DEFECT MONITORING APPARATUS AND DETECTION METHOD FOR TRANSFORMER OIL CONSERVATOR BASED ON EDGE COMPUTING

Abstract

Disclosed are a defect monitoring apparatus and detection method for a transformer oil conservator based on edge computing. The monitoring apparatus includes: a sensing terminal, an edge intelligent gateway, a monitoring terminal, an oil conservator defect monitoring cloud platform, and a client terminal. The edge intelligent gateway and the monitoring terminal are both disposed in a substation. The monitoring terminal is configured to access the edge intelligent gateway. The sensing terminal includes: an airflow sensor, an oil temperature sensor, an ambient temperature sensor, and a microwave oil level measurement assembly. The sensing terminal sends detection data information to the edge intelligent gateway, and the edge intelligent gateway stores and determines a category of the detection data information. The present disclosure can recognize capsule damage, partial blockage of a breathing circuit, leakage in the breathing circuit, moisture impurities in a breathing tube, and false oil level.

Description

TECHNICAL FIELD

The present disclosure relates to the technical field of oil conservators of electric power oil-immersed equipment, and in particular, to a defect monitoring apparatus and detection method for a transformer oil conservator based on edge computing.

BACKGROUND

An oil-immersed transformer is the most important device in transmission and substation equipment, and its safe and reliable operation is crucial. An oil conservator is one of the most important parts of the oil-immersed transformer, and is mainly used to buffer a volume change caused by an oil temperature change in an oil tank during the operation of the transformer. The oil conservator is disposed above the oil tank of the transformer, and is usually connected to a dehydrating breather and kept open to the outside atmosphere. When the oil temperature of the transformer increases, the oil volume expands, and the oil conservator exhales air to the outside; when the oil temperature of the transformer decreases, the oil volume decreases and air is inhaled from the atmosphere. This is the breathing of the oil conservator. In a commonly used capsule oil conservator, a capsule bag is added to an inner wall of the oil conservator, and the interior of the capsule bag is in communication with the atmosphere through a dehydrating breather and connecting tubes thereof. A bottom surface of the capsule bag floats on and is closely attached to the oil conservator, such that no air exists between the capsule bag and the oil level and the oil level is isolated from air, which prevents oil oxidation and prevents external moisture and impurities from entering.

In the power industry, due to the long operation of the transformer, failures often occur in the oil conservator, bringing great difficulty to transformer operation and maintenance, and even causing transformer damage and shutdown in serious cases. In a HVDC transmission project, there have been several cases of rupture of oil conservator capsule bags of the converter transformer, which seriously threatened the stable operation of the HVDC transmission project. Currently, the main problems in the operation of the oil conservator are as follows: First, the aging, cracking or damage of the capsule due to improper installation causes the capsule to lose its protective function, while it is difficult to observe whether the capsule is broken or folded because it is inside the oil conservator. Second, due to transformer vibration, poor installation quality and other factors, connection between the oil conservator and other parts as well as pipelines are prone to leakage, resulting in the entry of air moisture, affecting the quality of the oil and thus reducing the insulation effect. Third, the desiccant of the dehydrating breather is clumped with moisture and clogs the dehydrating breather, resulting in the blockage of the breathing circuit of the oil conservator, which causes oil spray and shutdown of the transformer in serious cases. Fourth, a false oil level occurs in the oil conservator, which is not easy to detect in the daily inspection, thus seriously affecting the safe operation of the transformer.

In most of the currently disclosed technologies, the method for oil conservator detection of a power transformer is to install a sensor inside the capsule. However, the sensor is prone to failure during long-term operation of the transformer and may cause other problems for the capsule. For example, in the European Patent Application No. 12741136.1 and the paper “Research on Device for Monitoring Rupture of Oil Pillow Capsule of Transformer” (Electronic World, 2015, No. 24), the capsule damage defect is monitored by installing a light source and a sensor inside the capsule to compare the light signal received by the sensor before and after the capsule damage, and the placement of sensors inside the capsule may bring other negative effects to the capsule in long-term operation. The Chinese Patent Application No. 202110178347.5 discloses a sensor group of an online intelligent monitoring device for an oil conservator and a health status value calculation method, in which a health status value of an oil conservator is obtained by calculation based on 9 sensors and corresponding parameter signals thereof. However, the process of calculating the health status value does not consider the operation principle of the oil conservator and involves as many as 9 sensors. These internal sensors induce other problems during long-term operation and lead to the failure of the oil conservator of the transformer, causing damage to the transformer.

SUMMARY

The present disclosure overcomes the shortcomings of the prior art and provides a defect monitoring apparatus and detection method for a transformer oil conservator based on edge computing, to effectively monitor and recognize defects of the oil conservator, such as capsule damage, partial blockage of a breathing circuit, leakage in the breathing circuit, moisture impurities in a breathing tube, and false oil level. According to a relationship between relative changes in an oil temperature and flow rates of air exhaled and inhaled from the oil conservator, oil level of the oil conservator is detected using microwaves, to detect and diagnose defects of the oil conservator in a non-contact manner. Once the defects of the oil conservator are detected and diagnosed, in the present disclosure, an alarm command is sent to a client terminal of a transformer operator in real time through a cloud platform, which gives the transformer operation and maintenance personnel valuable processing time for maintenance and repair, thereby greatly improving the efficiency and further ensuring the safe operation of power equipment.

To solve the above technical problems, the technical solutions adopted by the present disclosure are as follows:

A defect monitoring apparatus for a transformer oil conservator based on edge computing includes: a sensing terminal, an edge intelligent gateway, a monitoring terminal, an oil conservator defect monitoring cloud platform, and a client terminal. The edge intelligent gateway and the monitoring terminal are both disposed in a substation. The monitoring terminal is configured to access the edge intelligent gateway. The sensing terminal includes: an airflow sensor, an oil temperature sensor, an ambient temperature sensor, and a microwave oil level measurement assembly. The airflow sensor is disposed on a breathing tube, and is configured to measure a breathing airflow rate of an oil conservator. The oil temperature sensor is disposed at a lower portion of a side of the oil conservator, and a probe of the oil temperature sensor extends into the oil conservator and is immersed in transformer oil. The oil temperature sensor is configured to measure an oil temperature of the oil conservator. The ambient temperature sensor is disposed outside the oil conservator. The ambient temperature sensor is configured to measure an ambient temperature. The microwave oil level measurement assembly is disposed at a top of an inner wall of the oil conservator, and the microwave oil level measurement assembly is away from a capsule. The microwave oil level measurement assembly is configured to measure a height of an oil level of the oil conservator.

The sensing terminal is configured to send detection data to the edge intelligent gateway, and an original oil level gauge of the oil conservator is configured to send data information thereof to the edge intelligent gateway. The edge intelligent gateway is configured to store and evaluate the detection data, and send an evaluation result indicating a defect and corresponding data to the oil conservator defect monitoring cloud platform. The oil conservator defect monitoring cloud platform is configured to send an alarm command to the client terminal. The client terminal may be a personal computer (PC), a tablet computer, or a smart phone. The alarm command may be an email prompt alarm or a short message prompt alarm. Meanwhile, the transformer operation and maintenance personnel can use the client terminal to access the oil conservator defect monitoring cloud platform, to query specific data information. The monitoring terminal is configured to access the edge intelligent gateway. Operation and maintenance personnel in the substation can access data content in the edge intelligent gateway through the monitoring terminal at any time, which facilitates operation of local personnel.

Further, the microwave oil level measurement assembly includes: a microwave transmitting antenna and a microwave receiving antenna, where the microwave transmitting antenna and the microwave receiving antenna are both oriented towards the oil face, upper ends of the microwave transmitting antenna and the microwave receiving antenna are both fastened at the top of the inner wall of the oil conservator, and a transmitting end of the microwave transmitting antenna is located in a same horizontal plane with a receiving end of the microwave receiving antenna.

Further, the sensing terminal communicates with the edge intelligent gateway through a ZigBee network; the edge intelligent gateway communicates with the monitoring terminal through the ZigBee network, and the edge intelligent gateway communicates with the oil conservator defect monitoring cloud platform through a General Packet Radio Service (GPRS) network.

Further, the airflow sensor is disposed close to a dehydrating breather, such that data detection of the airflow sensor is more accurate.

Further, the airflow sensor is configured to adopt bidirectional airflow rate measurement, with a measurement range of −200 slpm to +200 slpm, a total error band less than or equal to 3% of a reading, and a response time less than or equal to 1 ms.

The oil temperature sensor and the ambient temperature sensor are both platinum resistance sensors, with a measurement range of −40° C. to 120° C. and accuracy of ±0.5° C.

The edge intelligent gateway is configured to adopt an ARM Cortex-A7 processor, with 1 GB RAM and 8 GB Embedded Multi Media Card (eMMC).

A defect detection method for a transformer oil conservator based on edge computing includes the following steps:

S1: monitoring, by a sensing terminal, the following data in real time: a breathing airflow rate of an oil conservator, an oil temperature of the oil conservator, an ambient temperature, a microwave transmitting power, and a microwave receiving power;

S2: determining whether the oil conservator has a false oil level defect: calculating an actual oil level height D₁ according to the microwave transmitting power and the microwave receiving power; sending, by an original oil level gauge of the oil conservator, data information to an edge intelligent gateway; calculating, by the edge intelligent gateway, a difference between the actual oil level height D₁ and a display value D of the original oil level gauge of the oil conservator; and when the difference is greater than 3 cm, determining that the oil conservator has a false oil level defect; otherwise, determining that the oil conservator has no false oil level defect;

S3: sampling real-time monitoring data, to form a breathing airflow rate curve; when a change in the ambient temperature is within 10° C. and a change in the oil temperature of the oil conservator is within 4° C., comparing a peak breathing airflow rate V_m of the oil conservator monitored in real time with an average peak breathing airflow rate V_a of the oil conservator in the passing month under the same condition; if a difference is greater than or equal to 15%, performing step S4; otherwise, determining that the following four types of defects do not exist: capsule damage, partial blockage of a breathing circuit, leakage in the breathing circuit, or moisture impurities in a breathing tube;

S4: when V_m≥1.2V_a, t_h≤0.9t_ha, and t_x≤0.9t_xa, determining that a capsule damage defect exists;

where t_h is a single expiration duration monitored in real time, t_x is a single inhalation duration monitored in real time, t_ha is an average single expiration duration in the passing month, and t_xa is an average single inhalation duration in the passing month;

when V_m≤0.8V_a, t_h≥1.1t_ha, and t_x≥1.1t_xa, determining that a defect of partial blockage of the breathing circuit exists;

when V_m≤0.8V_a, |t_h−t_ha|<0.05t_ha, and |t_x−t_xa|<0.05t_xa, determining that a defect of leakage in the breathing circuit exists; and

when V_m≤0.85V_a and an airflow rate zero-crossing number x₀ of a single expiration and inhalation process increases by 2 times or more compared with an average airflow rate zero-crossing number x_a0 in the passing month, determining that a defect of moisture impurities in a breathing tube exists;

S5: calculating a comprehensive evaluation coefficient for defect detection of the oil conservator, to obtain defect severity;

S6: when the defect severity is a general defect or a severe defect, sending, by the edge intelligent gateway, an evaluation result and corresponding data to an oil conservator defect monitoring cloud platform, where the evaluation result includes the comprehensive evaluation coefficient and defect information; and

S7: sending, by the oil conservator defect monitoring cloud platform, an alarm command to a client terminal, such that the transformer operation and maintenance personnel access the oil conservator defect monitoring cloud platform through the client terminal to check specific information.

Further, a formula for calculating the actual oil level height D₁ in step S2 is as follows:

D ₁ =H−d−h

where H is a height from the bottom to the top of the interior of the oil conservator, d is a distance from a transmitting end of a microwave transmitting antenna to an oil level, and h is a distance from the transmitting end of the microwave transmitting antenna to the top of the interior of the oil conservator;

$d=\sqrt{\frac{P_t{G}_t{G}_r{\lambda }^2}{4{\pi }^2{P}_r}-\frac{1}{4}{S}^2}$

where S is a distance from the transmitting end of the microwave transmitting antenna to a receiving end of a microwave receiving antenna, λ is a wavelength, P_t is a microwave transmitting power of the microwave transmitting antenna, G_t is a gain of the microwave transmitting antenna, P_r is a detected power of the microwave receiving antenna, and G_r is a gain of the microwave receiving antenna; microwaves transmitted by the microwave transmitting antenna are reflected by the oil level and then enter the microwave receiving antenna; the microwave receiving antenna detects a power and a gain; as a height of the oil level changes, the microwave power received by the microwave receiving antenna also changes, and thus d can be calculated based on the above data.

Further, a formula for calculating the comprehensive evaluation coefficient c_k for defect detection of the oil conservator is as follows:

$c_k=1.2p_k^2+0.4d_k^{\frac{1}{3}}+0.7x_k^{\frac{2}{3}}+1.5j_k^{\frac{2}{3}}+0.2s_k^{1.2}$

where when a capsule damage defect exists, a characteristic value p_k is 2; otherwise, p_k is 1;

when a defect of partial blockage of the breathing circuit exists, a characteristic value d_k is 2; otherwise, d_k is 1;

when a defect of leakage in the breathing circuit exists, a characteristic value x_k is 2; otherwise, x_k is 1;

when a defect of moisture impurities in a breathing tube exists, a characteristic value s_k is 2; otherwise, s_k is 1;

when a defect of false oil level of the oil conservator exists, a characteristic value j_k is 2; otherwise, j_k is 1; and

the defect severity is determined according to a value of the comprehensive evaluation coefficient c_k for defect detection of the oil conservator:

when c_k=4, the defect severity is no defect;

when 4<c_k≤4.8, the defect severity is the general defect; and

when c_k>4.8, the defect severity is the severe defect.

Compared with the prior art, the present disclosure has the following beneficial effects:

The airflow sensor, oil temperature sensor and ambient temperature sensor used in the present disclosure are simple to install and highly practical. By monitoring relative changes of the oil temperature of the transformer and analyzing changes of parameters related to breathing of the oil conservator through comparison with historical breathing data of the transformer, in combination with microwave detection of the oil level of the oil conservator, common defects of the oil conservator, such as capsule cracking and false oil level, can be comprehensively studied and judged, and the defect severity of the oil conservator can be evaluated. The present disclosure is an original technology, and the determination method is clear and operative.

The present disclosure adopts a non-contact detection method, which has no negative effect on the operation of the transformer, and the detection is safe, reliable and effective, solving the sensing failure during long-term operation of multiple built-in sensors and other quality problems induced by improper installation in the conventional method. The oil conservator monitoring apparatus of the present disclosure is based on network architecture of edge computing, which enables the operation and maintenance personnel to grasp the operation status of the oil conservator remotely at any time, and also solves the problems of high latency, unstable network and low bandwidth under the traditional Internet of Things cloud computing.

Through the present disclosure, the breathing condition of the transformer is monitored in real time, and the circulation of the oil in the transformer body is known in time, to avoid damage to the transformer caused by local overheating due to poor oil circulation. The present disclosure greatly saves manpower and material resources, and at the same time strongly improves the level of safe operation of the transformer as well as intelligent operation and maintenance of the substation, and has broad application prospects.

The present disclosure adopts the edge intelligent gateway with a computing function. The edge intelligent gateway can realize storage and processing of detection data information. Moreover, the edge intelligent gateway can be directly accessed by the existing monitoring terminal in the substation, which greatly reduces the cost of transformation.

BRIEF DESCRIPTION OF THE DRAWINGS

The present disclosure is described in detail below with reference to the accompanying drawings.

FIG. 1 is a structural diagram of the present disclosure.

FIG. 2 is a schematic diagram of installation of a sensing terminal according to the present disclosure.

FIG. 3 is a diagram of an airflow rate curve when a capsule damage defect occurs according to the present disclosure.

FIG. 4 is a diagram of an airflow rate curve when a defect of partial blockage of a breathing circuit occurs according to the present disclosure.

FIG. 5 is a diagram of an airflow rate curve when a defect of leakage in a breathing circuit occurs according to the present disclosure.

FIG. 6 is a diagram of an airflow rate curve when a defect of moisture impurities in a breathing tube occurs according to the present disclosure.

FIG. 7 is a schematic diagram of a microwave oil level measurement assembly according to the present disclosure.

In the drawings: 1 is an oil conservator, 2 is a capsule, 3 is a breathing tube, 4 is a dehydrating breather, 5 is an airflow sensor, 6 is an oil temperature sensor, 7 is an ambient temperature sensor, 8 is an oil level gauge, 9 is a microwave transmitting antenna, and 10 is a microwave receiving antenna.

DETAILED DESCRIPTION

The present disclosure will be further described below in conjunction with specific embodiments.

As shown in FIG. 1 and FIG. 2, a defect monitoring apparatus for a transformer oil conservator based on edge computing includes: a sensing terminal, an edge intelligent gateway, a monitoring terminal, an oil conservator defect monitoring cloud platform, and a client terminal. The edge intelligent gateway and the monitoring terminal are both disposed in a substation. The monitoring terminal is configured to access the edge intelligent gateway. The sensing terminal includes: an airflow sensor 5, an oil temperature sensor 6, an ambient temperature sensor 7, and a microwave oil level measurement assembly. The airflow sensor 5 is disposed on a breathing tube 3, and is configured to measure a breathing airflow rate of an oil conservator 1. The oil temperature sensor 6 is disposed at a lower portion of a side of the oil conservator 1, and a probe of the oil temperature sensor 6 extends into the oil conservator 1 and is immersed in transformer oil. The oil temperature sensor 6 is configured to measure an oil temperature of the oil conservator. The ambient temperature sensor 7 is disposed outside the oil conservator 1. The ambient temperature sensor 7 is configured to measure an ambient temperature. The microwave oil level measurement assembly is disposed at a top of an inner wall of the oil conservator 1, and the microwave oil level measurement assembly is away from a capsule 2. The microwave oil level measurement assembly is configured to measure a height of an oil level of the oil conservator.

The sensing terminal sends detection data to the edge intelligent gateway, and an original oil level gauge 8 of the oil conservator also sends data information thereof to the edge intelligent gateway. The edge intelligent gateway stores and evaluates the detection data. The edge intelligent gateway sends an evaluation result indicating a defect and corresponding data to the oil conservator defect monitoring cloud platform. The oil conservator defect monitoring cloud platform sends an alarm command to the client terminal.

The microwave oil level measurement assembly includes: a microwave transmitting antenna 9 and a microwave receiving antenna 10, where the microwave transmitting antenna 9 and the microwave receiving antenna 10 are both oriented towards the oil face, upper ends of the microwave transmitting antenna 9 and the microwave receiving antenna 10 are both fastened at the top of the inner wall of the oil conservator 1, and a transmitting end of the microwave transmitting antenna 9 is located in a same horizontal plane with a receiving end of the microwave receiving antenna 10.

The sensing terminal communicates with the edge intelligent gateway through a ZigBee network; the edge intelligent gateway communicates with the monitoring terminal through the ZigBee network, and the edge intelligent gateway communicates with the oil conservator defect monitoring cloud platform through a GPRS network.

The airflow sensor 5 is disposed close to a dehydrating breather 4.

The airflow sensor 5 adopts bidirectional airflow rate measurement, with a measurement range of −200 slpm to +200 slpm, a total error band less than or equal to 3% of a reading, and a response time less than or equal to 1 ms.

The oil temperature sensor 6 and the ambient temperature sensor 7 are both platinum resistance sensors, with a measurement range of −40° C. to 120° C. and accuracy of ±0.5° C.

The edge intelligent gateway adopts an ARM Cortex-A7 processor, with 1 GB RAM and 8 GB eMMC.

The client terminal may be a personal computer (PC), a tablet computer, or a smart phone, and supports an email prompt alarm of the PC and the tablet computer, and a short message prompt alarm of the smart phone.

A defect detection method for a transformer oil conservator based on edge computing includes the following steps:

S1: A sensing terminal monitors the following data in real time: a breathing airflow rate of an oil conservator, an oil temperature of the oil conservator, an ambient temperature, a microwave transmitting power, and a microwave receiving power.

S2: Determine whether the oil conservator has a false oil level defect: calculate an actual oil level height D₁ according to the microwave transmitting power and the microwave receiving power; an original oil level gauge of the oil conservator sends data information to an edge intelligent gateway; the edge intelligent gateway calculates a difference between the actual oil level height D₁ and a display value D of the original oil level gauge 8 of the oil conservator; and when the difference is greater than 3 cm, determine that the oil conservator has a false oil level defect; otherwise, determine that the oil conservator has no false oil level defect.

S3: Sample real-time monitoring data, to form a breathing airflow rate curve; when a change in the ambient temperature is within 10° C. and a change in the oil temperature of the oil conservator is within 4° C., compare a peak breathing airflow rate V_m of the oil conservator monitored in real time with an average peak breathing airflow rate V_a of the oil conservator in the passing month under the same condition; if a difference is greater than or equal to 15%, perform step S4; otherwise, determine that the following four types of defects do not exist: capsule damage, partial blockage of a breathing circuit, leakage in the breathing circuit, or moisture impurities in a breathing tube.

S4: As shown in FIG. 3, when V_m≥1.2V_a, t_h≤0.9t_ha, and t_x≤0.9t_xa, determine that a capsule damage defect exists.

where t_h is a single expiration duration monitored in real time, t_x is a single inhalation duration monitored in real time, t_ha is an average single expiration duration in the passing month, and t_xa is an average single inhalation duration in the passing month;

as shown in FIG. 4, when V_m≤0.8V_a, t_h≥1.1t_ha, and t_x≥1.1t_xa, determine that a defect of partial blockage of the breathing circuit exists;

as shown in FIG. 5, when V_m≤0.8V_a, |t_h−t_ha|<0.05t_ha, and |t_x−t_xa|<0.05t_xa, determine that a defect of leakage in the breathing circuit exists; and

as shown in FIG. 6, when V_m≤0.85V_a and an airflow rate zero-crossing number x₀ of a single expiration and inhalation process increases by 2 times or more compared with an average airflow rate zero-crossing number x_a0 in the passing month, determine that a defect of moisture impurities in a breathing tube exists.

S5: Calculate a comprehensive evaluation coefficient for defect detection of the oil conservator, to obtain defect severity.

S6: When the defect severity is a general defect or a severe defect, the edge intelligent gateway sends an evaluation result and corresponding data to an oil conservator defect monitoring cloud platform.

S7: The oil conservator defect monitoring cloud platform sends an alarm command to a client terminal, such that staff access the oil conservator defect monitoring cloud platform through the client terminal to check specific information.

A formula for calculating the actual oil level height D₁ in step S2 is as follows:

D ₁ =H−d−h

where H is a height from the bottom to the top of the interior of an oil conservator, d is a distance from a transmitting end of a microwave transmitting antenna to an oil level, and h is a distance from the transmitting end of the microwave transmitting antenna to the top of the interior of the oil conservator;

$d=\sqrt{\frac{P_t{G}_t{G}_r{\lambda }^2}{4{\pi }^2{P}_r}-\frac{1}{4}{S}^2}$

As shown in FIG. 7, S is a distance from the transmitting end of the microwave transmitting antenna to a receiving end of a microwave receiving antenna, λ is a wavelength, P_t is a microwave transmitting power of the microwave transmitting antenna, G_t is a gain of the microwave transmitting antenna, P_r is a detected power of the microwave receiving antenna, and G_r is a gain of the microwave receiving antenna.

A formula for calculating the comprehensive evaluation coefficient c_k for defect detection of the oil conservator is as follows:

$c_k=1.2p_k^2+0.4d_k^{\frac{1}{3}}+0.7x_k^{\frac{2}{3}}+1.5j_k^{\frac{2}{3}}+0.2s_k^{1.2}$

where when a capsule damage defect exists, a characteristic value p_k is 2; otherwise, p_k is 1;

when a defect of partial blockage of the breathing circuit exists, a characteristic value d_k is 2; otherwise, d_k is 1;

when a defect of leakage in the breathing circuit exists, a characteristic value x_k is 2; otherwise, x_k is 1;

when a defect of moisture impurities in a breathing tube exists, a characteristic value s_k is 2; otherwise, s_k is 1;

when a defect of false oil level of the oil conservator exists, a characteristic value j_k is 2; otherwise, j_k is 1; and

the defect severity is determined according to a value of the comprehensive evaluation coefficient c_k for defect detection of the oil conservator:

when c_k=4, the defect severity is no defect;

when 4<c_k≤4.8, the defect severity is the general defect; and

when c_k>4.8, the defect severity is the severe defect.

The above embodiments are merely intended to exemplarily illustrate the principles and effects of the present disclosure, rather than to limit the present disclosure. Any person skilled in the art can make modifications or alterations to the foregoing embodiments without departing from the spirit and scope of the present disclosure. Hence, all equivalent modifications or alterations made by those of ordinary skill in the art without departing from the spirit and technical teachings disclosed in the present disclosure shall fall within the scope defined by appended claims to the present disclosure.

Claims

1. A defect monitoring apparatus for a transformer oil conservator based on edge computing, comprising: a sensing terminal, an edge intelligent gateway, a monitoring terminal, an oil conservator defect monitoring cloud platform, and a client terminal, wherein the edge intelligent gateway and the monitoring terminal are both disposed in a substation; the monitoring terminal is configured to access the edge intelligent gateway; the sensing terminal comprises: an airflow sensor (5), an oil temperature sensor (6), an ambient temperature sensor (7), and a microwave oil level measurement assembly; the airflow sensor (5) is disposed on a breathing tube (3), and is configured to measure a breathing airflow rate of an oil conservator (1); the oil temperature sensor (6) is disposed at a lower portion of a side of the oil conservator (1), and a probe of the oil temperature sensor (6) extends into the oil conservator (1) and is immersed in transformer oil; the oil temperature sensor (6) is configured to measure an oil temperature of the oil conservator; the ambient temperature sensor (7) is disposed outside the oil conservator (1); the ambient temperature sensor (7) is configured to measure an ambient temperature; the microwave oil level measurement assembly is disposed at a top of an inner wall of the oil conservator (1), and the microwave oil level measurement assembly is away from a capsule (2); the microwave oil level measurement assembly is configured to measure a height of an oil level of the oil conservator; and the sensing terminal is configured to send detection data to the edge intelligent gateway, and an original oil level gauge (8) of the oil conservator is configured to send data information thereof to the edge intelligent gateway; the edge intelligent gateway is configured to store and evaluate the detection data, and send an evaluation result indicating a defect and corresponding data to the oil conservator defect monitoring cloud platform; and the oil conservator defect monitoring cloud platform is configured to send an alarm command to the client terminal.

2. The defect monitoring apparatus for a transformer oil conservator based on edge computing according to claim 1, wherein the microwave oil level measurement assembly comprises: a microwave transmitting antenna (9) and a microwave receiving antenna (10); the microwave transmitting antenna (9) and the microwave receiving antenna (10) are both oriented towards the oil face, upper ends of the microwave transmitting antenna (9) and the microwave receiving antenna (10) are both fastened at the top of the inner wall of the oil conservator (1), and a transmitting end of the microwave transmitting antenna (9) is located in a same horizontal plane with a receiving end of the microwave receiving antenna (10).

3. The defect monitoring apparatus for a transformer oil conservator based on edge computing according to claim 1, wherein the sensing terminal communicates with the edge intelligent gateway through a ZigBee network; the edge intelligent gateway communicates with the monitoring terminal through the ZigBee network; and the edge intelligent gateway communicates with the oil conservator defect monitoring cloud platform through a General Packet Radio Service (GPRS) network.

4. The defect monitoring apparatus for a transformer oil conservator based on edge computing according to claim 1, wherein the airflow sensor (5) is disposed close to a dehydrating breather (4).

5. The defect monitoring apparatus for a transformer oil conservator based on edge computing according to claim 1, wherein the airflow sensor (5) is configured to adopt bidirectional airflow rate measurement, with a measurement range of −200 slpm to +200 slpm, a total error band less than or equal to 3% of a reading, and a response time less than or equal to 1 ms.

6. The defect monitoring apparatus for a transformer oil conservator based on edge computing according to claim 1, wherein the oil temperature sensor (6) and the ambient temperature sensor (7) are both platinum resistance sensors, with a measurement range of −40° C. to 120° C. and accuracy of ±0.5° C.

7. The defect monitoring apparatus for a transformer oil conservator based on edge computing according to claim 1, wherein the edge intelligent gateway is configured to adopt an ARM Cortex-A7 processor, with 1 GB RAM and 8 GB Embedded Multi Media Card (eMMC); and the client terminal is a personal computer (PC), a tablet computer, or a smart phone.

8. A defect detection method for a transformer oil conservator based on edge computing, comprising the following steps: S1: monitoring, by a sensing terminal, the following data in real time: a breathing airflow rate of an oil conservator, an oil temperature of the oil conservator, an ambient temperature, a microwave transmitting power, and a microwave receiving power;S2: determining whether the oil conservator has a false oil level defect: calculating an actual oil level height D1 according to the microwave transmitting power and the microwave receiving power; sending, by an original oil level gauge (8) of the oil conservator, data information to an edge intelligent gateway; calculating, by the edge intelligent gateway, a difference between the actual oil level height D1 and a display value D of the original oil level gauge (8) of the oil conservator; and when the difference is greater than 3 cm, determining that the oil conservator has a false oil level defect; otherwise, determining that the oil conservator has no false oil level defect;S3: sampling real-time monitoring data, to form a breathing airflow rate curve; when a change in the ambient temperature is within 10° C. and a change in the oil temperature of the oil conservator is within 4° C., comparing a peak breathing airflow rate Vm of the oil conservator monitored in real time with an average peak breathing airflow rate Va of the oil conservator in the passing month under the same condition; if a difference is greater than or equal to 15%, performing step S4; otherwise, determining that the following four types of defects do not exist: capsule damage, partial blockage of a breathing circuit, leakage in the breathing circuit, or moisture impurities in a breathing tube;S4: when Vm≥1.2Va, th≤0.9tha, and tx≤0.9txa, determining that a capsule damage defect exists;wherein th is a single expiration duration monitored in real time, tx is a single inhalation duration monitored in real time, tha is an average single expiration duration in the passing month, and txa is an average single inhalation duration in the passing month;when Vm≤0.8Va, th≥1.1tha, and tx≥1.1txa, determining that a defect of partial blockage of the breathing circuit exists;when Vm≤0.8Va, |th−tha|<0.05tha, and |tx−txa|<0.05txa, determining that a defect of leakage in the breathing circuit exists; andwhen Vm≤0.85Va and an airflow rate zero-crossing number x0 of a single expiration and inhalation process increases by 2 times or more compared with an average airflow rate zero-crossing number xa0 in the passing month, determining that a defect of moisture impurities in a breathing tube exists;S5: calculating a comprehensive evaluation coefficient for defect detection of the oil conservator, to obtain defect severity;S6: when the defect severity is a general defect or a severe defect, sending, by the edge intelligent gateway, an evaluation result and corresponding data to an oil conservator defect monitoring cloud platform; andS7: sending, by the oil conservator defect monitoring cloud platform, an alarm command to a client terminal, such that the transformer operation and maintenance personnel access the oil conservator defect monitoring cloud platform through the client terminal to check specific information.

9. The defect detection method for a transformer oil conservator based on edge computing according to claim 8, wherein a formula for calculating the actual oil level height D1 in step S2 is as follows: D1=H−d−h wherein H is a height from the bottom to the top of the interior of the oil conservator, d is a distance from a transmitting end of a microwave transmitting antenna to an oil level, and h is a distance from the transmitting end of the microwave transmitting antenna to the top of the interior of the oil conservator;

10. The defect detection method for a transformer oil conservator based on edge computing according to claim 8, wherein a formula for calculating the comprehensive evaluation coefficient ck for defect detection of the oil conservator is as follows: