Abstract:Collector lines in wind farms mostly contain multiple branches,and generally only one measuring device is set at the head of the line. If single-end fault locating is carried out based on the device after a fault,only multiple suspected fault points can be obtained on different branches,and the actual fault branch cannot be determined. Although the current multi-terminal algorithm can locate the fault branch,the number of measuring devices required is too much,which is expensive. In this paper,the topological structure of a collector line is analyzed under the most common single-phase grounding fault. It is pointed out that when faults occur on different branches,the same data may be collected by the single-end measuring device,making the fault branch cannot be identified only by the single-end electrical volume,so it is necessary to add measuring devices to identify the fault branch. In order to reduce the investment,the minimum number and location of additional measuring devices required to identify the fault branch on lines with different topologies are determined. Combined with the new measurement information and the existing single-terminal method,an algorithm is proposed to locate the fault branch and achieve fault location based on multi-terminal fault recording information. The effectiveness and practicability of the proposed method are verified by PSCAD/EMTDC simulation. The proposed algorithm is improved from the single-terminal impedance method to solve the problem of fault branch location,and it can be realized by adding a few measuring devices on the basis of the current configuration. The cost is low,which means a good engineering application prospect.

Abstract:Aiming at the ineffective suppression results of sub-synchronous oscillation under grid connection of doubly-fed wind turbines process with traditional method,a virtual resistance-based fractional-order proportional integral (FOPI) control strategy is proposed. Firstly,the sub-synchronous oscillation mechanism of wind power systems is analyzed,and the grid-connected current of the wind turbine is selected as the input signal to design the sub-synchronous broadband current filter. To improve the electrical damping of the wind turbine system,a virtual resistance controller is constructed. After that,the influence of the integral order on the dynamic characteristics of current tracking is analyzed. Where the fractional integral operator is improved by employing Oustaloup algorithm,and the FOPI current control loop is reconstructed by choosing the grid-side converter's control equation as the bottom model. Then,the non-linear iterative weights and learning factors are introduced to improve the traditional gray wolf algorithm,as well as optimize the parameters of control policy. Finally,the simulation verification is carried out under Matlab/Simulink platform. The results show that the proposed strategy can suppress the sub-synchronous oscillation effectively under different compensation levels and exhibit strong robust stability.

Abstract:The modular multilevel matrix converter (M3C) is the core equipment for fractional frequency transmission system (FFTS). The AC-AC conversion of M3C leads to direct coupling of the ac electrical quantities at different frequencies,causing complex harmonic distribution. In order to analyze the harmonic characteristics of the M3C,the analytical expression of the sub-module capacitor ripple voltage is derived based on the operating principle of M3C at first. On this basis,the analytical formulas of the nine bridge arms currents harmonics are derived,taking into account the coupling of all four frequency components of the capacitor voltage. The relationships between the multi-frequency harmonic currents of bridge arm and the system currents on two sides are analyzed,as well as the key factors affecting the amplitudes of ripple voltage/harmonic current are discussed. The results show that in steady-state,the currents at frequencies ω₁ and ω₂ flow into the ac systems as the positive-sequence fundamental currents; the currents at frequencies 3ω₁ and 3ω₂ flow into the ac system as the zero-sequence components;the remaining harmonics are circulated in the converter. A zero-sequence current mitigation control strategy for the M3C is proposed. The accuracy of the theoretical harmonic analysis and the effectiveness of the control strategy are verified by simulations in Matlab/Simulink.

Abstract:Low-frequency transmission system (LFTS) based on modular multilevel matrix converter (M3C) has great application potential in offshore wind power transmission,urban power grid and long-distance power transmission. In order to study the startup process and the control strategies of LFTS and connect the converter station to the low-frequency transmission line without voltage or current impulse of M3C,a startup method of undisturbed grid connection is proposed. Firstly,the structure and operating principle of the LFTS are analyzed. Then,the selection principle of sub modules and the charging process of M3C are analyzed with two-terminal LFTS as the target. Taking the advantage that the blocked M3C does not exchange power with the low-frequency power grid when the capacitor is at rated voltage,the problems of over-voltage and over-current that tend to occur during LFTS startup can be solved by flexible control of M3C and reasonable design of startup strategy,without additional control devices and detecting the synchronization of the breaker. Finally,a real time digital simulation system (RTDS) is built to verify the proposed startup method. The simulation results show that converter stations connect to the low-frequency transmission line without voltage or current impulse. Also,the proposed strategy is correct and has a great prospect of the engineering application.

Abstract:The fault transient analysis of the modular multilevel converter based high voltage direct-current (MMC-HVDC) system is an important basis for the refined design of protection strategies and the selection of equipment parameters. To truly reflect the current traveling wave process on the overhead line after the fault,it is necessary to introduce a distribution parameter model of the transmission line in the analysis. Firstly,the MMC in the fault stage is represented equivalently. Then,the equivalent circuit of the pole-to-pole fault of the DC system is established in the complex frequency domain. The fault current is solved by the nodal analysis method,and the analysis results obtained in the complex frequency domain are converted into the time domain responses by the numerical inverse Laplace transform (NILT). Finally,the NILT-based calculation results are compared with the electromagnetic transient simulation results for verification. The results show that the proposed method has high accuracy and can effectively reflect the propagation characteristics of fault traveling waves,which is of positive significance for the protection design of MMC-HVDC systems.

Abstract:When the onshore accurate current (AC) grid connected to the offshore wind power voltage source converter-based high voltage direct current (VSC-HVDC) transmission system fails,the surplus power continuously output by the wind farm causes DC overvoltage. The most effective way to solve the surplus power is to use the energy dissipation device to discharge the surplus power. A centralized DC energy dissipation device scheme based on half-bridge sub-modules is proposed to solve the problems of large switching power shock when power devices directly connected in series,difficulty in switching devices at the same time and high cost of distributed DC energy dissipation device. The parameter design method and control strategy of the proposed topology are studied. The solution can overcome the technical difficulties and risks of direct series connection of large-scale power devices,and can gradually reduce the resistance voltage change rate by switching sub-modules,thereby reducing the impact of power discharge on the DC system and the difficulty of manufacturing unloading resistors. Compared with the distributed DC energy dissipation device,the proposed scheme only uses the centralized resistance and omits the bleeder branch in the sub-modules,which greatly reduces the equipment cost. The fault ride-through simulation of the offshore wind power VSC-HVDC transmission system is carried out on the PSCAD platform. And the results show that the proposed centralized DC energy dissipation device topology and control scheme have good performance.

Abstract:Combined heat and power (CHP) system has the outstanding advantages of environmental protection,economy and flexible operation mode,and has a better development prospect. Based on the different characteristics of electricity and heat in transmission and storage,a two-stage optimal scheduling method for CHP system considering thermal inertia is proposed. In the first stage,considering the structure and operation characteristics of the heating network,the optimal scheduling model of the CHP system based on model predictive control is established,and the strategy of intra-day controllable equipment output and grid interactive power is optimized. In the second stage,the goal is minimize the output adjustment of each unit in CHP system,and the real-time prediction error of renewable energy and load is taken into account to dynamically adjust the economic scheduling strategy in the first stage. The example shows that the two-stage optimal scheduling model can improve the economy of system operation and make up for the imbalance of supply and demand. Combined with the thermal inertia of the system,the building can store or release heat according to load demand and time-of-use electricity price,which can reduce the impact of renewable energy and load uncertainty on scheduling,smooth power fluctuations,and promote heat and power complementation.

Abstract:With the increasing installed capacity of renewable energy,the problem of wind and solar power abandonment in systems with high proportion of renewable energy is serious. It is of great significance to carry out the research on the problem of renewable energy consumption and accurately position the resistance of renewable energy consumption to improve the formulation of renewable energy consumption measures. In this paper,a refined evaluation method of renewable energy consumption resistance for power system is proposed. Firstly,based on the power flow tracking algorithm,the power flow of network nodes is divided into three parts,namely conventional unit injection,renewable energy injection and load outflow. Secondly,from the point of view of network nodes,the mathematical relationship models of peak modulation,frequency modulation,node voltage deviation constraint and renewable energy consumption are established,and the coupling relationship among the three is considered to analyze the power abandonment of network nodes. Then,the evaluation model of renewable energy consumption resistance based on network nodes is constructed and solved,the key nodes of renewable energy consumption in the system are located. The restriction factors of renewable energy consumption of each node are defined and the renewable energy consumption resistance is quantified. Finally,an example is designed based on improved IEEE 39-bus system to verify the effectiveness of the evaluation method. The proposed evaluation method can provide a basis for the power sector to formulate measures to improve the consumption of renewable energy.

Abstract:Analyzing the development trend of the scale of new energy vehicles is of great significance to the government regulation,the development direction of vehicle enterprises and the decision-making of the energy department. A new energy vehicle scale prediction method with nonlinear residuals is proposed in this paper. Firstly,the interval prediction method is used to study the uncertainty of economic policy. Secondly,considering the residual of scale prediction,the combined prediction model and support vector regression (SVR) model are used to predict the linear component and nonlinear residual respectively. Finally,the range of new energy vehicles scale in the future is obtained by taking the scale of national new energy vehicles as an example. By comparing different prediction methods,the effectiveness of the proposed method is verified,and the impact of different policy factors on the scale evolution of new energy vehicles is analyzed. The proposed method provides a corresponding reference for later charging facilities and other related planning.

Abstract:With the improvement of China's energy demand and environmental protection awareness,the installed capacity of green energy represented by intermittent renewable energy such as wind power and photovoltaics is visible in the upward trend. Large-scale intermittent renewable energy consumption is affected by many factors such as power supply,grid and load. In order to improve the development and utilization rate of renewable energy,appropriate consumption assessment methods should be used to analyze the consumption situation. The analytical method reflects the essence of the intermittent renewable energy consumption problem. The mathematical model of the analytical method is clarified firstly,and the method is assessed in combination with the application of the analytical method. Then,the principle and application of the time series production simulation method and the probabilistic production simulation method are sorted out. These two methods are based on the principle of the analytical method,and they are respectively proposed for the insufficient consideration of the time series characteristics and probabilistic characteristics of intermittent renewable energy consumption during modeling. Although the application of these two methods still has their own limitations,compared with the analytical method,the solution accuracy of the two methods has been greatly improved. Finally,combined with the existing technology,the future research on the assessment method of intermittent renewable energy consumption is prospected,which provides a reference for the modeling,application and improvement of the assessment method.

Abstract:The line fault of the modular multilevel converter based high voltage direct current (MMC-HVDC) grid seriously threatens the safe and stable operation of the power system,and the MMC-HVDC grid topology has a significant impact on the fault current. The main flow paths of the discharge current are analyzed when the single-pole grounding fault and the inter-pole short-circuit fault occur in the MMC-HVDC grid with different wiring methods. Then,based on the frequency domain analysis method,it is concluded that the fault current is mainly affected by the topological equivalent inductance parameters,and a simplified index k is proposed to evaluate the topology range and the number of converter stations that significantly affect the fault current. When the single-pole grounding fault and the inter-pole short-circuit fault occur in the power grid,the discharge degree of each converter station in the topology depends on the index k. The larger the k,the greater the discharge degree of the converter station. The smaller the k,the smaller the discharge degree of the converter station. Finally,the five-terminal pseudo-bipolar and five-terminal true bipolar systems are built on the PSCAD simulation platform for verification. The results show that the proposed high-frequency analysis method and simplified index of fault current are correct and effective.

Abstract:To explore the capacity value of a concentrating solar power (CSP) station,a method to optimize the thermal collection area and thermal storage capacity of a CSP station based on the credible capacity and levelized cost of energy (LCOE) is proposed in this paper. Firstly,generating efficiency model and economical model of CSP station are established in this paper. Secondly,the generation reliability of power generation system is calculated based on sequential Monte-Carlo method,and the particle swarm optimization (PSO) is utilized to search the credible capacity of the CSP station. Tthen the influences of solar multiple and thermal storage time on capacity credibility and LCOE of CSP station are studied. With credible capacity and levelized cost of energy as the optimization goal,weighted ideal point solution and entropy weight method are utilized to create a single objective optimization function and to determine the index weight respectively. Taking a CSP planning in the northwest as an example,using the region's real solar irradiation data,a model is established. The results show that the capacity credibility of CSP station increases monotonically with the increase of solar multiple and thermal storage time. At the same time,the levelized cost of energy decreases at the first stage and then increases at the second stage. The optimal solar multiple and thermal storage time is obtained under the constraints.

Abstract:Power load forecasting is one of the basic tasks power system research,and time series analysis is currently the most widely used forecasting method. Aiming at the fluctuation and the characteristics of peak and thick tail of user daily load time series,the generalized autoregressive conditional heteroskedasticity-in-mean (GARCH-M) family model is proposed to predict user load. Firstly,the autoregressive conditional heteroskedasticity (ARCH) effect of load series is examined by using the Lagrange multiplier (LM) test according to the distribution of user daily load time series. Secondly,under three different distributions of Gaussian distribution,t-distribution and generalized error distribution (GED),the GARCH-M family model is established according to the different forms of fluctuation compensation terms. Finally,combined with the loss function,the prediction analysis results show that the GARCH-M family model with different distributions improves the accuracy of short-term user load prediction compared with the traditional time series analysis model.

Abstract:The large-scale grid connection of distributed generation (DG) not only brings the problem of consumption,but also poses great challenges to the economic and safe operation of the AC/DC network. Based on this,an optimal operation method for AC/DC hybrid distribution network based on DG location and coordinated control of multi-terminal voltage source converter (VSC) is proposed. For DG location,a node loss sensitivity index is proposed through the sensitivity analysis method. The DG location on the AC network side is carried out by using the law that load nodes at different locations have different sensitivities to network loss.Then,a multi-objective optimization model aiming at minimizing network active total loss,node voltage deviation and DG surplus is established to coordinate the active power output of DG and the port power and voltage variables under different control modes of multi-terminal VSC. The simulation results show that the proposed optimal operation method can improve the economy and security of network operation while taking into account the DG consumption level of distribution network,and provides an important reference for decision-makers in practical engineering.

Abstract:The direct current (DC) load switch is the key equipment for constructing the DC distribution network. Aiming at the shortcomings of the existing DC load switch,such as much additional equipment,large size and complex control,an assembly DC load switch topology based on pre-charged capacitors is proposed. The working principle,control strategy and parameter design of the proposed topology are analyzed. The load switch utilizes pre-charged capacitors to form a source side part and a load side part. The load switches on the outgoing lines of multiple DC buses share the source side part,which reduces the size and cost of the equipment. In the process of load shedding,through the cooperation of source and load side capacitors,a low voltage and zero current breaking condition is provided for the mechanical switch. Finally,a simulation model is built in Matlab/Simulink to verify the feasibility and effectiveness of the proposed topology. The results show that the proposed DC load switch can reliably turn off the load current without arc which ensures the safe operation of DC distribution network. The cost of equipment required is low,which improves the economic efficiency of DC distribution network.

Abstract:In traditional microgrid with droop control,the bus frequency decreases with the increase of load,and a corresponding frequency regulation control strategy is taken to keep frequency around the rated value. Control strategies based on proportional integral (PI) are commonly used to recover the bus frequency at present. Because of the possibility of changes in structure and system parameters of microgrid,the quick response and quick restoration of frequency may not be fulfilled with PI control. To solve this problem,the single neuron adaptive PI control method is used as frequency recovery algorithm in microgrid central controller in the second control layer to realize non-error control. In order to accelerate frequency recovery and improve the robustness of microgrid,the proportionality coefficient of the single neuron PI controller is optimized online by fuzzy controller. By comparing this method with original single neuron adaptive PI control with fixed neuron proportionality coefficient in simulation. It is proved that this modified method can improve transient performance of frequency recovery control and accelerate frequency recovery of microgrid.

Abstract:Microgrid (MG) can coordinate the dispatch of distributed generation,energy storage devices and loads to maximize the use of renewable energy. In this paper,the energy trading method among MGs is studied. Aiming at the dynamic stochastic factors of MG,a multi-time-scale method is adopted,including daily scheduling and intraday optimization,and the energy management framework of MG is proposed. Then,the energy management model of MG is established,including the daily scheduling model and the intraday optimization model,and the objective functions and constraint conditions are determined. Particle swarm algorithm is used to achieve the lowest economic cost and improve the economic performance of MGs. The proposed dynamic stochastic model,energy trading method and energy management model are verified and analyzed by taking three MGs as examples. The simulation results show that the proposed energy trading method can effectively improve the economy and reliability of the multi-microgrids.

Abstract:The damping capacitor is core component of converter valve. Effectively evaluating life of damping capacitor is of great significance for the evaluation of the converter valve state and the stady of the localization of the damping capacitor. First of all,combined with the application principle of the damping capacitor of the converter valve,a multi-frequency point equivalent temperature rise test method is proposed. By equivalently calculating the loss of damping capacitor during actual operation and flexibly selecting the frequency and current for the test,the internal temperature and thermal resistance of damping capacitor during actual operation can be accurately evaluated. Then,considering the actual operating environment and working temperature of the damping capacitor of the converter valve,the life-influencing factors are analyzed.A life prediction method combining aging test and curve fitting is proposed based on the classical life prediction model. The proposed method can shorten time and cost of the test. Finally,the internal temperature rise and thermal resistance of the capacitor during operation are analyzed in combination with the actual engineering parameters,and the expected life of the damping capacitor sample is further evaluated by the life prediction test. The obtained results meet the requirements of ultra high voltage direct current projects.

Abstract:The gas insulated switch (GIS) vibrates during operation. In order to analyze the difference between the vibration signals for loosed bolts or misaligned conducting rods at the connection of GIS shell and the normal operation without defects,a 3D model of 252 kV three-phase box-type GIS is established in this paper. Computational modal analysis is firstly performed on the model in the field of solid mechanics. Secondly,the natural frequency is determined,and the representative sixth mode shapes are obtained. By applying different loads and changing the parameters of the model,the time-domain signals of the vibration acceleration of GIS shell under the state of no fault,different number of bolts loose and different cases of conducting rod misalignment are solved in the solid mechanics field. Then the frequency-domain signals are obtained through spectrum analysis. By comparing and analyzing the difference of vibration acceleration signals of GIS shell in several cases,the characteristics of vibration acceleration of GIS shell in case of bolt loosening fault and conducting rod misalignment fault are obtained. Finally,the criteria for the above faults are obtained. Some certain reference value about the follow-up GIS equipment running condition monitoring,mechanical fault identification and vibration research have been provided in the conclusion of this paper.

Abstract:An accurate estimation model of winding hot spot temperature is the key to assess the thermal state and insulation life of oil-immersed transformers. Based on the hot spot temperature and load current monitored by the substation,the genetic programming algorithm is applied to train the basic structure of the hot spot temperature estimation model. Then,the parameter identification of the hot spot temperature estimation model is performed by the normalized least square mean (NLMS) algorithm. Finally,an explicit prediction model of the hot spot temperature is established for oil-immersed transformers. The explicit winding hot spot temperature estimation model can effectively reflect the relationship between the load factor and the winding hot spot temperature. Moreover,the goodness of fit of the model under the test set is 0.998 8,and the maximum absolute error is only 1.36 ℃,which verify the correctness and effectiveness of the model. Furthermore,the strong generalization performance of the proposed model is proved by estimating the winding hot spot temperature for oil-immersed transformers with the same capacity and model in the same area.

Abstract:The partial discharge fault identification method based on the fusion of multi-dimensional information sources can greatly improve the accuracy and fault tolerance in the fault identification of power equipment. In this paper,four typical partial discharge models,namely corona discharge,suspended discharge,floating discharge and air-gap discharge are prepared. The partial discharge signals generated by different discharge models are collected by ultrasonic (Ultra),very-ultra high frequency (V-UHF) and pulse current method (PCM) sensors. Firstly,the deep convolutional neural network (CNN) algorithm is used to train the measurement data of different sensors,and then the Dempster-Shafer (D-S) evidence theory is used to perform fusion calculation on the recognition results of multi-dimensional information sources. Finally,according to the fusion calculation results,the identification conclusion is made. The results show that the fault identification model based on multi-dimensional information sources constructed in this paper has higher accuracy than that based on single information source. When a misjudgment occurs in one of the multi-dimensional information sources,the model can still correctly identify the type of discharge,which indicates that the model has better fault tolerance for the information sources and the recognition effect is good.

Abstract:The current information exchange interface and human-computer interaction interface of secondary equipment operation and maintenance in substation are not unified,which seriously restricts the efficient operation and maintenance of substation. In view of this situation,the research on the efficient operation and maintenance technology of secondary equipment in substation based on power system general service protocol (GSP) is carried out. Firstly,the architecture of integrated operation and maintenance of secondary equipment in substation is designed,and the information modeling method of integrated operation and maintenance of secondary equipment in substation based on substation operation and maintenance configuration language (SOMCL) is proposed. Then,the service interface and information interaction mechanism of integrated operation and maintenance of secondary equipment in substation are designed based on object-oriented technology (OOT),which achieves the goal of plug and play information interaction between operation and maintenance tools and the object to be transported. Finally,the integrated operation and maintenance tool of secondary equipment in substation is developed to realize the standardization and visualization of operation and maintenance of intelligent secondary equipment in substation,and achieves the goal of integrated and efficient operation and maintenance of secondary equipment in substation. The proposed method has been piloted in several new smart substations,which verifies the feasibility of the integrated operation and maintenance technology of secondary equipment in substation.

Abstract:As the voltage level is relatively low under medium and low voltage distribution network,the number of modular multilevel converter (MMC) sub-modules is also small. Therefore,the modulation mode of MMC with small quantities of sub-modules has an important influence on the performance of the MMC system. In order to improve the output quality and output capability of the MMC with small quantities of sub-modules,a hybrid modulation strategy suitable for MMC with small quantities of sub-modules based on level step switching at a high modulation degree is proposed. On the basis of the nearest level modulation (NLM),the MMC is controlled to switch between NLM and the carrier phase shift pulse width modulation (CPS-PWM) at the output level step point in real time. At the same time,the circulation control and the sub-module voltage equalization control are combined to further ensure the normal operation of MMC with small quantities of sub-modules under the hybrid modulation. Then,the simulation model of MMC hybrid modulation with four sub-modules is established,and a simulation study is carried out on the MMC hybrid modulation strategy of the MMC with small quantities of sub-modules. The results show that the proposed hybrid modulation strategy not only takes into account the characteristics of low output harmonics and low switching loss,but also improves its output quality. And its operation in high-profile system increases the DC voltage utilization rate and improves the output capacity of the MMC with small quantities of sub-modules.

Abstract:After occurrence of natural disasters,active distribution network (ADN) can promptly restore power supply to some critical loads through tie-line switching and flexible distributed generation (DG),and thus the fault risk is effectively mitigated. A data-driven multi-dimensional intelligent forecast approach for the fault risk levels in ADNs is proposed in this paper. Firstly,a feature selection method based on Chi-square test (χ²) and Pearson correlation coeffects is developed to analyze the strength of fault correlation factors from multiple dimensions and the optimal set of fault features is obtained. Then,an optimal network reconfiguration model is established for the damaged ADNs considering DG integration,and consequently the heterogeneity of the line importance can be taken into account which provides a solid foundation for the classification of fault risks. Furthermore,an intelligent forecast model for ADN fault risk levels is established based on extreme gradient boostig (XGBoost) algorithm. Finally,the numerical tests on IEEE RBTS Bus6 distribution network demonstrate that the proposed approach achieves a predication accuracy 3.17% higher than back propagation (BP) neural network does. The proposed approach has good generalization capability,thus providing an important basis for the fault risk management in ADNs to effectively reduce the fault loss.

Abstract:Permanent magnet synchronous motor is widely used in industrial sites. In some cases,it is started softly by variable frequency converter and then switched to operate normally with power frequency. In order to analyze such motor operating characteristic with voltage sag,a method for evaluating the low voltage ride through capability of permanent magnet synchronous motors is presented in this article. Based on the dq coordinate system,the analytical formula among electromagnetic torque,power angle and terminal voltage is established. With the relationship between slip and motor stability in the transient process,the slip based motor transient stability analysis criterion is established. It can be used to determine the safety and stability critical voltage of the permanent magnet synchronous motor. The analytical formula of critical cutting power angle is deduced by area law. Then,the critical cut-off time of fault can be got. Critical voltage and fault removal time is selected as the characteristic value to evaluate the motor's low voltage ride through capability. The transient stability can be judged by the negative slip value in the transient process. The smaller the motor load rate,the stronger the low voltage ride through capability. Simulation results verify the effectiveness of the evaluation method.

Abstract:There is abundant information in power equipment nameplates. Extracting information from nameplates through image and text recognition technology enables the high effective and quick works on statics and account check of power equipment,which is also beneficial to improve the equipment management level of power system. Considering the great difference of text recognition between power equipment nameplates and ordinary images,an algorithm of automatic recognition of power equipment nameplates based on deep learning is proposed in this paper. This algorithm consists of three parts,namely nameplate detection,text detection and text recognition. By improving the design of loss functions,adding the correction of text recognition,and synthesizing text images,the mean average precision of the nameplate detection model on the test set reaches 92.2%,the F₁ of the text detection model on the test set reaches 91.2%,and the character recognition accuracy rate of the text recognition model reaches 94.0%,the text line recognition accuracy rate of the text recognition model reaches 82.3%.

Abstract:In order to increase the transmission capacity of the new energy grid-connected transformer,a two-voltage three-winding transformer with star-angle-angle connection form is adopted. The sensitivity of the differential protection is insufficient when an inter-turn fault occurs on the low-voltage side of the transformer. At the same time,the transformer inrush current blocking criterion blocks the differential protection and extends the fault removal time when the transformer is closed at no-load during the inter-turn fault. Based on the zero-sequence equivalent loop of the new energy grid-connected transformer,the inter-turn fault identification criterion is proposed by using the characteristics of the out-of-area fault and the inter-turn fault. The proposed criterion which is not affected by the magnetizing inrush current has higher reliability and quicker action than the principle of differential protection does. The electromagnetic transient real time simulation system (RTDS) is used to perform fault simulation,which verifies the validity and reliability of the new criterion. Therefore,the new criterion can be used to identify transformer inter-turn faults and make up for the deficiency of differential protection.

Abstract:Decomposing the output signal of fiber optical current transformer (FOCT) with time-frequency conversion is the key step to obtain the gradual fault characteristics. Aiming at the characteristics of FOCT gradual fault signal with large time domain span and random deterioration process,the output signal is sampled across intervals,and the wavelet packet decomposition algorithm is used to extract fault signal features according to the frequency band of the fault signal. The characteristic parameters are screened to obtain the optimal characteristic parameters that characterize the FOCT degradation trend.The principal component analysis method is proposed to reduce the dimensionality of high-dimensional features,the problem of high signal feature dimensionality is solved,and the need for fast fault feature identification is met. Experiment results show that the wavelet algorithm can decompose the signal into various frequency bands and obtain 64 sub-sequences containing signals of different frequency bands. The operating status of the transformer is determined by the energy ratio of each frequency band of the wavelet signal,and the gradual fault identification is realized.

Abstract:Polypropylene (PP) is an environmentally friendly cable insulation material with relatively great insulation properties and has the advantages of being recyclable as a thermoplastic material. PP is easy to brittle fracture due to its poor toughness at low temperatures. Hence,it is necessary to improve the toughness of PP. In this paper,three modified PP samples are studied,namely PP blended with 10% mass fraction of polyolefin elastomer(POE),PP blended with 10% mass fraction of styrene ethylene butylene styrene (SEBS),and copolymerization modified PP specimen synthesized with PP as matrix and polyethylene (PE) in the reactor. The PP samples before and after modification are tested to obtain their physical properties,chemical properties,mechanical properties and electrical properties. It is concluded that the toughness,tensile strength and elongation at break of PP increase after the modification,while the melting temperature decreases and the dielectric loss increases. For the modification methods,the initial melting temperature and AC breakdown field strength at 20-80 ℃ of the copolymerization modified PP specimen are higher than those of the blending modified PP specimens. Instead,the blending modified PP specimens have better mechanical properties and lower dielectric loss at high temperature than the copolymerization modified PP specimen does. After comprehensive consideration,the copolymerization modification may be more suitable to enhance the toughness of PP than the blending modification.

Abstract:The access of ultra high voltage (UHV) gas-insulated transmission line (GIL) affects both the transient recovery voltage (TRV) amplitude and the rate of rise of TRV (RRRV) of circuit breakers,which may endanger the breaking capacity of circuit breakers. Based on the double-circuit UHV accurate current (AC) overhead line on the same tower and the double-circuit GIL hybrid transmission line,EMTP-ATP simulation is used to study the influence of GIL length,access position and GIL lead station measurement equipment on the TRV amplitude and RRRV of the circuit breaker when a three-phase short-circuit fault occurs at the end of the circuit breaker in the substation. The simulation results show that the access of the GIL has a greater impact on the TRV amplitude and RRRV of the circuit breaker. The increase in the length of the GIL has a greater impact on the TRV amplitude,but the impact on RRRV is small. In order to limit the RRRV,the GIL should be installed in the middle of the overhead line,and the metal oxide varistor (MOV) should be installed to limit the TRV amplitude. The measurement equipment of the GIL terminal station has little effect on the TRV amplitude and RRRV of the circuit breaker. The research results can provide a reference for the TRV calculation and parameter design of circuit breakers in UHV lines with GIL.