This study proposes an enhanced particle swarm optimization algorithm designed to overcome the limitations of the traditional particle swarm optimization (PSO) in reactive power optimization, including premature convergence and insufficient search capability. The proposed enhancements enhance global exploration by integrating a competition mechanism with adaptive inertial weights, while optimizing particle distribution through hierarchical congestion and perturbation strategies to improve accuracy. A multi-objective optimization model is established to simultaneously minimize network losses and operational costs. Benchmark function verification, IEEE 14-node case studies, and IEEE 30-node and IEEE 33-node studies have demonstrated the superior performance of the hierarchical mechanism multi-objective particle swarm optimization (LMMOPSO) algorithm: it reduces power losses, accelerates convergence compared to standard PSO, and optimizes voltage distribution and power flow to maintain them within reasonable unit ranges. The algorithm’s robustness in reactive power optimal dispatch confirms its practical applicability to power system optimization.
In recent years,the utilization of renewable energy and the proportion of distributed generation in power grid systems have been increased significantly,and the trend of power electronics in power systems is becoming more and more evident.The problem that arises is that when the grid voltage is unbalanced,the performance of grid-connected inverters decreases which shall be eliminated in the power grid,otherwise it can seriously affect the stability of the power system.Therefore,the control of grid-connected inverters under unbalanced grid conditions is particularly critical.The output power and current quality of the system under unbalanced grid voltage conditions are the main concerned data of grid-connected inverters.In this paper,a three-phase three-wire LCL-type grid-connected inverter is taken as an example,the issues of phase detection inaccuracies and power fluctuations are analyzed,and an improved phase-locked loop(PLL)and power-current cooperative control strategy is proposed.This method can refer to the reference current formula and coordination coefficients to achieve smooth switching among grid-connected current,active power,and reactive power.The feasibility and effectiveness of the proposed strategy are verified through Matlab/Simulink simulations and experiments on the RT-LAB platform.
With the continuous expansion of the application range of self-healing dry metallized film capacitor, its heat dissipation mode and internal temperature-rising have become important theoretical and engineering problems. In view of this, the paper is based on the heating rule of the cylindrical element of the DC-link capacitor. Furthermore, combined with the principle of electro-thermal duality, the thermal calculation model of the cylindrical element is established, and the thermal resistance calculation formula of a cylindrical element is obtained through the integration of the differential cells. Finally, the correctness of the thermal calculation model of cylindrical element is verified through the test, and the verification method of thermal calculation of cylindrical element is proposed, which will provide theoretical and experimental reference for the heating of cylindrical element of DC-link capacitor.
In view of the long period of the current offshore wind power project construction, a self-starting wind turbine grid-connected method is proposed, and a set of wind turbine grid-connected self-starting device is developed based on this method. Through this device, it is available to provide a stable power supply for the wind turbine through the submarine cable before the grid power on the land is delivered to the offshore substation. After the grid power is delivered, the wind turbines are directly connected to the grid, which greatly shortens the construction period of offshore wind power projects, creates considerable economic value. The technology had been applied in many offshore wind farms in Rudong, Jiangsu, and the field test data verified the advancement and effectiveness of the developed system and the proposed method.
研究±500 kV换流站TILP型交流滤波器电容器接线柱频繁发热原因,提出具体的预防解决措施.本文研究新型特殊线夹,为该型设备能够继续良好的运行提供了有力技术支持.通过对不同措施方案进行测试分析和数据研究,对比目前运行情况,得出TILP型交流滤波器电容器频繁发热是由于接线柱结构设计不合理供导流的接触面过小且接触面容易氧化引起,介绍了解决该问题的方案.
针对 220 kV变电站高峰负荷和低谷负荷期间功率因数越限和无功倒送问题,提出一种220 kV变电站无功补偿优化计算方法.建立了变电站功率因数诊断模型,对 220 kV变电站运行数据进行功率因数诊断评估,筛选得到不合格的运行数据,然后构建容性无功优化和感性无功优化模型,对高峰负荷和低谷负荷时间段的不合格运行数据进行无功补偿容量计算.采用容性/感性功率因数检验模型对无功补偿容量计算值进行检验计算,得到了最优的无功补偿配置方案.该方法简单易用、计算速度快,精度高,有效解决现有变电站无功设备配置不合理的技术问题.
为了使分布式补偿装置的配置和对这些装置的实时控制两个不同阶段的结合达到最优化的效果,本文提出了一种 10 kV配电网分布式电压无功优化系统构架,结合了补偿装置的配置优化和补偿装置的实时协调优化控制技术从而真正实现了系统的整体最优化.分析并提出了分布式智能补偿装置的必备功能并在其智能控制器中实现了这些功能.采用 4G/5G加密无线通信技术实现了实时优化控制主站和分布式补偿装置之间的信息交互.最后介绍了一个实施案例以及该案例的运行统计数据总结.
大电流罗式线圈电流传感器频率响应特性主要取决于配套的积分电路参数.从监测并联电容器组投切电流应用需求出发,提出一种适合测量电容器母排暂态电流的新型罗氏线圈传感器积分电路.在罗氏线圈电流传感器工作原理分析基础上,研究不同类型积分器对暂态电流频率响应特性,由频响互补出发计算了两级复合积分器电路参数,并利用雷击电流模拟暂态特性对设计参数进一步优化.测试结果表明该罗氏线圈传感器对暂态大电流测量具有优良性能,适用于测量并联电容器组母排电流谐波,以及投切电流监测及过电流故障分析.
The power electronic converter in the distribution network is required to have sufficient response speed and,at the same time,stable operation. The existing control method has certain contradiction between the rapidity and stability. In view of this issue,the AC grid connected power electronic converter is taken as the research object in this paper,a kind of passive control method with high performance is proposed,which sets the controlled object as passive system and achieves essential stable operation. Moreover,based on the influence of control parameters on the output performance of power electronic converter and with the overshoot and transition time in the process of dynamic regulation as control objectives,the optimization design of each parameter in the control part is performed to solve the contradiction between stability and dynamic property. The PSCAD simulation results proves that the proposed control method can ensure that the power electronic converter has sufficient stability margin and dynamic property,and the overshoot and transition time in the process of dynamic regulation are significantly reduced,which can meet the requirements of high performance and stable operation of the power electronic converter.
The process of frequent switching of parallel capacitor to compensate for power grid reactive power produces serious inrush current and harmonics. The transient over-voltage and large current pose a great threat to the safe operation of capacitor unit. A kind of power capacitor condition monitoring program based on wireless sensing technology is proposed. Through the analysis of core magnetic field distribution,material characteristics and coil output power,the double CT sensor structure with closed-circuit core for high-voltage energy extraction and open-circuit core for measurement is determined. The built-in A/D and the CC2530 chip of the ZigBee communication protocol stack as well as the dedicated signal processing circuit are used for current and temperature rise sensing of the capacitor unit;The low-power IEEE802.15.4standard protocol encrypts the terminal node information for wireless networking,and the aggregated data is comprehensively judged in the background machine analysis. The compact design of the dual-core structure sensor solves the problem of high energy extraction and accurate measurement of linear zone under the magnetic saturation of the iron core,and achieves the overall performance optimization of the sensor’s onsite high-voltage insulation,anti-interference and wireless networking design.
For improving the energy utilization rate and bus voltage stability of distributed power generation in DC microgrid,this paper starts directly from the bus voltage,and proposes the hierarchical control method based on DC bus voltage. The action thresholds are set on the converters corresponding to energy storage units,photovoltaic units,and controllable loads. Each unit is at different operation mode under different voltage interval. The energy storage unit adopts fuzzy droop control based on SOC（state of charge）deviation and DC bus voltage deviation,while,the photovoltaic unit adopts dual-mode switching control.Finally,a DC microgrid model is set up based on Matlab/Simulink,which simulates 8 kinds of operation conditions of coordination and hierarchical control of photovoltaic,storage and load based on fuzzy droop and verifies its superiority.
In view of such issues that two units are equipped with controllers from different excitation manufacturers in unit group wiring,the coordination setting of low excitation limit of different excitation manufacturers is studied. Firstly,the model of two units parallel infinite system is set up,and the small interference model of two units system when the low excitation limit action is not synchronous is deduced. The coordination principle of low excitation limit and the corresponding calculation method of two units parallel system are given. According to the results of simulation and theoretical calculation,it is concluded that different under excitation limit models will lead to the uneven damping distribution. The research results provide a good calculation basis for the parameter setting of under excitation limit of two units parallel system with different excitation manufacturers.
With the increasing precision of industrial automation,a large number of sensitive loads（SL）are connected to the distribution network. As for the SL with different sensitivity levels,the optimized configuration scheme of distributed wind turbines（DWTs）is constructed for improving the voltage recovery capability of SL nodes in case of voltage sag. Firstly,the weighted voltage recovery capability（WVRC）index is adopted to quantitatively characterize the magnitude of voltage recovery capability of SL nodes;Secondly,STATCOM technology is adopted to provide fast reactive power compensation to support the reactive power required during the process of voltage recovery on SL nodes;And then,based on different WVRC index,different DWTs optimal configuration schemes are obtained under the goal of minimum total social cost,so as to meet the requirements of SL nodes with different sensitivity for voltage recovery capability. The second-order cone relaxation（SOCR）method is used in this paper to transform the nonconvex feasible region in the original mathematical model into a convex second-order cone feasible region,thus obtaining the optimal solution of the original mathematical model. Finally,the extended IEEE 33-node distribution network system is selected for simulation and the impacts of different WVRC values on the DWTs configuration scheme has been verified.
In view of the problem that the traditional single time section energy efficiency assessment can’t deal with the time series and index deterioration and can’t identify the weak points of energy efficiency under the strong fluctuation of flexible interconnected distribution network parameters,a kind of combined variable weight based energy efficiency assessment method for flexible interconnected distribution network operation is proposed in this paper. Firstly,the state simulation and parameter calculation of the flexible interconnected distribution network are carried out,and the influencing factors of system level energy efficiency are analyzed from the three aspects of source,network and load,taking into account the influence of operation performance of key equipment,and then the operation energy efficiency index system of the flexible interconnected distribution network is constructed. After that,in view of the strong correlation and discreteness between the indexes,the improved AHP-improved CRITIC-minimum identification information principle is used to solve the comprehensive weight. In order to reflect the deterioration degree of the index time series value,the three-stage variable weight function is introduced to realize the comprehensive weight reward and punishment,and then the energy efficiency assessment value at each time and all day is obtained.Finally,an example shows that the proposed method can effectively reflect the degradation degree of index,identify the weakness of energy efficiency and guide the proposal of loss reduction measures.
In case of fault of both cyber system and physical system of distribution network,the coordination between the two systems may achieve a better recovery effect. A kind of fault coordination recovery method of cyber physical system of distribution network is proposed by way of comprehensive considering such factors as gain,risk and control cost in the process of recovery. Firstly,adjacent matrix of cyber physical system is set up according in accordance with topology structure of distribution network in combination with related knowledge of graph theory. Secondly,by mapping the recovery of cyber system to the recovery of physical system,an optimization model of the coordinated recovery of cyber physical system based on the monetary dimension is set up with the goal of maximizing the net recovery gain. On this basis,a coordinated recovery strategy of cyber physical system is proposed. Finally,an example is given to compare different restoration schemes,and the rationality and effectiveness of the proposed recovery method are verified.
The DC filter protection configuration for±800 certain converter station is introduced in this pa?per.It is found through analyzing an event of unbalance protection action of DC filter capacitor C 1 that the unbalance protection setting value of capacitor C 1 is inconsistent with the protection setting value used in the fault recording analysis,which will interfere analyst’s judgments on the correctness of protection action At the same time,it is also found through the analysis of no operation for a long time with disconnection of main wiring of the low?voltage side of the DC filter C 1 capacitor that the existing DC filter protection configuration is incomplete,and the problem of the main wiring disconnection of the DC filter could not be distinguished effectively.In response to above problems,some optimization measures are proposed.One of them is to optimize the capacitor C 1 unbalance protection logic and setting value to achieve the unification of the protection setting value and the fault recording analysis setting value.The other is to increase main wiring disconnection protection of DC filter,which can effectively judge the main wiring disconnection of DC filter.
Accurate estimation of system harmonic impedance is of great significance for power harmonic analysis and control. Adoption of the variable coefficient regression method to estimate the system harmonic impedance can not only obtain an accurate estimation value in case of large background harmonic fluctuation,but also adaptively track the time-varying trend of the system harmonic impedance. However,the computational efficiency of classical variable coefficient regression method is low,and the selection of window width of the kernel function has great influence on the estimation result. In order to make up the defects of this method and improve it’s engineering applicability,a method of system harmonic impedance estimation based on improved variable coefficient regression is proposed in this paper. The proposed method greatly improves the computational efficiency by avoiding high-dimensional matrix operation. Furthermore,adoption of generalized cross validation method to select the optimal window width makes the estimation result more accurate. The effectiveness of the proposed method is verified by the simulation analysis.
Modular multilevel-based static synchronous compensator（MMC-STATCOM）needs,in case of compensating unbalanced load,considering the current bearing capacity of the components of the device.In order to measure the unbalanced compensation capability of MMC-STATCOM,the ratio of the maximum arm current to the negative sequence current amplitude to be compensated can be used as the measurement standard,and the key factors affecting its compensation ability can be obtained through theoretical analysis.In addition,as for the internal circulating current existed in MMC-STATCOM,its DC component can be used when processing power balance and the inter-phase balance control,and its double frequency component can be eliminated. The current control strategy is based on the quasi-PR control,which can realize the fast tracking of the current to be compensated and the real-time suppression of the double frequency component of the circulating current. And then,the modulated wave equation containing the multiple resonance ring can be obtained through the voltage feedforward control. The DC side voltage of each sub-module is controlled in three layers,in which the inter-phase balance control can be implemented by power balance firstly,and then corrected by the DC component of the circulating current. The correctness of related analysis and the effectiveness of control strategy are verified in the simulation experiment.
The necessary condition for normal operation of modular multilevel converter（MMC）is to maintain stable operation condition at DC side,and the voltage equalization control of the bridge arm module is one of the important links of its control. In view of the disadvantages of traditional voltage equalization strategy in the aspects of calculation speed and switching loss,a kind of optimized voltage equalization strategy based on extreme index sorting algorithm is proposed. According to the voltage variation of the charge and discharge process of the sub-model,the reordering optimization is carried out on the basis of the extreme value index ordering and the number of switching operation is reduced. The Matlab/Simulink simulation software is used to model the MMC of 21-level. The obvious advantage of the extreme value index algorithm is reflected by way of simulating and verifying such four sorting algorithms as bubbling,mixing,extreme value index and optimized extreme value index,which reflects specifically in the low switching operation number and excellent speed of calculation. The C++ algorithm program is developed by using the Visual C++ 6.0 platform,and the simulation data operation is carried out,which has been applied to the control of the MMC reactive power generator.
Model predictive control（MPC）has received wide attention and has been applied gradually to high power converters due to its excellent performance,fast response and achievement of multi-objective control. Its application prospect is very broad especially in the control system of super conducting magnetics energy storage（SMES）with high power to weight ratio. The MPC-based SMES multi-objective control,which is proposed in this paper,can not only achieve high precision current tracking of SMES coil in energy storage,at the same time control the current change rate of SMES coil during charge and discharge,avoid large current impact,but also improve its response speed and accuracy. The simulation experimental platform is set up to verify the steady current performance,power balance stability performance,dynamic performance as well as the control performance of the charge and discharge rate of the SMES system proposed in this paper. The result shows through comparison with PI-based system that the SMES system based on model predictive control can not only achieve high-precision current tracking in steady state,but also can achieve the control of current change rate during current charge and discharge.
