulation is an important tool to analyze the dynamic characteristic, serious faults scenario and stability measures for this non-linear super large-scale power system, in which the electromagnetic transient, electro-mechanical transient, medium-term and long-term dynamic phenomena of power system are united for calculation. The present study and application status, current technical problems and prospects of Unified Dynamic Simulation technology were described, which mainly included the numerical integration method and modeling technologies involved in electromagnetic transient, electro-mechanical transient, medium and long-term simulation. It is pointed out that some key technologies, such as new numerical integration methods, DC transmission modeling method for multi-time scale simulation, parallel simulation software development with opening, interactive and intelligent methods, etc., are powerful simulation tools for the safe, stable and economical operation of large AC-DC power systems and the application researches of new technologies and new devices.
The electromagnetic-electromechanical transient hybrid real-time simulation platform called super-mixed real-time simulation （SMRT） has wide application prospect in the analysis of large-scale AC/DC power grid because of its advantages in simulation scale, simulation accuracy and easy use. At present, SMRT has been applied in the simulation analysis and research on the electromechanical transient behavior and stability characteristics of the large-scale AC/DC power grid, the rapid repetition of failure/accident in actual power grid based on typical operation modes, the production operation service of AC/DC system such as major operation risk assessment, and the system performance and control strategy, the role and influence of power electronic device of dynamic compensation for grid on the commutation failure recovery of grid after multi-DC fault, as well as the teaching and training of the dispatcher. After further development, SMRT can be applied in the system-level relay protection setting and verification and other problems.
There are three common used methods of dynamic equivalence, in which the coherent equivalent algorithm is the most practical in engineering. With the development of large-scale UHV AC/DC synchronous grid, the original outside equivalent schemes have been unable to meet the practical demand of engineering. From the current practical application, only electromagnetic transient simulation has equivalent simplification demand for power grid. With the improvement of real-time simulation technology and ability, the requirements of the simulation scale are also relaxed. According to the development of power grid and real-time simulation, the new inside equivalent principle was proposed. On the basis of it, this paper studies the equivalent level of the East China power grid in 2015. The equivalent research result shows that the East China power grid after using this equivalent principle can significantly reduce the grid scale and well maintain the load flow and transient performance of original grid, which can be applied in the electromagnetic transient real-time simulation study.
With the fast developments of UHV power grid, large-scale AC/DC interconnection causes the profound change of grid characteristic, which puts forward higher requirements for power system simulation. Power system digital-analog hybrid simulation is able to accurately simulate the operation characteristic and dynamic process of AC/DC system, which is the important method to research large-scale AC/DC power grid. Due to the characteristics of high simulation accuracy and high speed, the digital-analog hybrid simulation is a basic platform to acknowledge the operation mechanism of large power grid. But the modeling of hybrid simulation response is relatively complicated, and its scale has certain limitations. The digital-analog hybrid simulation of power system was deeply described in many aspects such as the simulation technology, development and application, etc. And its function and application range were introduced in detail. Finally, combined with the future development needs of UHV large power grid in China, the development direction of the digital-analog hybrid simulation of power system was proposed.
The electromechanical transient simulation model of double-fed wind power has complex structure and more control parameters. It is important prerequisite and basis for the actual application to obtain units model parameters based on the measured data. According to the basic structure characteristic of double-fed wind power model and the influence degree of different model parameters on output characteristics, this paper proposed classification determination method based on the influence degree of parameters. Firstly, based on the basic theory and model structure of double-fed wind power, this paper analyzed the influence of different model parameters on the active power and reactive power output characteristics of wind turbines. The results show that the control mode in normal status and the control mode, parameters in low voltage ride through status have great influences. Then, based on the actual data, this paper put forward classification processing method for different types of parameters including acquired from manufacturer, using typical parameters and parameter fitting from measured data. Finally, a double-fed wind power unit was used as the example to introduce the method process, as well as the result and effect of parameter fitting. The practice proves that the proposed parameter identification method can reflect the key characteristics of the actual wind power and has good practical application value.
As a new generation of HVDC technology, the theoretical study of VSC-HVDC control strategy has made a lot of achievements, but the research on its engineering implementation and application are relatively less. Therefore, an implementation scheme of VSC-HVDC engineering control strategy was proposed based on the analysis of its system model. The scheme was achieved on common control platform HCM3000/F. With existing valve system and primary system, a complete VSC-HVDC dynamic modeling system was formed to verify the functions and performance of control system. The dynamic simulation results show that the control system can complete a variety of control functions of VSC-HVDC, with excellent regulation performance in transient state and meeting the engineering requirements.
Static synchronous compensator （STATCOM） is the important component of modern power system due to its capability of flexible reactive power support. However, small time-step simulation is required for STATCOM, because it consists of a great amount of power electronics switches. The small time-step simulation causes a substantial reduction in the simulation speed, which is the great challenge of real-time simulation. In order to meet the simulation requirements of synchronous compensator, improve the speed of small time-step simulation and realize the real-time simulation functions, this paper studied the small time-step real-time simulation system of STATCOM. The small time-step simulator is the special power electronic equipment simulator based on FPGA （field programmable gate array）, which consists of core computation block, element block, interfaces of small and large time-step networks, commutation block and other core modules, to ensure the real-time simulation. The small time-step real-time simulation method of STATCOM was studied with using two-level convertor and double loop current control model. Then, the implementation of the model in the small time-step simulator was studied, and the small-step switch model was proposed for power electronic device simulation. Finally, the experimental results of small time-step real-time simulation system of STATCOM were presented.
Due to the characteristics of a large number of power connected into branches in power collection grid, the fault at the beginning of branches in wind-photovoltaic-battery power collection grid with radial and chain topologies will cause enormous power losses. To improve the reliability of wind-photovoltaic-battery generation system, the distribution network in meshed grid mode was proposed to construct wind-photovoltaic-battery power collection grid. On this basis, this paper establishes the optimization model of power collection grid scale with considering electricity revenue and cable cost constraint, as well as the optimization algorithm of power collection grid scale in meshed grid mode based on genetic algorithm theory. Then, this paper studies the power collection scale, electricity benefit and other problems of wind-photovoltaic-battery power collection grid in meshed grid mode. Finally, the research and example analysis prove the collection advantage of power collection grid in meshed grid mode and the effectiveness of the proposed optimization model and algorithm.
Economy and reliability are two indexes that must be considered in the optimal allocation process of energy storage system. However, they cannot achieve the optimum simultaneously. In order to coordinate the contradiction, a multi-objective optimal allocation model of battery energy storage system （BESS） with considering the economy and reliability was built. Firstly, according to the current situation that the reliability evaluation method of system with energy storage element was complex, and its computing speed was slow, a simple reliability evaluation method of distribution network with distribution generation （DG） and BESS was proposed, in which the contribution of energy storage configuration to the reliability improvement of distribution network was quantified, and two reliability indexes, continuous power supply time increment and power supply increment were obtained. Secondly, with the goal of maximizing the annual net income of distribution network, the continuous power supply time increment and power supply increment of energy storage system during distribution system fault, the multi-objective optimal allocation model of BESS was established. The improved strength Pareto evolutionary algorithm （ISPEA） was applied to solve the Pareto front of this multi-objective model. Simulation calculation was carried on IEEE-33 nodes distribution system, whose results verified the feasibility and effectiveness of the proposed model.
Vulnerability assessment of transmission line is an important part of the vulnerability researches of power system. The existing methods have problems such as single point of evaluation and lack of practicality, meanwhile cannot consider the impact of weather factors. Therefore, a new assessment method of transmission line vulnerability was proposed. The weighted betweenness indicators of lines were calculated based on the weighted betweenness model. Based the well-being risk tracking model, the well-being risk contribution degree indicators were proposed with considering the component sensitivity. At the same time, combining two-state weather model, the influence of bad weather on the line vulnerability was studied. The integrative vulnerability indicator considering system network structure and running state characteristics was put forward in combination with well-being risk contribution degree and line betweennes, and the process of vulnerability assessment for transmission line was presented. Finally, the rationality and practicability of the proposed method were demonstrated by the evaluation of the vulnerable lines of IEEE-RTS system. The proposed method can identify the vulnerable line in the critical transmission line channel of power grid, which can provide a reference for the operation and maintenance of power system.
Aiming at the problem that the existing time-domain fault location algorithms used precise line model and fixed data window, this paper proposed a time-domain fault location algorithm based on picking out outliers. According to the characteristic that the voltage distributions obtained at both ends had minimum difference at the fault point, the proposed algorithm made full use of the post-fault transient data, and obtained the fault positioning results through fault location function when translating the redundant data window. Taking the positioning results as dataset, the distanced-based method in statistics was adopted to pick out the outliers of the dataset, thus the final fault location result could be obtained. The Yunnan-Guangdong ±800 kV bipolar DC transmission system model was established in PSCAD/EMTDC to verify the proposed algorithm. The results show that the proposed algorithm is simple in principle, can realize accurate fault location within the scope of whole length of line, make full use of post-fault transient data, and effectively improve the accuracy and reliability of fault location.
The determination of key transmission line in power system can effectively improve the stability of the system. Therefore, this paper proposes the integrated importance indicators of lines. This method only uses the basic parameters of the system, and can overcome the disadvantage that the calculation of betweeness indicator needs the current operating mode of the system and the distribution of the power flow. The indicators can identify the key lines of the system from two aspects of the system structure and probability. According to power distribution theory, the power distribution betweeness of transmission lines was proposed as the structure importance index of the system, which took into account the line weight of directly connection node between generator and load, and could reflect the utilization degree of transmission lines in each generation and load of power transmission. And then combining with the risk index of line failure rate, the integrated importance index of the branch was put forward. The simulation analysis was carried out according to the standard examples of IEEE 39-bus, and the data obtained by the method were compared with those obtained by other criteria to verify the effectiveness of the proposed method.
One of the most effective methods to shield a power grid from the geomagnetic storms is the reasonable planning and scientific arrangement of the number and direction of the feeder lines connected to substations, where the precise GIC （geomagnetic induced current） calculation serves as the foundation. According to the measured geoelectric field data from Ziwu Project and the structure of Guangdong 500 kV power grid, this paper established the full node model to calculate the GIC of Guangdong power grid, and completed the theoretical GIC calculation in this grid during the geomagnetic storm on September 12th, 2014. By comparing the calculated result with measured data, it shows that the measured geoelectric field data serves better than geomagnetic data for GIC calculation, suggesting enhanced geoelectric field surveillance so as to provide service and data for geomagnetic storm disaster prevention in the power grid designing phase.
As oil chromatography online-monitoring data is unlabeled during power transformer failure, project sites tend to get a large number of unlabeled fault samples. However, traditional diagnosis methods often fail to make full use of those unlabeled fault samples in judging transformer fault types. Based on deep learning neural network （DLNN）, a corresponding classification model was established, whose classification performance was analyzed and tested by typical datasets. On this basis, a new fault diagnosis method of power transformer was further proposed, in which a large number of unlabeled data from oil chromatogram on-line monitoring devices and a small number of labeled data from dissolved gas-in-oil analysis （DGA） were fully used in training process. It could generate fault diagnosis result in the form of probabilities, and provide more accurate information for the maintenance of power transformer because of its better performance in fault diagnosis. Testing results from engineering example indicate that the proposed method is correct and feasible, and its diagnosis performance is better than that of three radio, BP neural network and support vector machine, which is suitable for the fault diagnosis of power transformer.
Reliability is an index, of great importance, to evaluate the level of power grid. Traditionally, the idea to improve the reliability of power grid is to use technical method and improve the specifications of power grid construction, which ignore the role that the government plays in this issue. In order to study the relationship between grid reliability and government actions such as the implementation of demand side management （DSM） and the incentive on transmission and distribution price, firstly, a simplified reliability evaluation index was presented based on the capacity-load ratio, an evaluation model of grid reliability was built based on system dynamics, and the causal loop diagram and the system stack flow chart were presented as well. Then, the simulation was carried out on the model by VENSIM under different scenarios, which revealed the relationship of the grid reliability index, demand-side resource permeability and the incentive on transmission and distribution price. The results show that the increase of demand-side resources can bring economic benefits without influencing the power grid reliability；there is an obvious positive correlation between the incentive strength of government on transmission and distribution price and the grid reliability index of power grid；and the government can keep the reliability index at a high level by slightly incenting the transmission and distribution price. Finally, some suggestions on the reliability improvement of power grid were proposed for the government, based on the example results.
According to the long-term development planning of Chinas energy, the proportion of non-fossil energy will reach to 15%. In order to develop effective investment incentive policies for renewable energy, this paper applied IOSLAB （Individual, Organization and Society Laboratory） platform to carry out choice experiment （CE） study, analyzed personal willingness to pay for renewable energy investment, and comprehensively evaluated the renewable energy investment from aspects of economy, environment and society. IOSLAB system was used to survey the local residents through field experiments. Moreover, a multinomial Probity model （MNP） was applied to solve the willingness to pay for renewable energy investment. MNP assumed that all respondents had the same preferences for the attributes being valued. Through the analysis on the research results, it is found that the public pay related fees to cover part of the extra cost brought by the development of renewable energy, because renewable energy can reduce the air pollution and increase social employment. The research results can help decision makers to develop the investment policies of renewable energy.
