Abstract:The ever-increasing coupling relationship between cyber and physical systems makes cyber-attacks become an important factor affecting the reliability of power system operations. First, this article conducts the analysis of the network security risk from the attacker’s perspective, infers the available resources of the attacker based on the identity of the attacker, analyzes the purpose of the attack to be achieved, and infers the possible penetration and intrusion path and damage modes. The guidance can be provided to develop the specific protection methods based on the above analysis. Then this paper analyzes the deficiencies of the trusted computing, hierarchical protection, security situation awareness and other defense mechanisms being implemented in the power industry, and this paper points out the potential supply chain security threats in the security detection of software and hardware systems. Considering the difference of the risk levels and the harmful consequences caused by attacks on different power monitoring systems, the power system risk matrix is constructed from the aspect of the possibility of successful attack and the harmful consequences, and it is pointed out that the multi-target coordinated attack will increase the risk compared to the single-point attack. Finally, from the available resources of the state-supported cyber-attacks and the purpose of the attack, two high-risk potential cyber-attack damage modes are proposed, and the attack realization process and damage mechanism are summarized.

Abstract:With the fast development of the economy and society, the reliable supply of electricity has triggered a wide range of concerns. On 15th, February, 2021, a large-scale power outage occurred in Texas, USA due to the extremely cold weather. More than 4.8 million users are affected. This accident has a serious impact on industrial production and people's lives. Firstly, the general situation of the power system in Texas is briefly introduced, and then the specific process of the blackout is described in chronological order. After that, the causes of the blackout event are analyzed from three aspects: grid adequacy, regional support capability, and the market mechanism. Finally, from the perspective of the construction of China's new power system, some inspirations are proposed for the meteorological forecasting technology applications, the power supply structure of the new power system, the flexible resource and integrated operation of the source-grid-load-storage system, and the construction of the grid structure of the power grid. This work provides a reference to help guarantee the adequacy of the power supply in China's new power system.

Abstract:As a distributed energy storage with high flexibility, high reliability, and a green economy, the energy storage system with phase change materials can provide flexible and reliable auxiliary services for the grid as a kind of adjustable load. For the purpose of further investigating the application potential of distributed energy storage systems with phase change material, a kind of regional-joint peak-load shifting strategy is presented based on the aggregated system with shared phase change energy storage. Firstly, an aggregated controllable model for units of distributed phase change material energy storage is proposed. After that, a feasible regional-joint control architecture based on shared phase change energy storage is designed. Then, a joint peak-load shifting strategy of intra-regional and inter-regional joint peak-shaving and valley-filling control strategies is established under different business models such as the non-cooperative game and the cooperative game. Finally, the proposed control strategy is effectively solved by the quantum particle swarm algorithm. The simulation results verify the feasibility and effectiveness of the proposed joint peak-shaving strategy in reducing the load variance and regional economic cost. The proposed method is also a new approach to grid auxiliary service.

Abstract:Integrated Energy System (IES) covers a variety of energy such as electricity, heat and gas, which have complex operating states and diverse dynamic characteristics of the equipment. Accurate state estimation (SE) helps IES operate safely and reliably. In this case, a method is proposed for IES mixed state estimation regarding to the poor timeliness and accuracy of IES estimation. Firstly, the influence of the coupling measurement model on the redundancy of IES measurement is analyzed, and the effect of heat network state estimation is improved according to the variation of redundancy. Then, the distributed state estimation is introduced into IES to construct two regional partitioning scenarios on the shared node type and shared line type. Finally, the simulation analysis of IES coupling between a 30-node power network and a 17-node heat network is investigated to verify the effectiveness and practicability of the proposed method.

Abstract:Most of the existing planning strategies for the electricity gas-heat-integrated energy system are realized by cutting or transferring the load, and the influence of the dynamic coupling process is not considered. Therefore, the scheduling is not flexible enough. Under the background, a two-layer optimal scheduling model that considers the Dynamic Energy Conversion (DEC) energy storage and loads collaborative optimization strategy is proposed in this paper. In the upper layer, the dynamic conversion of energy flow is considered and the whole life cycle impact of the electric facilities, such as batteries, super-capacitors, and electricity to gas, is also concerned. Then the DEC load model is introduced. For the lower layer, based on the configuration of energy storage of the upper layer, an optimal scheduling model of an integrated energy system is established with the operating cost and net loss as the objective function. The scheduling results will be fed back to the upper layer. And then the DEC load can participate in the secondary scheduling and the flexibility of scheduling can be enhanced. In addition, the Second Order Oscillatory Particle Swarm Optimization algorithm is adopted to calculate the solution of the upper model, while the lower model utilizes the ε-constraint method to solve the nonlinear equations. Finally, an Integrated Energy System comprising the modified IEEE 39 Panel Point Power System, the Belgian 20 Panel Points Natural Gas System, and the 6 Panel Points Heat System is analyzed as an example. Four operation scenarios are discussed from the perspectives of economy, reliability, and flexibility to verify the rationality and superiority of the proposed model and algorithm.

Abstract:With the deep penetration of wind and photovoltaic power, the moment of inertia in the power system is reduced, resulting in the problem of frequency modulation for the automatic generation control of the power system. Energy storage is widely applied in the frequency modulation of power systems due to its fast reaction and accuracy. As a result, random simulation and empirical mode decomposition are combined to propose an energy storage capacity determination method in automatic generation control. Firstly, the empirical mode decomposition is utilized to decompose the regional control deviation into the high-frequency and low-frequency components. The energy storage frequency modulation is responsible for its high-frequency components, while traditional unit frequency modulation is in charge of its low-frequency components. Afterwards, the energy storage capacity is calculated via the high-frequency components of regional control deviation, and the energy storage is simulated considering the energy storage charge and discharge conditions. In the end, the frequency modulation effect is compared for the scenarios with different energy storage capacities and also without energy storage, and the energy storage capacity with the best frequency modulation effect is given consequently. The example simulation and results show that the proposed method is effective and feasible for obtaining the optimal energy storage capacity of automatic generation control in wind and solar power systems.

Abstract:The traditional distribution network planning method ignores of the power difference before and after the distributed generation is connected to the distribution network, which leads to a high construction cost. Therefore, a distribution network planning optimization method is proposed on the basis of the urban energy internet. Firstly, the power of the distributed generator is calculated before and after it is connected to the distribution network. And then, the load changes are compared in terms of the power change, so as to determine the influencing factors of the urban energy internet on the distribution network. Based on these influencing factors, a load forecasting model is established to predict the change of distribution network load under the influence of urban energy internet. Finally, the constraints are introduced for power supply and demand balance, power flow capacity, voltage and current, etc., and the distribution network planning optimization model is constructed successfully. An example is included for verification. It is shown that the regional load forecasting is precise, and the difference between the forecasted performance and the actual value is small. The proposed method can reduce the investment and operation cost of distribution network planning and construction to realize a reasonable distribution network planning.

Abstract:Single phase VSC has a wide range of applications, and studying its operating characteristics is of great significance for harmonic generation mechanism, harmonic suppression and stability improvement. Firstly, the second-order generalized integrator is used to construct the synchronous phase-locked loop for the main circuit structure of single-phase VSC rectifier. Secondly, according to the requirements of DC voltage outer loop control and unit power factor operation, the equation of inner current control loop is established using the quasi-proportional-resonant controller. Thirdly, based on the linearization principle of the running trajectory of the time-varying period system, the harmonic transfer function matrix of the main circuit and control system is constructed, and the fifth-order admittance matrix model of VSC is deduced, and the influence of related factors on admittance characteristics and system stability is analyzed. Finally, the electromagnetic transient simulation model is used to verify the correctness of the analytical model and theoretical analysis.

Abstract:The conventional frequency-controlled LLC resonant converter suffers from wide frequency regulation range and low conversion efficiency, so that it is unsuitable for wide voltage range applications with a wide voltage range. In order to solve these problems, a PWM control strategy of the full-bridge LLC resonant converter is proposed in this paper, which realizes the gain adjustment by adjusting the duty cycle of the switch The switches of two lower arms of the resonant converter are controlled by a fixed frequency, and the switching frequency is equal to the resonant frequency, while the switches of the two upper arms are controlled by PWM to adjust the output voltage. With the method, the frequency variation range of the resonant converter is significantly narrowed. In addition, the switches of primary circuit and the rectifier diodes of secondary circuit could achieve zero voltage switching (ZVS) and zero current switching (ZCS), respectively, which effectively reduces the switching losses of the converter. Finally, the 1.5 kW simulation and experimental prototype with 400 V input voltage, 250~500 V output voltage are established to verify the effectiveness of the proposed PWM control strategy.

Abstract:In view of the current situation that the existing line loss allocation methods do not reasonably consider the actual impact of distributed generator (DG) grid connection on the line loss of distribution network, this paper proposes a line loss allocation method which takes into account DG impact levels and weight characteristics. Firstly, a set of impact index system is designed in terms of the planning, operation and management of distribution network so as to comprehensively characterize the impact of DG grid connection on distribution network. On this basis, the line loss allocation strategy is developed to deduce the economic cost of distribution network. Secondly, the value of index weight is given by the order relationship analysis method (critical-g1) which is combined with the critical method. The index weight characteristics is comprehensively analyzed from the subjective and objective perspectives, thereby, the degree of influence of different indicators on line loss is emphasised. In addition, the relation relationship between each index and line loss is not definitely linear. Thus, the impact levels are classified and the value of impact factor is set. Afterwards, the index weight and influence factor scores are employed to calculate the coefficient of line loss allocation correction, and the DG allocation power is modified on the basis of power flow tracing method. Finally, a typical day of a distribution network with DG is included as an example to verify the effectiveness of the proposed method.

Abstract:The high zero-sequence inrush will be caused when the high-voltage built-in high-impedance transformer performs no-load closing, which may penetrate into the 20 kV distribution network with a grounding small resistance through the adjacent transformers, leading to the maloperation of zero-sequence overcurrent protection of distribution lines. This will result in the power loss of users. To solve this, the amplitude characteristics of high-voltage built-in high-impedance transformers are analyzed based on zero-sequence inrush formula. Then, the validity of the model and the possibility of protection maloperation are studied based on PSCAD/EMTDC simulation platform. At last, the calculation procedure for assessing risks of zero-sequence over-current protection in distribution networks is produced by MATLAB, where the actual parameter is used to quantitatively analyze the maloperation probability of zero-sequence overcurrent protection when a high-voltage built-in high-impedance transformer performs no-load energize. The experienmental results indicate protection operation situations. A protection blocking strategy based on 5G wireless communication technology is proposed and its feasibility is analyzed.

Abstract:The traditional measurement method of distribution network grounding capacitance is greatly affected by the neutral grounding mode, and cannot eliminate the influence of harmonic elimination resistance and internal impedance of voltage transformer on the measurement accuracy. For this reason, an accurate measurement method of distribution network grounding capacitance that is not affected by the neutral point grounding mode is proposed. By using the Y/△ connection grounding transformer with tap connected externally to the distribution network, the zero sequence voltage of the distribution network is ensured to change within the specified offset value by adjusting the grounding tap of the high voltage side winding of the grounding transformer to a lower gear, and the required zero sequence voltage and current are measured to obtain the system grounding capacitance. The proposed method is simulated and analyzed in PSCAD/EMTDC simulation environment. The analysis results show that the method has high measurement accuracy, and the measurement process of grounding capacitance is safe, simple and economical.

Abstract:In order to cope with the difficulty in locating the fault section when single-phase-to-ground fault occurs in the small current grounding system of a distribution network, this paper proposes a fault section location method based on the Hausdorff Distance algorithm according to the characteristics that the zero sequence currents on both sides of the fault point are opposite in the fault line. In this method, the zero sequence current is selected as the fault feature firstly, and is then filtered to ensure the wavelet approximate sequence of the zero sequence current of each detection node of the fault feeder can be extracted by wavelet packet transform. Next, the deviation matrix of the wavelet approximate sequence of the zero sequence current among each detection node is obtained by Hausdorff Distance algorithm. Finally, a deviation degree is defined to represent the difference between the two sides of each section, and the fault section can thus be determined by comparing the relevant deviation degree. The simulation results show that the method can achieve precise positioning under different fault conditions, and it is also suitable for more complex distribution network structures. Both of which are convenient for staff to quickly repair and maintain the fault lines, and ensure the safe and reliable operation of power systems.

Abstract:Many faults of single core cables are related to metal sheath. The simplified model that only considers the core cannot realize the sheath-related fault location. This paper proposes a novel fault location method considering the metal sheath of the cable basing on the distributed parameter model. First, based on the distribution parameter theory, the equivalent distribution parameter circuit of the fault cable of considering the metal sheath is analyzed, the voltage and current equations are established; then, the unknown parameters in the voltage and current equations are solved through boundary conditions, the voltage and the current value of both sides of the fault point are obtained; finally, the fault distance is obtained by iterative search of the ranging equation. The proposed method transforms the complex phase network into mutually independent sequence network, which makes the algorithm simple and effective. A large number of simulation results show that this algorithm can achieve the accurate distance measurement of sheath-related faults (sheath-to-ground faults, core-sheath faults, and core-sheath-to-ground faults), and the ranging error is within 1.5%. The results are not affected by the fault location, fault type and the transition resistance.

Abstract:At present, in the behavior prediction of relay protection equipment, the state transition model is used for state classification and probability prediction. The description of the influence of equipment aging and planned maintenance factors on the prediction model is not accurate enough, and it is thus difficult to accurately reflect the future behavior of the relay protection equipment. Based on the improved three-parameter Weibull distribution that characterizes equipment aging and maintenance, this paper uses the whale algorithm to further enhance the three-parameter Weibull distribution function, constructs a continuous Markov chain state transition model with a time-varying transition rate. In this model, planned maintenance nodes are used to characterize the service age regression of the protection equipment, the state observation node to simplify the calculation of the transition probability, and then an action behavior prediction algorithm that calculates the time-varying transition rate and planned maintenance can be proposed. The Weibull curve is calculated and fitted based on the case database, and the predicted actions are comprehensively analyzed through the simulated comparison experiments, which verifies the rationality of the state transition model and the prediction algorithm.

Abstract:The traditional one-directional neural network has some problems in the field of ultra-short-term power prediction in distribution networks, such as the out-of-shape curve prediction, the over-fitting phenomenon of the model, low prediction accuracy and slow convergence speed, etc. Thus, an improved bi-directional recurrent neural network model is proposed based on the wavelet transform and self-attention mechanism to overcome these problems. Firstly, the forward and reverse laws of the power data are studied by the bi-directional network to improve the prediction accuracy of the model. Afterward, the wavelet transform is employed to reduce the overall difficulty of power prediction. Consequently, the model overfitting is reduced, and the convergence speed is increased in the meantime. In the end, the self-attention mechanism is adopted to grasp the hidden layer dimensional relationship of the model to further improve the prediction accuracy. An example shows that the proposed improved model can eliminate the existing problems effectively. Compared with the traditional model, the MAE increased by 50.1%, MAPE increased by 43.3%, RMSE increased by 51.1%; in the reactive dataset, dataset MAE increased by 60.5%, MAPE increased by 63.8%, and RMSE increased by 60.1%.

Abstract:The traditional value evaluation of power users does not consider the high and low voltage levels generally. In fact, the applicable indicators of high voltage and low voltage users are different, especially, there is no corresponding indicator system for the current low voltage power user value evaluation. Therefore, this paper proposes a low voltage power user value evaluation method based on the AHP and BP Adaboost algorithms. Firstly, a set of low-voltage power user value evaluation index systems are established, and the scoring rules and grading rules are defined. Then the comprehensive scores and grades of low-voltage power user value evaluation are obtained by AHP. Finally, the calculation method combining AHP and BP Adaboost is proposed to evaluate the low-voltage power user value, and the comprehensive scores and grades of low-voltage power users are obtained after that. In addition, the two parameters of determination coefficient and accuracy are used for verification. The simulation results show that the method is correct and effective.

Abstract:Nowadays, the contradiction in the seasonal and periodic power supply and demand is gradually dominant. The orderly use of electricity via avoiding peak and shifting peak can alleviate the contradiction, delay the construction of transmission and distribution facilities and confirm the safe operation of the power system. Under the background, a decision-making method of orderly Power consumption is proposed with weight given by the entropy weight-grey relation. Firstly, since the constant weight is insensitive to abnormal indicators, the information entropy is combined with the grey correlation degree to decide the weight. After that, the potential is comprehensively evaluated for users participating in the orderly power consumption so as to determine the priority of enterprises to participate in orderly power consumption. Then, in order to reduce the profit loss of traditional orderly power consumption scheme by the manual decision, an auxiliary decision-making method of orderly power consumption is proposed to simulate the peak avoidance and peak staggering of users. In the end, the operation scheme of the Changsha Heishi line in summer 2020 is simulated as an example. It is shown that the proposed method can alleviate the peak load regulation pressure of the power grid during the peak time and it has good economic and practical value in engineering.

Abstract:With the development of intelligent distribution networks, the type and quantity of equipment in distribution networks show the trend of massive access. The explosive growth of heterogeneous and multi-source data information also puts forward higher requirements for the real-time and reliability of intelligent distribution network protection business. In order to better meet the delay requirements of distribution network protection services, this paper proposes a container-based distribution network protection microservice modeling and computing resource scheduling method. Firstly, the distribution network protection service is decomposed into multiple microservices, and the microservice timing logic model of the distribution network protection service and the container-based microservice architecture of the distribution network protection service are established. Furthermore, the computing resource scheduling model of microservice is established, and the microservice computing resource scheduling result with minimum service delay is obtained by improving the differential evolution algorithm. Finally, simulation examples compare the results of different microservice computing resource scheduling strategies and analyze the impact of CPU computing resources on computing resource scheduling results. The simulation results verify the effectiveness of this method.

Abstract:Due to the difficulty of the inspection of distributed photovoltaic (PV) electricity theft, the number of electricity theft samples collected by relevant departments is limited, which cannot meet the needs of data-driven electricity theft detection. This paper proposes a data augmentation method for distributed PV electricity theft using Wasserstein generative adversarial network (WGAN). First, WGAN can explicitly learn the time correlation that is difficult to model in the PV electricity theft data sequence. Furthermore, it can generate new electricity theft samples with similar distributions to the real ones through the confrontation training of the generator and discriminator networks. Then, according to the typical PV electricity theft model and data characteristics, the convolutional neural network (CNN) is selected for electricity theft detection. Finally, through the case analysis, it is shown that the electricity theft samples generated by WGAN can conform to the fluctuation law of authentic samples and the probability distribution characteristics of historical data, thereby effectively improving the detection performance.

Abstract:In order to study the influence of layered soil on the impulse current dispersing characteristics of grounding device. This paper proposes a simulation method for the impulse characteristics of the double-layer soil tower grounding device based on ATP-Draw. By using the Laplace method, the grounding resistance of the horizontal grounding body and the vertical grounding body in the double-layer soil are calculated. The conductivity parameters, capacitance parameters, and inductance parameters of each part of the chain circuit model of the grounding device under the double-layer soil structure are obtained. ATP-Draw simulation model of pole grounding device in complex soil is established. The proposed model can realize the simulation analysis of lightning impulse characteristics of transmission line pole grounding device in the sealed layer soil and the accurate calculation of impact grounding resistance. The contribution of this paper can be applied in the grounding system design of power transmission tower.

Abstract:In order to solve the problem that the resolution of the transformer windings ultrasonic detection system decreases with the increasing detection distance, this paper proposes a combined algorithm of ultrasonic synthetic aperture arc scanning, so that the resolution of the detection system is only related to the transducer itself. By applying the proposed method, faults can be quickly located, thus detection efficiency and detection accuracy are improved. Finally, a transformer winding ultrasonic synthetic aperture detection system is designed in this paper, and a 400 kHz ultrasonic transducer is used to conduct experiments on a transformer. The final experimental results show that the detection system can locate the deformation position of the transformer winding quickly online. The relative error of the detection is only 4.26%.

Abstract:Aiming at the problem that it is difficult to directly detect the temperature of the front-end terminal of the smart meter, this paper proposes an improved particle swarm optimization radial basis function neural network for the detection of the temperature terminal of the smart meter. First, the mapping relationship between the terminal temperature of the electric energy meter and its influencing factors based on the RBF neural network is established. The central position of the appropriate network core function is selected through the K-Means++ algorithm, and the recursive least square method is used to obtain the network core function connection weight. The improved particle swarm algorithm is used to optimize the width coefficient of RBF neural network and model training, and the optimized temperature mapping expression is derived. Then the terminal temperature detection of the smart electric energy meter is realized. The simulation results show that the improved particle swarm hybrid optimization radial basis function neural network algorithm proposed in this paper has high calculation accuracy, strong search ability, fast convergence speed, and the relative error of temperature detection is less than 0.17%. Compared with the existing detection methods, the proposed method has higher accuracy.

Abstract:The field inspection of the secondary circuit is an important step in the field inspection of the gateway electrical energy metering device. However, the method of finding a secondary circuit terminal to be tested was rather complicated in the past. In order to optimize the process of finding terminals to be tested, this paper proposes a terminal number detection method based on EAST. In this method, the training dataset is first established, and the EAST model is trained. The trained model is used to detect the text in the terminal block image and outputs the size and position information of the terminal number region. Then, the region coordinates were clustered by DBSCAN clustering to distinguish possible multi-column terminals, and the tilt angle of each column terminal was calculated by linear regression. Finally, combined with the tilt angle and the average width and height of the region, the corrected detection results of the terminal number region are obtained. Examples show that this method can accurately detect the terminal number in the image and effectively improve the efficiency of terminal detection, which lays a foundation for the subsequent secondary circuit inspection work.

Abstract:The unmatched disturbance is a common disturbance type in practical engineering problems, and traditional control methods are difficult to achieve ideal control performance. Under the background, a control method based on auto-coupling PID(Auto-Coupling Proportional-Integral-Differential, ACPID) control theory is utilized to solve the control problem of a class of nonlinear systems with non-matching disturbances. Firstly, the external disturbances and internal state of this method in the unmatched channel are defined as new unknown states. Meanwhile, the internal dynamics and external disturbances are defined as total disturbances. Then, the system can be transferred to an equivalent unknown linear system. After that, a controlled error system under the reverse phase excitation of the disturbances is constructed. The ACPID control method is employed to design the controller and the closed loop control system model is obtained successfully. The robust stability of the closed-loop control system is also analyzed afterward. In the end, a 2nd order nonlinear system is simulated for verification. It is shown that the ACPD control system not only has good dynamic and steady-state performances, but also has good anti-disturbance robustness with a fast response speed. It can be considered for the application of electric power, transportation, aerospace, and other extensive fields.