Online monitoring of dry-type transformers

Dry-type transformers are widely used in high-rise buildings, urban rail transit, data centers and other scenarios with stringent requirements for power supply safety due to their oil-free operation, excellent fire resistance, low maintenance requirements and other characteristics. ItsOnline monitoring technologyThrough real-time collection of key parameters during the operation of equipment, early identification and warning of hidden faults can be realized to avoid unplanned shutdowns, which is the core technical means to ensure the continuous and reliable operation of the power system. The following is a comprehensive description of the core parameters and purpose of monitoring, key monitoring technology principles, system composition, application value, development trends and application considerations.

I. Core monitoring parameters and purpose of monitoring

Faults in dry-type transformers are mostly generated withAbnormal increase in temperature, occurrence of partial discharges, deterioration of insulation properties, deviation of electrical parametersRelated, online monitoring needs to be carried out for the following core parameters in order to reach the goal of "early detection, accurate diagnosis and timely treatment":

Monitoring parameters	Core monitoring purposes	Associated Fault Types
Winding Temperature	Prevents overheating of the windings leading to accelerated aging of the insulating material and avoids turn-to-turn short-circuit failures.	Equipment overload, cooling fan failure, winding turn-to-turn insulation damage
Core temperature	Avoiding localized overheating caused by multi-point grounding of the core and abnormal increase of silicon steel sheet loss	Poor core grounding, loose or damaged core laminations
partial discharge	Identification of internal defects (e.g. air gaps, cracks) in the insulation system to prevent insulation breakdowns	Aging of insulation materials, winding surface dirt, manufacturing process defects
Insulation Resistance / Dielectric Loss	Evaluate the overall performance of the insulation system and determine the degree of moisture and aging of the insulation.	Dampness of insulation, surface creepage, deterioration of insulating material properties
Load Current / Voltage	Monitor the actual load condition of the equipment and analyze the impact of current imbalance on the equipment	Three-phase load imbalance, external short-circuit shock, overload operation
Ambient temperature and humidity	Correction of temperature monitoring data (ambient temperature directly affects the efficiency of equipment heat dissipation), early warning of insulation moisture caused by high humidity environments	Excessive environmental humidity triggers insulation creepage and insulation degradation

II. Key monitoring technology principles

Different monitoring parameters need to be combined with the structural characteristics of the dry-type transformer (oil-free, air-cooled or forced air-cooled) to choose the appropriate technology to ensure the accuracy and stability of the monitoring data, the specific technical principle is as follows:

1. Temperature monitoring: Combination of direct and indirect, eliminating temperature measurement blind spots

Dry-type transformer winding using epoxy resin package, the traditional contact temperature measurement method by the package structure limitations, need to use "indirect temperature measurement + direct temperature measurement" combination program:

The infrared temperature sensor receives the infrared energy radiated from the surface of the winding or the exposed part of the core, and calculates the corresponding temperature value according to the blackbody radiation law. This technology is easy to install without destroying the structure of the equipment; however, it is susceptible to environmental dust and light interference and requires regular cleaning and maintenance of the sensor lens.

The temperature sensing probe of the fluorescent fiber optic sensor is pre-buried or pasted in the hot spot area inside the winding, and the receiving end of the sensor is connected with the temperature demodulator. When working, the demodulator issued by the ultraviolet pulse excitation light through the fiber optic conduction to the temperature sensing probe, excitation probe within the fluorescent material to produce fluorescence; excitation light stops, the fluorescence afterglow through the fiber optic back to the demodulator, through the photoelectric conversion components converted to electrical signals, and then through the single-chip microcomputer calculation of fluorescence life, and ultimately based on the fluorescence life of the correspondence with the temperature of the temperature output temperature value. This technology has strong anti-magnetic interference ability (fiber is not conductive material), temperature measurement accuracy up to ± 0.5 ℃, fast response speed.

Synchronized acquisition of forced air cooling (AF) system fan operation current, speed signals, through the current value changes, speed abnormalities to determine whether the fan blocking, failure and other faults, to ensure that the cooling system is normal to play the role of heat dissipation.

2. Local discharge monitoring: focusing on anti-interference issues

Dry-type transformer running in a high-voltage environment, local discharge signal amplitude is weak (usually tens to thousands of picocuries), and susceptible to grid harmonics, switching operations and other electromagnetic interference, the need for "signal acquisition + anti-interference processing" combination of technologies to achieve effective monitoring:

At the hardware level, band-pass filters and shielded cables are used to suppress the interference signals; at the software level, the collected signals are processed through algorithms such as wavelet transform, threshold denoising, and singular value decomposition to eliminate the interference components and extract the effective localized discharge signals.

3. Insulation performance monitoring: assessment of insulation ageing trends

Dry-type transformer insulation materials (epoxy resin, glass fiber) long-term temperature, electric field, humidity will occur aging, need to be monitored by the following parameters to assess the state of the insulation system:

Using an on-line insulation resistance tester, when the transformer is out of power or in low load operation (to reduce electric field interference), apply a specified DC high voltage (e.g. 10kV) to the windings, and measure the insulation resistance value of the windings to ground and between the windings. When the insulation resistance value drops below 1/3 of the standard value, it indicates that the insulation system may have moisture or aging problems.

Through the high-voltage dielectric loss tester to apply AC high voltage, measure the insulation material in the electric field under the action of the energy loss, expressed in tanδ value. tanδ value is larger, indicating that the insulation loss is more serious, the higher the degree of aging; room temperature dry-type transformer tanδ value usually need to be controlled in 0.005 or less.

4. Electrical parameter monitoring: real-time control of equipment operating loads

Three-phase current and voltage signals are collected through current transformer (CT) and voltage transformer (PT), and power, power factor, load ratio and other parameters are calculated by combining with intelligent power collection module:

III. Components of the online monitoring system

The complete online monitoring system for dry-type transformers consists ofSensing layer, transport layer, analysis layer, application layerIt consists of four layers, forming a closed-loop operation mechanism of "data collection - transmission - analysis - early warning":

1. Perceptual layer: the basis for data collection

It consists of various types of sensors and data acquisition modules, and needs to meet the insulation requirements and anti-interference performance under high-voltage environment:

2. Transport layer: data transmission channels

It is responsible for transmitting the raw data collected in the sensing layer to the analyzing layer, and the stability and security of data transmission should be guaranteed:

3. Analytical layer: data-processing core

It consists of an edge computing gateway or a cloud server, which processes and analyzes the raw data through algorithms to determine the operating status of the device:

4. Application layer: user interaction interface

Display monitoring results and provide operation functions to users through local monitoring terminals, Web platforms and mobile APPs:

IV. Application value of online monitoring

Compared with the traditional "periodic inspection" mode, dry-type transformer online monitoring technology can bring significant safety and economic benefits:

Early identification of insulation aging, partial discharge and other hidden dangers (such as partial discharge surge usually indicates that the insulation breakdown may occur within 1~3 months), to reserve time for maintenance work, to avoid production interruptions due to sudden equipment failures (such as data center power outages can reach hundreds of thousands of dollars in economic losses per hour).

Traditional inspection relies on manual experience, and is prone to "overhaul" (e.g. replacement of components that have not reached the end of their life) or "leakage"; online monitoring is based on the actual operating status of the equipment to formulate an inspection plan (e.g. extension of the inspection cycle when the insulation resistance index is normal), which reduces the number of inspections and the cost of investment. The online monitoring can reduce the number of inspections and cost investment.

Through real-time monitoring of load and temperature, it avoids long-term overload or overheating operation of equipment, slows down the aging of insulation materials (research data shows that for every 10℃ reduction in winding temperature, the insulation life can be extended by 1 times), extends the service life of transformer by 5~8 years, and improves the efficiency of asset utilization.

Although dry-type transformer has no oil leakage fire risk, but the insulation aging may cause short-circuit fire; online monitoring system can be in the early stage of the fault linkage circuit breaker to cut off the power supply, to avoid fire, explosion and other safety accidents.

V. Technology development trends

With the power system to "intelligent, digital" transition, dry-type transformer online monitoring technology presents the following development direction:

A single parameter can not fully reflect the state of equipment operation, the future will realize the "temperature + local discharge + insulation + vibration" multi-parameter fusion analysis, through the AI algorithm to build equipment "health index", to improve the accuracy of fault diagnosis.

Traditional wired sensors have high installation complexity. In the future, passive wireless sensors (e.g., sensors based on electromagnetic induction and vibration energy collection) will be used more often to reduce installation costs, and will be suitable for old transformer retrofit projects.

Construct a digital twin model of dry-type transformer, combine online monitoring data with physical model, simulate the running status of the equipment under different loads and environmental conditions, and realize the whole life cycle management of "fault simulation - early warning - optimization of inspection and repair scheme".

Adopting the mode of "edge computing gateway preprocessing data + cloud big data analysis", it reduces the amount of data transmission (the edge side only uploads abnormal data), improves the real-time response capability (the edge side can realize the local equipment linkage control), and at the same time, uses the computing power of the cloud to realize the collaborative diagnosis of multiple equipments (e.g., the comparative analysis of the status of multiple transformers in the region).

VI. Application considerations

High-voltage side sensors need to meet the corresponding insulation level requirements (such as 10kV transformer sensor insulation level ≥ 35kV), to avoid safety accidents caused by insufficient insulation; outdoor sensors need to have IP65 and above protection level, to ensure that the waterproof, dustproof performance.

Sensor cables need to be shielded, avoid parallel laying with high-voltage cables (spacing ≥ 0.5m); partial discharge monitoring equipment needs to be far away from frequency converters, static reactive generators (SVG) and other harmonic sources to reduce the impact of electromagnetic interference.

Sensors need to be calibrated every 1~2 years (e.g., fluorescent fiber optic sensors are calibrated against standard thermometers) to avoid distortion of monitoring data due to sensor drift; clean the infrared sensor lens regularly and check the signal strength of wireless communication.

Encryption algorithms such as AES should be used in the data transmission process, and the cloud platform should be set up with graded access privileges (e.g., administrator and operation and maintenance personnel privilege differentiation) to prevent data leakage or malicious tampering.

In summary, dry-type transformer online monitoring technology is the key support for realizing the "state maintenance" of power equipment, which can significantly improve the reliability and economy of the power system by accurately collecting key parameters and intelligently analyzing the state of the equipment. As the technology continues to iterate, its application in the smart grid, new power systems will be more extensive.