Abstract:In consideration of the different charging needs of users,the charging pile with multiple charging modes appears,increasing the complexity of electric vehicle charging control. Aiming at a certain number of charging piles with multiple charging modes,an optimized control strategy of charging piles is proposed. Firstly,according to the matching results of the charging piles and electric vehicles,the load state vector of the charging pile is established. Secondly,according to whether the control system needs to consider the demand response of the peak shaving,different control objectives and strategies are proposed. Finally,an actual industrial park is taken as an example. Considering the PV output,battery storage and time-of-use electricity price in the park,the results under the three conditions of uncontrolled,control without considering demand response,control considering demand response are compared,and the effectiveness of the proposed strategy is verified by simulation.

Abstract:Under the background of the global energy crisis,electric vehicles have become increasingly popular. Disordered charging of large-scale electric vehicles has brought new electric pressure to the power system. It is particularly important to plan orderly charging of electric vehicles and solve the problem of peak-valley difference widening caused by charging. Based on the game theory,evaluating the peak shaving level by variance is proposed firstly. Users' waiting time for charging is considered. A game charging model based on revenue from selling power stations and user satisfaction of electric vehicles is established. Then,particle swarm optimization (PSO) algorithm is used to obtain the optimal strategy of both sides to achieve the game equilibrium. Finally,simulation analysis is carried out with a residential area of 600 households as an example. Simulation results show that the built model and solution strategy can optimize the charging behavior of electric vehicles,adjust the grid load variance to a large extent,and prevent the generation of new charging peaks. At the same time,on the basis of ensuring the power grid's peak and valley adjustment needs,the income of selling power station is increased and the travel needs of users are satisfied.

Abstract:Large-scale electric vehicles (EVs) scheduling models are complex and require high calculation capacity. To solve these problems,machine learning methods have attracted more and more attention in electric vehicle charging and navigation scheduling. For the photovoltaic-storage-charging integrated energy station,a scheduling strategy of the energy stations based on deep reinforcement learning (DRL) is proposed in this paper. Firstly,the operation strategy of energy station and the basic theory of deep reinforcement learning are analyzed. Secondly,the users psychological state of time and cost for different charging schemes are described based on regret theory,and the agent perception model of user-EV-station state environment is established. To improve the convergence speed of the algorithm,time varying ε-greedy strategy is introduced as action selection method of agent. Finally,multi-scenario simulations are designed based on the actual road network and energy stations in Nanjing. The results show that the proposed method effectively improves the photovoltaic consumption rate of the energy station under the condition of considering the psychological effect of various users. The proposed method provides a new idea for electric vehicle charging scheduling.

Abstract:Flexible district which interconnects several AC districts through the flexible devices is able to make full use of capacity of transformers and cope with large-scale EV charging loads. The coupled districts could be co-operated. Hence,it enhances both the security and economy. The security-economy operational boundary model,as well as the concept of security margin percentage,is proposed for the operaiton of flexible district. Furthermore,a convex-hull based two-dimensional security-economy region is constructed to promote the co-operation among numerous charging stations. Numerical results demonstrate both the extensibility and the practicality of the proposed security-economy operational boundary. Besides,numerical results reveal that economic index has dramatic influence on the shape of security-economy operational boundary,and the boundary shapes for AC charging station and DC charging station are different.

Abstract:Since charging stations are aggregation centers which provide charging service for electric vehicles and adjustment flexibility for power grid,it is necessary for charging stations to give consideration to interests of all partners with respect to users' independent choices. Focused on multi-element charging stations (MCS) which contain conventional load,energy storage,distributed renewable energy generation (DRG),charging price mechanism is firstly designed in this paper to improve participation motivation of users in providing charging flexibility. Then,both rational model based on expected utility (EUT) theory and bounded rational model based on prospect theory (PT) are established to describe users' selections. Model predictive control (MPC) is adopted to deal with uncertainties. Both operating economy and adjustment flexibility of charging stations have been taken into account in the comprehensive scheduling strategy. Finally,double-layer optimization is converted into single-layer through dummy variables to reduce calculation time and make the strategy suitable for online application. Advantages of the bounded rational model in predicting users' behavior are verified by experiments towards real people. In the simulation experiment,the proposed scheduling strategy integrating social-physical methods shows good performance in following requirements of power grid and reducing operating costs.

Abstract:During the long-term operation of cable accessories,the interface silicone grease is in direct contact with the insulation silicone rubber,so that it is gradually absorbed by the silicone rubber,resulting in the performance degradation of silicone rubber,and even the interface problems of cable accessories. Based on the weight increasing test of silicone rubber absorbing silicone grease for accessory insulation,the diffusion rules of silicone grease/silicone oil in silicone rubber at different temperatures are studied and the corresponding diffusion parameters are obtained. The influence of silicone grease on the electrical properties of silicone rubber,such as volume resistivity,relative dielectric constant,dielectric loss angle tangent value and power frequency breakdown field strength,is investigated. The results show that the diffusion law of silicone grease/silicone oil in silicone rubber follows the Langmuir diffusion model at different temperatures. The absorption capacity of silicone rubber to silicone grease/silicone oil increases with the increase of temperature,and the desorption process of silicone oil molecules in silicone rubber is stronger. With the increase of the absorption amount of silicone rubber to silicone grease,the volume resistivity of silicone rubber decreases gradually,the relative dielectric constant and the tangent value of the dielectric loss angle increase gradually,but the breakdown field strength of its power frequency first increases and then decreases.

Abstract:Aiming at the problem that massive data in distribution Internet of Things is vulnerable to network attack,a data security protection method of distribution Internet of Things based on blockchain is proposed. Firstly,a collaborative security protection architecture is constructed,and the abnormal traffic identification mechanism is designed through adaptive traffic monitoring. Secondly,the improved practical Byzantine fault tolerance (PBFT) is used to establish a trust mechanism to share the attack detection model between the distribution cloud master station and each distribution edge agent. Then,based on the blockchain smart contract,the attack detection model is dynamically updated,and the deep reinforcement learning is used to train and fuse each attack detection model to obtain the attack detection fusion model. Finally,a simulation platform based on Mininet is built to demonstrate the proposed method. The results show that the comprehensive performance of the proposed attack detection model is better than that of centralized and distributed models.

Abstract:Cross-linked polyethylene (XLPE) is widely used in the field of cable insulation due to its excellent dielectric and physical and chemical properties. During the service of the cable,space charges accumulate inside the cable insulation,which can cause electric field distortion in severe cases,leading to cable breakdown accidents. Especially for DC XLPE cables,the accumulation and influence of space charge cannot be ignored. Aiming at the space charge accumulation effect generated in DC XLPE cable insulation,the current academic circles mainly adopt methods such as blending modification,polymer segment grafting polar groups,nano-doping modification and preparation of high-purity insulating materials. The modified DC XLPE cable insulation has an improved suppression effect on space charge generation. The suppression methods of space charge in the above-mentioned DC XLPE cable insulation are described in a review,and their suppression principles and corresponding inhibitory effect are introduced. Then,the advantages and disadvantages of different space charge suppression methods are compared and summarized. Finally,the future research and development of space charge suppression methods in DC XLPE cable insulation are prospected.

Abstract:Aiming at the demand of evaluating voltage sag severity considering both electric utilities and users,an evaluation method based on power system and equipment sensitivity analysis is proposed. Initially,based on the focus of voltage sag assessment on system and user-side,the comprehensive attribute set is formed. The system-side index set is constructed with the average influence degree index as well as the voltage sag frequency index,further forms a comprehensive evaluation index set with the failure rate index. The average influence degree index is formed based on cumulative distribution density functions,statistical table and severity indices. The failure rate index is formed based on the comprehensive voltage tolerance curve and energy loss calculation. Then,the indices are weighted and Mahalanobis distance-based technique for order preference by similarity to ideal solution (TOPSIS) is used to reflect the voltage sag severity levels. The simulation model of the IEEE 30-bus system is built so that the proposed method could be used to evaluate the voltage sag severity. The result could help find out the nodes which suffer from greater impacts of voltage sag events. Finally,the fuzzy C-clustering method is used to analyze the evaluation results. It is verified that the proposed evaluation method is helpful to overcome the one-sided problem of single index evaluation.

Abstract:The electromagnetic transient process of flexible direct current (DC) grid is complex. It easily causes overvoltage when a single-phase-to-ground fault occurs in converter station,which seriously affects the safe and stable operation of the system. The converter station is divided into alternating current (AC) fault area and DC fault area by single-phase-to-ground fault. Firstly,the compound sequence network is used to quantitatively analyze the mechanism of AC overvoltage and characteristics of fault zero-sequence components. Then,the propagation law of fault sequence components from AC side to DC side is deduced based on the control equation of converter,which reveals that the fault zero-sequence component is the main factor leading to overvoltage at the DC side and other converters. In addition,a DC overvoltage control strategy is proposed to suppress DC side overvoltage,and the selection method of controller parameters is given. At last,a test system is built on the basis of PSCAD simulation. The simulation results show the correctness of fault mechanism deduction and the effectiveness of overvoltage suppression strategy which improves the safety and stability of power system after single-phase-to-ground fault in the converter station.

Abstract:The parameter design and fault analysis of high voltage direct current (HVDC) system depend on modeling and simulation. The simulation of traditional detailed model usually needs real-time digital simulator and other equipments. However,it is difficult to ensure the accuracy of fault state in the existing simplified model. Therefore,it is necessary to propose a new modeling method to reduce the transient error of simplified model,and then reduce the simulation difficulty and design cost of HVDC system. In this paper,an ordinary differential equation (ODE) model of HVDC based on line-commuted converter (LCC-HVDC) is proposed,and the fault state of the switching function is modified. Firstly,based on the theory of switching function,the ODE model of LCC is proposed. Then,the influences of natural commutation point offset and commutation failure on the switching function are analyzed,and a switching function modulation strategy considering fault state is proposed. Finally,the simulation error and operation efficiency of ODE mode under various fault conditions are calculated with simulation examples. The simulation results show that the ODE model proposed in this paper accurately reflects the various fault states of the converter valve,with small error and high efficiency.

Abstract:SiC devices have advantages in high frequency and high temperature occasions. Applying SiC devices to wireless charging systems can effectively improve the efficiency of wireless charging systems. Firstly,the characteristics of SiC material and Si material are compared firstly. Based on this,a set of power supply device for wireless charging system based on SiC device is developed. The device consists of rectifier module and inverter module. The input terminal is connected to mains electricity. The rectifier module has a power factor correction function. Then,the design method of rectifier bridge,filter circuit,Boost circuit and switch tube control circuit in rectifier module is given. The design method of switch tube drive circuit and switch tube protection circuit in inverter module is also given in this paper. Finally,the experimental results verify the effectiveness and feasibility of the scheme. High power factor is achieved on the input side. The voltage oscillation of the inverter switch tube is suppressed. The peak efficiency of the experimental prototype is 98.2%.

Abstract:Due to the volatility of photovoltaic power generation,the existing photovoltaic power generation prediction technology has problems such as incomplete consideration of meteorological factors and insufficient feature extraction. In order to improve the accuracy of photovoltaic power generation prediction,an improved typical meteorological year method (TMY Method) is proposed to generate typical meteorological year data,and this method is combined with the generalized regression neural network (GRNN) to predict photovoltaic power generation. First of all,six kinds of historical meteorological indicators are selected,and Finkelstein-Schafer statistical method is used to select typical meteorological week and generate typical meteorological year data. Then,the factor analysis method is used to filter out the meteorological indicators that affect the photovoltaic power generation,and the selected meteorological indicators and daily photovoltaic power generation are standardized as the initial input of the GRNN model to obtain the predicted daily photovoltaic power generation. Finally,the designed model is trained and predicted by historical weather data and daily power generation data from Nanjing,Jiangsu Province. The results show that the prediction method proposed in this paper has better prediction accuracy than the original TMY-GRNN prediction method dose.

Abstract:The DC power distribution system usually adopts bipolar power supply mode,which is suitable for occasions with independent positive and negative load and high reliability requirements. In order to realize the application of power electronic transformer in real bipolar low voltage DC system, a power electronic transformer topology based on switching multiplexing sub-module is proposed,and its working principle and control strategy are analyzed. The proposed power electronic transformer control includes low-voltage port voltage control and voltage equalization control. The voltage equalization controller and the low-voltage port voltage controller are mutually decoupled to facilitate independent design. Compared with the traditional true bipolar wiring form,the effective multiplexing of power switches between double active bridges and voltage equalization circuits in the proposed sub-module effectively reduces system cost and improve power density. The simulation results verify the correctness and effectiveness of the proposed control strategy.

Abstract:The safe operation of power grid is affected seriously by transmission line galloping. Based on the influencing factors of transmission line galloping,the fuzzy mathematics theory is applied in the early warning of transmission line galloping. An early warning model of transmission line galloping based on the improved fuzzy hierarchical comprehensive evaluation method is proposed. Firstly,five major impact indicators including wind speed,rainfall,relative humidity,air temperature and wire tension are screened out to construct an early warning evaluation factor set. Four evaluation target sets including level IV,level III,level II and level I are constructed. Then the monitoring data is substituted into the model calculation. The early warning information is issued according to the fuzzy comprehensive evaluation results,which provides guidance for staff to perceive line risk events and prepare for disaster prevention. Finally,the early warning auxiliary decision-making system of the transmission line galloping based on ArcGIS Engine is developed to achieve the visualization of early warning information,and the actual line data is selected to verify the effectiveness of the proposed method. The results show that the proposed early warning method provides decision support for the safe operation of transmission lines.

Abstract:The paramount reason for the malfunction of a center connecting joint,where considers to be the major fault point of power cables,is construction deficiencies. As a result,a three-dimensional finite element model is generated in order to evaluate hazards of the construction deficiency towards the intermediate joint of the cable. Furthermore,this article analyzes the patterns of electric field distribution when three kinds of construction-deficiencies occurs including uneven application of silicone grease,scratches on the surface of major insulation and moisture into the intermediate connector. The last but not least,the article explores the relation of the fault location to the field strength. Additionally,a platform for partial discharge test in the context of alternating voltage is established for verifying samples. The results show that the most significant electric field distortion appears when the silicone grease is unevenly applied and the air gap emerges near the stress cone. While oppositely,the field strength decreases gradually when the air gap position is away from the stress cone. Besides,the most serious electric field distortion is visible while the major insulation has scratches. Also the field strength is maximized at the cut-off of the outer semi-conducting layer. Moreover,the most intense electric field distortion arises from when water membrane is close to the stress cone due to moisture ingress into the joint. In addition,numbers of discharge increase successively in the partial discharge test conducted at the same voltage and under the circumstances of above mentioned three kinds of construction deficiencies.

Abstract:To understand the partial discharge characteristics of the insulating material of high-voltage insulated gate bipolar transistor (IGBT) devices under the impact of the broadband pulse electric stress,a platform for partial discharge measurement is established. The effects of voltage amplitude,pulse repetition frequency,rising edge time and pulse width on the partial discharge characteristics of polyimide film are systematically studied. The results show that as the voltage amplitude increases from 3 kV to 11 kV,the amplitude of the partial discharge signal at the rising edge of the pulse increases,while the discharge delay decreases. When the pulse repetition frequency increases from 50 Hz to 100 kHz,the discharge delay of the rising edge increases but the amplitude of local amplified signal keeps almost unchanged. When rising edge time increases from 136 ns to 300 ns,the discharge delay increases and becomes more dispersed,and the amplitude of the partial discharge signal gradually decreases. Compared to the case of the pulse width of 1 μs,the amplitude of the partial discharge signal becomes larger and the discharge delay becomes smaller in the case of the pulse width of 500 ns. This result provides an important scientific basis for the insulation aging and condition monitoring of solid-state switches in high-voltage high-power electronic devices.

Abstract:In order to investigate the reliability of the continuous operation of the high voltage cross-linked cable which reaches the design life limitation,a 180-day pre-qualification test method is carried out on a section of high-voltage cross-linked cable which has been in operation for 32 years. By comprehensively analyzing and comparing the differences of electric field strength and temperature field in the inner,middle and outer layers of cable insulation during the test,the testing conditions of each insulation layer are converted into the surface of conductor core,and the corresponding equivalent testing conditions are obtained. It is found that equivalent testing conditions of the medium and outer positions are close to the cable practical operation condition. Several diagnostic measurements are conducted to analyze the micro-structure and aggregation structure among the samples. The results show that the inner layer produces more small molecular chains and oxygen-containing groups in the high-strength test condition. The medium layer insulation has a certain oxidative degradation,but the crystalline structure is still improved after the moderate condition. The crystalline form of the outer layer insulation has been improved under the equivalent testing condition. Therefore,it is expected that the rest of retired cables still have the potential to reuse in long-term actual operation condition.

Abstract:Motors comprise the major electricity loads in offshore oil projects and motor starting may lead to instantaneous impacts. The impacts probably deteriorates system performance if DC distribution networks are employed. Based on the idea of analytical expression,the impacts of such loads are analyzed. Firstly,by simplifying the differential algebraic equations of the system,the analytical expression of DC bus voltage is derived. Then the impacts of instantaneous active load on DC distribution network in offshore oil platform are researched. Finally,a four terminal DC distribution network model based on the PSCAD/EMTDC platform is built and different instantaneous loads are set for simulation. The DC bus voltage under different instantaneous loads is analyzed. The correctness of the analytical expression is verified. The results show that the analytical expression effectively reflects the adaptability of instantaneous load on DC distribution network in offshore oil platform.

Abstract:As more and more flexible AC transmission devices are used in modern power systems,the research of optimal power flow (OPF) with thyristor controlled phase shifting transformer (TCPST) is facing new challenges. In order to ensure the security of the system,the equivalent power injection model of the TCPST access system is combined to establish the optimal power flow model with TCPST with the objective function of the minimum active power loss. In order to improve the convergence speed,the multicenter correction interior point method is improved. The iteration step of the affine direction is improved and the key mapping parameters is reconfigured. Based on the improved multicenter correction interior point method,the optimal power flow calculation test is completed in IEEE 14,IEEE 30 and IEEE 118 node networks,and the control ability of TCPST to line power flow distribution is tested. The test results show that the improved interior point method proposed in this paper is feasible and practical.

Abstract:The integrated energy system contains a large number of uncertain factors,so it is necessary to consider the operating state of the integrated energy system in an uncertain environment. Based on the analytical method and the approximate method,a rapid calculation method for the probability power flow of a radial heating network is proposed in this paper. Firstly,the mean and variance of the mass flow rate within a pipe are obtained through the digital characteristics of the normal distribution function. Then,the mean and variance of the node temperature are calculated by the properties of continuous random variables. When the probability distribution of the heat output of the combined heating and power (CHP) unit is calculated,the probability distribution of the electric output of the unit is obtained. Finally the probability distribution of each state variable of the power system is calculated. The method proposed in this paper is able to calculate steady state power flow and probability power flow without iteration,and there is no convergence problem. The method greatly improves the calculation speed while ensuring the calculation accuracy,which provides a reference for the uncertainty analysis and risk assessment of the integrated energy system.

Abstract:Recognition of transformer bushing oil level based on infrared image suffers from over-reliance on temperature information and low efficiency on manual processing. In order to solve the existing problems,an intelligent bushing oil level recognition method based on improved single shot detection (SSD) with the combination of target detection technique is proposed in this paper. The bushing area in the infrared image is detected through introducing the SSD target algorithm. And SSD algorithm is improved by adding the center loss function. Furthermore,oil level detection of bushing is achieved through the application of simple linear iterative clustering (SLIC) without relying on the temperature information. Comparative results among the image-based oil level recognition algorithm proposed in the paper and the manual oil level detection method,show that the efficiency of the proposed one is prior to that of the traditional temperature-based method. Moreover,the relative error of the algorithm is only 0.08%. Therefore,the proposed algorithm greatly improves the detection efficiency while ensuring the detection accuracy,so as to enhance the efficiency of bushing fault diagnosis and the degree of intelligence.

Abstract:As the key equipment of flexible DC power grid,DC circuit breaker is used for fault fast isolation and recovery,which greatly improves the security and reliability of DC power grid. Based on the hybrid high voltage DC circuit breaker,the control,protection architecture and control protection strategy are studied in detail. First of all,the ‘redundancy dual’ control system architecture and the ‘two out of three’ protection system architecture are designed,which have the characteristics of clear architecture level and clear function allocation. Then,the opening and closing control strategies of the hybrid DC circuit breaker are given,and the overcurrent protection strategy,the main equipment protection strategy and the auxiliary equipment protection strategy are designed in detail. Finally,a set of control and protection system of 500 kV/25 kA hybrid high voltage DC circuit breaker is developed and simulated by real-time digital simulation system (RTDS). The results show that the control and protection system and its strategy are correct and effective.

Abstract:The electromagnetic induction relationship among gas insulated line(GIL) circuits is different from overhead conductors. The existence of GIL in the line affects the calculation results of induced voltage and current,thus impacting the selection of line grounding knife. The double-circuit overhead line on the same tower and the GIL hybrid UHV line are taken as examples to study the calculation of electromagnetic induction between loops. Firstly,the electrical parameters of GIL are calculated. The calculation formula of relationship between induced voltage and current of transmission line with GIL is analyzed theoretically. Secondly,the theoretical analysis results are verified by EMTP-ATP simulation,and the induced voltage and current change law of the GIL in the busbar position,in the middle of the line and without GIL are simulated separately,and the reason for the large magnitude of induced current when the GIL is in the middle of the line is analyzed. Finally,the influence of difference length proportion of GIL in line on the induced voltage and current is studied. The research results provide a theoretical calculation reference for the selection of ground switches on mixed lines with GIL.

Abstract:Power grid fault disposal preplan is an important reference for power grid fault disposal. Hence,extracting fine-grained key entity information such as power equipments,name and number from the preplan is an important basis for the computer to understand the content and further support the intelligent disposal. A named entity recognition technology for power grid fault disposal preplan is proposed based on deep learning. Firstly,the word vector is used to represent the preplan text. Then the word vector features are extracted by combining the attention mechanism and the bidirectional long short-term memory network. Finally,the optimal serialization annotation is solved by the conditional random field. The example shows that the proposed entity recognition model can automatically and efficiently extract text features,thus accurately identifying entity words in the preplan. It proves that the model meets the requirement of extracting key entity information in the preplan better than another commonly used model dose.

Abstract:It is significant for the safe and stable operation of power system to accurately diagnose and evaluate the insulation status of oil-paper-insulated power equipment. Frequency-domain dielectric spectroscopy has the advantages of abundant insulation information and strong anti-interference ability. However,it takes a lot of time for low-frequency test process of frequency domain spectroscopy (FDS),which has a negative influence on field testing,diagnostic and evaluation. In addition,the FDS curve is very vulnerable to external interference and result in test error due to the low current amplitude in the low frequency band. In order to solve this problem,a mixed solution of time domain method and frequency domain method is adopted. In the low frequency band,the Fourier Transform is used to transform the time domain polarized current signal into the frequency domain to obtain the loss factor of the low frequency band. Meanwhile,a set of insulation diagnosis devices is designed in this paper based on hybrid time and frequency domain method. The linear cascade amplifier has the advantages of high bandwidth,high voltage and low harmonic distortion. A symmetric differential structure is adopted in the micro current measurement circuit,which suppress temperature and power frequency interference effectively. Finally,the accuracy of the device is verified by the experimental test on the oil-paper insulation bushing model of the transformer.

Abstract:Renewable energy sources (RES) such as distributed photovoltaic (PV) and wind turbine (WT) provide uncertain and stochastic active power supply. Then the large amount of RES brings great challenges to the operation of distribution network. A distributionally robust optimal dispatch method considering multiple source-storage coordinated interaction for distribution network is proposed in this paper. Based on the distributionally robust optimization theory,the dispatch decisions of traditional discrete devices,RES and energy storage are optimized coordinately to improve the economy and safety of distribution network operation,and thus realizing the effective balance of conservativeness and robustness for distribution network dispatch decision. Firstly,with consideration of multiple source-storage coordinated interaction,the optimal distribution network dispatch model is constructed and reformulated as the mixed integer second-order cone programming form. Secondly,the RES output scenarios are utilized to describe the uncertainty while the distributionally robust optimal dispatch model of distribution network considering multiple source-storage coordinated interaction is constructed,after which the column and constraint generation algorithm are employed for model solution. Finally,it is carried out based on the PG&E 69 node system to verify the feasibility and accuracy of the proposed method.