Power grid technical improvement and maintenance optimization method and system based on fault prediction

Abstract

The invention discloses a power grid technical improvement and maintenance optimization method and system based on fault prediction, and relates to the technical field of power grid technical improvement and maintenance optimization, and the method comprises the following steps: monitoring equipment state data in real time; acquiring historical fault data of the power grid equipment; establishing an equipment attenuation model according to the historical fault data of the power grid equipment, and generating an equipment aging coefficient according to the established equipment attenuation model; according to the equipment state data and the equipment aging coefficient, a relation between the equipment state data and power grid equipment historical fault data and power grid technical improvement fault prediction is captured based on machine learning, a power grid technical improvement fault prediction model is constructed, and a power grid technical improvement fault prediction coefficient is generated; and performing fault prediction on the power grid technical improvement equipment according to the power grid technical improvement fault prediction coefficient. Through fault prediction, the problem that potential faults cannot be found in advance by fully utilizing operation data and state information of equipment is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid technical renovation and maintenance optimization, and more specifically, to the technical field of a power grid technical renovation and maintenance optimization method and system based on fault prediction. Background Art

[0002] The power system is an important part of the infrastructure of modern society, with the characteristics of high load, high reliability and large scale. With the continuous increase in electricity demand and the gradual emergence of the problem of aging of power equipment, how to operate the power system efficiently and reliably and ensure the safety and stability of power supply has become an important issue in the field of power engineering. In the operation and management of the power grid, traditional power grid maintenance work mainly relies on regular inspections and maintenance. Although this method can effectively prevent some sudden failures, it often lags in the prediction, detection and response of equipment failures, and fails to make full use of the equipment's operating data and status information to detect potential failures in advance, resulting in unnecessary power outages, maintenance and waste of resources. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method and system for optimizing technical transformation and maintenance of a power grid based on fault prediction, which solves the problem that there is often a lag in the prediction, detection and response of equipment failures through fault prediction, and fails to fully utilize the operating data and status information of the equipment to discover problems raised by potential faults in advance.

[0004] To achieve the above object, the present invention provides the following technical solutions: A method and system for optimizing technical transformation and maintenance of power grid based on fault prediction, which monitors equipment status data in real time, wherein the equipment status data includes abnormal data of equipment operation status, temperature, and vibration intensity; obtains historical fault data of power grid equipment, wherein the historical fault data includes initial performance value of equipment insulation resistance, initial value of mechanical wear, and service life and maintenance history data of equipment; establishes an equipment attenuation model according to the historical fault data of power grid equipment, and generates an equipment aging coefficient according to the established equipment attenuation model; captures the relationship between equipment status data and historical fault data of power grid equipment and power grid technical transformation fault prediction based on machine learning according to the equipment status data combined with the equipment aging coefficient, constructs a power grid technical transformation fault prediction model, and generates a power grid technical transformation fault prediction coefficient; performs fault prediction on power grid technical transformation equipment according to the power grid technical transformation fault prediction coefficient and repairs equipment that is about to fail.

[0005] In a preferred embodiment, the abnormal data of the equipment operation status includes an abnormal power coefficient of the equipment. The specific acquisition method of the abnormal power coefficient of the equipment is as follows: real-time monitoring and collection of voltage data of equipment in the power system, data analysis based on preset voltage stability data to obtain power equipment with abnormal fluctuation data; collecting historical harmonic data of power equipment, and analyzing the historical harmonic data based on standard harmonic data to obtain current harmonic influence factors and voltage harmonic influence factors; obtaining harmonic current, fundamental current, harmonic voltage and fundamental voltage of the power equipment with abnormal fluctuation data; calculating the current harmonic abnormal coefficient based on the current harmonic distortion rate according to the harmonic current and fundamental current combined with the current harmonic influence factor; calculating the voltage harmonic abnormal coefficient based on the voltage harmonic distortion rate according to the harmonic voltage and fundamental voltage combined with the voltage harmonic influence factor; calculating the current harmonic abnormal coefficient and the voltage harmonic abnormal coefficient based on the equipment operation status abnormal formula to obtain the equipment power abnormal coefficient.

[0006] In a preferred embodiment, the abnormal vibration intensity data is specifically acquired by the following method: acquiring the vibration signal of the power equipment in real time, denoising the vibration signal and extracting characteristic data with obvious vibration; acquiring the vibration signal of the normal operation of the equipment and obtaining the overall change law of the vibration signal on the time axis based on time domain analysis, and performing data analysis on the overall change law of the vibration signal on the time axis to obtain vibration intensity benchmark data; obtaining the vibration signal frequency domain based on Fourier transform of the characteristic data with obvious vibration, and analyzing the vibration signal frequency domain to obtain vibration signal frequency data; performing comparative analysis on the vibration signal frequency data in combination with the vibration intensity benchmark data to obtain abnormal vibration intensity data.

[0007] In a preferred embodiment, the aging coefficient of the equipment is specifically obtained by the following method: obtaining the initial performance value of the insulation resistance of the equipment, the initial value of mechanical wear and the service life of the equipment; obtaining the attenuation coefficient of the equipment performance by collecting and analyzing historical performance data; establishing an equipment attenuation model based on the initial performance value of the insulation resistance, the initial value of mechanical wear and the service life of the equipment in combination with the attenuation coefficient; collecting maintenance history data, performing feature extraction on the maintenance history data, and obtaining the number of historical maintenance times and historical fault types of the equipment; determining the historical fault components of the equipment based on the historical fault types, and obtaining the aging rate of the equipment for the historical fault components based on the equipment attenuation model.

[0008] In a preferred embodiment, the device status data is combined with the device aging coefficient based on machine learning to capture the relationship between the device status data and the historical fault data of the power grid equipment and the prediction of the power grid technical transformation fault, build a power grid technical transformation fault prediction model, and generate a power grid technical transformation fault prediction coefficient. The specific steps are as follows: obtain historical device status data and historical equipment aging coefficients, preprocess and normalize the historical device status data and historical equipment aging coefficients to obtain characteristic data that affect the power grid technical transformation fault; divide the characteristic data that affect the power grid technical transformation fault into a training set and a validation set, use the training set to train the power grid technical transformation fault prediction model based on a machine learning algorithm, find the association between the historical device status data and the historical equipment aging coefficient and the power grid technical transformation fault prediction model, and automatically adjust, assign weights to the device status data and the device aging coefficient respectively, and build a power grid technical transformation fault prediction model; input the validation set into the power grid technical transformation fault prediction model, and evaluate the prediction accuracy of the model based on cross-validation; input the newly acquired device status data and equipment aging coefficient into the model to generate a power grid technical transformation fault prediction coefficient.

[0009] In a preferred embodiment, the power grid technical transformation equipment is fault predicted and the equipment that is about to fail is repaired based on the power grid technical transformation fault prediction coefficient. The specific steps are as follows: the power grid equipment is divided into multiple node areas, multiple sections of power grid technical transformation fault prediction coefficients are obtained, and the multiple sections of power grid technical transformation fault prediction coefficients are compared and analyzed based on a preset power grid equipment safety factor; if the power grid technical transformation fault prediction coefficient is greater than or equal to the preset power grid equipment safety factor, the area is marked as a node area where the equipment will fail; if the power grid technical transformation fault prediction coefficient is less than the preset power grid equipment safety factor, the equipment in the area is not prone to failure.

[0010] In a preferred embodiment, the specific steps of the data analysis are as follows: by collecting the voltage data, power and frequency of the equipment in the power system, the voltage data is compared and analyzed with the preset voltage stability data, the equipment with abnormal fluctuations whose voltage data is higher or lower than the preset voltage stability data is marked, and data is collected for the marked equipment.

[0011] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Through fault prediction, potential faults can be identified in advance, the power grid technical transformation and maintenance plan can be optimized, the equipment operation reliability can be improved, the downtime can be reduced, the maintenance cost can be reduced, and the overall efficiency of the power grid can be improved.

[0012] 2. Building a fault prediction model through machine learning can automatically extract patterns from historical data, accurately predict the occurrence of faults, improve prediction accuracy and reduce human intervention, thereby improving system reliability and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A schematic structural diagram of a power grid technical renovation and maintenance optimization method based on fault prediction provided in an embodiment of the present application.

[0014] Figure 2 A schematic diagram of the structure of a power grid technical renovation and maintenance optimization system based on fault prediction is provided in an embodiment of the present application. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Embodiment 1, Figure 1 A schematic diagram of a power grid technical renovation and maintenance optimization method based on fault prediction provided in an embodiment of the present application includes the following steps: S1, real-time monitoring of equipment status data, wherein the equipment status data includes abnormal operation status data, temperature, and vibration intensity data of the equipment.

[0017] In this example, real-time monitoring of equipment status data has many benefits for power grid technology transformation and fault prediction. These monitoring data can help the power grid operation and maintenance team to promptly understand the health status of the equipment, predict potential faults and make effective technical improvements. The abnormal operating status data generally refers to data that deviates from normal operating standards during the operation of the power system. These anomalies may include voltage fluctuations, frequency anomalies, current overloads or load imbalances. This type of data is usually an early warning of system problems and needs to be detected and analyzed in a timely manner to prevent larger-scale failures or outages.

[0018] Obtaining abnormal data on equipment operation status has the following advantages for analyzing power grid equipment failure prediction: Early warning to reduce downtime: Through real-time monitoring and data analysis of the operating status of power grid equipment, potential equipment failure risks can be discovered early. For example, abnormal changes in key parameters such as voltage, current, and temperature are often precursors to equipment failure. Identifying these abnormal data in advance allows operation and maintenance personnel to take effective repair measures before the failure occurs, thereby avoiding serious damage to equipment, reducing downtime, and ensuring the normal operation of the power grid.

[0019] Improve the reliability and safety of power grid equipment: By continuously tracking the operating status of power grid equipment, the health status of the equipment can be dynamically evaluated. Abnormal data can reveal the potential failure mode of the equipment, help identify equipment aging, wear and tear or other hidden problems, reduce the risk of safety accidents caused by sudden equipment failures, and improve the overall reliability of the power grid system.

[0020] Optimize operation and maintenance resources and costs: Traditional power grid equipment fault detection often relies on regular inspections and manual inspections, which may result in missed or delayed fault detection. However, through abnormal data analysis, intelligent fault diagnosis and prediction can be achieved, which reduces the frequency and investment of manual inspections, optimizes the use of operation and maintenance resources, and reduces operation and maintenance costs.

[0021] Fault root cause analysis and precise positioning: Obtaining abnormal operating status data can not only help predict faults, but also dig deep into the root causes of faults through big data analysis and machine learning technology. For example, deviations in certain sensor data of equipment may be caused by electrical connection problems, changes in environmental factors, equipment aging, etc. Accurate fault diagnosis can help operation and maintenance personnel locate problems, avoid unnecessary replacement or repair, and improve fault handling efficiency.

[0022] Extending equipment life: Abnormal data analysis can provide important support for equipment health management. By continuously monitoring the equipment's operating status, minor problems can be discovered and repaired in a timely manner, thus preventing these minor problems from further developing into major failures, helping to extend the equipment's life and improve return on investment.

[0023] Improve the level of intelligent power grid: Abnormal data analysis technology combines cutting-edge technologies such as big data, the Internet of Things and artificial intelligence, and can intelligently monitor, analyze and predict the operation of the power grid. This intelligent operation and maintenance mode not only improves the automation level of the power grid, but also enables more accurate adaptive scheduling and optimization in the future, improving the overall operation efficiency of the power grid.

[0024] Accurate dispatch and load management: The collection and analysis of abnormal data not only helps predict equipment failures, but also helps predict problems such as grid load fluctuations and uneven loads on generator sets. This helps to reasonably adjust the grid's load distribution and operation strategy when the grid is heavily loaded or equipment is about to fail, thereby avoiding overloads or other sudden failures and ensuring stable operation of the grid.

[0025] Improve emergency response capabilities: Obtaining real-time abnormal data can speed up emergency response. For example, when a large-scale power grid failure or equipment failure occurs, abnormal data of the equipment can be obtained in real time, the scope and impact of the failure can be quickly assessed, and repair teams and emergency plans can be prepared in advance to shorten recovery time.

[0026] The abnormal data of the equipment operation state includes an abnormal power coefficient of the equipment. The specific method for obtaining the abnormal power coefficient of the equipment is as follows: Real-time monitoring and collection of voltage data of equipment in the power system, and data analysis based on preset voltage stability data to identify power equipment with abnormal fluctuation data; The specific steps of the data analysis are as follows: By collecting voltage data, power and frequency of equipment in the power system, the voltage data is compared and analyzed with preset voltage stability data, equipment with abnormal fluctuations whose voltage data is higher or lower than the preset voltage stability data is marked, and data of the marked equipment is collected; Collect historical harmonic data of power equipment, and analyze the historical harmonic data based on standard harmonic data to obtain current harmonic impact factors and voltage harmonic impact factors; Obtaining the harmonic current, fundamental current, harmonic voltage and fundamental voltage of the power equipment with abnormal fluctuation data; The current harmonic anomaly coefficient is calculated based on the harmonic current and fundamental current combined with the current harmonic influence factor and the current harmonic distortion rate; The voltage harmonic anomaly coefficient is calculated based on the voltage harmonic distortion rate according to the harmonic voltage and fundamental voltage combined with the voltage harmonic influence factor; The current harmonic abnormality coefficient and the voltage harmonic abnormality coefficient are calculated based on the equipment operation status abnormality formula to obtain the equipment power abnormality coefficient.

[0027] The specific calculation formula of the current harmonic abnormal coefficient is as follows:

[0028] The voltage harmonic abnormal coefficient is specifically calculated as follows:

[0029] The specific calculation formula of the equipment power abnormality coefficient is as follows:

[0030] In the formula, Y is the abnormal coefficient of equipment operation status, is the current harmonic anomaly coefficient, is the voltage harmonic anomaly coefficient, T is the temperature, is the frequency, P is the power, is the i-th harmonic current, where i=1,2,3,...,R, R is an integer, is the average harmonic current, N is the number of harmonic current measurements, is the fundamental voltage, is the current harmonic influence factor, is the jth harmonic voltage, where j=1,2,3,...,M, M is an integer, is the average harmonic voltage, J is the number of harmonic voltage measurements, is the fundamental voltage, is the voltage harmonic impact factor.

[0031] It should be noted that the abnormal coefficient of current harmonics and voltage harmonics can cause heating of power transformers, abnormal fluctuations in DC voltage and current, and abnormal operation of low-voltage power distribution equipment. Specifically, when the grid voltage or current contains harmonics, it will cause abnormal fluctuations in DC voltage and current, leading to failures of low-voltage power distribution equipment (such as circuit breakers, leakage protectors, contactors, thermal relays, etc.). Excessively high or low temperatures can also affect the power or frequency of power equipment, leading to failures of the entire power equipment.

[0032] Abnormal vibration intensity data means that during the operation of power equipment (such as generators, transformers, motors, switchgear, etc.), the vibration intensity of the equipment may gradually increase over time, indicating that some mechanical parts may be worn, such as imbalance, looseness, wear, bearing failure, etc., and need to be replaced in advance. Abnormal vibration may mean that the mechanical part of the equipment is worn, loose or faulty.

[0033] Obtaining abnormal vibration intensity data has the following advantages for analyzing fault prediction of power equipment: Vibration data can provide real-time feedback on the operating status of power equipment. Power equipment will generate a certain amount of vibration during normal operation, and the vibration pattern will change significantly when a fault occurs (such as bearing wear, motor imbalance, gear damage, etc.). By monitoring changes in vibration intensity, anomalies can be identified in the early stages of a fault, thereby achieving timely warning. This helps avoid more serious damage or downtime of equipment due to long-term undetected faults.

[0034] Non-contact monitoring: Compared with traditional mechanical temperature monitoring, oil pressure monitoring and other methods, vibration monitoring can usually be achieved remotely through sensors. There is no need to directly contact the equipment, which is a huge advantage for some equipment that operates in high temperature, high pressure environment or is difficult to access. For example, the vibration monitoring of power transformers, generator sets, transmission lines and other equipment can be achieved through sensors placed on the surface of the equipment to obtain data in real time without interfering with the normal operation of the equipment.

[0035] Real-time monitoring and data-driven analysis: Abnormal vibration intensity data can be monitored in real time, and through modern sensors and Internet of Things (IoT) technology, data can be continuously uploaded to the monitoring system, allowing operation and maintenance personnel to obtain real-time equipment operating status. Combined with big data analysis and machine learning technology, vibration data can be further used for modeling and analysis to accurately identify potential failure risks.

[0036] Accurately locate the source of the fault: By analyzing the characteristics of the vibration signal, especially the changes in parameters such as frequency, amplitude and phase, the source of the fault can be located more accurately. For example, when problems occur in different parts of the equipment (such as bearings, gears, rotors, etc.), the changes in their vibration frequency and amplitude will be different. Using frequency domain analysis, vibration signals of different fault modes can be effectively separated, thereby helping engineers quickly determine the type and location of the fault.

[0037] Reduce the frequency of manual inspections: Vibration monitoring of power equipment can reduce reliance on manual inspections. Through automated monitoring, equipment status can be checked around the clock and without interruption. This not only improves monitoring efficiency, but also reduces possible omissions in manual inspections. For some inaccessible equipment, vibration data can help remotely monitor equipment status and reduce labor costs and risks.

[0038] Multi-factor fusion analysis: Vibration intensity data can be combined with other monitoring data such as temperature, current, oil temperature, etc. to perform multi-dimensional data fusion analysis. This comprehensive analysis can further improve the accuracy of fault prediction. For example, the vibration analysis and temperature monitoring of the motor can be combined to determine whether the motor is vibrating abnormally due to overheating or mechanical problems, thereby fully understanding the health status of the equipment.

[0039] Diversity of fault types: The failure modes of power equipment are complex, and vibration data analysis can monitor and predict many different types of faults. For example, internal faults in power transformers may cause electromagnetic vibrations, while bearing faults manifest as mechanical vibrations. Different types of faults can be distinguished and identified through differences in vibration signals.

[0040] The specific method for obtaining the abnormal vibration intensity data is as follows: Acquire vibration signals of power equipment in real time, perform denoising on the vibration signals and extract characteristic data with obvious vibration; Obtain the vibration signal of the normal operation of the equipment, obtain the overall change law of the vibration signal on the time axis based on the time domain analysis, and perform data analysis based on the overall change law of the vibration signal on the time axis to obtain the vibration intensity benchmark data; The characteristic data with obvious vibration is transformed into the frequency domain of vibration signal based on Fourier transform, and the frequency domain of vibration signal is analyzed to obtain the frequency data of vibration signal; The vibration intensity abnormal data is obtained by comparing and analyzing the vibration signal frequency data with the vibration intensity benchmark data.

[0041] The specific calculation formula for the abnormal vibration intensity data is as follows:

[0042] In the formula, Z is the abnormal data of vibration intensity, is the Kth vibration signal frequency data, where k=1,2,3,...,k, k is an integer, L is the number of vibration signal frequency data, is the vibration intensity benchmark data, and t is the time.

[0043] It should be noted that time domain analysis can be used to monitor the overall vibration level of the equipment. Frequency domain analysis helps to identify failure modes at specific frequencies, such as mechanical friction in the equipment, gear failure, bearing damage, etc.

[0044] S2, obtaining historical fault data of power grid equipment, the historical fault data including the initial performance value of equipment insulation resistance, the initial value of mechanical wear, and the service life and maintenance history data of the equipment; establishing an equipment attenuation model based on the historical fault data of power grid equipment, and generating an equipment aging coefficient based on the established equipment attenuation model. Obtaining historical fault data of power grid equipment provides strong support for fault analysis, which can help improve the accuracy of fault prediction, optimize maintenance strategies, enhance the scientific nature of equipment management, and improve the safety and stability of the power grid system.

[0045] Obtaining historical equipment failure data has the following advantages for analyzing power equipment failure prediction: Improve prediction accuracy: Historical failure data provides detailed information about the failure modes, frequency, severity, and other aspects of the equipment in the past. Analyzing this data can help build more accurate failure prediction models. These models can identify potential failure risks and predict problems that may occur with equipment in the future. Analyzing historical data can reveal long-term trends in equipment failures. For example, the probability of failure of certain equipment may increase as the equipment is used for a longer period of time, and these trends can help predict when the equipment is most likely to fail.

[0046] Help formulate maintenance strategies: Predicting the time and type of equipment failures through historical data can help the maintenance team prepare in advance and conduct necessary inspections and repairs, thereby reducing sudden failures and downtime and reducing operation and maintenance costs. Combined with equipment failure data, the operation and maintenance team can formulate a more scientific maintenance plan. For example, the maintenance cycle of the equipment can be adjusted according to the failure data to avoid premature or late maintenance, improve maintenance efficiency and equipment life.

[0047] Reduce unexpected downtime and losses: Through historical data analysis, potential problems can be discovered before the equipment shows obvious signs of failure, thus avoiding the impact of equipment failure on the power system. For example, by monitoring the historical overload, current instability and other data of electrical equipment, it is possible to predict that some equipment may be about to fail and intervene in time. Fault prediction can help reduce unplanned downtime, maintain the stability of power supply, and improve system reliability.

[0048] Improve resource allocation and decision support: Historical data provides data support for equipment managers, making decisions more objective and scientific. For example, the health status of equipment can be evaluated based on historical fault records, and a basis can be provided for equipment replacement or upgrade. By analyzing historical fault data, the allocation of maintenance personnel, spare parts, and funds can be optimized. Especially for old equipment or equipment with frequent failures, resources can be allocated in a timely manner to focus on solving key problems.

[0049] Assisted intelligence and automation: The use of historical fault data and machine learning algorithms can further improve the intelligence level of power equipment fault prediction. Machine learning models can automatically extract features from historical data, identify complex fault modes, and provide more accurate predictions. By combining real-time data with historical data, the equipment management system can achieve automatic early warning, reduce manual intervention, and improve work efficiency.

[0050] The aging coefficient of the equipment is specifically obtained as follows: Obtain the initial performance value of the equipment's insulation resistance, initial value of mechanical wear, and service life of the equipment; The attenuation coefficient of equipment performance is obtained by collecting and analyzing historical performance data; Establish an equipment attenuation model based on the initial performance value of insulation resistance, initial value of mechanical wear, and service life of the equipment combined with the attenuation coefficient; Collect maintenance history data, extract features from the maintenance history data, and obtain the historical maintenance times and historical fault types of the equipment; The historical fault components of the equipment are determined according to the historical fault types, and the aging rate of the equipment is obtained for the historical fault components of the equipment based on the equipment attenuation model.

[0051] The aging rate of the equipment is calculated as follows:

[0052] In the formula, is the aging rate of the equipment, is the initial value of mechanical wear, is the initial performance value of insulation resistance, is the attenuation coefficient, t is the service life of the equipment, and c is the number of historical maintenance times.

[0053] It should be noted that by extracting features from the maintenance history data, the historical fault types of the equipment are obtained to determine which faults the equipment has experienced, and based on these faults, the equipment aging rate is calculated first for equipment that has undergone fault maintenance and has experienced failures, thereby increasing the accuracy and localization of equipment failure prediction.

[0054] S3, based on the equipment status data combined with the equipment aging coefficient based on machine learning, capture the relationship between the equipment status data and the historical fault data of the power grid equipment and the power grid technical transformation fault prediction, build a power grid technical transformation fault prediction model, and generate a power grid technical transformation fault prediction coefficient. The specific steps are as follows: Acquire historical equipment status data and historical equipment aging coefficients, perform preprocessing and normalization operations on the historical equipment status data and historical equipment aging coefficients, and obtain characteristic data that affects power grid technical reform faults; The characteristic data affecting the power grid technical transformation fault is divided into a training set and a validation set, and the training set is used to train the power grid technical transformation fault prediction model based on a machine learning algorithm, and the correlation between the historical equipment status data and the historical equipment aging coefficient and the power grid technical transformation fault prediction model is found, and the equipment status data and the equipment aging coefficient are automatically adjusted, and weights are respectively assigned to build a power grid technical transformation fault prediction model. The validation set is input into the power grid technical transformation fault prediction model, and the prediction accuracy of the model is evaluated based on cross-validation; The newly acquired equipment status data and equipment aging coefficient are input into the model to generate the power grid technical transformation failure prediction coefficient.

[0055] The specific calculation formula of the power grid technical transformation fault prediction model is as follows:

[0056] In the formula, is the fault prediction coefficient of power grid technical transformation, Y is the abnormal coefficient of equipment operation status, Z is the abnormal data of vibration intensity, is the aging rate of the equipment, is the weight coefficient of the abnormal coefficient of the equipment operation status, is the weight coefficient of abnormal vibration intensity data, is the aging rate weight coefficient of the equipment.

[0057] S4, predicting the faults of the power grid technical transformation equipment according to the power grid technical transformation fault prediction coefficient and repairing the equipment that is about to fail, the specific steps are as follows: Dividing the power grid equipment into multiple node areas, obtaining multiple sections of power grid technical transformation fault prediction coefficients, and performing comparative analysis on the multiple sections of power grid technical transformation fault prediction coefficients based on preset power grid equipment safety factors; If the grid technical reform failure prediction coefficient is greater than or equal to the preset grid equipment safety factor, the area is marked as a node area where the equipment will fail. It is necessary to notify relevant personnel to inspect and repair the equipment and conduct regular physical inspections of the equipment to check whether there are damaged, corroded, or loose parts, especially electrical contact parts and important mechanical parts. For parts that are prone to wear or damage, such as motors, pumps, sensors, etc., special tests can be carried out to ensure that their working performance meets the requirements. Conduct electrical inspections such as grounding resistance, insulation, etc. and mechanical tests such as balance and vibration tests to ensure the reliability of equipment operation. Check the quality and stability of the power supply to ensure that the voltage fluctuation does not exceed the tolerance range of the equipment. If necessary, consider using UPS power supply, filter or voltage stabilizer to ensure the stability of equipment operation. If the equipment generates abnormal vibration or noise during operation, it may be a sign of mechanical failure. Check whether the equipment installation and support structure are firm and make necessary adjustments. Develop a detailed maintenance plan for the equipment, including regular cleaning, lubrication, and parts inspection. Early warning can be provided by implementing condition monitoring and predictive maintenance to avoid sudden failures.

[0058] If the grid technical transformation failure prediction coefficient is less than the preset grid equipment safety factor, the equipment in this area is less likely to fail.

[0059] Embodiment 2, Figure 2 A schematic diagram of the structure of a power grid technical renovation and maintenance optimization system based on fault prediction provided in an embodiment of the present application includes an equipment status data acquisition module, a historical data acquisition and analysis module, an attenuation model construction module, a fault prediction model construction module, and a power grid fault analysis module, and there are connections between the modules: The device status data acquisition module is used to monitor the device status data in real time, and the device status data includes abnormal operation status data, temperature, and vibration intensity data of the device; A historical data acquisition and analysis module is used to acquire historical fault data of power grid equipment, wherein the historical fault data includes initial performance value of insulation resistance of equipment, initial value of mechanical wear, and service life and maintenance history data of equipment; An attenuation model building module is used to establish an equipment attenuation model based on historical fault data of power grid equipment, and generate an equipment aging coefficient based on the established equipment attenuation model; A fault prediction model building module is used to capture the relationship between the equipment status data and the historical fault data of the power grid equipment and the power grid technical transformation fault prediction based on machine learning according to the equipment status data combined with the equipment aging coefficient, build a power grid technical transformation fault prediction model, and generate a power grid technical transformation fault prediction coefficient; The power grid fault analysis module is used to predict the faults of the power grid technical transformation equipment according to the power grid technical transformation fault prediction coefficient and to repair the equipment that is about to fail.

[0060] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0061] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0062] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0063] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0064] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0065] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for optimizing power grid technical renovation and maintenance based on fault prediction, characterized in that: The steps include: Real-time monitoring of equipment status data, including abnormal equipment operation status data, temperature, and vibration intensity data; Acquire historical fault data of power grid equipment, wherein the historical fault data includes initial performance value of insulation resistance of equipment, initial value of mechanical wear, and service life and maintenance history data of equipment; Establish an equipment attenuation model based on historical fault data of power grid equipment, and generate equipment aging coefficients based on the established equipment attenuation model; According to the equipment status data combined with the equipment aging coefficient based on machine learning, the relationship between the equipment status data and the historical fault data of the power grid equipment and the power grid technical transformation fault prediction is captured, a power grid technical transformation fault prediction model is constructed, and a power grid technical transformation fault prediction coefficient is generated; According to the power grid technical transformation fault prediction coefficient, faults of power grid technical transformation equipment are predicted and equipment that is about to fail is repaired.

2. A method and system for optimizing power grid technical renovation and maintenance based on fault prediction according to claim 1, characterized in that: The abnormal data of the equipment operation state includes an abnormal power coefficient of the equipment. The specific method for obtaining the abnormal power coefficient of the equipment is as follows: Real-time monitoring and collection of voltage data of equipment in the power system, and data analysis based on preset voltage stability data to identify power equipment with abnormal fluctuation data; Collect historical harmonic data of power equipment, and analyze the historical harmonic data based on standard harmonic data to obtain current harmonic impact factors and voltage harmonic impact factors; Obtaining the harmonic current, fundamental current, harmonic voltage and fundamental voltage of the power equipment with abnormal fluctuation data; The current harmonic anomaly coefficient is calculated based on the harmonic current and fundamental current combined with the current harmonic influence factor and the current harmonic distortion rate; The voltage harmonic anomaly coefficient is calculated based on the voltage harmonic distortion rate according to the harmonic voltage and fundamental voltage combined with the voltage harmonic influence factor; The current harmonic abnormality coefficient and the voltage harmonic abnormality coefficient are calculated based on the equipment operation status abnormality formula to obtain the equipment power abnormality coefficient.

3. A method and system for optimizing power grid technical renovation and maintenance based on fault prediction according to claim 1, characterized in that: The specific method for obtaining the abnormal vibration intensity data is as follows: Acquire vibration signals of power equipment in real time, perform denoising on the vibration signals and extract characteristic data with obvious vibration; Obtain the vibration signal of the normal operation of the equipment, obtain the overall change law of the vibration signal on the time axis based on the time domain analysis, and perform data analysis based on the overall change law of the vibration signal on the time axis to obtain the vibration intensity benchmark data; The characteristic data with obvious vibration is transformed into the frequency domain of vibration signal based on Fourier transform, and the frequency domain of vibration signal is analyzed to obtain the frequency data of vibration signal; The vibration intensity abnormal data is obtained by comparing and analyzing the vibration signal frequency data with the vibration intensity benchmark data.

4. A method and system for optimizing power grid technical renovation and maintenance based on fault prediction according to claim 1, characterized in that: The specific method for obtaining the equipment aging coefficient is as follows: Obtain the initial performance value of the equipment's insulation resistance, initial value of mechanical wear, and service life of the equipment; The attenuation coefficient of equipment performance is obtained by collecting and analyzing historical performance data; Establish an equipment attenuation model based on the initial performance value of insulation resistance, initial value of mechanical wear, and service life of the equipment combined with the attenuation coefficient; Collect maintenance history data, extract features from the maintenance history data, and obtain the historical maintenance times and historical fault types of the equipment; The historical fault components of the equipment are determined according to the historical fault types, and the aging rate of the equipment is obtained for the historical fault components of the equipment based on the equipment attenuation model.

5. A method and system for optimizing power grid technical renovation and maintenance based on fault prediction according to claim 1, characterized in that: The specific steps of capturing the relationship between the equipment status data, the historical fault data of power grid equipment and the power grid technical transformation fault prediction based on machine learning based on the equipment status data combined with the equipment aging coefficient, building a power grid technical transformation fault prediction model, and generating a power grid technical transformation fault prediction coefficient are as follows: Acquire historical equipment status data and historical equipment aging coefficients, perform preprocessing and normalization operations on the historical equipment status data and historical equipment aging coefficients, and obtain characteristic data that affects power grid technical reform faults; The characteristic data affecting the power grid technical transformation fault is divided into a training set and a validation set, and the training set is used to train the power grid technical transformation fault prediction model based on a machine learning algorithm, and the correlation between the historical equipment status data and the historical equipment aging coefficient and the power grid technical transformation fault prediction model is found, and the equipment status data and the equipment aging coefficient are automatically adjusted, and weights are respectively assigned to build a power grid technical transformation fault prediction model. The validation set is input into the power grid technical transformation fault prediction model, and the prediction accuracy of the model is evaluated based on cross-validation; The newly acquired equipment status data and equipment aging coefficient are input into the model to generate the power grid technical transformation failure prediction coefficient.

6. A method and system for optimizing power grid technical renovation and maintenance based on fault prediction according to claim 1, characterized in that: The specific steps of predicting the faults of the power grid technical transformation equipment according to the power grid technical transformation fault prediction coefficient and repairing the equipment that is about to fail are as follows: Dividing the power grid equipment into multiple node areas, obtaining multiple sections of power grid technical transformation fault prediction coefficients, and performing comparative analysis on the multiple sections of power grid technical transformation fault prediction coefficients based on preset power grid equipment safety factors; If the grid technical reform fault prediction coefficient is greater than or equal to the preset grid equipment safety factor, the area is marked as a node area where equipment failure may occur; If the grid technical transformation failure prediction coefficient is less than the preset grid equipment safety factor, the equipment in this area is less likely to fail.

7. A method and system for optimizing power grid technical renovation and maintenance based on fault prediction according to claim 2, characterized in that: The data analysis specifically includes the following steps: By collecting the voltage data, power and frequency of the equipment in the power system, the voltage data is compared and analyzed with the preset voltage stability data, and the equipment with abnormal fluctuations whose voltage data is higher or lower than the preset voltage stability data is marked, and data is collected for the marked equipment.

8. A method and system for optimizing power grid technical renovation and maintenance based on fault prediction according to claim 2, characterized in that: The specific calculation formula of the current harmonic abnormal coefficient is as follows: The voltage harmonic abnormal coefficient is specifically calculated as follows: The specific calculation formula of the equipment power abnormality coefficient is as follows: In the formula, Y is the abnormal coefficient of equipment operation status, is the current harmonic anomaly coefficient, is the voltage harmonic anomaly coefficient, T is the temperature, is the frequency, P is the power, is the i-th harmonic current, where i=1,2,3,...,R, R is an integer, is the average harmonic current, N is the number of harmonic current measurements, is the fundamental voltage, is the current harmonic impact factor, is the jth harmonic voltage, where j=1,2,3,...,M, M is an integer, is the average harmonic voltage, J is the number of harmonic voltage measurements, is the fundamental voltage, is the voltage harmonic impact factor.

9. A method and system for optimizing power grid technical renovation and maintenance based on fault prediction according to claim 3, characterized in that: The specific calculation formula for the abnormal vibration intensity data is as follows: In the formula, Z is the abnormal data of vibration intensity, is the Kth vibration signal frequency data, where k=1,2,3,...,k, k is an integer, L is the number of vibration signal frequency data, is the vibration intensity benchmark data, and t is the time.

10. A system using a power grid technical renovation and maintenance optimization method based on fault prediction as claimed in any one of claims 1 to 9, characterized in that: It includes equipment status data acquisition module, historical data acquisition and analysis module, attenuation model construction module, fault prediction model construction module, and power grid fault analysis module. There are connections between the modules: The device status data acquisition module is used to monitor the device status data in real time, and the device status data includes abnormal operation status data, temperature, and vibration intensity data of the device; A historical data acquisition and analysis module is used to acquire historical fault data of power grid equipment, wherein the historical fault data includes initial performance value of insulation resistance of equipment, initial value of mechanical wear, and service life and maintenance history data of equipment; An attenuation model building module is used to establish an equipment attenuation model based on historical fault data of power grid equipment, and generate an equipment aging coefficient based on the established equipment attenuation model; A fault prediction model building module is used to capture the relationship between the equipment status data and the historical fault data of the power grid equipment and the power grid technical transformation fault prediction based on machine learning according to the equipment status data combined with the equipment aging coefficient, build a power grid technical transformation fault prediction model, and generate a power grid technical transformation fault prediction coefficient; The power grid fault analysis module is used to predict the faults of the power grid technical transformation equipment according to the power grid technical transformation fault prediction coefficient and to repair the equipment that is about to fail.

Images