Generator transformer unit protection device performance detection system and method

Abstract

The application provides a generator transformer unit protection device performance detection system and method, relates to the generator transformer unit protection technical field, and solves the problem of single monitoring data dimension in traditional technology through multi-dimensional data collection and fusion. The system collects various state parameters of the generator and the transformer, can reflect the equipment operation state in all directions, accurately identifies electrical, acoustic, mechanical and thermal abnormalities through modular feature extraction, extracts key features from the signal by using fast Fourier transform and other technologies, generates the transformer abnormality index and the power generation abnormality index by establishing a mathematical analysis model, and the comprehensive analysis capability can more accurately evaluate the overall health condition of the equipment. The performance detection module combines the transformer abnormality index and the power generation abnormality index, calculates the comprehensive abnormality index, compares the comprehensive abnormality index with the preset threshold value, and the system can timely identify the potential fault state of the equipment.

Description

Technical Field

[0001] This invention relates to the field of generator-transformer protection technology, specifically to a performance testing system and method for generator-transformer protection devices. Background Technology

[0002] Generator-transformer units are core equipment in power systems, primarily responsible for converting electrical energy generated by generators into voltage and current forms suitable for transmission requirements. Their operational stability and reliability are crucial to the security of the entire power grid. These devices typically operate under high loads and complex environments; therefore, real-time monitoring and preventative maintenance are key to ensuring the efficient operation of the power system.

[0003] Currently, monitoring technology for generator-transformer units mainly relies on monitoring single or a few parameters. For example, temperature monitoring is used to detect overheating, vibration monitoring to identify mechanical faults, and voltage monitoring to determine electrical anomalies. These monitoring systems typically use simple threshold-based judgment methods to determine the equipment status. Once a parameter exceeds a set safety range, the system will issue an alarm. However, this method struggles to capture the complex interactions between multiple signals, which can lead to false alarms or missed alarms in practical applications.

[0004] The main shortcomings of existing technologies lie in their limited monitoring dimensions and analytical capabilities. Due to the inability to effectively integrate and analyze various operational data, potential early fault signals in equipment may be overlooked. Furthermore, relying on human experience for fault diagnosis increases the uncertainty of the diagnosis. Therefore, improving the comprehensive analytical capabilities of fault diagnosis and monitoring systems, integrating multi-source data, and achieving a holistic assessment of equipment status is an important direction for current technological development. This will help improve the accuracy and timeliness of fault early warning, ensuring the stable operation of the power system.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a performance testing system and method for generator transformer protection devices to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A performance testing system for generator-transformer set protection devices, comprising:

[0009] The data acquisition module is used to collect the power generation status data of the generator and the voltage transformation status data of the transformer over the past k time period. The power generation status data includes the vibration signal of the generator bearing housing and the temperature signal of the stator winding housing surface. The voltage transformation status data includes the noise signal of the transformer operation and the voltage signal of the primary side.

[0010] The first processing module is used to extract features from the voltage signal on the primary side of the transformer, obtain the harmonic content and distortion rate in the voltage signal, and perform a fast Fourier transform on the noise signal to obtain the peak frequency of the noise signal from the transformed frequency signal.

[0011] The second processing module is used to extract the root mean square value and peak factor of the vibration signal from the vibration signal, and to extract the maximum value, minimum value and temperature change rate of the temperature from the temperature signal.

[0012] The data analysis module is used to analyze and process the harmonic content, distortion rate and noise peak frequency in the voltage signal to generate the transformer anomaly index. It also establishes a mathematical analysis model based on the vibration mean square value, peak factor and the maximum, minimum and temperature change rate of the vibration signal to obtain the generator power generation anomaly index.

[0013] The performance detection module is used to calculate a comprehensive anomaly index by combining the transformer anomaly index and the power generation anomaly index, and compare the comprehensive anomaly index with the anomaly threshold to determine the operating status of the equipment.

[0014] Furthermore, both the power generation status data and the transformer transformation status data are continuous time-series data. The time interval for the collected power generation status data and transformer transformation status data is [t0-k, t0], where t0 represents the current time value and k represents the length of the time interval for the collected power generation status data and transformer transformation status data as time-series data.

[0015] Furthermore, the method for obtaining the harmonic content and distortion rate in the voltage signal is as follows:

[0016] A Fourier transform is performed on the voltage signal on the primary side of the transformer to convert the voltage signal from the time domain to the frequency domain. The Fourier transform transforms the voltage signal into the form of a series of complex numbers. The number of complex numbers is represented as the number of frequency components within the voltage signal. The imaginary part of the frequency component of each complex number is represented as the phase of the corresponding frequency component within the voltage signal. The real part of the frequency component of each complex number is represented as the amplitude of the corresponding frequency component. This process identifies the fundamental and harmonic frequency components in the voltage signal.

[0017] The harmonic content and distortion rate are generated by comparing the amplitudes of each harmonic in the voltage signal with the fundamental amplitude. The formula used is as follows:

[0018]

[0019] Where Hd represents the harmonic content, T hd V represents the distortion rate in a voltage signal. i V represents the amplitude of the i-th harmonic in the voltage signal. J represents the amplitude of the fundamental wave in the voltage signal, v represents the number of harmonics in the voltage signal, and i represents the harmonic number in the voltage signal.

[0020] Furthermore, the method for obtaining the dominant peak frequency in the noise signal is as follows:

[0021] The noise signal is sampled by setting a sampling rate to obtain the discrete-time signal of the noise signal. The Fast Fourier Transform is applied to the sampled discrete signal to convert the discrete-time signal of the noise signal to the frequency domain, thereby obtaining the spectrum of the noise signal and the frequency resolution of the noise signal.

[0022] The amplitude of each frequency component of the noise signal is obtained from the spectrum of the noise signal. The location of the maximum amplitude is identified, and the frequency location corresponding to the maximum amplitude is multiplied by the frequency resolution of the noise signal to obtain the peak frequency of the noise signal.

[0023] Furthermore, the specific steps for extracting the root mean square value and peak factor of the vibration signal from the vibration signal are as follows:

[0024] The continuous vibration signal is discretized at equal time intervals to obtain discrete samples of the vibration signal: x1, x2, ..., x j ... x N ,x j The amplitude of the vibration signal at the j-th sampling point in the discrete sample of the vibration signal is represented by N, where N represents the total number of discrete points in the discrete sample of the vibration signal, and j represents the number of the discrete point.

[0025] The root mean square number and peak factor of the vibration signal are calculated based on the amplitude of all discrete points in the discrete samples of the vibration signal. The formulas used are as follows:

[0026]

[0027] Among them, R z and Bl z Let x(t) represent the root mean square value and peak factor of the vibration signal, respectively, and let t be the time variable.

[0028] The specific steps for extracting the maximum, minimum, and rate of change of temperature from the temperature signal are as follows:

[0029] Tw max =max[Tw(t)]

[0030] Tw min =min[Tw(t)]

[0031]

[0032] in, ΔTw and ΔTw represent the maximum, minimum, and rate of temperature change in the temperature signal, respectively, and Tw(t) represents the temperature signal, t max and t min These represent the time values ​​corresponding to the maximum and minimum temperatures, respectively.

[0033] Furthermore, the formula used to generate the transformer's transformer anomaly index is as follows:

[0034]

[0035] Among them, By y The transformer anomaly index is represented by Hd, which represents the harmonic content, and T represents the transformer anomaly index. hd Fp represents the distortion rate in the voltage signal, and Fp represents the peak frequency of the noise signal.

[0036] Furthermore, the specific power generation anomaly index of the generator is obtained as follows:

[0037]

[0038] Among them, Fd y R represents the power generation anomaly index. z and Bl z These represent the root mean square value and peak factor of the vibration signal, respectively. ΔTw and ΔTw represent the maximum, minimum, and rate of temperature change in the temperature signal, respectively.

[0039] Furthermore, the specific logic for calculating the comprehensive anomaly index is as follows:

[0040]

[0041] Where ZH represents the comprehensive anomaly index, α, β and γ represent the first weight coefficient, nonlinear coefficient and second weight coefficient respectively, and β>α>1>γ>0;

[0042] If ZH≤ZH y If ZH > ZH, then the generator-transformer unit is considered to be in normal operating condition. y If this is the case, then it is determined that there is an abnormality in the generator-transformer unit, and further inspection is required. y This indicates the abnormal threshold.

[0043] The present invention also provides a method for testing the performance of a generator-transformer set protection device. The testing method is performed by the aforementioned generator-transformer set protection device performance testing system, and the specific steps include:

[0044] Step 1: Collect the generator's power generation status data and the transformer's voltage transformation status data over the past k time periods. The power generation status data includes the vibration signal of the generator's bearing housing and the temperature signal of the stator winding housing surface. The voltage transformation status data includes the noise signal of the transformer's operation and the voltage signal on the primary side.

[0045] Step 2: Extract features from the voltage signal on the primary side of the transformer, obtain the harmonic content and distortion rate in the voltage signal, and perform a fast Fourier transform on the noise signal to obtain the peak frequency of the noise signal from the transformed frequency signal.

[0046] Step 3: Extract the root mean square value and peak factor of the vibration signal from the vibration signal, and extract the maximum value, minimum value and rate of change of temperature from the temperature signal;

[0047] Step 4: Analyze and process the harmonic content, distortion rate, and peak frequency of the noise main frequency in the voltage signal to generate the transformer anomaly index. Based on the vibration mean square value, peak factor, maximum and minimum temperature values, and temperature change rate of the vibration signal, establish a mathematical analysis model to obtain the generator power generation anomaly index.

[0048] Step 5: Combine the transformer anomaly index and the power generation anomaly index to calculate the comprehensive anomaly index, compare the comprehensive anomaly index with the anomaly threshold, and determine the operating status of the equipment.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] This invention solves the problem of single-dimensional monitoring data in traditional technologies by acquiring and fusing multi-dimensional data. The system collects multiple state parameters of generators and transformers, such as vibration, temperature, noise, and voltage signals, which can comprehensively reflect the operating status of the equipment. Through modular feature extraction, the system accurately identifies electrical, acoustic, mechanical, and thermal anomalies. Using techniques such as fast Fourier transform, key features such as harmonic content, root mean square value of vibration, and temperature change rate are extracted from the signals, which improves the accuracy of anomaly identification.

[0051] This invention establishes a mathematical analysis model, integrates various feature information, and generates transformer anomaly index and power generation anomaly index. This comprehensive analysis capability can more accurately assess the overall health status of the equipment. The performance detection module combines the transformer anomaly index and the power generation anomaly index to calculate the comprehensive anomaly index. By comparing the comprehensive anomaly index with a preset threshold, the system can promptly identify potential fault states of the equipment. Attached Figure Description

[0052] Figure 1 This is a schematic diagram of the overall system structure of the present invention;

[0053] Figure 2 This is a schematic diagram of the overall method flow of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0055] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0056] Example:

[0057] Please see Figure 1 The present invention provides a technical solution:

[0058] A performance testing system for a generator transformer protection device includes a data acquisition module, a first processing module, a second processing module, a data analysis module, and a performance testing module, wherein:

[0059] The data acquisition module is used to collect the generator's power generation status data and the transformer's voltage transformation status data over the past k time periods. The power generation status data includes the vibration signal of the generator's bearing housing and the temperature signal of the stator winding housing surface. The voltage transformation status data includes the noise signal of the transformer's operation and the voltage signal on the primary side.

[0060] The generator bearing housing refers to the exterior of the generator bearing. The generator bearing housing is used to support and maintain the rotation of the rotor. Vibration sensors installed on the bearing housing monitor the vibration of the bearing to determine its operating status and health condition.

[0061] The stator winding housing surface refers to the outer surface of the stator winding. The stator is the stationary part of the generator, and the winding is its key component, generating an electromagnetic field. Temperature sensors are installed on the stator winding housing surface to monitor temperature changes and prevent overheating from damaging the winding.

[0062] Noise signals are collected through acoustic sensors or microphones, which are installed on the transformer casing or near the core windings and other parts that are prone to noise generation, to capture and monitor noise signals. The data acquisition module acquires signals through sensors installed on key parts of the generator and transformer. Vibration sensors are installed on the generator bearing housing to detect mechanical vibration signals, and temperature sensors are located on the surface of the stator winding housing to monitor temperature changes. The transformer's noise signals are captured by acoustic sensors, while voltage sensors record the voltage signals on the primary side.

[0063] The purpose of collecting this data is to monitor the operating status of generators and transformers in real time. Vibration and temperature data can be used to determine the mechanical health of the generator and avoid potential faults. Noise and voltage signals are used to assess the electrical and mechanical performance of the transformer. For generators, bearing vibration signals are key to early identification of mechanical problems. Abnormal vibration may indicate bearing wear, imbalance, or misalignment. Timely monitoring helps prevent serious mechanical failures. In addition, stator winding temperature signals are an important health indicator of the generator. Excessive temperature may lead to insulation damage and electrical faults. Therefore, temperature monitoring can effectively avoid these risks.

[0064] For transformers, noise signals are an important method for detecting problems with the core and internal structure. The magnetostrictive effect of the core generates noise, and abnormal noise signals may indicate core faults or other mechanical problems. Therefore, noise monitoring helps to detect and correct potential structural faults early. In addition, the primary side voltage signal directly affects the electrical performance of the transformer. Voltage fluctuations may indicate load instability or electrical faults, affecting the quality of power supply and the safety of equipment.

[0065] In this embodiment, the power generation status data and the transformer transformation status data are both continuous time-series data. The time interval for the collected power generation status data and transformer transformation status data is [t0-k, t0], where t0 represents the current time value and k represents the length of the time interval for the collected power generation status data and transformer transformation status data as time-series data.

[0066] In this embodiment, generator status data and transformer status data are collected and analyzed as continuous time-series data. Time-series data means that the data is arranged in chronological order, with each data point having a timestamp. By monitoring this data, trends and patterns of change over time can be observed, which is crucial for predicting equipment status and detecting anomalies. Data acquisition covers an interval extending k time units backward from the current time t0. This interval length k allows us to review and analyze data changes over a period of time, rather than just data at a single moment. This time-series analysis helps identify long-term trends, periodic changes, and short-term fluctuations, resulting in a more comprehensive understanding of equipment status. Collecting and analyzing generator status data and transformer status data as continuous time-series data allows for more accurate prediction of equipment status changes and potential faults. Compared to single-moment data analysis, time-series data provides richer operational information about the generator-transformer unit.

[0067] The first processing module is used to extract features from the voltage signal on the primary side of the transformer, obtain the harmonic content and distortion rate in the voltage signal, and perform a fast Fourier transform on the noise signal to obtain the peak frequency of the noise signal from the transformed frequency signal.

[0068] The first processing module extracts features from the transformer primary voltage signal, including harmonic content and distortion rate, and performs a Fast Fourier Transform on the noise signal. This enables in-depth signal analysis. Compared to existing technologies, this method can more accurately identify the harmonic characteristics and distortion levels in the voltage signal and quickly capture the main frequency peaks of the noise signal. This process improves the resolution and accuracy of signal analysis, helping to more comprehensively assess the equipment's operating status. By accurately identifying and analyzing abnormal features in the signal, it provides reliable data support for subsequent fault identification and maintenance decisions, promoting the intelligence and efficiency of the overall solution and ensuring the stable operation of the system.

[0069] In this embodiment, the method for obtaining the harmonic content and distortion rate in the voltage signal is as follows:

[0070] A Fourier transform is performed on the voltage signal on the primary side of the transformer to convert the voltage signal from the time domain to the frequency domain. The Fourier transform transforms the voltage signal into the form of a series of complex numbers. The number of complex numbers is represented as the number of frequency components within the voltage signal. The imaginary part of the frequency component of each complex number is represented as the phase of the corresponding frequency component within the voltage signal. The real part of the frequency component of each complex number is represented as the amplitude of the corresponding frequency component. This process identifies the fundamental and harmonic frequency components in the voltage signal.

[0071] By performing a Fourier transform on the voltage signal on the primary side of the transformer, the time-domain signal can be converted into a frequency-domain signal. The Fourier transform decomposes the original voltage signal into a series of complex numbers, which represent different frequency components in the signal. The amplitude (real part) of each complex number represents the intensity of that frequency component, while the imaginary part represents its phase. This transformation allows us to identify the fundamental and harmonic frequency components in the voltage signal, and thus analyze the harmonic content and distortion rate of the signal.

[0072] In practice, the output of the Fourier transform helps us identify the main frequency components of a signal and their corresponding amplitudes, which is crucial for judging the quality of a voltage signal. By analyzing these frequency components, we can determine which components are the fundamental frequency and which are harmonics. The logic behind this step is to gain a deeper understanding of the internal composition of the voltage signal through frequency and amplitude analysis, thereby providing data support for subsequent signal quality assessment and anomaly detection.

[0073] The fundamental frequency is the standard frequency of a generator-transformer system. For example, in my country's AC power system, the fundamental frequency is usually 50Hz. Harmonics are integer multiples of the fundamental frequency. For instance, if the fundamental frequency is 50Hz, the harmonics could be 100Hz, 150Hz, etc. In a frequency spectrum diagram, after finding the fundamental frequency, identify the peaks of other frequencies. If these frequencies are integer multiples of the fundamental frequency, they are harmonics.

[0074] By following the steps above, the fundamental and harmonic components in a signal can be effectively distinguished. The amplitudes of each harmonic in the voltage signal are compared with the fundamental amplitude to generate the harmonic content and distortion rate. The formula used is as follows:

[0075]

[0076]

[0077] Where Hd represents the harmonic content, T hd V represents the distortion rate in a voltage signal. i V represents the amplitude of the i-th harmonic in the voltage signal. J represents the amplitude of the fundamental wave in the voltage signal, v represents the number of harmonics in the voltage signal, and i represents the harmonic number in the voltage signal.

[0078] Harmonic content reflects the overall intensity of harmonic components in a voltage signal. Harmonics are integer multiples of the fundamental frequency and affect the purity of the signal. When the harmonic content increases, it indicates that there are more harmonic components in the voltage signal, which means that the quality of the power signal decreases. This is because an ideal power signal should be a pure sine wave. Higher harmonic content will lead to increased heat loss in generator transformer sets, reduced equipment efficiency, and may cause equipment failure. When harmonic content is generated, the independent variable is the amplitude of each harmonic. An increase in these values ​​will increase the harmonic content. A voltage with high harmonic content means that there may be a problem with the generator transformer set, leading to a decrease in power generation quality.

[0079] Distortion rate indicates the overall degree of distortion of a voltage signal caused by harmonics. It measures the extent to which the signal deviates from an ideal sine wave. An increase in distortion rate indicates that the signal distortion is more severe and the signal deviates from the ideal state, which means that the power quality is deteriorated. High distortion rate can lead to overheating, reduced efficiency, or even damage to generator transformer sets. An increase in the amplitude of harmonics will increase the distortion rate, while an increase in the amplitude of the fundamental wave will decrease its value. A high distortion rate means that there may be defects in the generator transformer set, such as improper regulation or load imbalance, which leads to a decline in power generation quality.

[0080] When calculating harmonic content, the amplitudes of all harmonics are first summed to obtain a value representing the overall harmonic intensity. This part reflects the total amount of harmonics in the signal. To make the result more stable, the formula uses a logarithmic transformation. This transformation not only compresses the range of values ​​but also reduces the influence of extreme values, making the result smoother and easier to interpret. The logarithmic form provides a monotonically increasing relationship, allowing us to intuitively understand how changes in the total amount of harmonics affect the overall signal quality.

[0081] The distortion rate is calculated by first squaring the amplitude of each harmonic and then summing them. This step emphasizes the influence of larger harmonic amplitudes, highlighting their contribution to signal distortion. This emphasis on large harmonics ensures that their effect is not ignored when evaluating distortion. Subsequently, the square root of the sum of squares is taken to align with the original amplitude scale. Finally, the formula standardizes the distortion rate by dividing the result by the fundamental frequency amplitude, making it a relative value. This relative measurement facilitates comparisons between different signals and systems, clearly reflecting the degree of harmonic impact on overall signal quality. This design allows for a better understanding and control of signal distortion when assessing power quality.

[0082] Furthermore, the method for obtaining the dominant peak frequency in the noise signal is as follows:

[0083] The noise signal is sampled by setting a sampling rate to obtain the discrete-time signal of the noise signal. The Fast Fourier Transform is applied to the sampled discrete signal to convert the discrete-time signal of the noise signal to the frequency domain, thereby obtaining the spectrum of the noise signal and the frequency resolution of the noise signal.

[0084] The amplitude of each frequency component of the noise signal is obtained from the spectrum of the noise signal. The location of the maximum amplitude is identified, and the frequency location corresponding to the maximum amplitude is multiplied by the frequency resolution of the noise signal to obtain the peak frequency of the noise signal.

[0085] The sampling rate determines how many signal samples are collected per second. In order to accurately capture the frequency components in the signal, the sampling rate usually needs to be at least twice the highest frequency in the signal. Converting a continuous noise signal into a discrete time signal means dividing the signal into a series of equally spaced sampling points on the time axis.

[0086] The Fast Fourier Transform (FFT) algorithm converts discrete signals in the time domain into signals in the frequency domain. FFT is an efficient Fourier transform calculation method that can quickly calculate the signal's spectrum. The spectrum is the signal representation in the frequency domain, where each point represents the amplitude of a specific frequency. The frequency resolution is determined by the sampling rate and the number of FFT points, typically the sampling rate divided by the number of FFT points. It represents the frequency interval between each point in the spectrum. Each frequency component in the spectrum is examined to obtain its corresponding amplitude. The amplitude reflects the intensity of that frequency component in the signal. The frequency component with the largest amplitude is found in the spectrum. The frequency corresponding to this position is the signal's dominant peak frequency, usually the most prominent frequency component in the signal. Multiplying the spectral position corresponding to the largest amplitude by the frequency resolution yields the signal's dominant peak frequency. This step maps the position in the spectrum to the actual frequency value, and the result is the dominant frequency of the noise signal.

[0087] The peak frequency of a transformer noise signal can reflect the health status of the equipment. Abnormal frequency components may indicate internal faults, such as loose windings or core vibration. By monitoring the changing trend of the peak frequency, potential problems can be identified in advance, preventive maintenance can be carried out, and greater losses can be avoided. The higher the peak frequency, the more high-frequency components are contained in the noise. High-frequency noise may originate from rapidly vibrating parts of the transformer or electromagnetic interference. In some cases, high-frequency signals may indicate wear of mechanical parts or electrical faults.

[0088] Extracting the dominant peak frequency from the noise signal is a crucial step in transformer operation. This frequency reveals the equipment's operating status and potential problems, such as loose windings or abnormal core vibration. Determining the dominant peak frequency helps to identify and resolve these issues promptly, ensuring normal equipment operation and extending its service life.

[0089] A higher peak frequency usually indicates a greater presence of high-frequency components in the signal, which may reflect rapid movement of mechanical parts or electromagnetic interference. Compared to traditional methods, this analysis technique offers more accurate diagnostic capabilities, enabling more precise predictive maintenance and effectively reducing unplanned downtime and maintenance costs.

[0090] In this solution, acquiring and analyzing the dominant peak frequency in the noise signal is a core component of the entire monitoring system. This step improves the diagnostic accuracy of the system and provides important reference data for other monitoring data. This not only enhances the reliability of transformer operation but also strengthens the effectiveness of the entire maintenance and management strategy.

[0091] The second processing module is used to extract the root mean square value and peak factor of the vibration signal from the vibration signal, and to extract the maximum value, minimum value and rate of change of temperature from the temperature signal.

[0092] In this embodiment, the root mean square value and peak factor of the vibration signal are extracted from the vibration signal as follows:

[0093] The continuous vibration signal is discretized at equal time intervals to obtain discrete samples of the vibration signal: x1, x2, ..., x j ... x N ,x j The amplitude of the vibration signal at the j-th sampling point in the discrete sample of the vibration signal is represented by N, where N represents the total number of discrete points in the discrete sample of the vibration signal, and j represents the number of the discrete point.

[0094] The root mean square number and peak factor of the vibration signal are calculated based on the amplitude of all discrete points in the discrete samples of the vibration signal. The formulas used are as follows:

[0095]

[0096] Among them, R z and Bl z Let x(t) represent the root mean square value and peak factor of the vibration signal, respectively, and let t be the time variable.

[0097] The root mean square (RMS) vibration reflects the energy level of the signal and is a comprehensive indicator of vibration intensity. A higher RMS vibration value indicates a higher overall vibration intensity. A high RMS vibration value may suggest abnormal vibration in the equipment, requiring further investigation. The gusset factor represents the ratio of the maximum amplitude of the signal to the RMS vibration. A higher gusset factor indicates more significant peaks in the signal. A large gusset factor may indicate transient impacts or abnormal peaks, suggesting potential mechanical faults.

[0098] The root mean square (RMS) value of vibration is the square root of the mean of the sum of the squares of all vibration amplitudes. The magnitude of each amplitude directly affects the RMS; the larger the amplitude, the larger the RMS root. The peak factor is determined by the maximum amplitude and the RMS root; the larger the maximum amplitude or the smaller the RMS root, the larger the peak factor.

[0099] By analyzing the mean square fraction and peak factor of vibration signals, this method provides a precise assessment of vibration intensity and anomalies. Compared to existing technologies, it can identify potential faults, such as mechanical loosening or imbalance, earlier, efficiently capture and quantify key features in vibration signals, making maintenance more accurate and timely. By extracting key indicators from vibration signals, continuous monitoring of equipment status can be achieved. Compared to traditional methods, it can detect abnormal trends more quickly, improving the accuracy of fault diagnosis, especially in its ability to identify instantaneous impacts and abnormal peak values.

[0100] The specific steps for extracting the maximum, minimum, and rate of change of temperature from the temperature signal are as follows:

[0101] Tw max =max[tw(t)]

[0102] Tw min =min[Tw(t)]

[0103]

[0104] in, ΔTw and ΔTw represent the maximum, minimum, and rate of temperature change in the temperature signal, respectively, and Tw(t) represents the temperature signal, t max and t min These represent the time values ​​corresponding to the maximum and minimum temperatures, respectively.

[0105] The maximum and minimum temperatures reflect the operating status of the stator windings under extreme conditions. High temperatures may indicate equipment overload or insufficient heat dissipation, while low temperatures may be caused by environmental factors or equipment not being operated. The rate of temperature change represents the rate of temperature fluctuation over a short period of time. A larger rate of change indicates drastic temperature changes, which may indicate equipment abnormalities or potential faults.

[0106] Extracting the maximum, minimum, and rate of change of temperature helps to effectively monitor the generator's operating status. These indicators provide a scientific basis for predicting equipment failures, support preventative maintenance strategies, and reduce the likelihood of unexpected downtime. Furthermore, analyzing temperature signals can improve the stability and efficiency of equipment operation, extend equipment lifespan, and ensure the reliability and safety of the power generation process.

[0107] The data analysis module is used to analyze and process the harmonic content, distortion rate and noise peak frequency in the voltage signal to generate the transformer anomaly index. It also establishes a mathematical analysis model based on the vibration mean square score, peak factor and maximum, minimum and temperature change rate of the vibration signal to obtain the generator power generation anomaly index.

[0108] In this embodiment, the formula used to generate the transformer's transformer anomaly index is:

[0109]

[0110] Among them, By y The transformer anomaly index is represented by Hd, which represents the harmonic content, and T represents the transformer anomaly index. hd Fp represents the distortion rate in the voltage signal, and Fp represents the peak frequency of the noise signal.

[0111] The transformer anomaly index integrates the harmonic content, distortion rate, and peak frequency of the noise signal in the voltage signal, reflecting the overall state of the transformer voltage quality. The larger the index, the higher the degree of voltage signal distortion, indicating that the equipment may be subjected to greater electrical stress and abnormal operating conditions. It helps to detect and diagnose transformer voltage problems, supports decision-makers in taking measures for maintenance and optimization, and ensures the stability and safety of the power system. In the data processing on the transformer side, although the transformer has not yet been actually completed, the analysis of the harmonic content and distortion rate of the voltage signal is to assess the transformer's performance under load changes or power system anomalies in advance.

[0112] Harmonic content and distortion rate are important components of voltage signal quality. Higher harmonic content can lead to equipment overheating and damage, while higher distortion rate indicates that the voltage waveform deviates from normal, increasing electrical equipment losses. Fp represents the most significant frequency component in the signal. When Fp increases, it means there is a strong specific frequency component in the signal, which may be related to some anomalies. Therefore, a larger Fp indicates a higher degree of potential anomalies. y The larger the value, the greater the combined effect of the three factors: harmonic content and distortion rate. Increased harmonic content and distortion rate lead to a higher transformer anomaly index, indicating a decline in voltage signal quality and transformer malfunction. Therefore, the formula reflects the direct relationship between the independent and dependent variables, aiding in the identification and handling of electrical anomalies.

[0113] An increase in the peak frequency of the main frequency indicates the presence of high-frequency noise in the transformer, which may signify partial discharge or other abnormalities. These high-frequency components can cause aging or damage to the insulation materials. An increase in high-frequency components may reflect wear or vibration problems in mechanical parts. These factors can affect the normal operation of the transformer. High-frequency harmonics may originate from instability in the external power grid or the nonlinear characteristics of the load, which can lead to additional losses and temperature rise in the transformer. Therefore, when the peak frequency of the noise signal increases, the transformer needs to be carefully inspected to prevent potential faults.

[0114] It is a Euclidean norm form representing the combined magnitude of harmonic content, distortion rate, and the peak frequency of the noise signal. By taking the square root of the sum of squares, the overall impact of these three factors on voltage quality can be determined. The sum of squares ensures that the influence of each factor is fully considered and that there is no cancellation due to signs. Reflecting the relative relationship between harmonic content and distortion rate, this study emphasizes the proportional relationship between the two, thus providing more information about the direction of voltage distortion. The arctangent function can be used to convert the proportion into an angle, helping to identify the main sources of distortion.

[0115] Furthermore, the specific power generation anomaly index of the generator is obtained as follows:

[0116]

[0117] Among them, Fd y R represents the power generation anomaly index. z and Bl z These represent the root mean square value and peak factor of the vibration signal, respectively. ΔTw and ΔTw represent the maximum, minimum, and rate of temperature change in the temperature signal, respectively.

[0118] The generator anomaly index integrates the generator's vibration signal (root mean square value and peak factor) and temperature signal (maximum value, minimum value, and rate of temperature change). The larger the generator anomaly index, the greater the vibration and temperature fluctuations that the generator may have. This may mean that the equipment is malfunctioning or the risk of failure is increasing. It can help monitor and diagnose the generator's operating status, support timely maintenance and optimization, and ensure the generator's safety and efficiency.

[0119] The root mean square (RMS) vibration value and peak factor reflect the vibration status of the generator. The RMS vibration value represents the vibration intensity, while the peak factor reflects the sharpness of the vibration. The temperature index reflects the thermal state of the generator. The extreme values ​​and rate of change of temperature are important factors in the thermal performance of the equipment. Abnormal changes in vibration and temperature will increase the power generation anomaly index, indicating possible operational problems of the equipment. The influence of vibration intensity and sharpness on the power generation anomaly index is positively correlated, which means that an increase in vibration will directly lead to an increase in the index, and the greater the temperature change, the larger the index.

[0120] Vibration signals are characterized by the root mean square value and peak factor. The logarithmic transformation is used to reduce the values ​​of different magnitudes to a more comparable range. The logarithmic transformation can effectively smooth numerical changes, reduce the impact of extreme values ​​on the overall calculation, make the anomaly index more stable, avoid the excessive dominance of a single feature in the anomaly index, and ensure that the influence of each vibration feature is reasonably considered.

[0121] The square root part of the formula adopts the mathematical form of Euclidean distance to comprehensively measure vibration characteristics, taking vibration intensity and sharpness as a whole. In this form, the formula can fully reflect the multifaceted characteristics of the vibration signal in the calculation, ensuring that the anomaly index can accurately represent the overall abnormal condition of the vibration, thereby providing a warning of potential problems of the generator.

[0122] Temperature signals are described by maximum, minimum, and rate of temperature change. The arctangent transform in the formula converts the relationship between temperature fluctuation amplitude and rate of change into an angular expression. This design emphasizes the trend and magnitude of temperature changes, using angles to represent the possibility of temperature anomalies. The arctangent function not only provides a means to compare different temperature characteristics but also maintains the stability of exponential calculations in extreme cases. Combining these characteristics, the formula effectively identifies and quantifies abnormal generator states, supporting reliable operation and maintenance decisions.

[0123] The performance detection module is used to calculate a comprehensive anomaly index by combining the transformer anomaly index and the power generation anomaly index, and compare the comprehensive anomaly index with the anomaly threshold to determine the operating status of the equipment.

[0124] The specific logic for calculating the comprehensive anomaly index is as follows:

[0125]

[0126] Where ZH represents the comprehensive anomaly index, α, β and γ represent the first weight coefficient, the nonlinear coefficient and the second weight coefficient, respectively, and β>α>1>γ>0.

[0127] The comprehensive anomaly index represents the degree of abnormality in the overall operating status of the generator-transformer unit. By combining the transformer anomaly index and the power generation anomaly index, it provides a comprehensive indicator to measure the health status of the equipment. The higher the comprehensive anomaly index value, the more the overall operating status of the generator-transformer unit deviates from the normal range, and there may be potential faults or anomalies.

[0128] The transformer anomaly index and the power generation anomaly index reflect the overall abnormal state of the generator-transformer unit. Considering both factors comprehensively allows for a more holistic assessment of the equipment's overall health. They are combined using a sum of squares to represent their combined impact on equipment anomalies. Weighting coefficients and nonlinear coefficients adjust their contributions to ensure the stability and accuracy of the calculation. An increase in both the transformer and power generation anomaly indices leads to a larger overall anomaly index.

[0129] This part directly affects the base value of the comprehensive anomaly index, reflecting the cumulative effect of the sum of squares, γ*|By y -Fd y The absolute value of the difference between the two is partially reflected. Although γ is small, it balances the impact of different types of anomalies, ensuring accurate assessment of equipment under various abnormal conditions. Transformer anomalies and generator anomalies are placed within the same framework, and their impact is balanced using a squared average. The squared approach amplifies larger anomalies, increasing sensitivity to extreme anomalies. The exponential β adjusts the degree of nonlinearity, ensuring that the calculation results are closer to actual anomalies.

[0130] When By y -Fd y When the values ​​are similar, this contribution is small, emphasizing the consistency of anomalies; while when there is a large difference between the two, this contribution increases. The difference between the two provides a measure of the inconsistency between the two anomaly indices. When there is a significant difference between the two, this part is given extra weight, indicating that extra attention needs to be paid to a certain aspect of the anomaly, thus enhancing the comprehensiveness of the overall assessment. The formula can not only accurately reflect the degree of anomaly of the current equipment, but also provide more detailed analysis and early warning according to the anomaly type. It balances the sensitivity to anomalies and the stability of the calculation, and avoids the anomaly index being too biased towards a certain type of anomaly.

[0131] If ZH≤ZH y If ZH > ZH, then the generator-transformer unit is considered to be in normal operating condition. y If this is the case, then it is determined that there is an abnormality in the generator-transformer unit, and further inspection is required. yThe anomaly threshold is a pre-set standard value used to distinguish between normal and abnormal states of equipment. It is usually determined based on historical data, equipment characteristics, and empirical values. When the comprehensive anomaly index is less than or equal to the threshold, it indicates that the equipment is operating within the normal range. When the comprehensive anomaly index exceeds the threshold, it indicates that the equipment is experiencing a possible anomaly and requires further inspection and verification. By setting the threshold reasonably, false alarms and unnecessary maintenance operations can be reduced, anomalies can be identified and handled early, and the overall reliability and safety of the equipment can be improved.

[0132] This embodiment solves the problem of single-dimensional monitoring data in traditional technologies by acquiring and fusing multi-dimensional data. The system collects multiple state parameters of generators and transformers, such as vibration, temperature, noise and voltage signals, which can comprehensively reflect the operating status of the equipment. Through modular feature extraction, the system accurately identifies electrical, acoustic, mechanical and thermal anomalies. Using techniques such as fast Fourier transform, key features such as harmonic content, root mean square value of vibration, and temperature change rate are extracted from the signals, which improves the accuracy of anomaly identification.

[0133] This embodiment establishes a mathematical analysis model, integrates various feature information, and generates transformer anomaly index and power generation anomaly index. This comprehensive analysis capability can more accurately assess the overall health status of the equipment. The performance detection module combines the transformer anomaly index and the power generation anomaly index to calculate the comprehensive anomaly index. By comparing the comprehensive anomaly index with a preset threshold, the system can promptly identify potential fault states of the equipment.

[0134] Please see Figure 2 The present invention also provides a method for testing the performance of a generator-transformer set protection device. The testing method is performed by the aforementioned generator-transformer set protection device performance testing system, and the specific steps include:

[0135] Step 1: Collect the generator's power generation status data and the transformer's voltage transformation status data over the past k time periods. The power generation status data includes the vibration signal of the generator's bearing housing and the temperature signal of the stator winding housing surface. The voltage transformation status data includes the noise signal of the transformer's operation and the voltage signal on the primary side.

[0136] Step 2: Extract features from the voltage signal on the primary side of the transformer, obtain the harmonic content and distortion rate in the voltage signal, and perform a fast Fourier transform on the noise signal to obtain the peak frequency of the noise signal from the transformed frequency signal.

[0137] Step 3: Extract the root mean square value and peak factor of the vibration signal from the vibration signal, and extract the maximum value, minimum value and rate of change of temperature from the temperature signal;

[0138] Step 4: Analyze and process the harmonic content, distortion rate, and peak frequency of the noise main frequency in the voltage signal to generate the transformer anomaly index. Based on the vibration mean square value, peak factor, maximum and minimum temperature values, and temperature change rate of the vibration signal, establish a mathematical analysis model to obtain the generator power generation anomaly index.

[0139] Step 5: Combine the transformer anomaly index and the power generation anomaly index to calculate the comprehensive anomaly index, compare the comprehensive anomaly index with the anomaly threshold, and determine the operating status of the equipment.

[0140] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0141] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0142] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0143] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A performance testing system for generator-transformer set protection devices, characterized in that, include: Data acquisition module, the data acquisition module is used to collect past... The power generation status data of the generator and the voltage transformation status data of the transformer within a time period, wherein the power generation status data includes the vibration signal of the generator bearing housing and the temperature signal of the stator winding housing surface, and the voltage transformation status data includes the noise signal of the transformer operation and the voltage signal of the primary side. The first processing module is used to extract features from the voltage signal on the primary side of the transformer, obtain the harmonic content and distortion rate in the voltage signal, and perform a fast Fourier transform on the noise signal to obtain the peak frequency of the noise signal from the transformed frequency signal. The second processing module is used to extract the root mean square value and peak factor of the vibration signal from the vibration signal, and to extract the maximum value, minimum value and temperature change rate of the temperature from the temperature signal. The data analysis module is used to analyze and process the harmonic content, distortion rate and noise peak frequency in the voltage signal to generate the transformer anomaly index. It also establishes a mathematical analysis model based on the vibration mean square value, peak factor and the maximum, minimum and temperature change rate of the vibration signal to obtain the generator power generation anomaly index. The performance detection module is used to combine the transformer anomaly index and the power generation anomaly index to calculate a comprehensive anomaly index, compare the comprehensive anomaly index with the anomaly threshold, and determine the operating status of the equipment. The formula used to generate the transformer anomaly index is: in, This indicates the abnormal voltage index. Indicates harmonic content, This represents the distortion rate in a voltage signal. Indicates the peak frequency of the noise signal; The specific power generation anomaly index of the generator is as follows: in, Indicates the abnormal power generation index. and These represent the root mean square value and peak factor of the vibration signal, respectively. , and These represent the maximum, minimum, and rate of temperature change in the temperature signal, respectively. The specific logic for calculating the comprehensive anomaly index is as follows: in, Indicates the comprehensive abnormality index. , and Let represent the first weighting coefficient, the nonlinear coefficient, and the second weighting coefficient, respectively. ; like If so, the generator-transformer unit is considered to be in normal operating condition. If this is the case, it indicates that there is an abnormality in the generator-transformer unit, and further inspection is required. This indicates the abnormal threshold.

2. The performance testing system for a generator-transformer set protection device according to claim 1, characterized in that: Both the power generation status data and the transformer transformation status data are continuous time-series data, and the time interval for collecting the power generation status data and transformer transformation status data is [time range missing]. , Indicates the current time value. This indicates the length of the time interval for collecting power generation status data and transformer status data as time-series data.

3. The performance testing system for a generator-transformer set protection device according to claim 2, characterized in that: The method for obtaining the harmonic content and distortion rate in a voltage signal is as follows: A Fourier transform is performed on the voltage signal on the primary side of the transformer to convert the voltage signal from the time domain to the frequency domain. The Fourier transform transforms the voltage signal into the form of a series of complex numbers. The number of complex numbers is represented as the number of frequency components within the voltage signal. The imaginary part of the frequency component of each complex number is represented as the phase of the corresponding frequency component within the voltage signal. The real part of the frequency component of each complex number is represented as the amplitude of the corresponding frequency component. This process identifies the fundamental and harmonic frequency components in the voltage signal. The harmonic content and distortion rate are generated by comparing the amplitudes of each harmonic in the voltage signal with the fundamental amplitude. The formula used is as follows: in, Indicates harmonic content, This represents the distortion rate in a voltage signal. Indicates the first voltage signal The amplitude of each harmonic. This represents the amplitude of the fundamental wave in a voltage signal. This indicates the number of harmonics in a voltage signal. This indicates the harmonic number in the voltage signal.

4. The performance testing system for a generator-transformer set protection device according to claim 2, characterized in that: The method for obtaining the peak frequency of the dominant frequency in a noise signal is as follows: The noise signal is sampled by setting a sampling rate to obtain the discrete-time signal of the noise signal. The Fast Fourier Transform is applied to the sampled discrete signal to convert the discrete-time signal of the noise signal to the frequency domain, thereby obtaining the spectrum of the noise signal and the frequency resolution of the noise signal. The amplitude of each frequency component of the noise signal is obtained from the spectrum of the noise signal. The location of the maximum amplitude is identified, and the frequency location corresponding to the maximum amplitude is multiplied by the frequency resolution of the noise signal to obtain the peak frequency of the noise signal.

5. The performance testing system for a generator-transformer set protection device according to claim 2, characterized in that: The specific steps for extracting the root mean square value and peak factor of a vibration signal are as follows: Discretize the continuous vibration signal at equal time intervals to obtain discrete samples of the vibration signal: , In a discrete sample representing a vibration signal, the first... The amplitude of the vibration signal at each sampling point. This represents the total number of discrete points in a discrete sample of a vibration signal. Indicates the number of discrete points; The root mean square number and peak factor of the vibration signal are calculated based on the amplitude of all discrete points in the discrete samples of the vibration signal. The formulas used are as follows: in, and These represent the root mean square value and peak factor of the vibration signal, respectively. Indicates vibration signal, It is a time variable; The specific steps for extracting the maximum, minimum, and rate of change of temperature from the temperature signal are as follows: in, , and These represent the maximum, minimum, and rate of temperature change in the temperature signal, respectively. Represents a temperature signal. and These represent the time values ​​corresponding to the maximum and minimum temperatures, respectively.

6. A method for testing the performance of a generator-transformer set protection device, characterized in that: The detection method is performed by the performance testing system for a generator transformer protection device according to any one of claims 1-5, and the specific steps include: Step 1: Collect past data The power generation status data of the generator and the voltage transformation status data of the transformer within a time period, wherein the power generation status data includes the vibration signal of the generator bearing housing and the temperature signal of the stator winding housing surface, and the voltage transformation status data includes the noise signal of the transformer operation and the voltage signal of the primary side. Step 2: Extract features from the voltage signal on the primary side of the transformer, obtain the harmonic content and distortion rate in the voltage signal, and perform a fast Fourier transform on the noise signal to obtain the peak frequency of the noise signal from the transformed frequency signal. Step 3: Extract the root mean square value and peak factor of the vibration signal from the vibration signal, and extract the maximum value, minimum value and rate of change of temperature from the temperature signal; Step 4: Analyze and process the harmonic content, distortion rate, and peak frequency of the noise main frequency in the voltage signal to generate the transformer anomaly index. Based on the vibration mean square value, peak factor, maximum and minimum temperature values, and temperature change rate of the vibration signal, establish a mathematical analysis model to obtain the generator power generation anomaly index. Step 5: Combine the transformer anomaly index and the power generation anomaly index to calculate the comprehensive anomaly index, compare the comprehensive anomaly index with the anomaly threshold, and determine the operating status of the equipment.

Images