Abstract:Aiming at the output of photovoltaic arrays appears multipeak caused by the complexity of the environment in practical applications, the traditional control methods have difficulty in tracking the maximum power point of multipeak targets. This paper combines the optimized multipeak support vector machine function with the cat group algorithm with fast convergence characteristics, and proposes an improved cat group algorithm. The algorithm is used to solve the multipeak optimization problem in the case of partial and hence realize the maximum power tracking function of the solar power system. By constructing a mathematical model of photovoltaic cells, the multipeak output characteristics of photovoltaic (PV) cells under local shading conditions are analyzed. By analyzing the optimized structure of adaptive weights and mutation operators in IMCGA intelligent algorithm, this paper elaborates the implementation process of the IMCGA algorithm in maximum process point tracking of the shading photovoltaic system. Finally, the superiority of the proposed control algorithm is verified by simulation experiments.

Abstract:In order to realize the optimal control of modular multilevel converter (MMC), a model predictive control strategy based on voltage level for MMC is designed. Based on the sum and difference of arm voltages, the mathematical model of MMC system is derived. Then a model predictive controller with hierarchical structure is designed by the mathematical model. The cost function design in the controller takes into account the AC side current control, the circulation current suppression and the arm energy balance. The cost function also selects the optimal voltage level to provide a voltage reference for the pulse width modulator. In addition, submodule capacitor voltage balancing is accomplished with a separate control loop. Finally, the hardwareintheloop simulation test system for MMC is built. The steadystate and dynamic test results verified the effectness of the new control strategy.

Abstract:The active distribution network has a large of distributed generator (DG), therefore, the operation mode and control strategy of the traditional distribution network are not suitable for the active distribution system. The selfhealing control of active distribution network can prevent, diagnose, analyze of the fault and recover power supply. Under this background, a intelligent control terminal is designed for active distribution networks in this paper. Then, the fault isolation, fault phase selection, fault location are designed. Finally, the simulation model is constructed in Matlab/Simulink, and it is shown that the terminal system has a good ability to deal with the fault of active distribution network.

Abstract:The AC distribution network has threephase unbalanced characteristics and can be connected with DC networks to form a AC/DC hybrid grid. This paper proposes a new type of πmodel for the AC/DC converter and establishes a set of power balance equations to describe threephase AC and DC sections. The AC and DC network is described by admittance matrix. The node admittance matrix of the whole network is then set up since the AC/DC converter can be included in the matrix by a π equivalent model. Combining the converter model and the network equations, a power flow calculating method including threephase AC line and DC line is proposed and verified with cases.

Abstract:The gridconnected DC microgrid plays an important role in improving the power supply reliability and power quality in the end of power grid. The traditional optimal method pays more attention to economic of enterprise and little considerations to the utility of the customers. Under the market environment，the development of gridconnected DC microgrid must concern to the interests of users. This paper focuses on a typical gridconnected DC microgrid with windPVdieselstorage and proposes a bilevel optimal model of microgrid configuration, which takes into account economic efficiency of enterprises and customer electricity experience. Considering the initial installation cost，operation and maintenance cost，fuel cost，replacement cost and energy exchange cost of the microgrid，an upper objective model minimizes the comprehensive economic cost, and a lower objective model minimizes the customer power shortage expectations. By constructing the power supply reliability correlation model, a KKT condition is utilized to realize decoupling of bilevel optimal model. The GAACO is then employed to solve the model，and the results of the case study demonstrate the validity and rationality of the model and algorithm.

Abstract:Large scale integrations of EVs brings challenges on the security operation of the distribution networks. The Soft Normally Open Point (SNOP) is commonly utilized to replace tieswitch. It not only improves the power flow of both two lines, but also provides the accurate controlling for power flows of the feeders connecting to the SNOP and increase the control flexibility of whole systems. In this paper, sequential probability output models of EVs and loads are established based on the considerations of uncertain characteristics. Furthermore, a multiobjective model of the optimal allocation for SNOPs is proposed. Meanwhile, a corresponding solution is also presented by considering the system losses and uncertain outputs of EVs. Simulation results of an IEEE 33 nodes system are presented to verify the feasibility and validity of the proposed model and algorithm.

Abstract:According to the characteristics of distribution network and distributed generation，this paper establishes a multiobjective optimal model by minimizing the distribution network loss，distributed power costs and maximizing voltage stability. A multiobjective particle swarm optimization algorithm based on improved adaptive weight strategy for optimizing DG configuration is proposed. Compared with traditional multiobjective algorithms that are readily trapped at the local minimum, it guarantees to provide an solution closer to the global minimum. Meanwhile the algorithm offers a series of alternative solutions in better fitness to the Pareto optimal boundary in comparison with singleobjective algorithms. Finally, a numerical example of an IEEE 69bus system is employed to verify the reliability, adaptability and practicability of the proposed algorithm.

Abstract:The noncommutation operation mode is commonly utilized in distribution network, but there exist threephase unbalanced loads resulting in line parameter imbalance. This situation brings some difficulties into the fault location. In this paper, a new fault location algorithm is proposed for radial distribution networks. Just singleended fault information is considered to realize the accurate fault location. Firstly, a network matrix is employed to study the distribution network structure and the calculation complexity is reduced. Then, the Fibonacci (Fibonacci) sequence search method is adopted to speed up the fault point searching process. Finally, PSCAD/ EMTDC is utilized to model a typical distribution network under different fault types. It is shown that the presented algorithm is able to identify fault locations with high accuracy.

Abstract:With the continuous upgrading of rural power grids and rising living standards of rural residents, the power grid structure, operating level, load types and load operating characteristics of rural power grids have undergone significant changes. The traditional rural power grid comprehensive evaluation index system and evaluation methods have been unable to meet the current needs of rural power grids. This paper presents a comprehensive evaluation system for rural power network that takes into account the characteristics of distributed power access and load operation. Firstly, a comprehensive evaluation index system covering the structure and operation characteristics of sourcegridload reaction rural power grids is constructed. Then, the existing evaluation methods of single rural power grids are comprehensively reviewed, and a combinatorial weighting method based on game theory is proposed. In order to reflect the adaptability of the evaluation system, we choose the horizontal contrast between different regions in our country to verify the validity of the algorithm. Under the national strategic background of rural rejuvenation, the comprehensive evaluation system of rural power system proposed in this paper has strong practicality and can provide a reference for rural power grid investment.

Abstract:Aiming at the problem of serious abandonment of wind and high carbon dioxide emission in the current power system，a lowcarbon economic dispatching model for a combined electricitygas integrated energy system with power to gas(P2G) is proposed. The paper introduces the economic conversion coefficient to convert carbon dioxide emission into economic dimension and counts them in the objective function. The objective function is to minimize the sum of carbon dioxide emission cost and system operating cost while the energy flow model and safety constraints of electric power system containing powertogas and natural gas system are considered. The following optimization is realized by using the interior point method(IPOPT). The modified IEEE 30 node power system and Belgium 20 node natural gas network are used for example analysis. By comparing the comprehensive cost and carbon dioxide emissions under four different models, it is verified that the proposed model can effectively balance the low carbon and economy of the system operation, and greatly improves the ability to absorb wind power. When different economic conversion coefficients are taken, the optimal scheduling results with different emphasis on low carbon and economy are obtained.

Abstract:As the increasing awareness of environmental protection for people and the rising penetration ratio of wind power， power sectors have put forward higher requirements for the dispatch of power system integrated with wind power. In order to reduce the carbon emissions and promote wind power consumption while minimizing power generation costs, this paper firstly establishes an environmental economic dispatch model of power system integrated with wind power considering environmental and economic indicators. The objective function in the model considers the cost of thermal power generation, the cost of rotating spares, the trading mechanism of carbon emission rights, the cost of wind power and the cost of discarding wind. Aiming at the oscillation problem that is common to occur in the traditional firefly algorithm optimization process, a variable step glowworm swarm optimization is proposed to solve the model. Then, this paper carries out the simulation of the modified IEEE 39bus to verify the feasibility of the proposed model and the effectiveness of the algorithm. The results verify the feasibility of the proposed model and the effectiveness of the variable step glowworm swarm optimization to solve environmental economic dispatch problems.

Abstract:In this paper, the reliability of distribution network with distributed generation (DG) under different customer sectors is evaluated. First of all, the daily load characteristics of residents, business and industry are studied. A time series load model is established to reflect customer sectors characteristics. The model of the coordinated operation of the energy storage is established through the output of DG and load demand. Finally, based on the sequential Montecarlo simulation algorithm, the reliability of the IEEE RBTS BUS6 feeder one system is calculated. The results show that the access of DG can effectively improve the reliability of distribution network. The same output of DG has different effects on the reliability of different customer sectors.

Abstract:During the operation of small current grounding system, unbalanced voltages would be induced in neutral points as the unbalanced parameters of transmission lines. And this unbalanced voltage produces an unbalanced current in feeder system. Therefore, studying the distribution characteristics of the unbalanced current is helpful for analyzing system faults. This article theoretically analyzed the unbalanced current in feeder systems. Starting from the feeder, the equivalent circuit of the system is obtained based on Thevenin's theorem while the expressions for unbalanced currents in feeders are also obtained. Then, under the condition of the singlephase disconnection fault, the influence of unbalanced currents on the amplitude and phase of zero sequence current on feeder lines is revealed. Finally, the simulation model is built in MATLAB to verify the analytical results.

Abstract:In order to improve calculation speed for the electromagnetic transient of VSCHVDC system without lossing accuracy，the Parareal method is applied to the electromagnetic transient simulation of VSCHVDC. At the beginning，the VSCHVDC time domain dynamic model is established. Then the model is calculated by employing the Parareal method. This method can decomposed the whole simulation time into a series of subintervals and a coarse approximation of the trajectory is utilized to supply the initial conditions at each subinterval. Secondly, using this set of initial values solves the evolution problems at each subinterval concurrently and separately. Then a fine solution is obtained by using an estimatecorrection method. The experimental results show that the algorithm can obtain a more effective speed ratio with a high parallel efficiency. It can improve the calculation speed of VSCHVDC system in electromagnetic transient numerical simulations.

Abstract:With the increasing of the wind power penetration, the network loss and voltage fluctuation of the wind farm is becoming more and more serious. In this paper, the cable selection based on the location of Doubly Fed Induction Generator (DFIG) is studied, and the DFIG's reactive power limit is analyzed. DFIG is commonly utilized to supply reactive power for power grid, thus, the reactive power of DFIG, the number of fixed reactive compensation device, the reactive power output of Static Var Compensator (SVC) and the onload tap changing transformer tap are taken as the control variables. Then, the minimum objective function is established as the sum of total voltage deviation and active power loss of wind farm. The discrete variable is analyzed by employing the Linear decreasing Weight Particle Swarm Optimization (LinWPSO), and the optimal solution of reactive power optimization is successfully obtained. Finally, a case study is included for verification. It is shown that the decentralized and coordinated control of DFIG which considering cable selection can effectively reduce network losses and voltage fluctuations.

Abstract:Aiming at the photovoltaic generation and energy storage devices connected to the distribution network, considering the problem of voltage optimal control problem based on voltage and reactive power coordination, this paper studies the optimal voltage control strategy under “SourceGridLoad” interaction. For photovoltaic power generation and energy storage distribution network, LQRbased optimal voltage control model is built. Then ADPbased online optimization method is proposed in order to deal with the variety of the steadystate operation points and the unknown or inaccurate model and parameters. Simulations are accomplished to verify the effectiveness of the proposed method under different operation conditions of the network. The results show that the proposed control law is effective in voltage improvement. Therefore the proposed method is beneficial to power system online control problems.

Abstract:In order to solve the unsatisfactory section location problem of single phase grounding fault in distribution network， a novel method of fault section location based on phase current increment is presented according to the similarities between features of threephase phase current increment waveform in healthy lines and the fault line after the fault point, as well as great difference between healthy lines and the fault line before the fault point. In the method, the power coefficients of each detection point can be calculated by threephase current increment and sent back to host. The fault section is predicted by comparison with the power coefficients of its adjacent detection points. The advantage of this method is that it reduces the pressure of the communication system while only using current as the fault section location characteristic. Furthermore, simulation and data analysis indicate that this method can correctly predict fault section in high, medium and low resistance grounding under every kind of initial angles. It has good applicability under simple theory.

Abstract:The key problem of transformer differential protection method is the method to distinguish the inrush current from short circuit current in order to prevent misoperation. The traditional method based on two harmonics amplitude ratios has a single criterion and insufficient accuracy, and thus it may cause a misoperation in some conditions. On this background, the method of phase difference discrimination is improved, and the braking ratio is utilized to set up a comprehensive braking criterion. Simulation modules are modeled in Matlab/Simulink and the transformer is simulated and checked under different operating conditions. It is shown that the criterion not only distinguish short circuit faults in different situations, but also recognize inrush currents effectively, and make corresponding protection actions for different fault types of transformers.

Abstract:Since the traditional capacitive voltage transformer does not have good response characteristics in the high frequency band, and can not meet the requirements of voltage traveling wave measurement, the development of the voltage traveling wave fault location method is to some degree limited. Through coretype traveling wave sensor is designed according to the Rogowski coil principle in this paper. It can measure the current travelingwave from CVT to ground and transfer voltage travelingwave signals of transmission line effectively without high speed data acquisition. The designed sensor can be implemented in the fault location of voltage traveling wave. A variety of frequency response characteristics including frequency response, highfrequency transient traveling wave signal response, power frequency, withstand voltage and antiinterference of the designed sensor are tested in this paper. The testing results show that the sensor with excellent amplitude characteristics can efficiently transfer the traveling wave signal of frequency range from 10 kHz to 2 MHz. Therefore it can also further avoid maloperation of the protection by shielding the power frequency signal and preventing grounding leakage current transferring to the secondary of transformer.

Abstract:Traditional power forecasting models can not efficiently take various factors into account, neither to identify the relation factors. In this paper, mutual information in information theory and the artificial intelligence random forests algorithm are introduced into the medium and shortterm electricity demand prediction. Mutual information can identify the high relation factors based on the value of average mutual information between a variety of variables and electricity demand. Different industries may be highly associated with different variables. The random forests algorithm is used to build the different industries forecasting models according to the different correlation factors. The data of electricity consumption in Jiangsu Province is taken as a practical example. In the example, the above methods are compared with the methods without mutual information and the industries. The simulation results show that the above method is scientific, effective, and can provide higher prediction accuracy.

Abstract:Some key nodes in the power system may play a role in the expansion of faults when a largescale interlock failure occurs in the system. In order to improve the speed and accuracy of key node identification, this paper proposes a key node identification method based on PSNodeRank algorithm by improving the PageRank algorithm proposed by Google Company. This method selects the important evaluation index of the key nodes of the power grid, and establishes the directional weighted network model of the power system. Considering the network link direction and the characteristic for the weight of power system network, the PSNodeRank value is proposed to assess the importance of each node. And then the power system partitioning characteristics is utilized to improve the complicated calculation process for the importance of large power grid nodes. The speed of operation is greatly improved and the storage capacity required for the operation is also reduced. Finally, an IEEE 39node system is simulated for verification. It is shown that the proposed method can effectively and accurately identify the key nodes in the power grid and judge their roles in the critical evolution of ACDC power network. This method has a great significance to the critical state evolution of the system.

Abstract:The smart electricity meter is an important and sophisticated metrical equipment in smart grid. Its fault types have complicated characteristics of randomness and fuzziness, which have significant impacts on electrical safety of consumer as well as fairness and justness of measurement directly. Therefore, a classification diagnosis cloud model of fault information of smart electricity meter is proposed in order to analyze the randomness and fuzziness of fault data in smart electricity meter and their correlation. In this model, the clustering algorithm is applied, and the cloud theory is employed to conduct data mining. In addition, the weakness and key parts affecting reliable operation of smart electricity meter is also specified. Finally, the abnormal event record of smart electricity meter in a certain area in 2015 is chosen as the data source for analysis. The effectiveness and practicability of the proposed method is verified.

Abstract:Aiming at the heavy weight and inconvenience when carrying and installing copper earthing wires on the UHV transmission lines, this paper proposes a lightweight method for UHV earthing wire conductor by using the copper clad aluminum(CCA) as composite materials. Firstly, theoretical calculations and experiments of the fusing current in a short time for copper and CCA material are performed. Then, the transient electro thermal coupling process of the earthing wire shortcircuit of typical 1 000 kV doublecircuit line is simulated. The results show that the theoretical value of the earthing wire conductor's fusing current corresponds with the test value on condition of the conductor cross section greater than 4 mm2 as well as melting time less than 1.5 s. The CCA10 earthing wires present 71% cross section increase and 36.2% weight reduction compared with copper wires. In the case of singlecircuit poweroff maintenance of 1 000 kV doublecircuit line, it is suggested to select 50 mm2 copper wire or 95 mm2 CCA10 wire, and the CCA earthing wire has greater safety margin.

Abstract:Artificial live replacement for stockbridge damper is a laborintensive process with low efficiency and high danger. In order to overcome those shortages, a kind of living work robot with double operation ends is developed to replace stockbridge damper. In this paper, the robot's walking wheel, clamping mechanism and operation end are designed. The integration design of ontology control platform and telemanagement host control platform is completed. The replacing process of stockbridge damper by the robot is worked out. And an universal joint motion model of multiarm robot as well as its H Hinfinity control method are put forward to enhance the robustness and working environmentadaptive ability of the robot. Finally, the AC withstand voltage test and fielded experience test shows that the structural design of robot is reasonable. The robot action is coherent and fluent during the crossing, removing and installing the stockbridge damper. All key actions meet the positioning requirements.

Abstract:Automatic Plant Startup and Shutdown Control System(APS) is one of latest development directions of thermal automation technology in power plants. It is necessary to optimize the control scheme of startup boiler which is an auxiliary equipment. Taking the startup boiler of a certain large generating unit as the research target, the major equipment of this boiler is introduced and the startup and shutdown process of this boiler is analyzed. And the essential technical issues, including the frame of automatic startup and shutdown control system and the design of functional group, are emphatically discussed. After that, a scheme of automatic startup and shutdown control for the boiler is put forward. The scheme has been successfully applied to a startup boiler in a 1 000 MW ultrasupercritical unit. The real application results indicate that this scheme not only regulates operation, but also reduces the human intervention level. Both automation level and safety operation level of power plants are improved. The control scheme also provides references for the startup and shutdown control of other types of industrial boilers.