Search

With the rapid development of flexible DC transmission systems, the application of modular multilevel converter MMC is becoming increasingly widespread. Meanwhile, the nearest-level modulation is most applied. This paper establishes a simple and efficient harmonic model of MMC based on nearest-level modulation with the equidistant controlled firing scheme. First, based on the analysis of the working principle of MMC and the modulation principle, the Fourier decomposition of the stair wave is obtained. Then, according to the equivalent circuit diagram of MMC, the stair wave is exactly the differential-mode voltage. Moreover, the amplitude and phase of each differential-mode voltage harmonic content are obtained. So far, MMC can be regarded as a harmonic voltage source. Finally, simulations were performed on PSCAD to verify the accuracy of the MMC harmonic model proposed in this paper. Furthermore, the MMC harmonic model under the equidistant controlled firing scheme and real-time trigger control were also compared. The final results show that the model proposed in the paper can accurately reflect the harmonic characteristics of MMC in the existing power system after considering the equally-spaced trigger control.

Faulted line detection is used for locating faults and restoring service after the occurrence of a single line‐to‐ground fault in a non‐effectively earthed network. The traditional single faulted line detection method cannot adapt to the fault diagnosis of distribution systems under unfavourable conditions, while faulted line selection schemes based on neural network have poor generalisation ability when the structure of distribution network changes. A novel faulted line detection method based on multi‐feature integration is proposed to overcome the shortcomings of the previous methods. This study classifies the different fault characteristics of single line‐to‐ground faults in distribution systems, and analyses the applicable range of existing methods under multiple fault conditions. Subsequently, three detection methods that can be applied to different fault conditions and have complementarity are employed for feature extraction. Moreover, the extracted features of a feeder are processed by introducing the fault features of the remaining feeders, and the processed features are fused based on the fuzzy theory. Considering the selection of fault measure functions and weight coefficients in the fuzzy theory based on experience, an adaptive‐network‐based fuzzy inference system is employed for feature fusion. PSCAD and RTDS experiments confirmed the superior generalisation ability of the proposed method.

In order to improve the reliability and continuity of the power distribution system, power Internet of Things analysis monitoring system was constructed in this article, and the system was used for comprehensive applications such as data acquisition and fault analysis. Fault location methods considering distribution phasor measurement unit (D-PMU) information and network symmetry characteristics are proposed. With the measurement information provided by the D-PMU, the fault sections and the fault types are determined by the phase difference between the positive and negative sequence currents. Different location algorithms are obtained based on the load symmetry of the distribution network. The fault location method based on the line distribution parameters is used for the three-phase symmetric system, and the fault location method based on line impedance is proposed for the three-phase asymmetric system. Finally, the effect of different nodes with a D-PMU on the fault location accuracy is analyzed, and the configuration principle of the D-PMU was proposed. The PSCAD/MATLAB simulation software is used to analyze the example and verify the effectiveness of the proposed method. The simulation results demonstrate that high-precision fault location could be realized and that the practicality and economy are improved.

Data-driven and artificial intelligence based transformer protection has attracted increasing attention but not been widely applied in the power system owing to the poor generalisation ability. In this study, a feature transferring method is proposed for a knowledge-based artificial neural network (ANN) to develop a transformer protection with an improved generalisation ability. Normally, power experts can reliably identify the running states based on the professional knowledge of only focusing on the unsaturated parts of equivalent magnetisation curve (voltage of magnetising branch-differential current). In order to imitate the power experts, the images of equivalent magnetisation curves whose saturated parts are removed are defined as source domain and the original samples are target domain. An ANN named as S:ANN is firstly trained through the source domain where the extracted features are equivalent to the professional knowledge. Then another ANN with the same structure as S:ANN is trained through the target domain and named as T:ANN. It is specially designed for T:ANN that adaptive layers are employed between S:ANN and T:ANN to reduce the feature differences. Finally, simulations and experiments reveal that the knowledge-based ANN namely the determined T:ANN shows a better generalisation ability through paying more attention to the unsaturated parts.

The DC distribution network is becoming the basic supporting component of the urban distribution network. A variety of power supplies, loads, and energy storage devices can be directly integrated into a DC distribution network, which can improve the economical and energy efficiency of the power system. This study proposes two charging strategies, two-stage charging and one-by-one charging, to make the converter start up safely or quickly return to the normal operation state after the fault clearance. Owing to the existence of numerous DC capacitors in voltage source converters, the reactive power consumption in the start-up process needs to be explicitly calculated. Mathematical expressions of reactive power requirements for different charging strategies are investigated in detail, from where it can be seen that the current-limiting resistance can significantly affect the reactive power and the entire charging time. Simulation results show that the AC voltage can be seriously affected by the oscillation of the reactive power, which will lead to voltage drop and waveform distortion in mode switching.

Power system restoration after outages caused by natural disasters is difficult owing to the damage caused to transmission lines. To formulate an effective power system restoration scheme and avoid damage to power equipment, it is necessary to obtain the availability of transmission lines in real time. Due to the complexity and variability of natural disasters, the availability assessment of transmission lines that depends on statistical data cannot give effective availability, which will lead to the failure of recovery. Two correlation models between transmission line availability and information of relay action are established and then applied to different relays in this paper. Combined with the real time record from the sequence of event (SOE), the online availability assessment method for transmission lines is proposed, which can obtain the damage probability of transmission lines in real time. The proposed method is verified by testing on the IEEE 39-bus model, and the simulation results show that it can assess the availability of transmission lines in real time without needing to consider the type of disaster, thus, helping to avoid the use of damaged transmission lines in the recovery process.

A novel power-transformer protection algorithm based on the shape of the magnetization hysteresis curve is proposed. The magnetization hysteresis loop is studied and described by a voltage-differential current (V-DCurr) curve; furthermore, the characteristics of the V-DCurr curve are studied at different operating conditions. Three basic characteristics that can distinguish a fault from the other operating conditions are fed as the input to a support vector machine (SVM); the faults can be detected by the classifier of the SVM, based on the definite characteristics of the V-DCurr curve. Additionally, an optimization method for determining a suitable classifier, based on the grid search technique, is introduced to improve the reliability and efficiency of the proposed method. Digital simulation and dynamic-model experiments are conducted to verify the proposed method. Thus, the method is shown to have good application prospects in power systems.

The calculation of the short-circuit current is an important basis for fault detection and equipment selection in the DC distribution system. This paper proposes a linearized model for modular multilevel converter (MMC) considering different grounding methods and different failure scenarios. This model can be used in different fault conditions before MMC’s block. Under different fault forms, the model has different manifestations. This paper analyzes and models the DC distribution network with two types of faults: inter-pole short circuit and single-pole grounding short circuit. Among them, the modeling and analysis of single-pole grounding short-circuit uses the method of common- and differential-mode (CDM) transformation. To solve such a model, an analytical calculation method is proposed. As a mean of evaluating the effectiveness and accuracy of the proposed model, the analytical calculation solution is compared to the solution produced by PSCAD/EMTDC. A comparison of the results reveals the efficacy of the proposed model.

As an effective solution for managing the access of distributed generations, medium-voltage AC/DC distribution networks are emerging as prospective forms of future distribution systems under the circumstances of increasing energy crisis and environmental pollution. Compared with the transmission systems, the medium-voltage distribution network has a lower voltage level, and the two-level voltage source converter (VSC) has become the preferred topologies for the interconnected converter. This paper proposes a comprehensive small-signal model for two-level-VSC-based medium-voltage AC/DC distribution systems, considering AC and DC distribution networks, AC filters, and all the controllers in the VSCs. The complex transformation between two dq coordinate systems in current studies is avoided based on the characteristics of dq transformation and discipline of the linearization. The impact of various control parameters on the small signal stability of the AC/DC distribution system is analyzed, and the feasible region of each parameter is obtained based on the small signal model proposed. Numerical calculations by MATLAB programming and electromagnetic transient simulations on PSCAD are compared. The results indicate that the small-signal mathematical model proposed in this paper can accurately represent the dynamic process of corresponding AC/DC distribution system, and the analysis of impact of parameters on small-signal stability can also be proved.

Single line-to-ground (SLG) faulted line detection has always been a key constraint of the distribution system reliability. This paper focuses on practicability and generalization of detection methods, and therefore a novel SLG faulted line detection method based on sampled value of zerosequence current is proposed. The normalized sampled values used in the method don’t require complicated transformations while taking the correlation of lines into account. It uses gradient boosting decision tree (GBDT) as a classifier to tell whether a waveform is normal or not. Consequently, there is no need to retrain GBDT when the structure and parameters change in the operation of the distribution network. With much training work based on multiple data generated by PSCAD simulation, an adaptable GBDT structure is finally determined. The PSCAD and RTDS experiments verified its high performance of generalization ability and application prospects in different distribution systems.

A novel equivalent magnetization hysteresis (V-I curve)-based transformer protection scheme using convolutional neural network (CNN) is proposed in this paper. The characteristics of equivalent magnetization hysteresis demonstrating the relationship between excitation branch voltage and differential current can describe figuratively whether the power transformer is faulty. Due to the generality of V-I curves, the performance of artificial intelligence (AI)-based transformer protection methods can be improved. In this paper, the images of V-I curves are used as the input of CNN to identify the faults. The training samples are generated with typical 2-winding transformers through PSCAD simulations. And the test samples which are composed of field test data and PSCAD simulation data are obtained with different transformer parameters from that of training samples. With much work of adjusting CNN parameters, a CNN structure with better performance is decided. The results show that the generalization ability of proposed method is very well and the image cognition-based methods are free from the different sampling frequency. Despite there are still a lot of works for further improving the performance, the proposed method has shown a good application prospects.

A transformer protection scheme based on the deep forest is proposed to improve the performance of transformer protection. Considering that the magnetization curve can reflect the essential cause of the inrush current, but it is difficult to measure, the voltage and differential current sequences are selected as alternative input of the deep forest. The classification deviation of each layer is defined to automatically determine the depth of the deep forest. The protection acts based on the algorithm output. The deep forest is trained by the simulation sampling data from PSCAD/EMTDC simulation. The performance of the deep forest algorithm is verified through simulation data and dynamic model experimental data. The test results show that the deep forest algorithm can identify the operating states of transformer quickly and reliably. It has good generalization and no requirement of high sampling frequency, and has certain adaptability to CT saturation and over excitation.

The fault diagnosis technology of the distribution system is an important guarantee for the stable operation of the distribution network, and the fault line detection method is the premise of fault diagnosis. Based on the analysis of the characteristics of zero-sequence current when a single line-to-ground (SLG) fault occurs in the non-solidly earthed network, this paper proposes a fault line detection method that comprehensively uses the time-domain direction and spectral energy of the zero-sequence current, which can adapt to frequent changes in the operation mode of the distribution network. The effectiveness of the method is verified by a large number of simulation and field data.

Accurate and rapid detection of faulty line when single-phase-to-ground fault occurs is crucial to the reliability of power supply. This paper proposes a novel faulty line detection method based on artificial neural network, combining traveling wave and transient current phase information. Firstly, we introduce the existing faulty line detection methods and analyze the effective area of these methods, finding that effective area of the traveling wave method and the transient current phase method is complementary. After obtaining a large number of fault samples with PSCAD/EMTDC, we train and test the artificial neural network. The new faulty line detection method has excellent performance in faulty line detection and has good generalization ability.

Deep learning based transformer protection has attracted increasing attention. However, its poor generalization abilities hinder the application of deep learning in the power system owing to the limited training samples. In order to improve its generalization abilities, this paper proposes a knowledge- based convolutional neural network (CNN) for the transformer protection. In general, the power experts can reliably discriminate between faulty transformers and healthy transformers only through the unsaturated parts of equivalent magnetization curve (voltage of magnetizing branch-differential current curve) but deep learning intends to focus on the combined features of saturated and unsaturated parts. Inspired by the identification process of power experts, CNN adopted a specially designed loss function in this paper which is used to identify the running states of power transformers. Specifically, the presented Restrictive Weight Sparsity substitutes a special regularization term for the common L1 regularization. The presented Adaptive Sample Weight Adjustment endows the softmax loss of each sample with the optimizable weight the softmax loss of each sample with the optimizable weights to increase the impact of more-difficult-to-identify cases on the training process. With the modified loss function, the knowledge is abstractly introduced into the training process of CNN so as to successfully imitate the identification process of power experts. Accordingly, the proposed knowledge- based CNN will pay more attention to the unsaturated parts of equivalent magnetization curve even if only limited samples are included in the training process. The results of simulations and dynamic model experiments reveal that the knowledge-based CNN exhibits an improved generalization ability and the knowledge-based deep learning algorithm is a promising research direction.

The inrush current generated in the process of no-load closing of transformer is not only prone to misoperation of transformer relay protection, but also has adverse effects on transformer equipment and power grid. This includes adverse effects on the commutation process of the High Voltage Direct Current (HVDC) transmission system. The main manifestation is the influence of harmonic voltage and fundamental voltage changes caused by inrush current on the commutation process of inverter’s AC bus. In serious cases, it may lead to commutation failure of the HVDC system or even the continuous commutation failure. In this paper, at first, the influence mechanism of commutation failure of HVDC system caused by inrush current is discussed. On this basis, multi-principal elements model analysis method is proposed to analyze the distortion of the inverter’s AC bus voltage when the no-load transformer is closed at different positions. In this paper, the DC system model is built in PSCAD/EMTDC. For the built model, the no-load closing operation is carried out at different positions of the AC grid, and the influence degree of the bus voltage of the converter station caused by each harmonic wave is calculated, and is characterized the risk of commutation failure. In addition, the relevant measurements to suppress the inrush current are simulated and analyzed. It is of great significance to further ensure the safe and stable operation of the power grid.