Search

With the rapid development of new energy generation technology, the thermoelectric generation technology (TEG) can make good use of the waste heat generated in new energy generation. However, the change of temperature distribution will worsen the output characteristics and reduce the power generation efficiency of the TEG system. In this paper, a TEG array reconfiguration method based on the artificial bee colony (ABC) algorithm is proposed. In three different temperature distributions, ABC is used for dynamic reconfiguration of symmetric 9×9 and unsymmetric 10×15 TEG arrays. Three meta-heuristic algorithms, the genetic algorithm, the particle swarm optimization algorithm, and the bald eagle search are compared with the proposed method, and the temperature distribution of the TEG array reconfiguration by ABC is given. The results show that ABC can improve the output power of the TEG array, and the output power-voltage curves tend to show a single peak value. In addition, real-time hardware-in-the-loop (HIL) experiment based on the RTLAB platform is undertaken to verify the implementation feasibility.

In order to develop an accurate model of proton exchange membrane fuel cell (PEMFC), it is essential to exactly identify unknown parameters in PEMFC. However, parameter identification of PEMFC is a multi-variable, multi-peak, and strongly coupled nonlinear optimization problem, of which traditional parameter identification methods often fail to achieve satisfactory results. In addition, noises generated under different operation conditions will hinder meta-heuristic algorithms (MhAs) to obtain accurate parameters. To handle these thorny obstacles, extreme learning machine based MhAs (ELM-MhAs) are proposed for PEMFC parameter identification, which can achieve denoising through ELM. ELM is used to train data to reduce or eliminate noises and provide more accurate and reliable fitness functions for MhAs, thus ensuring the accurate identification of PEMFC parameters by MhAs. To verify the feasibility and effectiveness of this strategy, 25 groups of voltage-current data are processed without denoising, with Bayesian regularization neural network (BRNN) denoising or with ELM denoising under two conditions—low temperature and low relative humidity; high temperature and high relative humidity, respectively. Subsequently, parameter identification results of six MhAs and a Levenberg-Marquardt backpropagation of different data are thoroughly compared. The simulation results indicate that ELM can significantly reduce the impact of noise on the data, while effectively improving the parameter identification accuracy of MhAs, compared with no denoising and BRNN denoising.

Based on the relation between battery energy storage systems (BESSs) planning and operation, a multi-objective optimal allocation model that takes into account both economic and technical requirements is established, and a bi-level optimization structure is constructed to ensure effective planning and high-efficient operation of BESSs. In the inner layer, a peafowl optimization algorithm (POA) is employed to solve the BESSs charge-discharge operation strategy with the purpose of BESSs operation benefit maximization. In the outer layer, a multi-objective peafowl optimization algorithm (MOPOA) is devised to solve the Pareto solution set of BESSs siting and sizing scheme, which aims at minimizing BESSs cost, as well as voltage fluctuation and load fluctuation in distribution network. Furthermore, a typical scenario set is obtained via the clustering algorithm considering uncertain operating conditions. The simulation is performed based on the extended IEEE-33 bus system. The results show that the proposed algorithm achieves a trade-off between local search and global search, thus obtains a high-quality solution. It can obtain a more widely distributed and uniform Pareto front, which not only achieves the best investment benefit, but also improves voltage quality and power stability.

The ice-shedding may induce isolated ground wires temporary grounding during de-icing, which can cause the protective relay of direct current de-icing devices triggered frequently and unnecessarily. After analyzing the circuit of the grounding isolated ground wire, the conclusion can be obtained that the temporary grounding of isolated ground wires leads to the fluctuation of electric quantities. On the basis of above analysis, the permutation entropy is applied to describe the trend of the voltage waveform, which is able to overcome the influence of high frequency harmonics caused by converters. Based on the voltage trend acquired by the permutation entropy, the variance is employed to determine the fault and the disturbance. Then, parameters of the de-icing device in Kunbei substation is utilized for simulation. The simulation results verify the effectiveness of the proposed method, which can be performed to identify the disturbance induced by de-icing and the permanent fault.

In this paper, an optimal sizing and placement model for distributed generation (DG) is established, which includes active power losses, voltage profile, pollution emission, DG costs, and meteorological conditions. Since optimizing placement and sizing are discrete and continuous variables respectively, the model established is a highly nonlinear complex one with discrete optimization variables. Therefore, the adaptive manta ray foraging optimization (AMRFO) algorithm is applied to obtain the optimal Pareto front, which has a rich and diverse search mechanism, individual updating mechanism, and advanced Pareto solution selection mechanism. For this model, a better solution of high quality can be obtained. In order to avoid the influence of subjective setting of weight coefficient, the ideal point method based on Mahalanobis distance is used to make Pareto front decision. Finally, the simulation based on the IEEE 33, 69-bus distribution network and the IEEE 33, 69-bus distribution network in isolated network operation are implemented. The results show that compared with the traditional multi-objective intelligent optimization algorithm, AMRFO algorithm can obtain a more widely distributed and uniform Pareto front. While considering the economy, the optimized distribution network voltage profile and active power losses can be significantly improved.

DC electronic current transformer is the data source of the system control and protection device, so its measurement delay has always been concerned. In this paper, the DC current transformer is modelled equivalently and the key factors affecting the delay of shunt and the remote module are analyzed. A field test method for the delay characteristics of the DC electronic current transformer is proposed. Moreover, a complex multi-state large current generator and integrated test system are developed. On the basis of the IEC 60044-8 standard transmission protocol, the accuracy of standard signal acquisition with the shunt is better than 0.2%. The maximum output current of the established testing system is 600 A. Based on the field test in an actual HVDC project, transient step and various frequency signal components are applied for analyzing. [ABSTRACT FROM AUTHOR]

The phenomenon of wind abandoned caused by fault of transmission lines has occurred frequently. Since fault current output by wind turbine after fault may induce the frequency variation of the fundamental frequency, fault location based on the fundamental resistance is easily invalid. In order to locate fault accurately and rapidly after wind farm transmission line faults, a time‐domain‐based novel fault location method for transmission line of wind farm is introduced in this paper. The main idea of the proposed scheme is to discretise the distributed‐parameter model of transmission line that is given in the form of partial differential equations by using the finite‐difference time‐domain. In addition, the phase current and voltage at both ends of lines are regarded as boundary conditions. Then, the variational iteration method is utilised to calculate transmission line equations to obtain line length function. At last, fault location function is constructed by a minimised objective function. This paper builds wind power transmission line system simulation model and carries out simulation experiments on PSCAD/EMTDC. To verify the performance of the algorithm, several factors are taken into account for simulation including different fault distances, different fault types, different fault resistances, and different fault inception angles. Simulation results reveal the performance of the proposed method to locate fault for transmission lines connected to wind farm with high accuracy. [ABSTRACT FROM AUTHOR]

The vigorous development of renewable energy has induced multiple sources connected to regional distribution networks. However, the high impedance ground fault that appeared in the distribution network can induce an intermittent arc accompanied by extinction and reignition, which may damage renewable energy generation devices severely. Therefore, the detection of arcing faults is profoundly necessary for practice, but the high impedance arcing fault is easily confused with switches of motor or capacitor bank. In order to solve high impedance arcing fault detection, this paper proposes a high impedance arcing fault detection method that combines complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and Teager energy operator (TEO). Firstly, CEEMDAN algorithm is used to extract the intrinsic mode function (IMF) of transient zero‐sequence current (TZSC). According to the correlation coefficient and waveform similarity, which can select maximum correlation coefficient value as the characteristic component of TZSC; Then, TEO is utilized to track the change of signal in real‐time, meanwhile, wavelet singular entropy (WSE) is applied to calculate the characteristic component of TZSC. In particular, high impedance arcing fault (HIAF), motor switching (MS), and capacitor bank switching (CBS) can be effectively identified by comparing WSE. The effectiveness of CEEMDAN‐TEO detection method is verified by simulation, which owns advantages in extraction accuracy, calculation efficiency, and detection accuracy of feature components. [ABSTRACT FROM AUTHOR]