Power grid fault early warning and positioning method, device, medium and system

Abstract

The invention provides a power grid fault early warning and positioning method and device, a medium and a system. The method comprises the following steps: acquiring initial acquisition data of a current power grid node acquired by a current sensor, a temperature sensor and a vibration sensor, processing the initial acquisition data by at least adopting a weighted fusion algorithm and a Kalman filtering algorithm to obtain a final prediction result, and finally determining the actual measurement result according to the final prediction result and the actual measurement result of the last time. Compared with an existing scheme, a plurality of sensors are considered, and data acquired by the plurality of sensors are fused and judged, so that the accuracy of early warning and positioning is improved, and the problem that for early warning and positioning of power grid faults, simple judgment is carried out through voltage, current or a single sensor in the existing scheme, and the accuracy of early warning and positioning is poor is solved. And the abnormality of the equipment is difficult to find.

Description

Technical Field

[0001] The present application relates to the technical field of power grid fault detection, and in particular to a method, device, medium and system for early warning and locating power grid faults. Background Art

[0002] For the method of early warning and locating power grid faults, the existing technology is generally implemented through the following steps:

[0003] The first step is to build an intelligent sensor network. In order to achieve comprehensive monitoring of the real-time status of the power grid, multiple types of intelligent sensors are first deployed in the key equipment and nodes of the power grid. These sensors include current sensors, voltage sensors, temperature sensors, and vibration sensors, etc. Each sensor is responsible for collecting different types of data. The current sensor monitors the current changes in the line, the voltage sensor detects voltage fluctuations in real time, the temperature sensor is used to monitor the overheating of the power grid equipment, and the vibration sensor is used to detect abnormal vibration of mechanical equipment. These sensors transmit the collected data to the central processing unit in real time through wireless communication technologies (such as Wi-Fi, igBee, LoRa, etc.). For example, on a long-distance transmission line, sensors may be deployed on different transmission towers to ensure coverage of the operating status of the entire power grid. In this way, the real-time data of each power grid facility can be collected and transmitted to the monitoring center, forming a comprehensive perception network for the power grid.

[0004] The second step is the data acquisition and transmission module. In order to ensure the accuracy and real-time nature of the data, after the data is collected by the intelligent sensor, it is first standardized through the data acquisition module. These raw data are transmitted to the centralized central processing unit via a wireless network. This module is responsible for storing, preprocessing and preliminarily analyzing the sensor data. The data transmission adopts a low-power, high-reliability communication protocol to ensure that the transmission is not affected even in harsh environments. In practical applications, data transmission delays and packet loss are important issues that must be solved in power grid monitoring systems. For example, when the current sensor detects a sudden change in current, the data is immediately transmitted to the central control system through a low-latency wireless transmission link. If there is data loss or abnormality, the system will ensure the integrity of the data through an automatic retransmission mechanism.

[0005] The third step is big data analysis and fault detection. The system trains a large amount of historical data through machine learning algorithms (such as random forests, support vector machines, deep neural networks, etc.), so that the algorithm can make accurate fault judgments and warnings based on different fault modes.

[0006] The fourth step is fault location and accurate early warning. Once a fault is detected, the system will conduct further analysis through the fault location module. This module is based on sensor data and the topological structure of the power grid, using information such as fault propagation model and signal propagation speed to accurately locate the area where the fault occurs. When a fault occurs in the power grid, abnormal data such as current, voltage and temperature collected by the intelligent sensor will be transmitted to the central processing unit. The system accurately determines the coordinates of the fault point by comparing the distribution of abnormal data in each area. For example, if a short circuit occurs on a transmission line, the current sensor will detect a sharp change in current and transmit it to the central system through data. The system analyzes the data of each sensor node and combines it with the topological structure of the power grid to finally determine the specific location of the fault. For example, by analyzing the voltage changes of the sensors adjacent to the line section, the system can infer the exact location of the fault and give the positioning result of "the fault occurred at the 12th transmission tower".

[0007] Although smart sensor networks and big data analysis can improve the accuracy of fault detection, there is still a risk of false alarms and missed alarms, especially when the power grid operating environment is complex and changeable. For example, factors such as equipment aging, environmental noise or power grid fluctuations may lead to system misjudgment, thereby affecting the decision-making of power dispatchers and increasing response costs.

[0008] Currently, when a line fails, dispatchers and branch personnel conduct a comprehensive assessment based on the voltage and current detected on the line, the fault alarm signals reported by the automatic switches, and the alarm tripping signals from the substation. This is time-consuming and has low fault assessment accuracy, and there is a lack of an effective fault assessment auxiliary tool.

[0009] At the same time, the current monitoring system usually makes simple judgments through voltage, current or a single sensor, which makes it difficult to detect equipment abnormalities. Currently, equipment abnormalities are generally detected by manual inspections, and judgments are made through partial discharge tests, manual temperature measurements, etc., and it is even more difficult to detect abnormalities in equipment machinery.

[0010] That is, for power grid fault warning and location, the existing solutions use simple judgments based on voltage, current or a single sensor, which makes it difficult to detect equipment abnormalities. Summary of the invention

[0011] The main purpose of the present application is to provide a method, device, medium and system for early warning and positioning of power grid faults, so as to at least solve the problem that the existing solutions for early warning and positioning of power grid faults make simple judgments through voltage, current or a single sensor, and it is difficult to detect abnormalities of equipment.

[0012] In order to achieve the above object, according to one aspect of the present application, a method for early warning and locating a power grid fault is provided, the method comprising:

[0013] Acquire initial collected data of a current power grid node collected by multiple sensors, wherein the sensors include a current sensor, a temperature sensor, and a vibration sensor;

[0014] At least a weighted fusion algorithm and a Kalman filter algorithm are used to process the initial collected data to obtain a final prediction result;

[0015] According to the size of the final prediction result and the last measured result, it is determined whether to generate warning information, and the warning information is used to prompt the current power grid node failure.

[0016] Optionally, at least a weighted fusion algorithm and a Kalman filter algorithm are used to process the initial collected data to obtain a final prediction result, including:

[0017] Processing the initial collected data using the weighted fusion algorithm to obtain first collected data;

[0018] The first collected data is processed by using Flink's stream processing technology to obtain second collected data;

[0019] The second collected data is processed using the Kalman filter algorithm to obtain the final prediction result.

[0020] Optionally, in the process of using the Kalman filter algorithm to process the second collected data to obtain the final prediction result, the method further includes:

[0021] use

[0022] as well as

[0023] Updating the prediction result of the Kalman filter algorithm;

[0024] in, is the final prediction result or the current prediction result, For the second data collection, is the current forecast error covariance, P k is the current update error covariance, I is the unit matrix, H is the observation matrix, K k is the Kalman gain matrix, is the current updated prediction result, R is the observation noise covariance, z k is the observed value at the current time k.

[0025] Optionally, using the Kalman filter algorithm to process the second collected data to obtain the final prediction result includes:

[0026] according to

[0027] Determine the final prediction result, where A is the state transfer matrix, B is the control matrix, is the final prediction result or the current prediction result, For the second collected data, P k - is the current prediction error covariance, Q is the process noise covariance, P k-1 is the error covariance of the last update.

[0028] Optionally, at least a weighted fusion algorithm and a Kalman filter algorithm are used to process the initial collected data to obtain a final prediction result, including:

[0029] Processing the initial collected data using the weighted fusion algorithm to obtain first collected data;

[0030] The first collected data is processed using the Kalman filter algorithm to obtain the final prediction result.

[0031] Optionally, determining whether to generate warning information according to the size of the final prediction result and the last measured result includes:

[0032] Determine the prediction difference as the difference between the final prediction result and the last measured result;

[0033] When the absolute value of the predicted difference is within a preset range, determining to generate the warning information;

[0034] When the absolute value of the predicted difference is not within the preset range, it is determined not to generate the warning information.

[0035] Optionally, in the case of determining to generate warning information, the method further includes:

[0036] The warning information is broadcast in the form of voice.

[0037] According to another aspect of the present application, a power grid fault early warning and positioning device is provided, the device comprising:

[0038] An acquisition unit, used to acquire initial acquisition data of a current power grid node acquired by a plurality of sensors, wherein the sensors include a current sensor, a temperature sensor and a vibration sensor;

[0039] A first processing unit, configured to process the initial collected data by at least using a weighted fusion algorithm and a Kalman filter algorithm to obtain a final prediction result;

[0040] The second processing unit is used to determine whether to generate early warning information according to the size of the final prediction result and the last measured result, and the early warning information is used to prompt the current power grid node failure.

[0041] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the methods described.

[0042] According to another aspect of the present application, a power grid fault early warning and positioning system is provided, the system comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of the methods described.

[0043] By applying the technical solution of the present application, the initial collected data of the current power grid node collected by the current sensor, the temperature sensor and the vibration sensor are obtained, and the initial collected data is processed by at least a weighted fusion algorithm and a Kalman filter algorithm to obtain a final prediction result. Finally, according to the size of the final prediction result and the last measured result, it is determined whether to generate early warning information. Compared with the existing solution, multiple sensors are taken into consideration, and the data collected by multiple sensors are fused and judged, which improves the accuracy of early warning and positioning, thereby solving the problem that the existing solution for early warning and positioning of power grid faults makes simple judgments through voltage, current or a single sensor, and it is difficult to detect abnormalities of equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0045] Figure 1 A schematic diagram of a process of an early warning and location method for a power grid fault provided according to an embodiment of the present application is shown;

[0046] Figure 2 A schematic diagram of a process of another method for early warning and locating a power grid fault provided according to an embodiment of the present application is shown;

[0047] Figure 3 A structural block diagram of a power grid fault early warning and positioning device provided according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0048] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0049] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this application.

[0050] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0051] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0052] Flink's stream processing technology refers to a data processing technology provided by the open source stream processing framework Apache Flink. Flink supports high-performance, low-latency stream data processing, can process unbounded and bounded data streams, and provides a rich set of operators and APIs so that users can easily perform data conversion, aggregation, window operations, and other operations. Flink's stream processing technology also supports fault tolerance and state management, which can ensure the accuracy and reliability of data processing. Flink's stream processing technology is widely used in real-time data analysis.

[0053] As introduced in the background technology, after a line fault occurs, the dispatcher and the branch personnel conduct a comprehensive analysis based on the voltage and current detected by the line, the fault alarm signal reported by the automatic switch, and the alarm trip signal of the substation. This is time-consuming, the fault judgment accuracy is low, and there is a lack of an effective fault judgment auxiliary tool; at the same time, the current monitoring system usually makes a simple judgment through voltage, current or a single sensor, and it is difficult to find the abnormality of the equipment. At present, the abnormality of equipment is generally judged by manual inspection, and it is judged by means of partial discharge testing, manual temperature measurement, etc., and it is even more difficult to find the abnormality of the equipment. In order to solve the problem of early warning and positioning of power grid faults, the existing scheme makes a simple judgment through voltage, current or a single sensor, and it is difficult to find the abnormality of the equipment. The embodiments of the present application provide a method, device, medium and system for early warning and positioning of power grid faults.

[0054] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0055] In this embodiment, a method for early warning and locating power grid faults is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0056] Figure 1 FIG. 1 is a flow chart of a method for early warning and locating a power grid fault according to an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:

[0057] Step S101, acquiring initial collection data of a current power grid node collected by multiple sensors, wherein the sensors include a current sensor, a temperature sensor, and a vibration sensor;

[0058] Step S102, at least using a weighted fusion algorithm and a Kalman filter algorithm to process the initial collected data to obtain a final prediction result;

[0059] There are two ways to achieve this:

[0060] The first implementation method:

[0061] The weighted fusion algorithm is used to process the initial collected data to obtain first collected data;

[0062] The core idea of ​​the weighted fusion algorithm is to assign different weights to different types of sensors and integrate the data of various sensors to obtain a unified fault diagnosis value. Since the data provided by each sensor plays a different role in power grid fault judgment, the algorithm needs to consider the accuracy, reliability and relevance of the sensor in fault diagnosis.

[0063] Different types of sensors have different accuracy and fault relevance. For example, in current fault detection, the measurement results of current sensors tend to have higher reliability and can be given higher weights; in some cases, temperature sensors and vibration sensors may provide auxiliary information to help confirm whether the fault is related to equipment overheating or mechanical failure.

[0064] To dynamically assign the right weight to each sensor, the system adjusts the weight based on the sensor's historical accuracy, signal-to-noise ratio, and the sensor's performance in different failure scenarios. For example, a sensor may be more sensitive to certain types of failures and less likely to fail in other situations. The weight can be adjusted based on the sensor's contribution to different types of failures in historical data.

[0065] A commonly used weight adjustment method is based on the historical error analysis of the sensor. Assume that the average error of sensor i in the past m fault identifications is e i , n is the total number of sensors, is a measure of the reliability of the sensor, that is, the error is The larger it is, the more reliable the sensor is. The sum of the reciprocals of the reliability of all sensors is a normalization factor used to ensure that the sum of the weights is 1. The weight w i It can be calculated by the following formula: ; Consider a power grid node, where three sensors are deployed: current sensor x1=15 kA, temperature sensor x2=85℃, and vibration sensor x3=3mm / s. According to the historical performance of the sensors, the current sensor has the highest reliability in fault detection, so it is given a weight of w1=0.6; the temperature sensor is relatively sensitive, so its weight is w2=0.3; the vibration sensor is more auxiliary, with a weight of w3=0.1. Fault diagnosis value after weighted fusion for: ; The above-mentioned Kalman filter algorithm is used to process the above-mentioned first collected data to obtain the above-mentioned final prediction result.

[0070] Combining weighted fusion algorithm and Kalman filter technology, the system can process data from different types of sensors more effectively. In the weighted fusion step, the system fuses different types of sensor data into a comprehensive judgment value by assigning appropriate weights to each sensor. Kalman filtering further eliminates noise and reduces misjudgments caused by abnormal instantaneous data by smoothing these data.

[0071] Through multi-dimensional data fusion, the system can not only improve the accuracy of fault detection, but also provide more contextual information at the early stage of the fault. For example, when the grid load suddenly increases, relying solely on the current sensor may not accurately determine whether it is a normal fluctuation caused by excessive grid load or a short circuit in the line. At this time, combined with the readings of the voltage sensor, if the voltage also fluctuates abnormally, the possibility of load fluctuation can be preliminarily ruled out and it can be judged as a grid fault.

[0072] For example, if the current sensor detects a sudden increase in current at a certain moment, the increase in current may normally mean that the device is overloaded. However, if a temperature sensor is installed near the device and the temperature sensor shows that the temperature fluctuation is not large, it means that the device is overheating and not the cause of the fault. At this time, the system will combine data such as current, temperature and vibration to determine that the cause of the increase in current may be an electrical short circuit in a part of the device, rather than overload or overheating, thereby avoiding misjudging the fault as other reasons.

[0073] Kalman filtering technology is mainly used to smooth the real-time data of sensors and reduce the interference of noise on fault diagnosis, which is especially important in power grid systems with high real-time requirements. By recursively estimating and filtering the sensor observation values, Kalman filtering can effectively eliminate instantaneous anomalies caused by environmental or equipment failures and provide a more stable and accurate estimate.

[0074] The second implementation method:

[0075] The weighted fusion algorithm is used to process the initial collected data to obtain first collected data;

[0076] The first collected data is processed by using Flink's stream processing technology to obtain second collected data;

[0077] Specifically, based on Flink's real-time computing capabilities, the system can adaptively adjust the process noise Q and observation noise R parameters of the Kalman filter. These parameters can be dynamically adjusted according to the changing trends of sensor data and fault characteristics to optimize the estimation effect of the Kalman filter. For example, when the power grid experiences frequent fluctuations, the system can increase the process noise to better adapt to the volatility of the data.

[0078] The second collected data is processed using the Kalman filter algorithm to obtain the final prediction result.

[0079] Kalman filtering technology can also be used to perform joint filtering with other sensor data to reduce conflicts and redundant information between sensors and further improve the accuracy of fault detection. For example, when the measurement results of the temperature sensor and the current sensor are contradictory, the Kalman filter can jointly filter the data of these two sensors to obtain a more stable estimation result.

[0080] Joint filtering: By implementing multi-sensor Kalman filtering in Flink, data from multiple sensors such as current, temperature, and vibration are combined. The data from each sensor can be pre-processed through an independent Kalman filter, and then processed and finally updated through Flink's window merging technology (Flink's window merging technology is a method for merging multiple windows into a larger window. This can reduce computing and communication overhead and improve processing efficiency. In Flink, window merging can be implemented by using a window merger, which can merge adjacent windows into a larger window according to certain conditions. This can reduce unnecessary computing and communication overhead and improve processing efficiency).

[0081] Fault correlation analysis: While processing real-time data streams, Flink can perform fault correlation analysis on the output of the Kalman filter, and combine machine learning algorithms to identify abnormal correlations in sensor data, thereby better determining the type of fault. For example, abnormal increases in current and temperature may mean a short circuit, while abnormal changes in current and vibration may indicate a mechanical failure. Flink can combine this information to provide more accurate fault warnings.

[0082] Update steps:

[0083] use

[0084] as well as

[0085] Update the prediction results of the above Kalman filter algorithm;

[0086] in, is the above final prediction result or the current prediction result, that is, the a priori estimation result, the prediction result when there is no actual observation data, is the second collected data, P k - is the current forecast error covariance, P kis the current update error covariance, that is, the updated state covariance matrix, which represents the confidence of the corrected state estimate, I is the unit matrix, H is the observation matrix, which describes the relationship between the actual observation value (sensor data) and the state variable. For example, in power grid fault detection, the output of the current sensor may be a linear transformation of the system state (such as current), H describes this transformation, and K k is the Kalman gain matrix, which weighs the weights of the predicted value and the observed value. If the current prediction error covariance is large, the Kalman gain matrix will make the observed value more important, otherwise it will make the predicted value more important. is the current updated prediction result, i.e., the posterior estimation result, which is the final estimated prediction value after the prediction value is corrected through the update step. R is the observation noise covariance, which describes the degree of sensor measurement error. The larger the noise, the more the system tends to rely on the prediction value. k is the observed value at the current moment k, that is, the actual measurement data obtained by the sensor.

[0087] The prediction steps include:

[0088] according to

[0089] Determine the above final prediction results, where A is the state transfer matrix, which represents the dynamic change relationship of the system from the previous moment k-1 to the current moment k. For example, in the power grid system, the state can be the law of change of current or voltage over time, and A represents this law. B is the control input influence matrix, which describes the control variable u k Impact on the current system status, is the above final prediction result or the current prediction result, that is, the state prediction value of k at the current time, is the second collected data, i.e. the best estimated value of the system at the last moment, P k - is the current forecast error covariance, indicating the forecast status The uncertainty is also used to describe the relationship between multidimensional state variables and the degree of uncertainty in prediction. Q is the process noise covariance, which describes the degree of random noise changes within the system and reflects the imperfection of the model. In the power grid, process noise may come from environmental interference or random fluctuations of the system itself. k-1 is the error covariance of the last update, which means the accuracy of the state estimation at the last moment, u k is the control quantity at the previous moment, which represents the external control quantity applied to the system, such as the regulation input to the power grid.

[0090] For example, in a power transmission line, current sensors, temperature sensors, and vibration sensors are deployed at multiple key nodes. During a certain operation, the current sensor suddenly reported that the current was too large in a certain area, and the system's initial judgment was that a short circuit occurred there. However, after multi-dimensional data fusion processing, the temperature sensor data did not show a significant temperature increase, and the vibration sensor did not detect abnormal vibration. At this time, the fusion algorithm will adjust the weight of the sensor and re-judge whether the sudden increase in current is a short circuit fault. If there is only current fluctuation but no temperature and vibration abnormalities, the system will output the judgment that "current fluctuations are normal load changes" to avoid false alarms.

[0091] At the same time, if at another power grid node, the temperature sensor shows that the equipment temperature has risen significantly, the vibration sensor captures violent vibrations of the equipment, and the current also shows abnormal fluctuations, then the system will make a judgment that "the failure may be caused by equipment damage or aging" based on the fusion of multi-dimensional data and issue a timely warning.

[0092] The core of Kalman filtering lies in the two steps of state prediction and state update. In traditional Kalman filtering, this process is calculated offline, and there will be a certain time delay in warning and positioning, which is not fast enough. In order to warn and locate faults more quickly, Apache Flink technology of big data can be combined to form a streaming Kalman filter to quickly predict and update sensor data at every moment.

[0093] Apache Flink is an open source streaming data flow execution engine for distributed computing, mainly for stateful computing of unbounded and bounded data streams. Flink can process real-time data streams and historical data streams, support batch processing and stream processing, and pays special attention to operation and maintainability, providing excellent fault recovery capabilities and runtime monitoring tools.

[0094] Streaming Kalman filter: Combine the Kalman filter algorithm with Flink's stream processing function, and use Flink's window technology, such as sliding window and rolling window, to update the system status in real time. Every time sensor data arrives, Flink will automatically trigger the prediction step and update step of the Kalman filter.

[0095] Consider a grid node where the system uses Kalman filtering to filter the temperature sensor data. Assume that the sensor's predicted value is The observed value is z k =80℃, process noise is Q=0.5, observation noise is R=1, and initial error covariance is P0=2.

[0096] Calculate the Kalman gain:

[0097] Then, update the state estimate:

[0098] Step S103, determining whether to generate warning information based on the size of the final prediction result and the last measured result, wherein the warning information is used to indicate the current grid node failure.

[0099] In the above steps, the initial collected data of the current power grid node collected by the current sensor, temperature sensor and vibration sensor is obtained, and the above initial collected data is processed by at least a weighted fusion algorithm and a Kalman filter algorithm to obtain the final prediction result. Finally, according to the size of the above final prediction result and the last measured result, it is determined whether to generate early warning information. Compared with the existing scheme, multiple sensors are considered, and the data collected by multiple sensors are fused and judged, which improves the accuracy of early warning and positioning, thereby solving the problem that for power grid fault early warning and positioning, the existing scheme uses voltage, current or a single sensor for simple judgment, and it is difficult to detect equipment abnormalities.

[0100] In one embodiment of the present application, determining whether to generate warning information according to the magnitude of the above-mentioned final prediction result and the last measured result includes:

[0101] Determine the prediction difference as the difference between the above final prediction result and the above last measured result;

[0102] When the absolute value of the predicted difference is within a preset range, determining to generate the warning information;

[0103] When the absolute value of the above-mentioned prediction difference is not within the above-mentioned preset range, it is determined not to generate the above-mentioned warning information.

[0104] Specifically, the preset range can be set accordingly according to the range of the absolute value of the predicted difference.

[0105] In one embodiment of the present application, when it is determined that warning information is generated, the method further includes:

[0106] The above warning information is broadcast in the form of voice to improve the efficiency of the warning.

[0107] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the power grid fault early warning and positioning method of the present application will be described in detail below in combination with specific embodiments.

[0108] This embodiment relates to a specific method for early warning and locating power grid faults, such as Figure 2 As shown, the following steps are included:

[0109] Acquire initial data of a current power grid node collected by multiple sensors, where the sensors include a current sensor, a temperature sensor, and a vibration sensor;

[0110] Using a weighted fusion algorithm to process the initial collected data to obtain first collected data;

[0111] The first collected data is processed by using Flink's stream processing technology to obtain second collected data;

[0112] The second collected data is processed using a Kalman filter algorithm to obtain a final prediction result;

[0113] Determine the prediction difference as the difference between the final prediction result and the last measured result;

[0114] When the absolute value of the predicted difference is within a preset range, determining to generate warning information;

[0115] When the absolute value of the prediction difference is not within the preset range, it is determined that no warning information is generated.

[0116] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0117] The embodiment of the present application also provides an early warning and positioning device for power grid failure. It should be noted that the early warning and positioning device for power grid failure in the embodiment of the present application can be used to execute the early warning and positioning method for power grid failure provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0118] The following is an introduction to the early warning and positioning device for power grid faults provided in an embodiment of the present application.

[0119] Figure 3 It is a structural block diagram of a power grid fault early warning and positioning device provided according to an embodiment of the present application.

[0120] like Figure 3 As shown, the device comprises:

[0121] An acquisition unit 31 is used to acquire initial acquisition data of a current power grid node acquired by multiple sensors, wherein the sensors include a current sensor, a temperature sensor, and a vibration sensor;

[0122] A first processing unit 32 is used to process the initial collected data using at least a weighted fusion algorithm and a Kalman filter algorithm to obtain a final prediction result;

[0123] The second processing unit 33 is used to determine whether to generate warning information according to the size of the above-mentioned final prediction result and the last measured result, and the above-mentioned warning information is used to prompt the above-mentioned current power grid node failure.

[0124] In the above-mentioned device, the initial collection data of the current power grid node collected by the current sensor, temperature sensor and vibration sensor is obtained, and the above-mentioned initial collection data is processed by at least a weighted fusion algorithm and a Kalman filter algorithm to obtain the final prediction result. Finally, according to the size of the above-mentioned final prediction result and the last measured result, it is determined whether to generate early warning information. Compared with the existing scheme, multiple sensors are considered, and the data collected by multiple sensors are fused and judged, which improves the accuracy of early warning and positioning, thereby solving the problem that for power grid fault early warning and positioning, the existing scheme uses simple judgments through voltage, current or a single sensor, and it is difficult to detect equipment abnormalities.

[0125] In one embodiment of the present application, the first processing unit includes a first processing module, a second processing module and a third processing module.

[0126] The first processing module is used to process the initial collected data using the weighted fusion algorithm to obtain first collected data;

[0127] The second processing module is used to process the first collected data using Flink's stream processing technology to obtain second collected data;

[0128] The third processing module is used to process the second collected data using the Kalman filter algorithm to obtain the final prediction result.

[0129] In one embodiment of the present application, the third processing module includes a first processing submodule,

[0130] The first processing submodule is used to process the second collected data using the Kalman filter algorithm to obtain the final prediction result.

[0131] use

[0132] as well as

[0133] Update the prediction results of the above Kalman filter algorithm;

[0134] in, is the above final prediction result or the current prediction result, is the second collected data, P k - is the current forecast error covariance, P k is the current update error covariance, I is the unit matrix, H is the observation matrix, K k is the Kalman gain matrix, is the current updated prediction result, R is the observation noise covariance, z k is the observed value at the current time k.

[0135] In one embodiment of the present application, the third processing module includes a second processing submodule,

[0136] The second processing submodule is used to

[0137] Determine the above final prediction results, where A is the state transfer matrix, B is the control matrix, is the above final prediction result or the current prediction result, is the second collected data, P k - is the current prediction error covariance, Q is the process noise covariance, P k-1 is the error covariance of the last update.

[0138] In one embodiment of the present application, the first processing unit includes a fourth processing module and a fifth processing module.

[0139] The fourth processing module is used to process the initial collected data using the weighted fusion algorithm to obtain the first collected data;

[0140] The fifth processing module is used to process the first collected data using the Kalman filter algorithm to obtain the final prediction result.

[0141] In one embodiment of the present application, the second processing unit includes a sixth processing module, a seventh processing module and an eighth processing module.

[0142] The sixth processing module is used to determine that the prediction difference is the difference between the final prediction result and the last measured result;

[0143] The seventh processing module is used for determining to generate the above-mentioned warning information when the absolute value of the above-mentioned prediction difference is within a preset range;

[0144] The eighth processing module is used to determine not to generate the above-mentioned warning information when the absolute value of the above-mentioned prediction difference is not within the above-mentioned preset range.

[0145] In one embodiment of the present application, the above-mentioned device also includes a third processing unit. When it is determined that the warning information is generated, the third processing unit is used to broadcast the above-mentioned warning information in the form of voice.

[0146] The above-mentioned early warning and positioning device for power grid faults includes a processor and a memory. The above-mentioned acquisition unit, the first processing unit, the second processing unit, etc. are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions. The above-mentioned modules are all located in the same processor; or, the above-mentioned modules are located in different processors in any combination.

[0147] The processor contains a kernel, which calls the corresponding program unit from the memory. One or more kernels can be set, and the kernel parameters can be adjusted to solve the problem of power grid fault warning and location. The existing solution uses voltage, current or a single sensor for simple judgment, which makes it difficult to find abnormal equipment problems.

[0148] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0149] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned power grid fault early warning and positioning method.

[0150] An embodiment of the present invention provides a processor, and the processor is used to run a program, wherein the program executes the above-mentioned power grid fault early warning and positioning method when running.

[0151] The embodiment of the present invention provides a device, which includes a processor, a memory, and a program stored in the memory and can be run on the processor. When the processor executes the program, at least the following steps are implemented: obtaining the initial collection data of the current power grid node collected by multiple sensors, the sensors including current sensors, temperature sensors and vibration sensors; using at least a weighted fusion algorithm and a Kalman filter algorithm to process the initial collection data to obtain a final prediction result; determining whether to generate an early warning message based on the size of the final prediction result and the last measured result, the early warning message is used to prompt the current power grid node failure. The device in this article can be a server, PC, PAD, mobile phone, etc.

[0152] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having at least the following method steps: obtaining initial collection data of a current power grid node collected by multiple sensors, the above sensors including current sensors, temperature sensors and vibration sensors; processing the above initial collection data using at least a weighted fusion algorithm and a Kalman filter algorithm to obtain a final prediction result; determining whether to generate early warning information based on the size of the above final prediction result and the last measured result, the above early warning information being used to prompt a fault in the above current power grid node.

[0153] The present application also provides a power grid fault warning and positioning system, the system includes: one or more processors, memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include methods for executing any of the above methods. The initial collection data of the current power grid node collected by the current sensor, temperature sensor and vibration sensor are obtained, and the initial collection data is processed by at least a weighted fusion algorithm and a Kalman filter algorithm to obtain the final prediction result, and finally determine whether to generate warning information based on the size of the final prediction result and the last measured result. Compared with the existing solution, multiple sensors are considered, and the data collected by multiple sensors are fused and judged, which improves the accuracy of warning and positioning, thereby solving the problem that the existing solution for power grid fault warning and positioning is difficult to find abnormal equipment by simple judgment through voltage, current or a single sensor.

[0154] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0155] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0156] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0157] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0159] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0160] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0161] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0162] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0163] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0164] 1) The early warning and positioning method of power grid faults of the present application obtains the initial collected data of the current power grid node collected by the current sensor, temperature sensor and vibration sensor, and processes the above initial collected data using at least a weighted fusion algorithm and a Kalman filter algorithm to obtain the final prediction result. Finally, based on the size of the above final prediction result and the last measured result, it is determined whether to generate early warning information. Compared with the existing scheme, multiple sensors are considered, and the data collected by multiple sensors are fused and judged, which improves the accuracy of early warning and positioning, thereby solving the problem that the existing scheme for early warning and positioning of power grid faults is difficult to detect equipment abnormalities through simple judgment through voltage, current or a single sensor.

[0165] 2) The power grid fault early warning and positioning device of the present application obtains the initial collection data of the current power grid node collected by the current sensor, temperature sensor and vibration sensor, and processes the above initial collection data using at least a weighted fusion algorithm and a Kalman filter algorithm to obtain the final prediction result. Finally, based on the size of the above final prediction result and the last measured result, it is determined whether to generate early warning information. Compared with the existing scheme, multiple sensors are considered, and the data collected by multiple sensors are fused and judged, which improves the accuracy of early warning and positioning, thereby solving the problem that the existing scheme for early warning and positioning of power grid faults uses simple judgments through voltage, current or a single sensor, and it is difficult to detect equipment abnormalities.

[0166] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for early warning and locating power grid faults, characterized in that: include: Acquire initial collection data of the current power grid node collected by multiple sensors, wherein the sensors include a current sensor, a temperature sensor, and a vibration sensor; At least a weighted fusion algorithm and a Kalman filter algorithm are used to process the initial collected data to obtain a final prediction result; According to the size of the final prediction result and the last measured result, it is determined whether to generate warning information, and the warning information is used to prompt the current power grid node failure.

2. The method according to claim 1, characterized in that At least a weighted fusion algorithm and a Kalman filter algorithm are used to process the initial collected data to obtain a final prediction result, including: Processing the initial collected data using the weighted fusion algorithm to obtain first collected data; The first collected data is processed by using Flink's stream processing technology to obtain second collected data; The second collected data is processed using the Kalman filter algorithm to obtain the final prediction result.

3. The method according to claim 2, characterized in that In the process of using the Kalman filter algorithm to process the second collected data to obtain the final prediction result, the method further includes: use as well as Updating the prediction result of the Kalman filter algorithm; in, is the final prediction result or the current prediction result, For the second collected data, P k - is the current forecast error covariance, P k is the current update error covariance, I is the unit matrix, H is the observation matrix, K k is the Kalman gain matrix, is the current updated prediction result, and R is the observation noise covariance.

4. The method according to claim 2, characterized in that: The second collected data is processed using the Kalman filter algorithm to obtain the final prediction result, including: according to and P k - =AP k-1 A T +Q, Determine the final prediction result, where A is the state transfer matrix, B is the control matrix, is the final prediction result or the current prediction result, For the second collected data, P k - is the current prediction error covariance, Q is the process noise covariance, P k-1 is the error covariance of the last update, z k is the observed value at the current time k.

5. The method according to claim 1, characterized in that At least a weighted fusion algorithm and a Kalman filter algorithm are used to process the initial collected data to obtain a final prediction result, including: Processing the initial collected data using the weighted fusion algorithm to obtain first collected data; The first collected data is processed using the Kalman filter algorithm to obtain the final prediction result.

6. The method according to claim 1, characterized in that Determine whether to generate warning information based on the difference between the final prediction result and the last measured result, including: Determine the prediction difference as the difference between the final prediction result and the last measured result; When the absolute value of the predicted difference is within a preset range, determining to generate the warning information; When the absolute value of the predicted difference is not within the preset range, it is determined not to generate the warning information.

7. The method according to any one of claims 1 to 6, characterized in that In the case of determining to generate warning information, the method further includes: The warning information is broadcast in the form of voice.

8. A power grid fault early warning and positioning device, characterized in that: include: An acquisition unit, used to acquire initial acquisition data of a current power grid node acquired by a plurality of sensors, wherein the sensors include a current sensor, a temperature sensor and a vibration sensor; A first processing unit, configured to process the initial collected data by at least using a weighted fusion algorithm and a Kalman filter algorithm to obtain a final prediction result; The second processing unit is used to determine whether to generate early warning information according to the size of the final prediction result and the last measured result, and the early warning information is used to prompt the current power grid node failure.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

10. A power grid fault early warning and positioning system, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of claims 1 to 7.

Images