Abstract:Photovoltaic field connected to the grid via voltage source converter based high voltage direct current (VSC-HVDC ) transmission is an important trend in the future development of the power system,but there is a risk that the interaction between photovoltaic and VSC-HVDC system may trigger system oscillations and instabilities. So,the open-loop and closed-loop interconnection models of the photovoltaic field integrated into grid through VSC-HVDC are established. The dynamic interaction between the photovoltaic subsystem and the VSC-HVDC subsystem based on the open-loop mode resonance theory are analyzed. When a strong interaction between two subsystems occurs,it may cause the corresponding closed-loop oscillation mode to enter the right side of the complex plane and trigger system oscillation instability. The risk of interaction instability between the photovoltaic subsystem and the VSC-HVDC subsystem can be mitigated by adjusting the control parameters of the dominant link. If the control parameters can not be adjusted,an additional damping controller for the photovoltaic unit is proposed as an inhibitory measure,which destroys the mode resonance phenomenon by adjusting the open-loop oscillatory modes of the photovoltaic subsystem away from the open-loop oscillatory modes of the VSC-HVDC subsystem in the complex plane,so as to stabilise the closed-loop system. The correctness of the above theoretical analysis results and the superiority of the proposed damping controller are verified by the simulation examples.

Abstract:DC side dynamics are ignored in most of the research on grid-connected converter,which is dealt as a constant voltage source. In this case,the analysis result of small-signal stability is affected to some extent. In this paper,the impedance model of grid-following converter is reconstructed,then followed by stability analysis of the system considering DC side dynamics. Firstly,the applicable conditions are analyzed,that is to say,DC side is equivalent whether to voltage source or controlled current source. It is demonstrated that the DC side dynamics must be considered in case of small-signal stability analysis. Then,the impedance model of the grid-following converter is established based on harmonic state-space (HSS). Secondly,the stability of the system with different DC side structures is analyzed by Bode criterion under different grid strengths. Subsequently,it is revealed that the influence mechanism of grid strength on the stability of grid-following converter. In addition,the influence of different links on impedance characteristics of the system is analyzed,including phase-locked loop,current loop and filter. Finally,the theoretical analysis and electromagnetic transient simulation results show that under the condition of weak grid,the interaction between phase-locked loop and grid is strengthened,which reduces the small-signal stability of the system. Moreover,the critical short-circuit ratio of the system considering the DC side dynamics is larger.

Abstract:The current reliability evaluation method of the security and stability control system (SSCS) is primarily based on static modeling,which fails to capture the dynamic aging and maintenance processes of individual devices within the system,thereby impacting the accuracy of evaluation results to a certain extent. Therefore,a dynamic reliability evaluation method of the SSCS based on Markov chain Monte Carlo (MCMC) is proposed in this paper. Firstly,the four-state non-homogeneous Markov model is constructed to simulate the aging process of the device,and the evaluation methods of each state are given. Secondly,according to the repair process,the influence of different maintenance strategies on the state transition of the device is analyzed to reflect the difference of state maintenance. Finally,the reliability of the SSCS is modeled dynamically by considering the temporal or conditional correlation of the state transition process of the security and stability control device. Taking the actual SSCS as an example,the reliability under different maintenance strategies is compared by simulation,and the sensitivity of model parameters is analyzed. The evaluation results show that this method can solve the time-varying availability of the SSCS,and can be used to guide the reasonable maintenance of the security and stability control device on site.

Abstract:Under inverter operating conditions,significant loss is incurred by the insulated gate bipolar transistor (IGBT) (referred to as T₂ tube) in the lower part of the half-bridge submodule. The reduction of loss is beneficial for the improvement of equipment operation reliability. At the same time,the suppression of capacitor ripple voltage has the advantage for reducing capacitor demand and enhancing power density. However,attention is not given by existing optimization control strategies to the contradiction between loss distribution optimization and capacitor ripple voltage,making it difficult to balance equipment operation reliability and power density. Therefore,a comprehensive optimization method that combines the reduction of T₂ transistor losses and the suppression of capacitor voltage ripple is proposed in this article. Firstly,the inherent contradiction between reducing the on-state loss of the T₂ transistor and suppressing capacitor voltage ripple is explained by analyzing the impact path of charge on device loss and capacitor ripple voltage. Then,by introducing a penalty function,a comprehensive objective function is established that takes into account T₂ transistor losses and capacitor voltage ripple. Subsequently,using the active bypass strategy as an example,a comprehensive optimization method based on the injection of second harmonic current and third harmonic voltage is proposed by analyzing the impact of the second harmonic current and third harmonic voltage injection on T₂ transistor loss and capacitor voltage ripple. Finally,a simulation model is built in MATLAB/Simulink and PLECS for verification. The simulation results suggest that the reliability and power density of the device increase by the comprehensive optimization method,considering both T₂ transistor losses and capacitor voltage ripple.

Abstract:Virtual synchronous generator (VSG) enhances the stability of distributed power sources during grid connection by simulating the operating principle of a synchronous generator and introducing the virtual inertia and damping coefficient. However the traditional VSG control strategy struggles to address abrupt changes in grid voltage phase and amplitude when the voltage drops in the grid. Therefore,a low voltage ride-through control method for VSG based on phase and amplitude compensation is proposed. Firstly,the different effects of voltage drops and recovery on the power grid are analyzed. Secondly,during voltage drops,the purposes of overcurrent suppression,rapid stabilization of output power,and reactive power compensation are achieved by controlling the phase and amplitude difference between VSG output voltage and grid voltage within the allowable range. Then,during voltage recovery,rapid compensation eliminates the phase difference and amplitude difference between VSG output voltage and grid voltage to suppress overcurrent and other problems caused by grid voltage jumps. Finally,the effectiveness of the proposed control strategy is verified by MATLAB/Simulink simulation. The simulation results show that it can effectively suppress the overcurrent and realize reactive power compensation.

Abstract:A coordinated control strategy for a variable-coefficient wind power full DC system to address problems such as reduced system inertia and insufficient frequency regulation abilities in the AC grid caused by the integration of wind power full DC systems is proposed. In terms of inertia response,the grid-side converter station adopts inertial synchronization control,while the DC-boost converter station adopts constant variable-ratio control to provide inertia support to the grid through DC capacitor and perceive the AC system frequency through the DC voltage on the low-voltage side. Additionally,a virtual inertia control scheme is introduced by adding a variable virtual inertia coefficient to the DC wind turbine (DCWT),enabling the wind power full DC system to have different equivalent inertia in different frequency response stages. As for frequency regulation,the DCWT operates in a load-reducing mode combining overspeed and variable-pitch control,adjusting its active power output through variable droop control. This approach fully utilizes the available capacity at different wind speeds,allowing the wind power full DC system to participate more effectively in frequency regulation. Simulation results demonstrate that this strategy improves the inertial response and primary frequency modulation of the power system after the integration of the wind-powered full DC system.

Abstract:Based on the new energy power prediction data with the highest probability of occurrence to check the security and stability of the deterministic dispatching plan,the conclusion is quite different from the actual situation of the power grid. In order to improve the adaptability of the results,a security and stability checking method of dispatching plan considering uncertainty of new energy is proposed. Based on the quantitative assessment results of the safety and stability of traditional deterministic dispatching plan,a safety and stability margin minimization model with confidence interval constraints on the predicted power of new energy stations is constructed to realize effective identification of high-risk planning modes. Based on the active power sensitivity,the equivalence aggregation and uncertainty variable dimensionality reduction of new energy stations are carried out to improve the efficiency of the security and stability margin minimization model. Based on the parallel processing platform,the multi-mode grouping parallel security check of the dispatching plan is carried out. The results of security and stability check considering the uncertainty of new energy power prediction are obtained. The proposed method can effectively improve the accuracy and timeliness of the security and stability check of the high-proportion new energy grid dispatching plan. The effectiveness of the proposed method is verified by an actual grid case.

Abstract:The battery system is the critical component to supporting the next generation of advanced power grids. However,parasitic lithium plating reactions can be triggered by improper charging and discharging strategies,leading to a significant compromise of the charge-discharge performance of batteries. Centered on ternary lithium-ion batteries,the degradation of charge performance due to lithium plating is elucidated through the utilization of a reference electrode-based approach in this study,followed by the implementation of measures to regulate safe charging currents. Diverse temperature-dependent charge-discharge cycling experiments are initially designed to evaluate batteries under both low-temperature and high-temperature cycling. Subsequently,the calibration of safe charging curves is conducted using a reference electrode,and the negative electrode potential of batteries undergoing high-temperature cycling is analyzed. The occurrence of lithium plating in batteries subjected to high-temperature cycling is identified,resulting in an average charging current reduction of 61.7% compared to pristine cells. Furthermore,a comprehensive charge state-temperature-current contour map is established for batteries with lithium plating. A reduction of 69.84% in the charging current region above 200 A is demonstrated through comparative analysis with the contour map of pristine batteries. A quantitative metric for assessing the degradation of charge performance in batteries with lithium plating is provided by this study,underscoring the necessity of considering these factors in the comprehensive lifecycle management of lithium-ion batteries.

Abstract:Within the context of carbon peaking and neutrality,carbon capture plants can effectively reduce carbon emissions in power systems. Yet,the regular integration of these units for peak shaving in grids harnessing renewable energy tends to hamper system efficiency. Thus,acknowledging the operation and carbon trading mechanism of adaptable carbon capture plants,firstly,the incorporation of pumped storage plants is suggested to aid peak shaving,thereby facilitating wind power utilization,optimizing carbon capture plants for carbon sequestration,and reducing overall system carbon emissions. Then,given the uncertainty following grid-integrated wind power,fuzzy theory is employed to model system power constraints as fuzzy parameters,representing wind power and load. The constraint is transformed into a credibility-based fuzzy chance constraint. This fuzzy chance constraint is clarified by the clear equivalence classes. Prioritizing maximum system net income as the objective function,elements such as unit online revenue, extraction income, carbon trading profits,operational costs,and system security constraints are considered. The carbon capture-pumped storage joint operation model is developed. Finally,the model is solved using CPLEX. Simulation results validate that the integration of pumped storage plant enhances the system's net income by 7.62% and reduces carbon emissions by 7.01%,reflecting a balanced consideration of both economic and environmental aspects.

Abstract:In order to reduce carbon emissions and the impact of source-load uncertainty on system operation,a multi-timescale low-carbon optimization scheduling strategy in day-ahead,intra-day and real-time operations for energy hub (EH) based on a reward-punishment ladder carbon price mechanism and distributed model predictive control (DMPC) is proposed. A reward-punishment ladder carbon price calculation method is introduced and a day-ahead low-carbon optimization scheduling model for EH is constructed. A feedback closed-loop optimization strategy based on DMPC for intra-day rolling and real-time adjustments is formulated. The optimization strategy reduces source-load prediction errors and improves the efficiency of traditional model predictive control (MPC) solving. In the intra-day stage,a rolling optimization model with the objective of minimizing the sum of the ladder carbon price cost,operational cost,and penalty cost for energy storage adjustment is constructed. In the real-time stage,the overall optimization problem is decomposed,and a multi-agent real-time adjustment model based on DMPC is established. The simulation results indicate that the proposed strategy is effective in enhancing the economic efficiency of the system,reducing the uncertainty of source and load,and achieving the low-carbon,economic,stable,and reliable operation for EH.

Abstract:With the development of the smart grid,power quality issues have been widespread in the power grid and it threaten the safety and stability of the power grid. The monitoring data of power quality disturbances (PQDs) increase rapidly,and it is of great significance to achieve deep feature extraction and intelligent classification of PQDs in large-scale systems for power system pollution detection and management. To this end,stacked sparse auto encoder (SSAE) and feedforward neural network (FFNN) based method for composite PQDs classification is proposed in this paper. Firstly,a PQDs simulation model is constructed based on IEEE standard. Then,a PQDs classification model based on SSAE-FFNN is established,and the scaled conjugate gradient (SCG) algorithm is used to optimize the model,in order to accelerate gradient descent and improve training efficiency. Next,to reduce the reconstruction loss of the stacked network and extract deep low-dimensional features,the layer-wise training and fine-tuning strategy of SSAE are constructed. Finally,the examples are used to verify the classification effect,robustness,generalization and applicable scenario scale of the proposed method. The results show that the method can effectively identify composite PQDs and it has a high accuracy even for both error-containing disturbances and 21 sets of measured disturbance data of a local municipal grid.

Abstract:With the widespread integration of distributed energy sources into low-voltage distribution networks,the demands for operational flexibility and absorptive capacity of distribution grids have been continuously increasing. The use of low-voltage flexible interconnection devices to interconnect independently operated low-voltage distribution substations helps to avoid frequent operations of traditional voltage regulation and reactive power compensation devices. Considering the high cost of flexible interconnection devices,a method is proposed to plan the siting and capacity of low-voltage flexible interconnection devices in coordination with traditional voltage-reactive power regulation devices. Firstly,the topology and operational mode of the low-voltage flexible interconnection devices are analyzed,and their power flow model is established. Subsequently,a dual-layer planning model is formulated for optimizing the configuration of low-voltage flexible interconnection devices. The upper-layer planning aims to minimize the annual comprehensive cost,while the lower-layer planning takes into account a time-series model for voltage-reactive power coordination control. The objectives of the lower-layer planning are to minimize operation costs and voltage deviations. The optimal solution for the distribution network system's flexible interconnection scheme and operational mode is obtained through alternating iterations of particle swarm optimization and mixed-integer second-order cone programming algorithms. Finally,a case study is conducted on the IEEE 33-node test system to validate the effectiveness of the proposed dual-layer planning algorithm. The results indicate that the proposed method can effectively reduce the excessive deployment of flexible interconnection devices while simultaneously decreasing the operational costs caused by frequent fluctuations in distributed energy sources. The method of convexifying and linearizing the model significantly enhances the solution efficiency.

Abstract:Aiming at the problem that traditional fixed-weight multi-objective reactive power optimization is unable to make the most suitable control decisions for real-time working conditions when dealing with the complex and changing working conditions of new power systems,an adaptive multi-objective reactive power optimization control strategy is proposed,which takes the weighted minimum of the deviation of the system active network loss and the voltage of the grid connection points as the objective function,and the weighting coefficients of the objective function are adaptively adjusted according to the deviation of the voltage of the grid connection points. The strategy takes the minimization of active network loss and the deviation of grid voltage as the objective function. Firstly,the relationship between voltage fluctuation at the grid-connected points of offshore wind farms and the active and reactive power outputs is analyzed to establish the corresponding reactive power allocation model,and the corresponding reactive power control model is established with respect to the input and output characteristics of the wind turbine and the static var generator (SVG). In addition,considering the power constraints and safe operation constraints of offshore operation,the variable inertia weight particle swarm optimization algorithm is used to solve the reactive power control strategy. Finally,the offshore wind farm model is built in MATLAB for simulation verification,and the simulation example shows that,compared with the traditional fixed-weight multi-objective reactive power optimization,the adaptive multi-objective reactive power optimization control strategy can quickly adjust the priority of each optimization objective according to the real-time working conditions of the grid,which can achieve the coordinated optimization of the active network loss and grid-connected point voltage.

Abstract:Aiming at the randomness and uncertainty in the spatio-temporal distribution prediction of electric vehicle charging load,a method for electric vehicle load prediction that integrates travel chain theory and actual geographic information is proposed. On the basis of road network integration and travel chain theory,a model for the spatio-temporal characteristics of electric vehicle charging demand is established to simulate the user's travel behavior characteristics. At the same time,by modeling the road network in the target area,dividing it by functional area,combining the user behavior characteristics of travel chain theory with target geographic information,and planning and designing the travel path of electric vehicle users through Floyd algorithm,the electric vehicle charging demand load can be predicted. The results of the case study show that the proposed model can predict the variation of electric vehicle charging load based on actual geographic information,and analyze the charging demand and load characteristics of electric vehicles in different functional areas and different administrative regions. The simulation results validate the effectiveness of the proposed model and method.

Abstract:The variation of electric vehicle (EV) charging load is constrained by the climbing performance of microgrids. Therefore,this paper considers the climbing characteristics of microgrid units and proposes a two-layer optimization strategy for EVs participating in microgrid scheduling considering dynamic electricity prices. The upper layer is the EV load model. The fast/slow charging characteristics of different types of EVs are analyzed and the guidance of microgrid electricity price on EV charging demand is considered,thereby establishing the EV load model with the maximum user satisfaction as the target. The lower layer is a multi-microgrid operation model. The dynamic electricity price strategy is formulated according to the net load of the microgrid,and the dynamic electricity price of each region is optimized considering the consumption of new energy of the microgrid by EV charging and the demand for power climbing. The multi-microgrid regional operation model is established with the objective of minimizing the net load fluctuation and operating cost of the microgrid. Finally,an example analysis of the microgrid and EV charging demand in an urban area is conducted to verify the results. The results show that compared with the fixed electricity price and the peak and valley time-of-use price,the proposed method can realize the orderly charging of EV loads in the microgrid area and smooth the net load fluctuation. Also,the proposed method can effectively reduce the influence of charging behavior on the safe and economic operation of the microgrid.

Abstract:Cluster division can effectively solve the problem of massive data analysis and a large number of equipment regulation caused by large-scale access of new energy to the distribution network. However,existing research on cluster partitioning algorithms exhibits low accuracy and may yield unreasonable outcomes. In order to solve the above problems,factors that should be considered in the cluster division strategy of distribution network are described when a large number of distributed power sources are connected,and the scale limit index is designed accordingly. By studying the process of genetic algorithm,the reason why genetic algorithm shows poor global optimization ability is found out,and then the algorithm is enhanced. Simulation results demonstrate that the proposed scale limit index successfully avoids unreasonable partitioning outcomes. The proposed improved genetic algorithm greatly improves the accuracy of the algorithm. Because the genetic algorithm has no convergence criterion,the reduction of the number of iterations can not directly reduce the experiment time. In summary,the research effectively improve the accuracy of genetic algorithm and enhance the efficiency of cluster division in distributed network.

Abstract:With the intensification of polar scientific research,the demand for energy is increasing rapidly,and the environmental pollution and carbon emissions caused by polar energy consumption dominated by fuel cannot be underestimated. It is urgent to build a microgrid system based on clean and renewable energy to ensure the green and sustainable development of polar scientific research. Based on the analysis of the resource characteristics of the Antarctic research station,the wind-solar microgrid structure is selected and the system capacity optimization configuration model is established in this paper. The polar environment is complex and diverse,which has a significant impact on the energy storage system. In order to ensure the future operation of the microgrid with better adaptability,the influence of energy storage capacity attenuation on the economic benefits of the microgrid system is considered in detail in the configuration stage,and the influence of temperature,charge-discharge rate and cycle times on the battery capacity is analyzed. A multi-factor coupling battery capacity attenuation model based on adaptive grey relational analysis is established,and it is incorporated into the optimal configuration model. An example of Antarctic scientific research station is solved and analyzed by using improved particle swarm optimization. The simulation results show that the microgrid configuration scheme considering energy storage capacity attenuation can reduce the probability of battery overcharge and overdischarge,extend the service life of the battery,and is more suitable for polar regions.

Abstract:In order to effectively improve the economy and feasibility of wind power grid access,an optimal configuration scheme of electric-thermal hybrid energy storage considering the probability of typical scenarios of wind power is proposed. Firstly,using scenario analysis and K-means clustering method,a large amount of wind power historical data is simplified into six typical output scenarios and the probability of each scenario is established. The number of clusters is determined by the elbow curve method and the Dunn index method. Secondly,a control strategy for electric-thermal hybrid energy storage system is proposed and a combined wind-storage system model applicable to multiple scenarios is established. Finally,a capacity configuration optimization model containing the integrated response of electric and thermal load with the objective of minimizing the economic cost and the amount of abandoned wind is established. The scenario probabilities are added to the objective function in the form of weights. The model is solved by particle swarm algorithm. Through simulation analysis and comparison with other energy storage configuration scenarios,it is verified that the proposed configuration strategy can improve wind power utilization by about 16.12% while reducing the overall system cost by about 43.76%.

Abstract:Given the frequent missing and false alarms of partial discharge insulation defects in on-site gas insulated substation (GIS),intermittent discharge characteristics tests are conducted in this article to study the discharge characteristics of common solid insulation defects inside GIS at different temperatures in the actual operating temperature range of GIS. In this paper,the platform of GIS electric thermal coupling intermittent discharge simulation test is built with the pulse current method,ultrahigh frequency (UHF) method,ultrasonic method and gas characteristic component detection method used in combination to obtain and analyze the intermittent discharge characteristic data of solid insulation defects at different temperatures. It is found out that both UHF detection method and pulse current detection method can effectively detect the intermittent discharge UHF signals of test defects at different temperatures,while ultrasonic and gas characteristic component detection methods are not applicable to effectively collect effective discharge data. The intermittent discharge voltage of metal pollution defects on the surface of solid insulation and internal air gap defects shows a negative correlation with temperature. The intermittent discharge voltage of pollution defects shows a significant trend of linear decrease,while the intermittent discharge voltage of air gap defects shows a significant decrease first and then a trend of slow-paced linear decrease. The average discharge capacity and UHF signal amplitude of intermittent discharge of pollution defects exhibit a positive correlation with the increase of temperature. The intermittency of pollution defect discharge increases with the extension of discharge time at different temperatures,while the interval of gap defect discharge can change from second level to microsecond level at 26 ℃,40 ℃ and 50 ℃,posing a risk of breakdown discharge. The research results obtained in this article further enrich the theoretical system of GIS intermittent discharge and improve the effective diagnosis rate of on-site GIS intermittent discharge.

Abstract:Accurately predicting the concentration trend of dissolved gas in oil has a positive effect on the evaluation of transformer status and life assessment. In order to improve the accuracy of dissolved gas in oil prediction,a dissolved gas in oil prediction method based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and time convolution network (TCN) is proposed in this paper. Firstly,the CEEMDAN method is used to decompose the original sequence of dissolved gas in oil into multiple intrinsic mode functions(IMFs),separating the stable and unstable IMFs. Secondly,TCN are established for IMFs,then predictions based on these trained TCNs are made. Finally,the prediction results of IMFs are overlaid to reconstruct the prediction results of the original sequence. Analysis in this paper shows that the root mean square error,mean absolute error and maximum error of the prediction method are 1.01 μL/L,1.53 μL/L,5.54 μL/L respectively,which are reduced by 53.47%,41.18%,13.36% compared to the case without using the CEEMDAN. When using CEEMDAN,the three errors are the smallest compared to commonly used recurrent neural networks. The proposed dissolved gas in oil prediction method has higher prediction accuracy and can provide more effective support for condition based maintenance strategy.

Abstract:Oil-immersed power transformer is an important part of power grid,and its reliable operation plays a vital role in pomler system security. Aiming at the problem that the insulation state of transformer cannot be assessed quantitatively after long-term operation,the accelerated aging and damp tests of oil-paper insulation model are carried out in this paper. The influence of aging and damp of oil-paper insulation on its recovery voltage curves is explored. The particle swarm optimization (PSO) is used to improve the parameter prediction method of extreme learning machine (ELM),which realizes the quantitative assessment of aging and moisture of oil-paper insulation based on the characteristic parameters of the recovery voltage curve. By comparing the physical and chemical performance analysis of oil-paper insulation models,it is shown that the prediction accuracy of PSO-ELM method is much higher than that of traditional ELM method. The absolute error range for predicting the moisture content of oil-paper insulation the degree of polymerization (DP) of pressboard is less than ±0.4% or ±30,respectively.

Abstract:Voiceprint detection technology can assist inspectors in assessing the state of transformers. A method for detecting and assessing transformer states based on voiceprint compression and cost-sensitive techniques is proposed. The method first extracts voiceprint features from transformer audio,then filters and compresses these features in the frequency domain. Subsequently,a convolutional neural network is employed to evaluate the transformer's state,incorporating a cost-sensitive loss function to enhance attention towards difficult samples. Using a 35 kV transformer as the experimental subject,transformer audio data is collected through on-site recordings,simulated experiments and sample augmentation. Test results demonstrate that the proposed method reduces the voiceprint dimensionality from 1 025 to 80,decreasing computational complexity and video memory usage to 8.1% and 7.7% of the original 1 025 dimensions,respectively. Simultaneously,the proposed method achieves a voiceprint recognition accuracy of 83.5% and improves the recall rate of the most challenging short-circuit current anomaly from 48.2% to 63.6%.

Abstract:Transformer live fault diagnosis is of great significance to ensure the safe and stable operation of power transformers. In response to the problem of complex working environment and limited fault types characterized by a single parameter,a method based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and feature entropy weights method (EWM) is proposed for fault diagnosis. The correlation coefficient weighted kurtosis (CCWK) principle is used to filter the CEEMDAN components and reconstruct the signal to achieve an improved characterisation of transformer vibration signal features while eliminating redundant components. The EWM is used to construct feature determination coefficients (FDC) to achieve a single data diagnosis of transformer fault types. The principal component analysis (PCA) is used to reduce the scale of mixed domain features and the chicken swarm optimization (CSO) algorithm is used to optimize support vector machine (SVM) model for fault diagnosis. The analysis is performed on a 110 kV three-phase oil-immersed transformer in a certain substation,and the results show that compared with other transformer fault diagnosis methods such as probabilistic neural network (PNN) and SVM,the proposed method not only provides early qualitative fault type identification but also improves the accuracy and efficiency of transformer fault diagnosis.

Abstract:In order to solve the problem of lacking velocity feedback information in the velocity closed-loop control system for linear induction motors (LIM) after cancelling the velocity sensor,a velocity observer based on the extended Kalman filtering algorithm has been implemented,considering the end edge effect of LIM. Firstly,based on the mathematical model of three-phase linear induction motor considering edge effect,an extended Kalman filter observer with appropriate gain and covariance update matrix is derived. Based on the vector control system of LIM,the speed parameters identified by the observer are fed back to the speed closed-loop system. Then,the vector control system model of linear induction motor with speed observer is built in Simulink,and the identification speed of observer is compared with the actual speed of motor. Finally,the results show that the closed-loop control using the identification speed can ensure the stable operation of the system. Under three kinds of loads,the error between the predicted speed and the actual speed of the observer is form 0.51% to 2.34%. The various dynamic performance of the system reveals that the LIM vector control system based on the extended Kalman filter observer increases prediction velocity error and decreases thrust error with increasing load. However,the magnetic flux amplitude error slightly increases. Therefore,considering the end-edge effect,the observer based on the extended Kalman filter can replace the velocity sensor to achieve three-phase LIM control under both unloaded and loaded conditions.

Abstract:Substation secondary equipment is monitored through inspection robots,providing an important means to enhance power equipment automation and intelligent management,thereby ensuring the safe operation of power engineering equipment. In this paper,a Mecanum wheeled omnidirectional mobile robot is developed for automatically inspecting secondary equipment in substations. It possesses autonomous navigation,positioning,and identification capabilities,significantly enhancing inspection efficiency and the accuracy of protection plate state identification. The Mecanum wheels enable the inspection robot to move flexibly and adjust its attitude within narrow working environments. Meanwhile,a multi-track lifting platform facilitates image acquisition and state identification of the secondary equipment pressure plate,covering a height range of 350-1 800 mm. The robot employs the lidar-based simultaneous localization and mapping (SLAM) method for autonomous positioning and navigation,supplemented by a vision-based path extraction and tracking algorithm for precise position correction at set points. Moreover,a color recognition-based image arrangement and state recognition method is proposed to accurately identify and assess the connection state of the secondary equipment protection plate. Experimental results demonstrate that the substation secondary equipment inspection robot,utilizing Mecanum wheels,achieves successful autonomous navigation and precise positioning,with maximum deflection angles and distances during path tracking processes being ±3° and ±8 mm,respectively. Additionally,the plate recognition method,combining machine vision and color recognition,achieves an outstanding recognition accuracy rate exceeding 95.80%,thereby elevating the level of robot automation in inspection operations.

Abstract:With the development of new energy,conventional units which are the main source of system inertia are constantly being replaced. Meanwhile,the load-side inertia becomes increasingly prominent with the increase of proportion. However,the existing load-side inertia estimation method is relatively simple,and the lower estimation accuracy cannot meet the needs of system operation management. Based on the principle of inertial quantitative statistics,a refined statistically modified method for load-side inertia estimation is proposed. Starting from the inertia analysis of the load-side basic elements and inertia modelling of the basic load units,an expression of load-side inertia estimation is given under typical load operating modes. Considering the scenario of distributed power supply access with hidden effect,the expression is modified in two cases of power supply with or without inertia equipment. According to IEEE 9-bus system,a simulation system is conducted,and simulation tests are carried out respectively with behind-the-meter power supply or not. The results show that the error of load-side inertia estimation is less than 5%,which verifies the accuracy and reliability of the proposed inertia estimation method.

Abstract:With the continuous penetration of renewable energies,thermal power plants are facing more frequent and wider variable load operation. The control of high order,large inertia superheated steam temperature processes in power plants is facing a great challenge. Therefore,a phase compensation based active disturbance rejection control (PC-ADRC) strategy for a class of high-order,large-inertia processes is proposed. Firstly,the working principles and control problems of SST system are illustrated. Then,the phase compensation (PC) network model is derived carefully by low-frequency approximation. A model simplification method is presented,in which the PC is adopted to compensate process dynamics and obtain reduced-order plant equivalently. To facilitate engineering applications,its simple implementation and equivalent model analysis for the PC-ADRC system are also given. At last,the stability and robustness of the PC-ADRC control system are theoretically studied. Theoretical analysis and simulation results show that the proposed controller can effectively improve the robustness and fast response ability of the high-order process control system.