2020, 44(6):1-8. DOI: 10.7500/AEPS20191108007
Abstract:The method of quantifying ancillary services as corresponding green certificates can motivate the enthusiasm of ancillary service providers, which is conducive to the consumption of renewable energy. After analyzing the additional increase of ancillary service caused by connecting renewable energy, the unfairness of current cost allocation way for ancillary service is pointed out. Then, based on the deviation theory of predicted power generation curve, a green certificate calculation method for compensating the ancillary service by transferring the deviation between predicted and actual power generation curves into the numbers of green certificates. Furthermore, a quantitative allocation method of green certificate for frequency regulation ancillary service is designed. The effects and possible problems of compensation method are discussed.
2020, 44(6):9-16. DOI: 10.7500/AEPS20190626005
Abstract:The vulnerability assessment of branch lines in power grid is an important step in the subsequent analysis and prevention of cascading outages. Existing vulnerability indices have problems including unclear physical meaning, lacking consideration of the impact of multiple outages on vulnerability and so on. To address the issues, this paper constructs a cooperative game framework of branch lines in power grid. According to the game result, the loss caused by a fault chain is properly distributed to each branch line as their vulnerabilities. The indices can reflect the contribution of the fault on branch lines to the power grid loss caused by the outage accidents, and its physical meaning is clear. To solve the problem of the non-superposition of the loss of mutiple fault chains, the corresponding multi-objective optimal programming is transformed into a two-stage optimization model with collective rationality and individual rationality. The two-stage optimization model uses genetic algorithm and linear programming, respectively. The case study of IEEE 39-bus system verifies the validity of the proposed index.
2020, 44(6):17-26. DOI: 10.7500/AEPS20190528009
Abstract:In order to further improve the prediction accuracy of transient stability for power system and give more refined evaluation results, the deep learning is combined with the transient stability of power system. A refined evaluation model of transient stability for power system based on ensemble deep belief network (DBN) with different structures is proposed based on the characteristics of the generator power angle "trajectory cluster" after fault removal. The base classifier DBN of the model can effectively utilize the feature extraction ability of the deep architecture and fully exploit the nonlinear mapping relationship between the input features and the evaluation results of transient stability. Experimental results on the New England 10-machine 39-node system show that this method is not only superior to the shallow learning framework, but also superior to the partial deep learning model. In addition, the ensemble DBN algorithm not only has higher prediction accuracy, but also can effectively evaluate the stability margin and instability level of the system. It shows strong robustness when some information of phasor measurement unit (PMU) is missing and contains noise.
2020, 44(6):27-36. DOI: 10.7500/AEPS20190217002
Abstract:In view of the fact that the traditional continuous power flow method is difficult to adapt to the AC/DC interconnected systems with access of large-scale wind power, this paper proposes a probabilistic continuous power flow calculation method for AC/DC systems with wind power. Based on the frequency regulation characteristics of conventional units and the static characteristics of voltage and frequency of loads, the randomness of wind power output, the DC control mode and the frequency fluctuation deviation constraints are taken into account. Then this paper establishes the probabilistic continuous power flow equation of AC/DC systems with wind power considering the frequency regulation of power sources and static characteristics of voltage and frequency of loads. Based on the Jacobian matrix of the modified equation, the load margin sensitivity index reflecting the frequency fluctuation, wind power output and control mode is derived. The power flow solution and load margin in multiple scenarios of N and N-1 states are obtained by combining Monte Carlo method and predictor-corrector method. The correctness and validity of the proposed method are verified by simulation results of the modified IEEE 39-bus system.
2020, 44(6):37-44. DOI: 10.7500/AEPS20190403005
Abstract:Compared with the conventional single-layer connection mode of ultra-high voltage direct current (UHVDC) system, the hierarchical connection mode improves the ability of AC system in acceptance and provides voltage support for multi-infeed UHVDC systems from the aspect of grid structure, and it makes power flow distribution and control more flexible and reasonable. In order to make full use of the fast power adjustment capability of UHVDC systems in the hierarchical connection mode, and avoid or reduce the economic loss caused by the load shedding during the medium- and long-term voltage instability process, this paper proposes a coordinated control method for AC/DC power systems with hierarchical connection mode of UHVDC systems. Firstly, based on the quasi-steady model of UHVDC system in the hierarchical connection mode, this paper deduces the analytical expressions of the sensitivities of converter bus voltages with different control modes of UHVDC systems to the inverter transmission power of UHVDC systems in hierarchical connection mode. Secondly, the voltage trajectory prediction model of AC/DC power systems is established based on the derived sensitivities. Finally, considering modulations of the DC current and the extinguishing angles of high and low ends of the inverter, a receding horizon optimization model of coordinated voltage control is constructed based on the predicted trajectory. The hierarchical DC injection power and the voltage control means of AC system are coordinately controlled. The simulation analysis of the planning system of Shandong power grid shows that the proposed method can effectively coordinate the distribution of DC transmission power in AC power grid, which enhances the system voltage stability and reduces the load shedding loss.
2020, 44(6):45-51. DOI: 10.7500/AEPS20190411007
Abstract:In order to reduce the impact of time-lag of water flow in the process of short-term generation optimization dispatch of cascade hydropower reservoirs, the mapping function between outflow of upstream reservoir, local inflow, water level and inflow of downstream reservoir is quantized based on the principle of linear space. Secondly, the artificial neural network, K-means and Relief methods in data mining are comprehensively applied to solve the mapping relationship. A short-term generation optimization dispatch model of cascade hydropower reservoirs considering time-lags of water flow is established. Finally, the analysis of case in the application of short-term generation optimization dispatch of Jinguan power package in Yalong River of China shows that, the established model could improve the accuracy of short-term generation dispatching plan.
2020, 44(6):52-58. DOI: 10.7500/AEPS20190227008
Abstract:In order to solve the power fluctuation caused by the change of micro-source output, loads and charging and discharging power of energy storage devices because of state of charge in hybrid AC/DC microgrid, a segmented coordination control strategy for hybrid microgrid is proposed. Aiming at the hybrid microgrid in the islanded state, the typical topology of the hybrid microgrid and the power relations in different operation modes are analyzed. The characteristic quantity which can represent the whole operation state of hybrid microgrid is obtained by the standardization method. According to the variation of the characteristic quantity, the control strategy is divided into segments. Then the working principle of each section control is analyzed in detail and the coordinated control of the converters in different control segments is studied. Aiming at the possible power oscillation in microgrid and frequent switching of operation modes for interlinking converter (ILC), the action criterion is compensated. Finally, a simulation model is established in PSCAD/EMTDC. The simulation results show that all converters can respond quickly in different operation conditions to ensure the stable operation of the system.
2020, 44(6):59-66. DOI: 10.7500/AEPS20190603011
Abstract:Traditional control strategy of high voltage ride through (HVRT) for photovoltaic (PV) grid-connected inverter (GCI) improves reactive power output at the cost of reducing active power, which is difficult to achieve a balance between grid-side current and DC bus voltage. And it suppresses the transient impact caused by sudden voltage and current changes after the fault removal. Based on the analysis on the transient power characteristics of high voltage, an HVRT control strategy that maintains active power output and adjusts the reference value of reactive current is proposed. Firstly, a small-signal model is used to analyze the transient power characteristics of HVRT, which reveals that the key of suppressing voltage recovery is the constant active power and reactive redundancy on grid side. Secondly, a method for estimating the reference value of reactive current is proposed according to the voltage surge amplitude. On this basis, by combining with active current control, the control capacity of the GCI with three different voltage surge amplitudes is discussed, and the control strategies of HVRT are given respectively. Finally, the simulation and experiments verify the effectiveness of the proposed control strategies.
2020, 44(6):67-74. DOI: 10.7500/AEPS20190201001
Abstract:In the renewable energy power system dominated by high proportion of wind power, configuring energy storage system (ESS) is an effective solution to suppress wind power fluctuations, achieve peak load shaving and improve wind power dispatchability. A multi-objective hybrid optimization simulation model for joint wind power and energy storage operation is designed, which integrates the impact of output level of the large-scale energy storage and the future output of wind power on the current energy storage operation. Markov prediction model is adopted to forecast the future output of wind power, and particle swarm optimization algorithm is used to optimize the grid-connected power of wind power and energy storage system in real time, then the optimal operation strategy of ESS can be obtained. The typical wind power data from a hundred-megawatt wind farm are used for simulation. Simulation results show that the proposed method has good smoothing performance, which avoids excessive charging and discharging of ESS and prevents it from entering dead zone, and further improves the reliability and the economy of the integrated system of wind power and energy storage.
2020, 44(6):75-83. DOI: 10.7500/AEPS20190804005
Abstract:The output harmonics and switching losses can be significantly reduced with the application of modular multilevel converter (MMC) in high voltage DC transmission system, while it also brings prominent stability problems. Impedance-based stability analysis method can be applied to analyze the stability problems when MMC system operates. In the existing studies, the sequence impedance model of MMC is derived, which is decoupled into a positive-sequence impedance and a negative-sequence impedance, and the system stability can be determined by single-in-single-out stability criterion. However, MMC exhibits significant frequency coupling characteristic in the low frequency band, making the positive-sequence impedance and negative-sequence impedance no longer decoupled, therefore the single-in-single-out stability criterion cannot accurately determine the system stability. In this paper, the frequency coupling mechanism of MMC is revealed by analyzing the interaction between perturbation and steady-state harmonics. Furthermore, the frequency coupling model of MMC with AC voltage control is derived based on harmonic transfer matrices. Based on the derived model, the influence of frequency coupling characteristic on system stability judgment and the main factor influencing frequency coupling characteristic are analyzed. Finally,the simulation results based on MATLAB/Simulink simulation system validate the correctness of the frequency coupling model and stability analysis results.
2020, 44(6):84-90. DOI: 10.7500/AEPS20190411002
Abstract:DC circuit breaker (DCCB) is a key device for clearing DC faults in voltage source converter based multi-terminal DC (VSC-MTDC) system. Most DCCB systems require to connect DC reactor in series to limit the rising rate of the fault current, but large-capacity DCCB connecting DC reactor in series could decrease the damping of DC system and reduce the stability of multi-terminal DC system. In this paper, a small-signal mathematical model for VSC-MTDC system with DC breakers is constructed and modal analysis is performed to illustrate the main factors which could affect the system stability. To solve the stability problem caused by the circuit breaker system, a damping controller based on H∞ hybrid sensitivity theory is designed to improve the weak damping characteristics of the system. Finally, with a four-terminal VSC-MTDC model implemented in MATLAB/Simulink, time-domain simulation and eigenvalue analysis results show that the proposed approach can efficiently suppress the DC system oscillations and improve the system stability. Compared with other traditional damping controllers, it is more applicable in scenarios with strong disturbances and uncertainties.
2020, 44(6):91-97. DOI: 10.7500/AEPS20190328001
Abstract:Medium-voltage multilevel converters usually adopt modular structures. They can be used in AC/DC distribution networks. As a result of the relatively low voltage in distribution networks, the number of converter sub-modules is small. Therefore, the carrier phase shift pulse width modulation (CPS-PWM) is usually adopted to increase the equivalent switching frequency. The modeling of CPS-PWM is one of the fundamentals of converter system modeling. The modulation stage is nonlinear, and thus, only the fundamental frequency component is usually considered. The existing fundamental approximation model of CPS-PWM is based on equivalent switching frequency, hence the low time delay and large errors exist when asymmetric regular sampling is adopted. In this work, the following three forms of fundamental approximation models of CPS-PWM are established, i.e. zero order holder (ZOH) delay approximation, ZOH no approximation, and delay Padé approximation. These models are developed by combining the outputs of sub-models on the basis of a sub-module. The proposed models reflect the frequency characteristics of CPS-PWM more accurately than the traditional one. A simulation study verifies the correctness of the theoretical analysis.
2020, 44(6):98-105. DOI: 10.7500/AEPS20190611005
Abstract:Considering the environmental demands of residents in the hard-hit areas of carbon emissions, an allocation scheme of comprehensive index carbon emission rights based on equilibrium of regional carbon emission is proposed. The allocation of carbon quota is obtained by entropy method with the indicators of the benchmark scheme and the regional carbon emissions of the plant. In order to study the impact of carbon price on carbon emissions in power system, a three-stage optimal tiered carbon price model is established. The first stage is unit commitment problem in normal conditions. The second stage is to find the worst
2020, 44(6):106-112. DOI: 10.7500/AEPS20190304008
Abstract:Iterative bidding mechanism can promote power generators to quote electricity price economically and rationally and organize production. It can also give purchasers the ability to negotiate, and promote the coordinated and efficient operation of the power market. However, current iterative bidding methods mainly settle accounts according to the quoted price or the uniform marginal clearing price, but seldom according to the locational marginal price of nodes. There may be also too many iterations to be applied. Therefore, an iterative bidding model for power market based on analytical target cascading theory is proposed. Power generators and independent system operators are regarded as different stakeholders, and their objectives are optimizing the production efficiency and minimizing the cost of purchasing electricity, respectively. Coupling and parallel solution of the two optimized models are realized by connecting through generation power. Considering the influence of network constraints, the model can effectively motivate generators to report their real costs and is conducive to the optimal allocation of social resources. Examples of IEEE 39-bus system and IEEE 118-bus system show that analytical target cascading can be applied to the iterative bidding mechanism of fully competitive power generation market, and the convergence and computational efficiency can meet the requirements.
2020, 44(6):113-120. DOI: 10.7500/AEPS20190321010
Abstract:Applying the new generation of artificial intelligence in smart grid and Energy Internet, to achieve high proportion renewable energy access to the power grid in a timely and effective manner, the deep deterministic policy gradient (DDPG) algorithm based on deep learning is applied in the optimized operation of active distribution network (ADN). Firstly， DDPG return function of optimization model for ADN with multiple microgrids is constructed, which can minimize the total node voltage deviation and line loss of ADN. The proposed function can also minimize the variation of the power regulation of microgrid to reduce the impact on operation of the microgrid, and maintain the balance of tie-line power blance to reduce the impact on the distribution network. Secondly, DDPG sample data processing, design of return function, model training and learning process of optimization control for ADN are analyzed. Finally, the effectiveness of the algorithm is verified by the improved IEEE 14-bus example simulation.
2020, 44(6):121-128. DOI: 10.7500/AEPS20190507001
Abstract:In order to improve the economy of the investment and operation of user-side battery energy storage and reduce the cost of users’ power consumption, a rolling optimization method of configuration and operation scheduling for user-side battery energy storage is proposed. Firstly, the benefit of users after installing energy storage and the constraints of energy storage operation are analyzed. Then the configuration optimization model of energy storage, the optimization model of energy storage before month and the rolling optimization model for intra-day operation are constructed and solved by CPLEX solver. Performance constraints of energy storage are added to the model, which can effectively reduce the number of transitions between charging and discharging states and prolong the life of energy storage. In the pre-month optimization, a predicted monthly demand defense value is determined. In the intra-day rolling optimization, a subsection optimization model of daily operation for energy storage and a renewal model of monthly demand defense value are constructed. The daily load data and monthly demand defense value are updated in real time, and the rolling optimization is carried out to continuously correct the impact of load forecasting error. Finally, an industrial user is simulated to verify the validity of the proposed optimization model.
2020, 44(6):129-137. DOI: 10.7500/AEPS20190308009
Abstract:The grid-connected power generation system integrated with photovoltaic (PV) and energy storage systems (ESSs) is taken as the research object, and multiple time scales are used in the modeling ideas. Focusing on the main time scale affecting the dynamic characteristics of system frequency—the DC voltage time scale, this paper develops a dynamic model of DC voltage time scale for the analysis of the system inertia and damping characteristics. On the basis of this model, the electric torque analysis method is used to analyze the main factors and their action laws on the inertia effect, damping capacity and synchronization characteristics from the physical mechanism level. The research results show that the dynamic characteristics of the system are determined by its control parameters, structural parameters and steady-state operation point parameters. The inertia effect and synchronization characteristics of the system are respectively affected by the proportional controller and integral controller in the DC bus voltage control loop. The damping capacity of the system is mainly affected by the frequency droop control in the energy storage device. The system dynamic characteristics will not be affected by the PV converter operating in the maximum power point tracking (MPPT) mode. In addition, the system dynamic characteristics are also affected by the structural parameters (such as line impedance and DC bus capacitance), the steady-state operation point parameters (such as the AC/DC bus voltage level of the system) and the steady-state operation power (power angle). Finally, the correctness of the above analysis is verified by the simulation and experiment results.
2020, 44(6):138-145. DOI: 10.7500/AEPS20190610002
Abstract:At present, the grid-connection process of most new-type synchronous condensers in China is achieved in the idle speed mode, which is uncontrollable and irreversible. Therefore, the improvement of the grid-connection success rate of the grid-connection process is the key to minimize the corresponding economic cost. Considering the complete idle speed process of the new-type synchronous condenser and the main loss during the process, this paper analyzes the change rule of frequency difference and phase angle difference between the synchronous condenser and the power grid. Then the influence of the phase angle difference between the synchronous condenser and the power grid on the grid-connection success rate at the start of idle speed is obtained. Based on the analysis of the influence of the condenser speed on the initial position of grid-connection success area, the adjustment method of the speed at the idle speed point is proposed, which is able to improve the grid-connection success rate with satisfaction of the strict grid-connection requirements. Finally, the example results validate the feasibility of the proposed idle speed point setting method, and the complete iteration times and the final idle speed point are predicted accurately by statistics, while the required information during the setting process is significantly reduced.
2020, 44(6):146-154. DOI: 10.7500/AEPS20190216002
Abstract:A compensation control strategy for doubly-fed induction generator (DFIG) based wind turbine is proposed based on disturbance compensation of grid voltage, which can solve the stability problem of grid-connected in weak grid. Firstly, a unified impedance model of DFIG based wind turbine is established in the synchronous reference frame including rotor side converter (RSC) and grid side converter (GSC). Then, based on the established impedance model, the transfer relationship between the voltage disturbance at point of common coupling and the controller output is analyzed, the voltage disturbance compensation is introduced into the current loop of RSC and GSC respectively to improve the converter control. The generalized Nyquist criterion (GNC) proves that the proposed method can effectively improve the grid-connected stability of DFIG based wind turbine in weak grid. Theoretical analysis shows that the compensation control of RSC and GSC based on the voltage disturbance compensation at point of common coupling can improve the impedance characteristics of DFIG so as to improve their stability under weak grid condition. Finally, the effectiveness of the proposed compensation control method is verified by simulation analysis.
2020, 44(6):155-163. DOI: 10.7500/AEPS20190513001
Abstract:Power electronic transformers (PETs) are the key equipment in AC/DC hybrid distribution network. In order to improve the operation efficiency of cascaded PET under the condition of light-load, this paper proposes an efficiency optimization strategy based on active unbalanced power control, which utilizes the unbalanced characteristics of hybrid pulse width modulation (HPWM) technology. By analyzing the power losses of both cascaded H-bridge converter with HPWM and isolated bidirectional DC/DC converter, the operation efficiency model of cascaded PET is established. Based on the efficiency model, the feasibility for improving the overall efficiency of PET through unbalanced power distribution of each module is analyzed. The power stable operation range of each unit with HPWM is derived, and the active unbalanced power control strategy suitable for system efficiency optimization under the condition of light-load is proposed. Finally, the validation of the proposed strategy in comparison with the traditional power balanced control is confirmed by experiments, and the results show that the proposed unbalanced power control can reduce the power loss under the condition of light-load and improve the overall efficiency of PET.
2020, 44(6):164-170. DOI: 10.7500/AEPS20190408002
Abstract:The calculation of voltage gain-frequency characteristic is vital in the design of LLC resonant converter. The LLC converter have various operation modes due to the existence of nonlinearity in the rectifier part of the converter. And the voltage gain-frequency characteristic is closely related to operation modes. The PO mode is a discontinuous conduction mode (DCM), which is beneficial to reduce switching losses of the converter. However, the voltage gain-frequency characteristic of converter in the PO mode is hard to be solved because it involves complex computations. On the basis of subinterval analysis method, this paper analyzes the operation characteristics in the PO mode of LLC resonant converter, and deduces a simplified and high-accuracy calculation formula of voltage gain. In addition, the parsing and concise expressions of power boundary conditions both in DCM mode and PO mode are provided according to the nonlinear characteristic in the rectifier part of the converter. Finally, a simulation model and an experimental prototype of LLC converter are developed. The simulation and experiment results verify the validity of the deduced voltage gain formula and power boundary condition.
2020, 44(6):171-177. DOI: 10.7500/AEPS20190409006
Abstract:Dynamic voltage restorer (DVR) is a power electronic device connected in series with source and load, which is used to rapidly compensate voltage fluctuation in the power system. However, due to the energy storage equipment, the conventional DVR based on voltage source inverters (VSIs) has some shortcomings. And the DVR based on direct AC/AC conversion with pulse width modulation (PWM) has the shortages of commutation problem and voltage balance of flying capacitor. Therefore, this paper proposes a single-phase DVR based on bipolar direct AC/AC conversion. The AC/AC converter topology with PWM used in the proposed DVR has the characteristics of common grounding between the input and output ports. In addition, the proposed DVR can achieve bipolar voltage regulation with a simple control strategy and effectively solve the commutation problem during operation. In order to verify the engineering value of the proposed DVR, a 1 kW experimental platform has been built to verify its rationality and effectiveness on the basis of theoretical analysis.
2020, 44(6):178-185. DOI: 10.7500/AEPS20190630005
Abstract:In order to solve the difficulties in the accurate prediction of DC current during fault transient period, a DC line transient mathematical model with flat wave reactor is presented. Considering the change of overlap angle during transient period, a prediction method of DC current variation is proposed for the accurate prediction of DC current variation. And a commutation failure prediction criterion based on DC current variation during commutation is constructed to realize the prediction of the first commutation failure. Meanwhile, a control strategy which can effectively prevent the first commutation failure is proposed. The DC system model of CIGRE high voltage direct current (HVDC) benchmark model is built based on PSCAD/EMTDC simulation platform. The simulation results of different fault situations reveal that the proposed method and control strategy are effective.
2020, 44(6):186-193. DOI: 10.7500/AEPS20190729004
Abstract:Some frequency oscillations occurred near the governor dead band in the test of asynchronous interconnection in Yunnan in China. To analyze this issue, a piecewise linear system model considering governor dead band is established. The effect of governor dead band on frequency stability is clarified through steady-state and dynamic behavior analysis. Research results show that the system may have no equilibrium point with step-response dead band and it will cause frequency oscillations near the dead band. Meanwhile, the stability of piecewise linear system is determined by the eigenvalue of the linear system corresponding to each frequency range, and the negative damping effect can also cause system frequency instability. Different frequency oscillations in the test are simulated by both the simplified system model and the actual model of Yunnan power grid.
2020, 44(6):194-199. DOI: 10.7500/AEPS20190521008
Abstract:Aiming at the problem that the spatial load forecasting accuracy is reduced due to the random fluctuation in the measured cellular load data, a method for ascertaining the reasonable maximum value of cellular load by using complementary ensemble empirical mode decomposition and runs test technique is proposed. The method decomposes each class Ⅰ cellular load sequence by complementary ensemble empirical mode decomposition technique. Each class Ⅰ cell obtains a set of intrinsic mode functions. Random test of each intrinsic mode function is carried out by using runs test technique, and the criterion for identifying the high-frequency component is established. The high-frequency intrinsic mode function that characterizes the random fluctuation of the cellular load is removed, and the remaining intrinsic mode functions that characterize the regularity and trend of the cellular load are reconstructed to obtain the main component, and the maximum value is taken as the reasonable maximum value of the class Ⅰcellular load. Finally, the reasonable maximum value is used to predict the space load based on class Ⅰ cells and class Ⅱ cells. The engineering example shows that the method is correct and effective.