Dîjîk Bîgêhkirin û Analîzkirin ê Malperên Mehanîkî yên Swîçekanê ya Nîştmanan de

Di daramên nîşanî yên elektrîkî ya modern, çalakanên bêgirînê yên hêvî gerîyê wateya herî mohemîn dikin. Wan tewa tevahî û amana xebitandina malperê yê elektrîkî an rêyanên herêmdekirina bêgirîn ê bi serparastî da. Kesêtina mekanîkî yên çalakanên bêgirînê yên hêvî gerîyê, wek pêşbûna xweherdina naverok, kesêtina aktuator, an gorînî yên parçeyên struktûrî, dikevin berbiyayîn îmara navbera sisteman elektrîkî.Rêzikên cihazî yên digelînê rengînên kesêt û derxistina herêmdekirina bêgirînê bikar anîn.

Ev rêzikên cihazî ne ji bo demê pir û karî yên zorîn nîşan dan, lê heke ne ji bo dema herêmdekirina serbest ya kesêt din çareserkirina herêmdekirina bêgirînê. Bi pêşrawî teknolojî, rêzikên derxistina awenî yên çareser kirin dikeve şopandin, ku piştî wan çareser bike şopandin û derxistina kesêt.

Rêzikên derxistina awenî, wek derxistina data yên sensor, pêşkirina data û analîz, analîzeya signala current motor, û pêşkirina tenzîyon, dikarin rastîn wergerandina çalakanên bêgirînê yên hêvî gerîyê, pêşkerdanî çareser bike kesêt, û derxistina serkarûbikin. Ev jî tişta amana û efektiyê operasyonan sisteman elektrîkî.

1. Cûrekên kesêt mekanîkî yên çalakanên bêgirînê yên hêvî gerîyê
1.1 Kesêt pêşbûna xweherdina naverok

Pêşbûna xweherdina naverok eke hatiye li ser oxidanê naverok, xweherdina naverok, an girtina xweherdina naverok. Li vir cûrekê, vegera xweherdina naverok, resistance dikare çêbek, ku bi rêya xweherdina çalakanên bêgirînê yên hêvî gerîyê. Ji bo pêşbûna xweherdina naverok, ji roja xweherdina naverok hêsan dibe, ku ji bo vegera xweherdina naverok girîn dibîne. Ev jî kesêtên termîkî yên seriusan, wek pêşbûna koyî, an girînên termîkî yên herêmdekirina bêgirînê.

Pêşbûna xweherdina naverok dikare stabîlîya voltage'ê werek birne, ku bi rêya amana xweherdina çalakanên bêgirînê yên hêvî gerîyê. Pêşbûna xweherdina naverok dikare amana xweherdina çalakanên bêgirînê yên hêvî gerîyê werek birne, ku bi rêya amana xweherdina çalakanên bêgirînê yên hêvî gerîyê. Ji ber vê yekê, destpêkî derxistina û çareser bike pêşbûna xweherdina naverok ê çendkariye ji bo amana û serparastî sisteman elektrîkî.

1.2 Kesêt pêşbûna aktuator

Kesêt pêşbûna aktuator dikare werek birne ji bo performansa çalakanên bêgirînê yên hêvî gerîyê. Li vir cûrekê, kesêt pêşbûna aktuator dikare gorînî yên mekanîkî, xweherdina naverok, an gorînî yên parçeyên aktuator. Gorînî yên mekanîkî dikare werek birne ji bo xasînên parçeyên aktuator, wek ballantîn û gear, li vir karên berdewamî yên operasyon.

Bi pêşrawî demê xebitandina, parçeyên aktuator dikare gorînî yên material û deformasyon biguherînin, ku bi rêya amana û serparastî çalakanên bêgirînê yên hêvî gerîyê. Heke ev kesêt ne bi destpêkî derxistin û çareser bike, dikare werek birne ji bo operasyonên çalakanên bêgirînê yên hêvî gerîyê, û ji bo seriusan dikare werek birne ji bo stabîlîya sisteman elektrîkî.

1.3 Kesêt pêşbûna gorînî û gorînî yên parçeyên struktûrî

Kesêt pêşbûna gorînî û gorînî yên parçeyên struktûrî dikare werek birne li vir tesîri stress mekanîkî û faktorên amûran. Parçeyên struktûrî, wek sûtûr, barên peyvend, û ballantîn, bi rêya gorînî yên material, ku bi rêya karên berdewamî yên operasyon, gorînî dikin. Bi pêşrawî demê, ev stress dikare werek birne li ser material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi rêya gorînî yên material, ku bi......

2. Rêzikên derxistina awenî yên kesêt mekanîkî yên çalakanên bêgirînê yên hêvî gerîyê
2.1 Sensor û derxistina data

Sensor dikare werek birne ji bo derxistina kesêt mekanîkî yên çalakanên bêgirînê yên hêvî gerîyê. Wan jî tişta herêmdekirina parametreyên fizîkî yên serpilîn, wek vibrasyon, ses, termîna, û current. Ji bo çalakanên bêgirînê yên hêvî gerîyê, wan sensorên pêşdest bi vîjanî vibrasyon, ses, û current bikar anîn.

Sensorên vibrasyon dikarin werek birne ji bo derxistina vibrasyon freqans û amplitûd li vir karên parçeyên cihaz. Bi analîzeya data vibrasyon, ev ê dibeke girîngîyan û kesêtan. Li vir norma operasyon, freqansa vibrasyon çalakanên bêgirînê yên hêvî gerîyê divê li navbera (genîra, threshold dikeve biguherînin da ku bi yekê ya freqansa operasyon). Heke ev navendekî, ev dikare nîşan bideke anomaliya.

Sensorên ses dikarin werek birne ji bo derxistina ses li vir sesên high-frequency yên material û gorînî yên struktûrî. Li vir karên çalakanên bêgirînê yên hêvî gerîyê, heke ev cracks û looseness hene, sensorên ses dikarin werek birne ji bo derxistina sesan li vir minor deformations û ruptures. Prinsipal sensorên ses dibeke şema Figure 2.

Sensorên current û voltage tişta derxistina guhertinên current û voltage li vir çalakanên bêgirînê yên hêvî gerîyê. Readings abnormal current û voltage ji wan sensoran dikare nîşan bideke problem û xebitandina electrical connections û functionality.

1 - Bolt holes; 2 - Foundation;3 - Piezoelectric crystals;4 - Electronic Amplifier;5 - Terminal Connector

Li vir derxistina data, tişta herêmdekirina data li vir data collected bi sensors. Sistem derxistina data tişta herêmdekirina:

• Data Acquisition Unit (DAU). DAU tişta herêmdekirina analog signals li vir sensors û convert analog signals to digital signals. DAU tişta herêmdekirina data li vir rate û precision û meet requirements processing.

• Data Transmission. Data collected transmitted to central server through communication network. This step often relies on wireless communication technologies such as Wi-Fi or 4G/5G networks, which can further increase the speed and efficiency of data transmission and reduce the complexity and cost of wiring.

• Data Storage and Management. After successful data transmission, effective data storage and management must be carried out on a server or in the cloud to establish a more complete database. Data storage needs to support fast access and large-scale data analysis, so high-performance databases are required to achieve data query and retrieval. A schematic diagram of database establishment is shown in Figure 3.

Bi sensors û derxistina data, real-time monitoring operating status and performance indicators of equipment can promptly detect potential defects, providing a necessary basis for intelligent diagnosis of mechanical failures, preventing the occurrence of failures, and ensuring the stable operation of the power system.

2.2 Data Processing and Analysis
2.2.1 Time-Frequency Analysis

Time-frequency analysis is an efficient data-processing method that can transform signals from the time domain to the frequency domain, thereby revealing the internal characteristics and changing trends of signals. Commonly used time-frequency analysis methods include Short-Time Fourier Transform (STFT), wavelet transform, and Wigner-Ville distribution.

STFT performs a local Fourier transform on the signal through a window of fixed size, making it suitable for analyzing signals whose frequencies change slowly over time. For example, when monitoring the actuator, STFT can effectively identify frequency drifts caused by friction or structural looseness.

The wavelet transform can provide windows of variable size, making it suitable for processing signals with instantaneous mutation characteristics. By adjusting the mother wavelet function, precise identification of abnormal vibrations within a specific frequency band can be achieved.

As an advanced time-frequency analysis tool, the Wigner-Ville distribution, despite generating cross-term interference, offers a more refined analysis of the signal's time and frequency, making it particularly suitable for fault detection in complex signal environments.

In practical applications, combining the above-mentioned time-frequency analysis methods with the original data measured by sensors can accurately monitor and diagnose the operating conditions of high-voltage disconnect switches. Under normal operating conditions, the frequency range of high-voltage disconnect switches can generally be maintained at 50-100 Hz; while in the case of poor contact, structural component fatigue, and damage failures, the frequency of high-voltage disconnect switches will shift significantly or new frequency components will appear.

2.2.2 Machine Learning and Pattern Recognition

First, after data collection, through a pre-processing stage such as noise elimination and feature extraction, input data is prepared for machine-learning algorithms. The data includes frequency components of vibration signals, waveform characteristics of electrical parameters, etc.

Second, supervised learning algorithms such as Support Vector Machines (SVM) and Random Forest can be used to classify the data obtained from sensors. These algorithms are trained to identify different types of fault patterns, such as the unique signal patterns caused by poor contact or actuator failures. In practical applications, thousands of data points are input into the algorithms for training to ensure that they can accurately identify fault states.

Finally, deep-learning techniques, especially Convolutional Neural Networks (CNN), are used for complex pattern recognition. Deep-learning techniques can extract useful information from large-scale multi-dimensional data through their automatic feature-learning capabilities, improving the accuracy of diagnosis. For example, in a specific CNN model, several convolutional layers and pooling layers are designed to process the collected video image data to identify typical fault features.

2.3 Drive Motor Current Signal Analysis

Real-time monitoring and analysis of the current signals generated during the operation of the drive motor can predict and diagnose potential mechanical failures. Drive motor current signal analysis generally focuses on detecting small changes in the current signal to determine the anomalies or wear of mechanical components.

If there are failures in the mechanical components of the high-voltage disconnect switch, such as bearing damage, gear wear, or imbalance, it will indirectly affect the load of the drive motor, thereby causing specific pattern variations in its current signal.

In terms of data analysis, a current sensor is used to record the current waveform under normal operating conditions around the motor's power-supply coil. The sampling frequency is usually set above 20 kHz to capture detailed information and ensure high-precision data parsing.

In terms of feature extraction, the Fourier transform is used to convert the time-domain current signal into a frequency-domain signal, which helps to identify harmonic anomalies caused by mechanical failures. For example, under fault-free conditions, the drive motor current signal mainly contains the fundamental frequency and its integer-multiple harmonics. If there is a fault, such as bearing failure, new peaks will be observed at specific frequencies.

In subsequent data processing, statistical methods can be used to analyze the extracted frequencies. For example, calculate the amplitude changes of each frequency point, and train a fault-identification model using a machine-learning algorithm. The input of the algorithm is the frequency characteristics of the current signal, and the output is the prediction of the fault type and severity.

By analyzing the current signal, the deviation of the current signal can be quantified. For example, in the initial stage of bearing failure, the amplitude of the current harmonic can increase by 5-10 A, while in the case of gear wear, the amplitude of the relevant harmonic can increase by 3-8 A. This enables the maintenance team to accurately determine the equipment status and plan maintenance work, thereby avoiding large-scale power outages caused by failures.

2.4 Application of Resistance Strain Measurement Technology

Resistance strain measurement technology can be used to monitor the structural stress and deformation of high-voltage disconnect switches. This technology is realized through resistance strain gauges installed on key components.

A resistance strain gauge is a sensor that converts mechanical deformation into an electrical signal. Its working principle is based on the property that the resistance value of a metal conductor changes when it is deformed under force. A schematic diagram of the resistance strain gauge structure is shown in Figure 4.

When selecting resistance strain gauges, high-precision metal foil resistance strain gauges can be chosen. These gauges have good linear characteristics and stable temperature response, and are usually installed at the positions where the high-voltage disconnect switch is most stressed and most prone to fatigue, such as the contact arm and the rotating shaft.

After the selection and installation of the resistance strain gauges are completed, the gauges are required to be connected to the data collection system through wires. The data collection system is responsible for recording the resistance changes transmitted from the resistance strain gauges and converting them into voltage signals for reading. The data collection system needs to have a high-speed sampling rate and high resolution to ensure that it can capture the rapid strain changes generated during the operation of the high-voltage disconnect switch. The sampling rate used is usually in the kilohertz range, and the resolution reaches the millivolt level.

Appropriate software is used to process the collected voltage signals. First, filtering is performed to remove possible noise interference, and then mathematical algorithms such as the Fast Fourier Transform (FFT) are used to analyze the signal spectrum and extract strain data. The strain data can be converted to obtain the actual stress state of the corresponding component.

The measured strain data is compared with the pre-established stress model of the high-voltage disconnect switch to evaluate the current health status of the equipment. When the monitored stress exceeds the design threshold, the data collection system will automatically issue a warning signal to remind the operation and maintenance personnel to conduct inspections or maintenance.

3 Conclusion

This article has in-depth explored the common types of mechanical failures of high-voltage disconnect switches and their intelligent diagnosis methods. Using intelligent diagnosis methods for mechanical failures of high-voltage disconnect switches can not only improve the reliability of equipment operation but also significantly reduce maintenance costs and optimize the maintenance decision-making process.

With the progress of science and technology and the increasing maturity of data analysis technology, relevant personnel need to increase research investment to improve the intelligent diagnosis level of mechanical failures of high-voltage disconnect switches, providing strong support for the stable operation of the power system.