Iet Science Measurement & Technology最新文献

Solar-blind ultraviolet (UV) discharge detection based on UV signals in the solar-blind band from 240 to 280 nm is an effective non-contact discharge detection method for electrical equipment, but there is still a problem that the detection results of different types of UV imagers are not comparable. To this end, by using standard light source and MODTRAN model, the relationship between the facular area and the radiant flux received by UV imagers was investigated by establishing a radiation transmission model for UV detection under different environmental conditions. Through analysis, it was found that the influence of humidity on the transmittance of the solar-blind band in atmosphere can be neglected, and the transmittance is slightly affected by atmospheric pressure, temperature, and detection distance under conditions suitable for UV detection. Moreover, a consistency formula for UV detection results was proposed, thus forming a consistency method for detection results suitable for various types of UV imager. Finally, the feasibility of the consistency method was verified by UV discharge detection tests with a rod-plate model and the absolute irradiance at discharge points of the rod-plate model measured by a grating spectrometer. The consistency method is conducive to the promotion of UV discharge detection technology.

A new equivalent circuit model for aged-polluted polymer outdoor insulators is proposed here. The structure of the model is constructed as such that it can be applied in investigating the effect of insulator aging and dry band discharge intensity on leakage current (LC) characteristics. To implement the capability of modelling aged insulators, the measured leakage currents from laboratory tests on field aged MV insulators with different level of aging and hydrophobicity were examined. Furthermore, the samples were artificially polluted and the corresponding measured leakage currents from laboratory tests were employed as references. The proposed circuit model can consider the effect of pollution by its topology and represents the aging phenomenon of insulator by the changes in the values of its elements. High similarity between test results and model outputs has confirmed model ability to represent aging effect on LC characteristics. Finally, mathematical equations for model parameters in terms of aging level are extracted using curve fitting, in order to predict the age and life-time of polymer insulator.

Permanent magnet (PM) machines have gained increasing attention from the industry because of their potential merits over conventional electrical machines. High-temperature level in electrical machines is a major factor that can adversely impact the performance and health conditions of these machines. These effects need to be minimized to make them competitive with currently operational machines in the industry. It can be done by performing a thorough thermal analysis of the prototype during the design process and proper online thermal monitoring of the machine during the operation stage. Thermal modelling is a powerful tool for achieving such goals in both cases. This paper aims to review the state-of-the-art of thermal modelling techniques and provide a detailed comparison of them in PM machines. The fundamental concepts of heat transfer and sources of heat in PM machines are first discussed. Then, different thermal modelling techniques, including thermal subdomain models (TSMs), lumped parameter thermal networks (LPTNs), finite element models (FEMs), computational fluid dynamics (CFD), reduced order models (ROMs), and hybrid thermal models (HTMs) are introduced and compared. Finally, the utilization of these models in the thermal design, control, and monitoring of PM machines is studied.

In this study, a portable Total Dissolved Solids (TDS) and Electrical Conductivity (EC) of water sensor using a disposable polymer strip and the readout circuit segments is presented. The strip consists of gold electrodes and a vent hole for water flow injection inside the measurement chamber. This strip reduces the cost of the measurement, decreases the measurement duration, and increases the repeatability of the system. Besides, there is no demand for electrodes cleaning and recalibration of the system. According to the electrical model of strip electrodes and their frequency responses, the EC value is in the range of 78.1–1230.8 (µs)/cm. The experimental tests are carried out based on the frequency method in the concentration range within 50–800 ppm. The results confirm the linear relation of TDS concentration on the EC voltage value which is 1.9 mV for 1 ppm resolution. Also, the effect of temperature variations on the EC value is investigated. Furthermore, the average amount of error for EC measurement exhibits an acceptable amount in comparison with other studies. Moreover, studying the drift variations for 20 days reveals the maximum drift of 1.6%. Finally, according to the present study, the sensor is a suitable candidate for practical applications.

Insulation failures caused by water drops deformation on insulator sheds edge greatly threaten the stable operation of power grid. However, only few studies have focused on water drops deformation characteristics on insulator sheds edge. In this paper, full-time domain deformation of water drop was defined. The rain tests were performed to study the deformation of water drops and obtain water drops parameters under DC voltage. The water drops deformation parameter was selected according to the state of water drops when arc is generated. A fluid-electric field coupling method is employed to establish simulation model to study the influence of water drops initial diameter, initial mass flow rate, DC voltage, and conductivity on water drops deformation. It is indicated that water drops maximum deformation length before first break in simulation is in good consistency with that obtained from rain tests. The water drops initial diameter, initial mass flow rate, and DC voltage significantly influence the deformation of water drops on sheds edge. Moreover, the water drops may even bridge adjacent sheds in some cases. In contrast, the effect of water drops conductivity on the water drops deformation is not apparent.

For an alternating-current (AC) shielding ball with a large radius of curvature, a three-dimensional (3-d) physical model for the corona onset voltage (COV) of the large-diameter shielding ball under different ambient conditions was established according to numerical calculation in concert with the theory governing gas discharge. At first, distribution of the spatial 3-d electric field around the shielding ball was calculated by using the ring charge simulation method. Then, a calculation criterion suitable for the COV of the shielding ball was derived and the area factor of the geometric absorption function for photons applicable to such structures was proposed in detail based on the theory of secondary photoelectron emission. Moreover, tests on COVs of shielding balls with different diameters were conducted in an extra-high voltage (EHV) AC test-base. The test results show that the model was found to show a favourable effect in calculating the COV of a large-diameter shielding ball under different ambient conditions. Finally, in combination with the thickness of the ionosphere and effective ionization coefficient, the influence and mechanism of action of various parameters (such as atmospheric pressure, humidity, and temperature) on the COV of the shielding ball were investigated.

The UHV DC tension insulator strings are often large-tonnage suspension porcelain or glass insulators with high mechanical strength. The single insulator is heavy, so two workers working at the same time are required to make a replacement on insulator strings in live working conditions. One of the key problems that must be solved in live working conditions is to select an appropriate way to reach the working position and realize the potential transfer between two workers. In this paper, the electric field calculation model of live working of ±800 kV tension strings was established. The process that two workers enter and leave the working position along the tension string and different conditions that two workers replace insulators were simulated and analyzed, and distribution characteristics of the electric field of insulator strings and live workers were obtained. According to the calculation results, when the two workers replace the tension insulators in different tension strings, it is easy to realize the potential transfer, so as to reach the same potential for work. The study results of this paper provide a basis for the selection of replacement method of large-tonnage single insulator by two workers, and a theoretical basis for UHV DC live working.

An insulation board is an important insulator used between damping capacitors in a converter valve. The potential creeping discharge phenomenon on the insulation board will affect the insulation between the damping capacitors and even the safe operation of the converter valve; therefore, a modified chaos particle swarm optimization multiple signal classification (MCSPO-MUSIC) localization algorithm based on a sparse array was proposed. The performance of the localization algorithm and the sparse array was analyzed by MATLAB simulation, and a test platform was established to detect the insulation board discharge position localization. The simulation results showed that the calculation time of this algorithm is about 1.5 s, which is an order of magnitude less than traditional MUSIC algorithm, and it is found that when the sparsity of the 4 × 4 array is 4 (the sparse array elements are 5, 9, 14 and 15), the localization accuracy remains high. Ten groups of experimental data were put into the MCSPO-MUSIC algorithm; the root mean square errors (RMSE) of the localization errors are 1.91° (non-sparse array) and 3.12° (sparse array), respectively. Finally, the blind source separation algorithm was used to remove the field noise, which verifies the algorithm and sparse concept accurate and economical in practical application.

This paper develops a new model with adjustable accuracy to evaluate performance of permanent magnet Vernier machine (PMVM) based on flexible magnetic equivalent circuit (MEC) method. The desired machine parameters such as number of flux modulation poles (FMPs), winding pole-pairs, and geometrical dimensions can be chosen in the presented model. By utilizing nonlinear core magnetization curve, saturation effect is taken into account. There are two main contributions in this study. First, the flexible MEC model is developed to predict the machine performance accurately. The second contribution is that the developed model can be used to evaluate the performance of the machine under no-load and loading conditions at the same time. The results of MEC model under no-load and loading conditions are compared with 2D-finite element method (FEM) and 3D-FEM in terms of accuracy and simulation time. The results indicate acceptable accuracy and suitable processing time of the proposed MEC model.

The lack of labelled data presents a common challenge in many fault diagnosis and machine learning tasks. It requires the model to be able to efficiently capture useful fault features from a smaller amount of labelled data. In this paper, a method to train multiple convolutional auto-encoders by self-learning method and integrate them using ensemble learning, called ensemble self-taught learning convolutional auto-encoders (STL-CAEs), is proposed, which can effectively extract features of bearing vibration signals. First, an ensemble learning strategy is proposed to obtain two auto-encoders that satisfy the strategy by optimizing the model parameters and structure. Then, a self-taught learning training method is proposed to solve the problem of little label data. Finally, ensemble learning and fault diagnosis is achieved by the SoftMax classifier. Applying the proposed method to the bearing data from Case Western Reserve University, the STL-CAEs have higher accuracy and generalization than common fault diagnosis methods such as CAE, CNN, SAE and EMD, and also have significant advantages in terms of diagnostic time and training time.