Zhongyong Zhao

Transformer winding deformation diagnostic system using online high frequency signal injection by capacitive coupling

Transformer winding deformation is common among all sorts of transformer failures. Cumulative deformation can eventually burgeon into catastrophic faults and result in entire network outage. It is possible to detect the early signs of faults with continuous online monitoring transformer. However, frequency response analysis (FRA) is considered to be a useful and accurate tool for sufficient detection. This paper aims at proposing a novel diagnostic system for online monitoring power transformer winding deformation based on FRA. In order to realize online monitoring transformer winding, the system uses the capacitive coupling method to inject controllable nanosecond pulses, which function as the excitation signal of winding, and to obtain the response signal. This proposed method may extend frequency range for analysis and perhaps could early detect minor winding movement and looseness. Transformer experiments show great prospect in the application of the system.

Impact of capacitive coupling circuit on online impulse frequency response of a power transformer

Detecting the early signs of mechanical failures of power transformer winding is necessary and is possible with online monitoring techniques. Online impulse frequency response analysis (IFRA) is a promising diagnostic method when a transformer is in service. This paper examines the unrevealed problem existing in the method, namely, the impact of bushing capacitive coupling circuit on online impulse frequency response. An equivalent electrical model of capacitive coupling circuit and transformer winding is established. The frequency response of the capacitive coupling circuit is obtained to study its influence on online impulse frequency response. The parameter variations of capacitive coupling circuit caused by coupling capacitance variation and bushing dielectric breakdown are simulated to investigate their influence on online impulse frequency response signatures. A few experiments are eventually performed to verify the theoretical analysis and simulation results. This paper contributes to the application of online IFRA.

Noninvasive method for online detection of internal winding faults of 750 kV EHV shunt reactors

Seven hundred fifty kilovolt extra-high voltage (EHV) shunt reactors are important power equipment that stabilize the normal operation of an entire 750 kV power grid. Internal winding faults of a reactor could be a difficult issue when the reactor is in service, and result in sudden equipment failure and significant economic losses. To prevent sudden catastrophic failure of equipment, online detection should be implemented to evaluate the state of the reactor properties when requested. In this paper, a noninvasive method based on online impulse frequency analysis was proposed and developed to detect internal winding faults of a 750 kV EHV shunt reactor. Initially, the safe implementation of the online detecting device was verified before field installation. Unlike in previous work, the online detecting device was modified and improved according to the specific situation of shunt reactors. Original fingerprints of 750 kV EHV shunt reactors were obtained and analyzed after site installation and normal operation of the detecting device. A fault simulation test was implemented to validate the method. This paper provides guidance on actual implementation of online impulse frequency response analysis on power equipment.

Improved Online Monitoring Method for Transformer Winding Deformations Based on the Lissajous Graphical Analysis of Voltage and Current

Power transformers have important roles as voltage converters in substations, which directly affect the safe operation of power grids. Online monitoring of winding deformation, which is one of the most common transformer faults, is crucial in protecting transformers from damage, particularly those induced by short-circuit current failures, and ensuring their safe operation. This study proposes an improved online monitoring method for winding deformations based on the Lissajous graphical analysis of voltage and current. A practical measurement system is also presented and discussed in this paper. A load normalization method is proposed to solve the problem of Lissajous diagrams changing as load varies. Finally, transformer and winding experiments are performed to verify the validity of the proposed load normalization method and the feasibility of the proposed monitoring method in distinguishing winding deformations with different degrees and locations.

Investigation of condition monitoring of transformer winding based on detection simulated transient overvoltage response

In order to keep the original connection of transformer, a novel method and apparatus is proposed for condition monitoring of transformer winding based on the detection of transient overvoltage signals. Firstly, a lumped parameter model of transformer winding is established to verify the equivalence in frequency response analysis between the overvoltage and sweep frequency voltage. Then experimental study is conducted on different types of winding faults, to conform the capability to identify the winding fault types. And the investigation of condition monitoring of 110 kV transformer is conducted. The over-voltage frequency response method is verified and can be applied to online monitoring of field windings.

Determination of nanosecond pulse parameters on transfer function measurement for power transformer winding deformation

Transfer function method is now a widely acceptable tool to diagnose transformer winding deformations. A sweep frequency sine wave generator is often used to excite the different modes of resonance and anti-resonances. However, it is time consuming. Nanosecond square wave pulse signal offers an alternative that can serve the same objective. However, as so far, there is no certain criterion for selecting pulse parameters. This paper provides a comprehensive method for the determination of nanosecond square wave pulse parameters for transfer function evaluation of power transformer for winding deformation studies.

High-voltage subnanosecond pulsed power source with repetitive frequency based on Marx structure

The electromagnetic radiation effect of high-voltage, subnanosecond pulsed electric fields attracts strong interest from researchers because of its significant development potential in biological treatments, particularly in noninvasive diagnosis and treatment. But up to now, high-voltage subnanosecond pulse generators are widely used for national defense and military, and they are difficult to implement successfully in the civil medical field. A compact, self-contained, repetitive frequency, high-voltage subnanosecond pulsed power source is proposed in this paper. It was designed, built, and tested successfully. Based on a four-stage, low-inductance Marx generator, the pulsed source produces subnanosecond rise-time pulses. A chopping switch was designed to cut off unformed signals and generate subnanosecond pulses. A measurement device, based on the principle of a capacitive voltage divider, was also constructed to determine both the amplitude and the rise-time of the pulse delivered by the source. Preliminary tests show that the source can produce repetitive frequency pulses with a peak value that exceeds 30 kV, as well as rise-time and pulse width (full wave at half maximum) within 1 ns under atmospheric pressure conditions. The pulse amplitude may be extended to hundreds of kilovolts by filling the switch system with an inert gas. The rise-time would be shortened as well. The pulsed power source shows optimistic prospects in the biological fields.

Understanding online frequency response signatures for transformer winding deformation: Axial displacement simulation

The power transformer is considered as the most critical and expensive device in substation, however, the irreversible transformer winding mechanical deformation can eventually develop into catastrophic failure if no further steps are taken in a proper way, which would cause the outage of transformer and the significant economic losses. Online frequency response analysis (FRA) has been proven to be a promising tool for condition monitoring and diagnosing of winding deformation. Online FRA relies on graphic comparison of signatures, but up to now, there is no standard and practical interpretation code for signatures classification and quantification. This paper particularly studies the characteristic of online FRA signatures under the winding axial displacement mode, in which the 3D finite element electromagnetic analysis and online transformer equivalent high frequency electrical model are established as auxiliary tools to precisely emulate winding axial displacement. Results of this simulation will provide guidance on understanding online frequency response signatures.

Online detection method for power transformer winding deformation and movement by impulse signal injection: Safety and feasibility validation

Power transformer assumes the important role in transportation of power energy; however, the initial faults inside transformer would eventually develop into catastrophic failure during its operation, which could result in the outage of equipment. Winding deformation and movement are considered to be the most common faults inside transformer. This paper proposed an online detection method for power transformer winding deformation and movement. As the key component, a capacitive coupling sensor mounting on the bushing was used to realize impulse signal injection. The safety of this sensor was primarily verified on a 110 kV bushing. In addition, the feasibility of the proposed method was validated through the experiment. The results show that the proposed method may reach the potential to detect winding deformation and movement when transformer is in service.

Practical online detection of internal winding faults of EHV shunt reactors based on impulse frequency response analysis

Internal winding faults of extra-high voltage (EHV) shunt reactors could be a serious threaten to the normal operation of power grid. Initial winding faults could develop into catastrophic failure, resulting in sudden outage of equipment and significant economic losses. It is necessary to implement online detection method to evaluate the status of properties. This paper proposed a practical method for online detecting the internal winding faults of EHV shunt reactors; it is based on impulse frequency response analysis. The basic principle and the detection devices of the method were introduced, respectively. Then the original frequency response curves of EHV shunt reactors were obtained through field online tests, at last, the offline winding fault simulation test was carried out to further investigate the proposed method. This paper is a contribution for the application of impulse frequency response analysis on high voltage equipment.