Automated deep learning system for power line inspection image analysis and processing: architecture and design issues

The continuous growth in the scale of unmanned aerial vehicle (UAV) applications in transmission line inspection has resulted in a corresponding increase in the demand for UAV inspection image processing. Owing to its excellent performance in computer vision, deep learning has been applied to UAV inspection image processing tasks such as power line identification and insulator defect detection. Despite their excellent performance, electric power UAV inspection image processing models based on deep learning face several problems such as a small application scope, the need for constant retraining and optimization, and high R&D monetary and time costs due to the black-box and scene data-driven characteristics of deep learning. In this study, an automated deep learning system for electric power UAV inspection image analysis and processing is proposed as a solution to the aforementioned problems. This system design is based on the three critical design principles of generalizability, extensibility, and automation. Pre-trained models, fine-tuning (downstream task adaptation), and automated machine learning, which are closely related to these design principles, are reviewed. In addition, an automated deep learning system architecture for electric power UAV inspection image analysis and processing is presented. A prototype system was constructed and experiments were conducted on the two electric power UAV inspection image analysis and processing tasks of insulator self-detonation and bird nest recognition. The models constructed using the prototype system achieved 91.36% and 86.13% mAP for insulator self-detonation and bird nest recognition, respectively. This demonstrates that the system design concept is reasonable and the system architecture feasible .

Stability Analysis on Power System with Large Power Source

In this paper, the stability problems of power systems with large power source are analyzed with viewpoint of energy balance. The phenomena are studied when the energy delivery are “blocked” by faults that occur on some key transmission lines of a large power source within a power system, on the transmission lines between the interconnected power systems, or on some key buses. The cases are based on a practical power system and its interconnected systems.

Dynamic physical characteristics of DC arc on arcing horn for HVDC grounding electrode line

The dynamic physical characteristics of a DC arc on an arcing horn for a high voltage direct current(HVDC)grounding electrode line are significantly different from those of the switching device arc,secondary arc,AC fault arc and pantograph-catenary arc.In this work,an experimental platform for the DC arc on the arcing horn was built,and mechanisms of the arc column short circuit and arc root movement were studied.This work further analyzes the characteristics and mechanisms of the arc motion when wind speed and direction,magnetic field and the expansion angle of the electrode are varied.Arc root movement is more likely to occur at the upper electrode.There is a competitive relationship between arc expansion and the transferring effect.The effect of wind on the arc column is greater than the effect on the arc root.The magnetic field has a significant driving effect on both the arc column and the arc root.The research results provide a comprehensive experimental basis for forther probing the method of DC arc suppression,and the improvement of the arcing horn.

Construction and application of knowledge graph for grid dispatch fault handling based on pre-trained model

With the construction of new power systems,the power grid has become extremely large,with an increasing proportion of new energy and AC/DC hybrid connections.The dynamic characteristics and fault patterns of the power grid are complex;additionally,power grid control is difficult,operation risks are high,and the task of fault handling is arduous.Traditional power-grid fault handling relies primarily on human experience.The difference in and lack of knowledge reserve of control personnel restrict the accuracy and timeliness of fault handling.Therefore,this mode of operation is no longer suitable for the requirements of new systems.Based on the multi-source heterogeneous data of power grid dispatch,this paper proposes a joint entity–relationship extraction method for power-grid dispatch fault processing based on a pre-trained model,constructs a knowledge graph of power-grid dispatch fault processing and designs,and develops a fault-processing auxiliary decision-making system based on the knowledge graph.It was applied to study a provincial dispatch control center,and it effectively improved the accident processing ability and intelligent level of accident management and control of the power grid.

FPGA-based acceleration for binary neural networks in edge computing

As a core component in intelligent edge computing,deep neural networks(DNNs)will increasingly play a critically important role in addressing the intelligence-related issues in the industry domain,like smart factories and autonomous driving.Due to the requirement for a large amount of storage space and computing resources,DNNs are unfavorable for resource-constrained edge computing devices,especially for mobile terminals with scarce energy supply.Binarization of DNN has become a promising technology to achieve a high performance with low resource consumption in edge computing.Field-programmable gate array(FPGA)-based acceleration can further improve the computation efficiency to several times higher compared with the central processing unit(CPU)and graphics processing unit(GPU).This paper gives a brief overview of binary neural networks(BNNs)and the corresponding hardware accelerator designs on edge computing environments,and analyzes some significant studies in detail.The performances of some methods are evaluated through the experiment results,and the latest binarization technologies and hardware acceleration methods are tracked.We first give the background of designing BNNs and present the typical types of BNNs.The FPGA implementation technologies of BNNs are then reviewed.Detailed comparison with experimental evaluation on typical BNNs and their FPGA implementation is further conducted.Finally,certain interesting directions are also illustrated as future work.

Dynamics of an impurity ion transport in oil-paper insulation under various electric fields

The ionic transport process is an important counterpart to charge transport in oil-paper insulation,and it significantly impacts oil flow electrification at the oil-paper interface.Despite this,the dynamics of this phenomenon and the underlying mechanisms remain unclear,particularly at the molecular level.To understand this fundamental aspect,we conduct Molecular Dynamics study on the transport behaviour of an impurity ion in different oil-paper insulation models under various external electric fields.Different influence factors,such as external electric fields,temperatures,and local structural characteristics,are investigated in relation to the corresponding ionic mobility in different polymer models.According to the simulations,ionic mobility and its response to electric fields are higher in weaker electrostatic models.As a result,mineral oil exhibits the highest ionic mobility and the most substantial enhancement of ionic mobility by external electric fields,followed by vegetable oil,oil-paper blends,and insulating paper.This significant deviation in ionic mobility between oil and paper leads to the formation of an electric double layer near the oil-paper interface.The underlying physical mechanisms of different ionic mobility and its response to the electric field in different polymer models could be explained by the different polymer structural influences in terms of interaction energy and coordination numbers in a static manner,as well as by the interactions between the impurity ion and its surrounding atoms in terms of lifetime correlation functions and velocity autocorrelation functions in a dynamic manner.In addition,the influence of temperature on ionic mobility in mineral and vegetable oil is examined,and their activation energies are calculated.Advancing in the fundamental understanding of the dynamics of the ion transport process in oil-paper insulation is vital to improving their insulating properties for oil-impregnated power transformers.

Analysis of Excitation Characteristics of Ultra High Frequency Electromagnetic Waves Induced by PD in GIS

The understanding of the excitation mechanism of ultra high frequency (UHF) electromagnetic waves (EW) is essential for ap- plying UHF method to partial discharge (PD) detection. Since the EW induced by PD in gas insulated switchgear (GIS) contains not only transverse electromagnetic (TEM) wave, but also high-order transverse electric (TE) and high-order transverse magnetic (TM) waves, we analyzed the proportions between the TEM wave and the high order waves, as well as the influence of the PD position on this proportion, using the finite different time domain (FDTD) method. According to the unique characteristics of the waves, they are separated only ap- proximately. It is found that the high-order mode is the main component, more than 70%, of the electric field around the enclosure of GIS, and that with the increasing distance between PD source and inner conductors, the low frequency ( below about 800 MHz) component of EW decreases, but the high frequency component (above 1 GHz) increases, meanwhile the proportion of high-order components in EW could reach 77% from 70%. It concluded that the closer the PD source to the enclosure is, the easier high order EW may be excited.

Study on Earth Surface Potential and DC Current Distribution around DC Grounding Electrode

DC magnetic biasing problem,caused by the DC grounding electrode, threatened the safe operation of AC power grid. In this paper, the characteristics of the soil stratification near DC grounding electrode was researched. The AC-DC interconnected large-scale system model under the monopole operation mode was established. The earth surface potential and DC current distribution in various stations under the different surface thickness was calculated. Some useful conclusions are drawn from the analyzed results.

Study on the Influence of Grounding Conductor Structure on Its Grounding Impedance Spectrum Response

The grounding diagnosis of pole and tower plays an important role, and the tower is characterized by its unique characteristics. It is necessary to develop a new method for the diagnosis of pole and tower. Frequency diagnosis is applied in many aspects, so it is necessary to analyze the frequency response characteristics of pole and tower. The influence of the frequency response of pole and tower on the radius of the pole and the soil resistivity is analyzed under three different grounding structures using CDEGS software, and the differences and similarities of spectral response under different structures are discussed.

Control Method for Additional Damper in Hydro-turbine Speed Governor of Hydro-dominant Power Systems

With the rapid development of the national interconnected power grid,fast automatic excitation regulators and fast turbine governors are being widely used.In the current systems,some power plants adjust their responses for turbine governor systems to a very sensitive level to meet the primary frequency regulation assessment index,which actively brings about negative damping.Under these circumstances,the ultralow frequency oscillation(ULFO)phenomenon emerged in hydrodominant power systems.In order to solve the problem of ULFO,the hydro-turbine governor additional damper(GAD)based on phase compensation,which adapts to ULFO,is developed.The design of the control system is relatively simple.As a phase compensation based on the field testing and modeling of the turbine governor,GAD can provide the power system with positive damping,which contributes to the stability and security of the system.According to the impact of the mechanism of the turbine governor on the dynamic stability of the power system and the research of modeling and parametric testing,the proposed method not only supports positive damping,but also eliminates the negative damping of the speed governing system,which can significantly suppress the ultralow frequency oscillation and make it feasible to configure the GAD parameters according to field testing.

Low-carbon Operation of Combined Heat and Power Integrated Plants Based on Solar-assisted Carbon Capture

Accelerating the development of renewable energy and reducing CO_(2)emissions have become a general consensus and concerted action of all countries in the world. The electric power industry, especially thermal power industry, is the main source for fossil energy consumption and CO_(2)emissions. Since solvent-based post-combustion carbon capture technology would bring massive extra energy consumption, the application of solar-assisted carbon capture technology has attracted extensive attention. Due to the important role of coal-fired combined heat and power plants for serving residential and industrial heating districts, in this paper, the low-carbon operation benefits of combined heat and power integrated plants based on solar-assisted carbon capture(CHPIP-SACC) are fully evaluated in heat and power integrated energy system with a high proportion of wind power. Based on the selected integration scheme, a linear operation model of CHPIP-SACC is developed considering energy flow characteristics and thermal coupling interaction of its internal modules. From the perspective of system-level operation optimization, the day-ahead economic dispatch problem based on a mix-integer linear programming model is presented to evaluate the low-carbon benefits of CHPIP-SACC during annual operation simulation. The numerical simulations on a modified IEEE 39-bus system demonstrate the effectiveness of CHPIP-SACC for reducing CO_(2)emissions as well as increasing the downward flexibility. The impact of different solar field areas and unit prices of coal on the low-carbon operation benefits of CHPIP-SACC is studied in the section of sensitivity analysis.

Research on vibration energy harvesting technology of power equipment based on alternating magnet array

Vibration with a frequency of 100 Hz is widely distributed in the power equipment,and it can provide a new way to supply energy for sensors by vibration energy harvesting.The vibration energy harvesting method based on electromagnetic induction principle was studied through the arrayed structure of magnets and coils.Static magnetic field models were established for four magnet array structures and it was found that the alternating magnet array has the largest magnetic flux and magnetic flux gradient.Based on the alternating magnet array,prototypes of energy harvester with vertical and parallel movement mode were proposed.Through structural parameter optimisation analysis,two different energy harvesters were fabricated and it was found that the energy harvester with a parallel movement mode has better output performances.The energy harvester could provide output voltage/current and power of 8.35 V/17.39 mA and 15.13 mW(matched resistance is 200Ω)at an acceleration of 5 m⋅s^(-2).The 100 mF capacitor could be charged to 2.72 V within 300 s,and the final voltage of the capacitor is greater than 3 V,which could sustainably drive commercial wireless temperature/humidity sensors.

Improved Identification Method and Fault Current Limiting Strategy for Commutation Failure in LCC-HVDC

Line-commutated converter based high-voltage direct-current(LCC-HVDC)transmission systems are prone to subsequent commutation failure(SCF),which consequently leads to the forced blocking of HVDC links,affecting the operation of the power system.An accurate commutation failure(CF)identification is fairly vital to the prevention of SCF.However,the existing CF identification methods cause CF misjudge or detection lag,which can limit the effect of SCF mitigation strategy.In addition,earlier approaches to suppress SCF do not clarify the key factor that determines the evolution of extinction angle during system recovery and neglect the influence.Hence,this paper firstly analyzes the normal commutation process and CF feature based on the evolution topology of converter valve conduction in detail.Secondly,the energy in the leakage inductance of converter transformer is presented to characterize the commutation state of the valves.Then a CF identification method is proposed utilizing the leakage inductance energy.Thirdly,taking the key variable which is crucial to the tendency of extinction angle during the recovery process into account,a fault current limiting strategy for SCF mitigation is put forward.Compared with the original methods,the proposed methods have a better performance in CF identification and mitigation in terms of detection accuracy and mitigation effect.Finally,case study on PSCAD/EMTDC validates the proposed methods.

Dynamic fragments awareness based virtual network mapping strategy of elastic optical networks

As virtual networks services emerge increasingly with higher diversification, the issue of spectrum fragments presents great challenge to the elastic optical networks(EON), especially under heaven services burdens. Aimed to solve this problem, this article proposes a dynamic fragments awareness based virtual network mapping(DFA-VNM) strategy of elastic optical network. In this proposed approach, the dynamic fragments awareness model of it is established, which takes available bandwidth demand and spectrum fragment degree into consideration. Moreover, the dynamic fragments awareness based virtual network mapping strategy makes full advantage of real-time fragments awareness result to conduct virtual network service mapping operation with less fragments and lower blocking rate. Testing results show that the proposed approach is able to improved services supporting ability of EON.

Enhancement of Soil Discharge Channels on Potential Surrounding Buried Cables Under Impulsive Currents

Substations have a large number of signal transmission cables beneath the ground.Both the insulation safety and signal reliability of the cables are affected severely by the electromagnetic field.Under high-amplitude impulsive currents,the dispersion of currents can cause soil discharge and thus cause unexpected distortions in an electromagnetic field.This paper focuses on the distortions of the electric field.In general,soil discharge channels occur in the vicinity of the independent rod.Closer development of the channel might enhance the electric field distribution and the potential surrounding the outer insulation of the cables(i.e.the surface potential on the cable).Therefore,this paper establishes a platform for observing the soil discharge channel and measuring the surface potential.Direction characteristic of the channel is extracted from the captured image of soil discharge channels and the surface potential is obtained by the measured coupling capacitive current on the shield experimentally.This paper also presents an improved model considering a dynamic growing discharge channel for the transient analysis of the grounding electrode.Study results show the surface potential increases as the discharge channel approaches the cable.To quantify this enhancement effect,the ratio of the highest to the lowest value of surface potential in different directions is taken as the multiple of the surface potential increase.The calculated multiples of the surface potential increase are in the range of 1 to 1.64 times under different conditions by the improved model.Therefore,taking the soil discharge channel into account is helpful to accurately analyze the impulsive interference of buried cables.

Self-sensing active magnetic bearing using real-time duty cycle

In a self-sensing active magnetic bearing (AMB) system driven by pulse width modulation (PWM) switching power amplifiers, the rotor position information can be extracted from coil current and voltage signals by a specific signal demodulation process. In this study, to reduce the complexity of hardware, the coil voltage signal was not filtered but measured in the form of a duty cycle by the eCAP port of DSP (TMS320F28335). A mathematical model was established to provide the relationship between rotor position, current ripple, and duty cycle. Theoretical analysis of the amplitude-frequency characteristic of the coil current at the switching frequency was presented using Fourier series, Jacobi-Anger identity, and Bessel function. Experimental results showed that the time-varying duty cycle causes infinite side frequencies around the switching frequency. The side frequency interval depends on the varying frequency of the duty cycle. Rotor position can be calculated by measuring the duty cycle and demodulating the coil current ripple. With this self-sensing strategy, the rotor system supported by AMBs can steadily rotate at a speed of 3000 r/min.