RU2709604C1 - Method of diagnosing electrical insulation in process of remote computer monitoring of process equipment - Google Patents

Abstract

FIELD: physics.SUBSTANCE: invention refers to technical diagnostics. Summary: using partial discharge detectors resistant to effects of network frequency currents, which are installed on conductors of electric power supply. Sensors comprise current transformers with magnetic conductors of nanocrystalline material and high-pass filters formed by windings of transformers, connected together with capacitors, reducing influence of network frequency amplitude of network frequency when selecting pulses of partial discharges. Piezoelectric acoustic emission sensors are also used, which are installed by piezocrystals on metal structural elements of electrical equipment in the immediate vicinity of the diagnosed object for simultaneous detection of acoustic oscillations caused by partial discharges and transient voltages arising as a result of electromagnetic fields from partial discharges to grounded metal structural elements of electrical devices. Automatic computer monitoring system is equipped with an expert system which is used to evaluate the technical state of insulation, the place, type and level of development of insulation defects by the criteria "Permissible", "Requires measures", "Inadmissible" and gives out the target-indicating instructions to service personnel.EFFECT: invention can be used for diagnosing state of insulation of electrical equipment, in particular asynchronous electric motors.1 cl, 12 dwg

Description

The invention relates to the field of technical diagnostics and can be used to diagnose the state of insulation of electrical equipment, in particular, asynchronous motors.

The prior art method [1] for determining the location and type of defects in the active part of an electric machine that is in operating mode, which consists in the fact that at least 4 measuring sensors are installed on opposite sides of the outer surface of the body of the controlled object (electric machine) partial discharges connected to the recording equipment. Each of the sensors is alternately moved within the boundaries of the surface of the housing on which it is installed, until the maximum signal is received at the output of the sensor, after which the sensor is fixed on the surface of the housing in this position. Using the recording equipment, the shape of the sensor output signal is stored. After fixing the first sensor, the following sensors are fixed provided that the shape of their output signals matches the shape of the output signal of the first sensor. The delay time of the signals received by the sensors relative to the first received signal is measured. According to their size, the location of the defective site is determined. The shape of the sensor output signal judges the type of discharge phenomenon.

The disadvantage of this method is the need to carry out manually a large number of works on the search for defects on an electric motor in operation, and to perform a manual analysis of the received signals in order to determine the alleged type of defect. Accordingly, the reliability of the diagnosis is reduced due to the influence of the human factor.

Also known from the prior art is a method [2] for measuring the characteristics of partial discharges by the electric method during testing and / or operation of high-voltage equipment in a three-phase system under the influence of industrial frequency voltage, which allows determining the maximum value of the apparent charge of a partial discharge.

The disadvantages of this method are the inability to determine the types of insulation defects of high-voltage equipment, which initiated the occurrence of partial discharges, their location and the degree of danger that they pose for insulation.

Also known from the prior art is a method for assessing the technical state of insulation [3] by the parameters of the pulses of high-frequency signals generated by the occurrence of partial discharges in the insulation of electrical equipment during its operation, based on periodic measurement of the amplitude and quantity parameters of partial discharges, for example, the frequency of their repetition, and comparing the obtained values with threshold values determined previously empirically in the form of a database containing threshold values of pairs partial discharge ameters.

The known method is used to assess the risk of failure of electrical equipment, in particular, switchgears of high and medium voltage, and consists in measuring such parameters of partial discharges as the amplitude of the pulses and the number of pulses for one period of the supply voltage, while the risk of equipment failure is estimated depending on values of the measured parameters, which are compared with threshold values. Threshold values are taken from the database on the values of the measured parameters of the partial discharges for equipment of the same type with the test.

The specified method has the following disadvantages:

- provides for not all possible risks of failure, since an incomplete set of combinations of the values of the partial discharge parameters is analyzed and the possibility of exceeding only one of the partial discharge parameters by the corresponding threshold value is not taken into account;

- does not allow, according to the diagnostic results, to separate insulation defects by type and place of their localization.

Closest to the proposed method is a device and method for remote monitoring of partial discharges in an electrical device [4], which includes one or more sensors for detecting partial discharges of a discharge in one or more conductors for supplying electric energy to a device, a programmable computer that analyzes the measured pulses and distinguishing between signals coming from the monitored device and from other places. Distinction is achieved by analyzing the shape of the pulse and by comparing the pulses with the partial discharge events measured locally on the sensors. The device allows you to monitor partial discharges in an electrical device without the need to install sensors on the equipment itself.

The specified device and method adopted for the prototype has the following disadvantages:

- there is no automatic expert system, which necessitates an external external data analysis to determine the type of defect and its localization;

- there is no possibility of operational individual adaptation of diagnostic algorithms, which makes it difficult to configure the process for use on a specific unit of electrical equipment;

- separate acoustic and capacitive sensors are used, which increases the number of measuring channels necessary for the diagnosis of a unit of electrical equipment;

- there is no possibility of simultaneously measuring the parameters of partial discharges in more than six measuring channels, which limits the possibility of using the device and method.

The aim of the invention is to develop a method and sensor aimed at reducing the complexity, increasing the reliability and efficiency of the diagnostic process in real-time operation of the state of electrical insulation of technological equipment, for example, asynchronous electric motors of pumps or compressors, their cable lines and supply cells, reducing the number of used at the same time measuring channels and the integration of the specified diagnostic procedure into the computer monitor system Inga of technological equipment, combining various methods of technical diagnostics, for example, methods of vibration diagnostics or signature analysis of consumption currents.

The technical result in the proposed method for diagnosing electrical insulation in the process of remote computer monitoring of technological equipment, which consists in using one or more sensors for detecting signals from partial discharges occurring in the electrical isolation of technological equipment devices, in one or more conductors for supplying electrical energy to the technological process electrical device equipment without installing sensors on the electrical devices themselves of partial discharges, the use of an automatic computer monitoring system for technological equipment to determine the characteristics of electric discharge activity in isolation, for example, the repetition rate of partial discharges, and determination of zones with the presence of insulation defects on their basis, is achieved by the fact that they are resistant to significant network currents, , with an rms value (RMS) of more than 1100 A, a partial discharge sensor, which is a high-frequency current transformer, and The piezoelectric acoustic emission sensor, which is mounted directly on the metal structural elements of the electrical equipment in the immediate vicinity of the diagnosed object with a piezocrystal, and its signal is used to simultaneously record acoustic vibrations caused by partial discharges, as well as transient voltages arising due to the influence of electromagnetic fields from partial discharges on grounded elements of electrical devices, while configuring algorithms about Partial discharge signal processing, partial discharge signal processing, calculation of their characteristics, for example, discharge repetition rate, is carried out taking into account the individual technical features of the diagnosed electrical devices of technological equipment and any used partial discharge sensors, and the results of the preliminary signal processing are input into an automated expert computer monitoring system, which provides information on the technical condition: insulation, location, type and level of def such as are for the isolation of the "admissibility" criteria "requires action", "unacceptable", as well as issues tseleukazuyuschie prescription service personnel.

The technical result is also achieved by the fact that the partial discharge sensor comprises: a current transformer, wherein the magnetic core of the current transformer is made of nanocrystalline material, for example, a tape of a soft magnetic nanocrystalline alloy based on iron, and the transformer windings connected together with a capacitor forming a high-pass filter for reducing the influence of the amplitude of the current of the network frequency during the allocation of pulses of partial discharges.

Analysis of the distinguishing features of the proposed method and sensor showed that:

- the use of a significant value of the frequency-stable current frequency network, for example, with an RMS of more than 1100 A, of a partial discharge sensor, in which the magnetic circuit is made of nanocrystalline material, which is a high-frequency current transformer, allows to increase the reliability of diagnosis due to the high value of saturation magnetic induction, which several times higher than the same value for ferrites;

- installation of a piezoelectric acoustic emission sensor directly by a piezocrystal on metal elements allows the use of an acoustic emission sensor for simultaneous registration of both acoustic vibrations produced by partial discharges and electromagnetic ones, which allows using two “TEV sensors” and acoustic, instead of the traditionally used two, for electrical diagnostics, one. This reduces the complexity of mounting sensors and implementing the proposed method;

- configuration of partial discharge signal processing algorithms and processing of partial discharge signals, calculation of their characteristics taking into account the individual technical features of the diagnosed electrical devices of the technological equipment and any used partial discharge sensors allows you to quickly adapt the diagnostic algorithms, which allows you to more accurately configure the diagnostic system to take into account the specific features units of electrical equipment;

- the presence of an automatic expert system implemented using a diagnostic controller (computer) eliminates the need for manual data analysis to determine the type of defect and its localization. Such an expert system based on an integrated knowledge base allows issuing instructions to operating personnel on-line, without the need for diagnosticians;

- the proposed sensor design based on a current transformer with a magnetic circuit, as well as using a high-pass filter made on the transformer windings together with a capacitor, increases the sensor's resistance to the effects of mains currents consumed by the diagnosed electrical equipment by filtering partial discharge signals from the current signals consumed technological equipment.

Thus, the proposed set of distinctive features, providing the result obtained, seems new at the existing stage of development of science and technology. The invention corresponds to the inventive step, since the achieved result is determined not only by the sum of the distinguishing features, but also by the synergistic effect of their close interaction with each other.

The essence of the claimed invention is illustrated in FIG. 1-12.

In FIG. 1 shows a schematic diagram of the proposed high-frequency transformer partial discharge (PD) sensor.

In FIG. Figure 2 shows the effect of the influence of a mains frequency current with a RMS of 150 A on a ferrite current sensor of the PD of the RFCT-4 model of a typical design.

In FIG. 3 shows a graph of test PD signals recorded using a PD current sensor of the proposed design when a current of a network frequency is applied to it and with a RMS of 1100 A.

In FIG. Figure 4 shows the installation diagram of a piezoelectric acoustic emission (AE) sensor in a high-voltage cell for simultaneous recording of both acoustic and electromagnetic oscillations produced by partial discharges.

In FIG. 5a) and b) shows the acoustic signals of the PD recorded by the piezoelectric AE sensor.

In FIG. 6 shows a schematic diagram of the localization of insulation defects of a diagnosed asynchronous electric motor, its cable line and high-voltage cell.

In FIG. 7 in a tabular form shows possible combinations of signals from PD sensors, allowing to localize the place of their occurrence

In FIG. 8 in a table form shows an example of a description of the parameters of the knowledge base of an expert system of insulation defects using an example of an electric motor.

In FIG. Figure 9 shows an example of installing a set of three PD current sensors proposed by the applicant and a piezoelectric AE sensor in a high voltage cell.

In FIG. 10 shows examples of graphs of the initial and processed signals from the PD current sensor and the amplitude-phase distribution of PD pulses.

In FIG. 11 shows the amplitude-phase distributions of PD recorded for the three phases of the electric motor supply.

In FIG. 12 shows traces of surface discharges detected on the insulation of a cable line during visual inspection.

The implementation of the invention is presented on the example of the use of the PD sensor model 5902 and the piezoelectric sensor model 5703.

Currently, current transformers with ferrite magnetic cores are used as current sensors for PD. The disadvantage of such sensors is their high sensitivity to network frequency currents, which can flow through the conductor on which these sensors are installed. High sensitivity is due to the low value of saturation magnetic induction inherent in ferrites. In practice, manufacturers of such sensors provide non-magnetic gaps that divide the magnetic circuit into parts and reduce its saturation. However, this measure is insufficient for the use of PD current sensors on power supply conductors of electrical equipment, therefore, such sensors are installed, as a rule, on grounding conductors of electrical equipment directly at the installation site. This installation method may be unacceptable when placing electrical equipment in hazardous areas.

In FIG. 1 is a schematic diagram of a PD sensor proposed by the applicant that is resistant to mains currents with a RMS of more than 1100 A, for example, model 5902, in which the magnetic circuit 1 is made of an iron-based magnetically soft nanocrystalline alloy GM24DS, the saturation magnetic induction of which is several times higher than the similar value for ferrites. The windings of the transformer 2, included in conjunction with the capacitor 5, form a high-pass filter, which also reduces the amplitude of the harmonics of the current of the mains frequency to separate out partial discharge pulses. The resistor 4 and the capacitor 5 form a divider used to correct the amplitude of the measuring signal and the amplitude-frequency characteristics of the proposed partial discharge sensor.

The parameters of the elements 1, 2, 3, 4 and 5 are selected based on the required frequency range and the amplitude of the pulses of the partial discharges recorded by the measuring system. Block 6 in FIG. 1 serves to meet the requirements of explosion protection and coordination of the communication line.

Shown in FIG. Figures 2 and 3 demonstrate the advantage of using the proposed sensor design over the design of a typical transformer partial discharge sensor, for example, RFCT-4 model with a ferrite core with a non-magnetic gap recommended by the manufacturer.

So in FIG. Figure 2 shows the amplitude-phase distribution of test pulses, the effect of the influence of a network frequency current with a RMS of 150 A on a ferrite current sensor of partial discharges of a typical design. On the amplitude-phase distribution along the horizontal axis, the phase angle of one period of the current consumption of electrical equipment is plotted, relative to which the amplitudes of the recorded pulses of partial discharges (vertical axis) are noted. In this case, the color intensity of the mark means the number of pulses registered at the moment. Sections 7 and 8 of the diagram correspond to the moments when the partial discharge test pulses that are not subject to distortion are recorded by the RFCT-4 sensor, since the values of the current of the mains frequency near the phase angles 0 ° and 180 ° are small. Sections 9 and 10 of the diagram correspond to the moments when the RMS values of the current of the network frequency reach 150 A in the vicinity of the phase angles 90 ° and 270 °. As can be seen from FIG. 2, the effect of this current on the sensor leads, in fact, to a 5-fold overestimation of the recorded amplitudes of the test pulses (amplitude change by 400%).

In FIG. Figure 3 shows a graph of PD test signals recorded with a PD PD current sensor of the proposed design when exposed to a network frequency current with a RMS of 1100 A, on which the minimum and maximum values of the amplitudes of the recorded partial pulse test pulses are indicated by dashed lines. From FIG. Figure 3 shows that the difference in the amplitudes of the recorded test pulses of the PD is no more than 20%, which confirms the effectiveness of the proposed design of the PD sensor when measuring partial discharge parameters under the influence of significant currents of the mains frequency consumed by the diagnosed electrical equipment. At present, for detecting acoustic radiation produced by partial discharges, AE sensors of various designs are used, installed, for example, inside a high-voltage cell. At the same time, to detect changes in the potential of metal elements of electrical equipment structures under the influence of electromagnetic fields emitted by partial discharges, other types of sensors are used, the so-called TEV sensors (transient earth voltage) [5].

In FIG. 4 shows a diagram of the proposed method for installing a piezoelectric acoustic emission sensor 11, using the AE 5703 sensor as an example, directly using a magnetic holder on a metal wall of a high-voltage cell 12, where the proposed PD sensors 13, for example sensors 5902, are mounted on the cable line 14 of the diagnosed electric motor.

This method of installation of the AE sensor, in which its sensitive element made of a piezocrystal, has electrical contact with the metal housing element of the high-voltage cell, allows simultaneous registration of acoustic and electromagnetic oscillations produced by partial discharges by one piezoelectric sensor of acoustic emission. This allows you to reduce the number of measuring channels required for the diagnosis of a unit of electrical equipment, which reduces cost, simplifies the installation of sensors and the implementation of the proposed method.

In FIG. 5a) and b) shows the acoustic signals of the PD recorded by the piezoelectric AE sensor installed by the proposed method. As can be seen, various types of the initial signal from partial discharges cause a different shape of the output signals from the piezoelectric acoustic emission sensor, which makes it possible to localize the place of occurrence of the partial discharge signal. Dotted lines for clarity highlight individual pulses.

Acoustic vibrations, the source of which is partial discharges, affect the metal structural elements of electrical equipment in the immediate vicinity of the diagnosed object, on which a piezoelectric acoustic emission sensor is mounted and recorded by it in the form of the signal shown in FIG. 5a). In the dotted line region, it can be seen that the acoustic pulse has a symmetrical shape.

Electromagnetic vibrations, the source of which is partial discharges, also affect the metal elements on which the piezoelectric acoustic emission sensor is mounted and recorded in the form of the signal shown in FIG. 5 B). In the dotted line region, it can be seen that the electromagnetic pulse has an asymmetric shape with a pronounced peak of negative polarity. In this case, the capacitance of a piezocrystal plays the role of a capacitive TEV sensor used for similar purposes.

As shown in FIG. 5a) and b) the signals are easily distinguishable, then one piezoelectric acoustic emission sensor can be used instead of two different sensors (acoustic sensor and TEV sensor).

The described phenomenon is used in the work of an automatic expert system, which eliminates the need for manual data analysis to determine the type of defect and its localization. An example of such an expert system for other diagnostic methods is described in [6].

In FIG. 6 shows a schematic diagram of the connection of a diagnosed asynchronous electric motor to a high-voltage cell, explaining the principle of propagation of signals from typical partial discharges occurring in an asynchronous electric motor and cable line. The above diagram shows that when partial discharges 16 occur in the insulating constructions of the electric motor 15, a positive current pulse + i will propagate along the conductors of the cable power line 14, which will be detected by the corresponding current sensor of the PD 13. A negative-polarity pulse symmetrical to it will propagate along structural elements of the motor will be scattered in the grounding device of the motor.

If the source of discharge activity 17 is in the insulation of the conductors of the cable power line 14, a current pulse of positive polarity + i 'will propagate along the conductors of the power line, and a symmetrical pulse of negative polarity -i' will propagate through the metal sheath (armor) of the cable power line 18. V In this case, the pulse + i 'will also be detected by the corresponding PD current sensor. The pulse -i ', in turn, will pass through the ground circuits 19 of the high-voltage cell 12, briefly changing the value of the potential of its metal structures.

Since the piezoelectric acoustic emission sensor 11 is installed near the cable termination and, accordingly, near the grounding location of its metal sheath (armor), such processes cause the piezoelectric sensor to record acoustic emission pulses of negative polarity, indicating the presence of discharge activity in the insulation of the conductors of the power cable 14 .

Similarly, partial discharges occurring inside the volume of a high-voltage cell are recorded by their acoustic and electromagnetic radiation.

Using the described phenomenon of the difference in the shape and polarity of the signals from the piezoelectric acoustic emission sensor depending on the place of occurrence of partial discharges, one can accurately find the localization of the place of occurrence of partial discharges.

In FIG. 7 in a tabular form describes the combinations of signals from PD sensors, which can be used in this case.

Filtering the pulses -i ', selected in the signals of the PD current sensors, allows them to be removed as not related to the insulation of the electric motor and taken into account as signals of partial discharges occurring in the insulation of the cable line of the diagnosed electric motor.

An expert system based on a built-in knowledge base allows you to automatically issue instructions to operating personnel in real time by analyzing values, for example, such parameters of partial discharges as the repetition rate and amplitude. An example of a description of the parameters of the knowledge base and the corresponding defects in the electrical insulation of technological equipment is given in table form in FIG. 8.

The following notation is used in the table:

- TPM threshold - threshold value “REQUIRES ACCEPTANCE OF MEASURES”;

- NDP threshold - threshold value “UNACCEPTABLE”.

In an automatic computer-aided monitoring system for technological equipment, messages about the degree of development of a defect are highlighted in the appropriate color (green - state is acceptable, yellow - state requires action and red - state is unacceptable).

Threshold values are determined empirically from statistical data for the diagnosed technological equipment.

In FIG. Figure 9 shows an example of installing a set of three current sensors CR 13 on the phases of a cable line 14 and a piezoelectric sensor AE 11 in a high-voltage cell 12 of an asynchronous electric motor of the VAO2-560-800 type with an operating voltage of 6000 V and a power of 800 kW as part of a technological unit for hydrotreating diesel fuel at the refinery.

In FIG. 10 shows examples of a graph of the initial signal 20 from one of the PD current sensors, a graph of the selected pulses 21 of the PD, and a graph of the amplitude-phase distribution of 22 pulses of the PD.

In FIG. 11 shows the amplitude-phase distribution of partial discharges recorded for phase A of the electric motor power, which, according to the source [7], indicates the presence of partial discharges in the insulating structures of the high-voltage cell.

In FIG. 12 shows photographs taken during a visual inspection of the insulation of a cable line with traces of surface discharges. In FIG. 12a) shows a white powder of nitrogen oxides resulting from the occurrence of partial discharges, and in FIG. 12b) is the trace of a surface discharge caused by delamination in the insulating structure.

The proposed method for diagnosing the state of insulation was successfully tested at one of the large oil refineries by integrating into the complex diagnostic system and monitoring the dynamic equipment of hazardous production facilities.

Thus, the application of the proposed group of inventions of the method and sensor can improve the quality of diagnosing the state of insulation of electrical devices of technological plants, and perform a comprehensive assessment of their condition at all stages of operation by automatically determining the technical condition, localization and determination of the types of detected defects.

Bibliographic list

1. Patent RU 2370784 C1 G01R 31/34 A method for determining the location and type of defects in the active part of an electrical machine in operating mode.

2. Patent RU 2374657 C1 G01R 31/12 Method for measuring the characteristics of partial discharges.

3. Publication of “Partial Discharge (PD) techniques for measuring the condition of aging HV / MV switchgear”, Neil Davies, EA Technology International (http://www.eatechnology.com.cn/wp-content/uploads/pdfs/PD_techniques_for_measuring_condition_of_HM_and_HV_switch_gear .pdf, https://www.cablejoints.co.uk/upload/PD-Techniques-For-Measuring-Condition-Of-MV-and-HV-Switchgear.pdf).

4. Patent US 9581632 Apparatus and method for remote control of partial discharge in electrical equipment.

5. Capacitive sensors for registering partial discharges (https://dimrus.ru/tev.html).

6. Kostyukov V.N. Production safety monitoring. M.: Mechanical Engineering, 2002.224 s.

7. GOST IEC / TS 60034-27-2-2015 Electric rotating machines. Part 27-2. Partial discharge measurements in the insulation of the stator windings of rotating electrical machines connected to the network. - M .: Standartinform, 2015.

Claims (1)

A method for diagnosing electrical insulation in the process of remote computer monitoring of technological equipment, which consists in using sensors for detecting signals from partial discharges occurring in the electrical isolation of technological equipment devices in one or more conductors of electrical energy supply and / or grounded elements of electrical devices of technological equipment, using automatic computer systems for technological monitoring requirements for determining the characteristics of electric discharge activity in insulation and determining on their basis zones with the presence of insulation defects, characterized in that they use partial discharge sensors that are resistant to the influence of currents of the network frequency and are installed on the conductors for supplying electric energy, containing current transformers with magnetic circuits made of nanocrystalline material and high-pass filters formed by transformer windings connected together with capacitors, reducing the influence of current amplitude with frequency during the extraction of pulses of partial discharges, and piezoelectric acoustic emission sensors mounted by piezocrystals on metal structural elements of electrical equipment in the immediate vicinity of the diagnosed object for simultaneous recording of acoustic vibrations and transient voltages caused by partial discharges arising from the influence of electromagnetic fields from partial discharges on grounded metal structural elements of electrical devices, at the same time, an automatic computer monitoring system is equipped with an expert system, with which they evaluate the technical condition of the insulation, location, type and level of development of insulation defects according to the criteria “Permissible”, “Requires action”, “Inadmissible” and issue targeting instructions to maintenance personnel.

Images