Substation monitoring method, system and apparatus

Abstract

This invention provides a substation monitoring method, system, and device, relating to the field of power system risk monitoring technology. The invention collects bus current and voltage waveform data in real time, performs time synchronization processing, constructs a discrete expression for electrical energy, and calculates the change in electrical energy between adjacent sampling intervals based on the trapezoidal integral method. It sets thresholds and combines them with adaptive sliding window statistical verification to identify electrical energy jump intervals and abnormal time windows. Then, it collects electrical parameters within the abnormal time window, constructs a mutation degree index using first-order and second-order differences, detects the set of abnormal mutation time points for electrical parameters, obtains the surface temperature distribution of equipment through infrared thermal imaging, identifies abnormal temperature regions by combining hotspot thresholds and connectivity region algorithms, statistically analyzes their characteristic indicators to construct an abnormal temperature function, extracts the set of abnormal temperature mutation time points, and takes the intersection of the electrical parameters and the set of abnormal temperature mutation time points to accurately determine the potential fault time.

Description

Technical Field

[0001] This invention relates to the field of power system risk monitoring technology, specifically to a substation monitoring method, system, and device. Background Technology

[0002] With the continuous expansion of the power system and the increasing complexity and variability of the operating environment, substations, as important nodes in the power grid, play a crucial role in ensuring the overall reliability of the power system through their safe and stable operation. Traditional substation monitoring often relies on single electrical parameters or manual inspections, which suffers from problems such as monitoring blind spots, response delays, and inaccurate fault location. With the development of sensor technology and big data analysis, the comprehensive monitoring of substation status using multi-source heterogeneous data has become a research hotspot. However, how to effectively integrate multiple monitoring data, accurately identify abnormal patterns, locate faults in a timely manner, and reduce false alarm rates remains a key challenge in the current technology.

[0003] In existing technologies, traditional substation monitoring technologies mainly rely on monitoring single electrical parameters or manual inspections, lacking the ability to fuse and analyze multi-source heterogeneous data. This results in asynchronous current and voltage waveform data, leading to inaccurate power analysis. Furthermore, traditional anomaly detection methods based solely on electrical parameters struggle to accurately identify power jump intervals and their start and end times. The unclear division of abnormal time limits the timeliness and accuracy of fault diagnosis, easily leading to monitoring blind spots and missed reports. Abnormal changes in electrical signals such as current and voltage are often difficult to capture and identify in a timely and accurate manner, especially in the early stages of fault development, where effective prediction and early warning methods are often lacking.

[0004] Therefore, it is necessary to provide a substation monitoring method, system, and device to solve the aforementioned problem.

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

[0006] The purpose of this invention is to provide a substation monitoring method, system, and device to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A substation monitoring method, comprising the following steps: Step 1: Real-time acquisition of current and voltage waveform data of the target substation bus, time synchronization processing of the acquired current and voltage waveform data, and energy analysis based on the processed current and voltage waveform data to construct a discrete expression of the energy of the target substation bus. Step 2: Calculate the change in electrical energy within adjacent sampling intervals in the discrete expression of electrical energy based on the trapezoidal integral method, set a threshold for electrical energy change, filter out the electrical energy jump interval based on the change in electrical energy, and detect the start and end points of the electrical energy jump based on the statistical test method of adaptive sliding window in order to determine the abnormal time window of the substation. Step 3: Collect electrical parameters of the target substation within the abnormal time window, and construct a mutation degree index by performing first-order and second-order difference calculations on the electrical parameters within the abnormal time window of the target substation to detect the set of abnormal mutation time points of electrical parameters. Step 4: Obtain thermal images of the surfaces of relevant equipment in the target substation, determine the temperature anomaly areas of the target substation, and statistically analyze the characteristic indicators of the temperature anomaly areas to construct an abnormal temperature function, so as to detect the set of abnormal temperature change time points of the target substation. Based on the set of abnormal change time points of electrical parameters and the set of abnormal change time points of substation temperature, the potential time of fault occurrence can be determined.

[0008] Furthermore, the collected current and voltage waveform data are processed for time synchronization to construct a discrete expression for the target substation bus power. The method used is as follows: First, a unified time axis is determined, and the sampling time points of the voltage signal are used as the reference time series. For data points of the current signal that do not exist at the reference time, the linear interpolation method is used to calculate the current value corresponding to these moments. Finally, a set of synchronous current and voltage data at exactly the same time points is formed, and the same sampling interval is set. The sampling interval is the reciprocal of the sampling frequency. The instantaneous power of the busbar in the three-phase AC system of the substation is calculated at each moment. Based on the instantaneous power of the busbar in the three-phase AC system of the substation at each moment, and combined with the preset sampling time interval, the cumulative electrical energy of the substation during the total sampling time is calculated.

[0009] Furthermore, the change in electrical energy within adjacent sampling intervals is calculated to screen out electrical energy jump intervals. The method used is as follows: The change in electrical energy within adjacent sampling intervals is calculated using the trapezoidal integral method. For the th For each sampling point, take the sampling points on both sides. and The total instantaneous power at a given point is estimated using two trapezoidal approximations. The energy approximation on the left is equal to the energy approximation in the interval [missing value]. The trapezoidal energy estimate is calculated by multiplying half the sum of the powers at both ends by the sampling time interval. The energy approximation on the right side is equal to the energy value within the interval. The trapezoidal energy estimation is that half of the sum of the power at both ends is multiplied by the sampling time interval, and the change in electrical energy between two adjacent intervals is taken as the absolute value of the difference between these two approximate energies. The change in power consumption between two adjacent sampling intervals is compared with a set power consumption change threshold. If the change in power consumption exceeds the power consumption change threshold, the two adjacent sampling intervals are marked as power consumption jump intervals.

[0010] Furthermore, a statistical test method based on an adaptive sliding window is used to detect the start and end points of power jumps in order to determine the abnormal time window of the substation. The method used is as follows: Within the selected energy jump interval, two adjacent sliding windows of equal length are chosen. and These are used to detect the first and second sampling intervals within the energy jump interval, respectively, to calculate the mean and variance of the energy data in the two windows, and then to perform statistical hypothesis testing to determine whether the energy data in the two windows come from the same distribution. The significance level was set based on hypothesis testing. And calculate the degrees of freedom between two adjacent sliding windows. In the standard distribution critical value table, based on the degrees of freedom and Find the corresponding critical value ; At this point, starting from the left end of the energy jump interval, move the sliding window point by point to the right. Select the first one that satisfies: The time point is taken as the starting point of the energy jump; starting from the right end of the energy jump interval, the sliding window is moved point by point to the left. Select the first one that satisfies: The time point is used as the termination point of the power jump, and the abnormal time window of the substation is determined based on the start and end points of the power jump.

[0011] Furthermore, first-order and second-order difference calculations are performed on the electrical parameters within the abnormal time window of the substation, based on the following method: Abnormal time window of the target substation Electrical parameters and their corresponding timestamps are collected, and the first-order and second-order differences of the electrical parameter sequences are calculated respectively. First, within the abnormal time window of the substation, the electrical parameter sequences are collected. And its corresponding timestamp, the first-order difference value is the electrical parameter at... The first difference at time t is equal to the parameter value at that time. Subtract the parameter value from the previous moment The first-order difference represents the change in the parameter between two adjacent sampling points, i.e., the instantaneous rate of change; the second-order difference value is the electrical parameter sequence at... The second difference value at time t is equal to the parameter value at that time. Subtract twice the parameter value from the previous moment In addition to the parameter values ​​from the previous two time points The resulting second-order difference represents the change in the first-order difference, i.e., the acceleration of the parameter change, where... This represents the index of a time step in the first-order difference sequence. Let be the index of time in the second-order difference sequence, and , .

[0012] Furthermore, a mutation rate index is constructed to detect the set of time points of abnormal mutations in electrical parameters. The method used is as follows: A mutation rate index is constructed based on the first and second difference values ​​of the electrical parameter sequence within the abnormal time window. A detection time window is established, and its length is defined as follows: The local standard deviation of the difference is calculated to measure the degree of abrupt change. Statistical historical normal data Based on the distribution, the 95th percentile was selected as the mutation threshold. By detecting time windows, each moment within a different time window is examined. If a problem exists... If a time point is detected as an abnormal change in electrical parameters, then that time point is determined as an abnormal change in electrical parameters. All abnormal change time points detected within the abnormal time window are grouped into a set and defined as the set of abnormal change time points for electrical parameters.

[0013] Furthermore, the abnormal temperature region of the target substation is identified, and an abnormal temperature function is constructed to detect the set of time points of abrupt temperature changes in the target substation, thereby determining the potential time of fault occurrence. The method used is as follows: Within the abnormal time window, infrared thermal imaging is performed on the relevant equipment in the target substation using the same sampling interval as the electrical parameters. The signal intensity corresponding to each pixel in the infrared image represents the infrared radiation intensity of that pixel, and the temperature of that pixel is obtained based on Planck's radiation law. Data was collected during the normal operation of the relevant equipment to statistically analyze the average temperature and standard deviation of the surface area of ​​the relevant equipment, and hot spot thresholds were set based on the three-standard-deviation rule. The temperature of each pixel in the infrared image is compared with a set hotspot threshold to filter out pixels that exceed the hotspot threshold and mark them as hotspot pixels. A two-pass connected region labeling algorithm is used to select temperature anomaly areas in the infrared image.

[0014] Furthermore, the abnormal temperature region of the target substation is identified, and an abnormal temperature function is constructed to detect the set of time points of abrupt temperature changes in the target substation, thereby determining the potential time of fault occurrence. The method used is as follows: The system collects characteristic indicators of abnormal temperature areas, including hot spot area ratio, average hot spot temperature, and maximum hot spot temperature. Based on these indicators, an abnormal temperature function is constructed. The system then detects the hot spot areas of the target substation within an abnormal time window according to a pre-set sampling interval. The abnormal index corresponding to each sampling time is obtained, and all abnormal indices are arranged in descending order. The top 5% of abnormal indices are selected as monitoring objects, and their corresponding sampling times are recorded as the set of abnormal temperature change time points in the substation. The intersection of the set of abnormal change times of electrical parameters and the set of abnormal change times of substation temperature is used as a potential time for monitoring substation faults.

[0015] The present invention also provides a substation monitoring system, the monitoring system being used to execute the above-described substation monitoring method, comprising: The power modeling module is used to collect the waveform data of current and voltage of the target substation bus in real time, perform time synchronization processing on the collected current and voltage waveform data, perform power analysis based on the processed current and voltage waveform data, and construct a discrete expression of the power of the target substation bus. The abnormal time window extraction module calculates the change in electrical energy within adjacent sampling intervals based on the trapezoidal integral method, sets a threshold for electrical energy change, filters out the electrical energy jump interval based on the change in electrical energy, and detects the start and end points of the electrical energy jump based on the statistical test method of adaptive sliding window to determine the abnormal time window of the substation. The electrical parameter mutation analysis module is used to collect electrical parameters of the target substation within an abnormal time window. It constructs a mutation degree index by performing first-order and second-order difference calculations on the electrical parameters of the target substation within the abnormal time window, so as to detect the set of abnormal mutation time points of electrical parameters. The potential fault timing determination module is used to acquire thermal images of the surfaces of relevant equipment in the target substation, identify temperature anomaly areas in the target substation, and construct an abnormal temperature function by statistically analyzing the characteristic indicators of the temperature anomaly areas to detect the set of abnormal temperature change time points in the target substation. Based on the set of abnormal change time points of electrical parameters and the set of abnormal change time points of substation temperature, the potential timing of the fault occurrence is determined.

[0016] The present invention also provides a substation monitoring device, wherein the monitoring device uses the above-described substation monitoring method for monitoring.

[0017] Compared with the prior art, the beneficial effects of the present invention are: This invention achieves accurate modeling of substation operating status by real-time acquisition and time-synchronous processing of bus current and voltage waveform data, combined with power analysis and trapezoidal integral method. It effectively identifies power jump intervals and clarifies abnormal time windows by using adaptive sliding window statistical verification method, further improving the sensitivity and accuracy of fault detection. In addition, it introduces first-order and second-order differential mutation degree index to achieve rapid detection of electrical parameter anomalies, significantly improving the response speed to sudden faults. Furthermore, this invention integrates thermal imaging technology. By comprehensively analyzing the temperature anomaly areas and characteristic indicators on the surface of relevant equipment, an abnormal temperature function is constructed, combining temperature anomalies with electrical anomalies to achieve multi-dimensional and multi-physical quantity data fusion monitoring. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the overall method flow of the present invention.

[0019] Figure 2 This is a schematic diagram of the system module flow of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0021] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0022] Example: Please see Figure 1 A substation monitoring method, the specific steps of which include: Step 1: Real-time acquisition of current and voltage waveform data of the target substation bus, time synchronization processing of the acquired current and voltage waveform data, and energy analysis based on the processed current and voltage waveform data to construct a discrete expression of the energy of the target substation bus.

[0023] In a specific embodiment of the present invention, by performing time synchronization processing on the current and voltage waveform data, and using linear interpolation to supplement the data at different sampling time points, the three-phase current and voltage data are completely aligned, ensuring the consistency and accuracy of the data. Based on the synchronized data, the total instantaneous power of the bus at each moment is calculated using the instantaneous power formula of the three-phase AC system. Then, the cumulative electrical energy is calculated by using the numerical integration method in combination with the sampling interval, thus constructing a discrete expression for the electrical energy of the substation bus. Time synchronization processing solves the analysis error problem caused by inconsistent timestamps in multi-channel sampling data, ensuring the accuracy and stability of power calculation. The precise discrete expression of power provides quantitative indicators for power jump detection and energy change analysis, improving the sensitivity and accuracy of abnormal event identification. The power model built based on this method provides solid data support for intelligent early warning systems and fault location, which helps to improve the safe operation level and maintenance efficiency of substations and promotes the development of power systems towards a more intelligent and efficient direction.

[0024] Furthermore, the collected current and voltage waveform data are processed for time synchronization to construct a discrete expression for the target substation bus power. The method used is as follows: First, a unified time axis is determined, and the sampling time points of the voltage signal are used as the reference time series. For data points of the current signal that do not exist at the reference time, the linear interpolation method is used to calculate the current value corresponding to these moments. Finally, a set of synchronous current and voltage data at exactly the same time points is formed, and the same sampling interval is set. The sampling interval is the reciprocal of the sampling frequency. The formula used to calculate the instantaneous power of the substation busbar at each moment in a three-phase AC system based on a substation is: in, Indicates that the substation is in Time of the first The total instantaneous power of each sampling point, , , These respectively represent the three-phase AC system Time of the first Collected from each sampling point Phase voltage value, Phase voltage value, Phase voltage value, , , These respectively represent the three-phase AC system Time of the first Each sampling point collected Phase current value, Phase current value, Phase current value, The index of the sampling point, and , This represents the total number of sampling points; Based on the instantaneous power of the busbar in the three-phase AC system of the substation at each moment obtained through calculation, and combined with the preset sampling time interval, the cumulative electrical energy of the substation during the total sampling time is calculated. The formula used is as follows: in, This is a discrete expression for the electrical energy of the substation busbar, used to represent the cumulative electrical energy of the substation during the sampling period. It is the sampling interval.

[0025] Step 2: Calculate the change in electrical energy within adjacent sampling intervals in the discrete expression of electrical energy based on the trapezoidal integral method, set a threshold for electrical energy change, filter out the electrical energy jump intervals based on the change in electrical energy, and detect the start and end points of the electrical energy jumps based on the statistical test method of adaptive sliding window to determine the abnormal time window of the substation.

[0026] In a specific embodiment of the present invention, this part accurately calculates the change in electrical energy between adjacent sampling intervals based on the trapezoidal integral method, and filters out the electrical energy jump intervals by combining the electrical energy change threshold. Then, it uses the statistical test method of adaptive sliding window to perform significance analysis on the electrical energy data, which can accurately identify the start and end points of the electrical energy change, thereby efficiently and accurately determining the abnormal time window of the substation. This technical means effectively improves the sensitivity and detection accuracy of abnormal changes in electrical energy, reduces the possibility of subjective human judgment and false alarms, and provides a reliable time basis and data foundation for subsequent anomaly tracing and fault location, significantly enhancing the intelligence and automation level of substation monitoring.

[0027] The reason for screening power jump intervals and determining abnormal time windows is that when a substation experiences operational anomalies or potential faults, it often causes a sudden change in power within a very short time. By calculating and comparing the power changes in adjacent sampling intervals, these abnormal energy jump intervals can be identified in a timely manner. Furthermore, statistical testing methods can be used to accurately locate the start and end points of the jumps. This helps to effectively extract the critical time periods of abnormal events from massive monitoring data, avoiding the omission of abnormal information and improving the pertinence and efficiency of subsequent electrical parameter and temperature anomaly analysis.

[0028] It should be noted that calculating the change in electrical energy within adjacent sampling intervals to screen out electrical energy jump intervals serves the following purpose: comparing the change in electrical energy between adjacent sampling intervals can sensitively capture instantaneous changes in electrical energy during substation operation, accurately reflect the short-term characteristics of anomalies, and avoid the impact of long-term trends or noise interference; at the same time, this method is simple to calculate, facilitates real-time online monitoring, and helps to accurately locate the start and end times of anomalies, improving the timeliness and accuracy of anomaly detection, thereby providing reliable data support for substation fault early warning and diagnosis.

[0029] Furthermore, the change in electrical energy within adjacent sampling intervals is calculated to screen out the energy jump intervals, and the start and end points of the energy jumps are detected based on the statistical test method of adaptive sliding window to determine the abnormal time window of the substation. The method used is as follows: The formula used to calculate the change in electrical energy within adjacent sampling intervals based on the trapezoidal integral method is as follows: in, This represents the change in electrical energy between two adjacent sampling intervals. Indicates that the substation is in Time of the first The total instantaneous power of each sampling point, Indicates that the substation is in Time of the first The total instantaneous power of each sampling point; The change in power between two adjacent sampling intervals is compared with a set power change threshold. If the change in power exceeds the power change threshold, the two adjacent sampling intervals are marked as power jump intervals. Within the selected energy jump interval, two adjacent sliding windows of equal length are chosen. and These are used to detect the first and second sampling intervals within the energy jump interval, respectively. The mean and variance of the energy data within each window are calculated, and then a statistical hypothesis test is performed to determine whether the energy data from the two windows come from the same distribution. The formula used is: in, It is a statistic used to test for significant differences between sample means in two sliding windows. , These represent the number of sample electrical energy data points within the first and second adjacent sliding windows, respectively. , These represent the sample mean values ​​of the electrical energy data points within the first and second adjacent sliding windows, respectively. , These represent the sample variances of the electrical energy data points within the first and second adjacent sliding windows, respectively. The significance level was set based on hypothesis testing. And calculate the degrees of freedom between two adjacent sliding windows. In the standard distribution critical value table, based on the degrees of freedom and Find the corresponding critical value ; At this point, starting from the left end of the energy jump interval, move the sliding window point by point to the right. Select the first one that satisfies: The time point is taken as the starting point of the energy jump; starting from the right end of the energy jump interval, the sliding window is moved point by point to the left. Select the first one that satisfies: The time point is used as the termination point of the power jump, and the abnormal time window of the substation is determined based on the start and end points of the power jump.

[0030] Step 3: Collect electrical parameters of the target substation within the abnormal time window. Construct a mutation degree index by performing first-order and second-order difference calculations on the electrical parameters within the abnormal time window of the target substation to detect the set of abnormal mutation time points of electrical parameters.

[0031] In a specific embodiment of this invention, by performing first- and second-order difference analysis on the electrical parameter sequence within the abnormal time window of a substation, constructing a mutation degree index by combining local standard deviation, and setting mutation criteria based on the high percentile of historical data, efficient and accurate detection of abnormal mutation time points of electrical parameters is achieved. This method can effectively identify key mutation moments in the abnormal process, automatically filter out specific time points that may have hidden dangers or faults, significantly improve the accuracy and granularity of anomaly identification, reduce false alarms and missed alarms, and significantly enhance the anomaly analysis capability of substation monitoring data. It should be noted that abrupt changes in electrical parameters often indicate potential equipment failures, operational anomalies, or safety risks. Timely detection of these abrupt changes is crucial for ensuring the safe and stable operation of substations. This method can accurately pinpoint critical moments in anomaly processes, providing essential time-series data for anomaly tracing, fault location, and emergency response.

[0032] Furthermore, first-order and second-order difference calculations are performed on the electrical parameters within the abnormal time window of the substation to construct a mutation degree index, which is used to detect the set of abnormal mutation time points of electrical parameters. The method used is as follows: Abnormal time window of the target substation Electrical parameters and their corresponding timestamps are collected, and the first and second differences of the electrical parameter sequences are calculated respectively, using the following formula: in, Electrical parameters are The first-order difference value at time t represents the change of the electrical parameter between two adjacent sampling points, i.e., the instantaneous rate of change. It is an electrical parameter sequence in The second-order difference value at time t represents the rate of change of the first-order difference of the electrical parameter, i.e., the acceleration of the electrical parameter. , They represent in , The electrical parameter sampling values ​​at time 10:00. , , They represent in , , The electrical parameter sampling values ​​at time 10:00. This represents the index of a time step in the first-order difference sequence. Let be the index of time in the second-order difference sequence, and , ; A mutation rate index is constructed based on the first and second difference values ​​of the electrical parameter sequence within the abnormal time window. A detection time window is established, and its length is defined as follows: The method for calculating the local standard deviation of the difference to measure the degree of abrupt change is as follows: in, Indicates in The degree of abrupt change calculated at any given time. Indicates the detection time window Inner first-order difference sequence standard deviation Indicates the detection time window Inner second difference sequence Standard deviation; Statistical historical normal data Based on the distribution, the 95th percentile was selected as the mutation threshold. By detecting time windows, each moment within a different time window is examined. If a problem exists... If a time point is detected as an abnormal change in electrical parameters, then that time point is determined as an abnormal change in electrical parameters. All abnormal change time points detected within the abnormal time window are grouped into a set and defined as the set of abnormal change time points for electrical parameters.

[0033] Step 4: Obtain thermal images of the surfaces of relevant equipment in the target substation, determine the temperature anomaly areas of the target substation, and statistically analyze the characteristic indicators of the temperature anomaly areas to construct an abnormal temperature function, so as to detect the set of abnormal temperature change time points of the target substation. Based on the set of abnormal change time points of electrical parameters and the set of abnormal change time points of substation temperature, the potential time of fault occurrence can be determined.

[0034] In a specific embodiment of the present invention, the temperature distribution of relevant equipment in the target substation is acquired in real time by infrared thermal imaging technology. Combined with the hot spot threshold based on the statistics of normal operation data, the temperature abnormal area is accurately identified by the two-pass connected region algorithm. An abnormal temperature function is constructed by multi-dimensional feature indicators such as hot spot area ratio, hot spot average temperature and maximum temperature to achieve quantitative evaluation and dynamic monitoring of temperature abnormal area. By sorting and filtering abnormal temperature function values, significant temperature anomaly abrupt change time points can be effectively extracted. Then, by combining the intersection analysis with the electrical parameter anomaly abrupt change time points, the accuracy and reliability of identifying potential fault moments are significantly improved. This method integrates temperature anomaly features in both spatial and temporal dimensions, achieving highly sensitive detection and precise location of temperature anomalies.

[0035] It should be noted that abnormal temperatures are often an important precursor to potential equipment failures, aging, or abnormal operation. Timely identification of abnormal temperature areas and their sudden change times can help to detect potential equipment problems in advance and prevent the fault from escalating to the point of power outage or equipment damage. By constructing an abnormal temperature function and combining it with cross-validation of abnormal electrical parameter time points, not only is the accuracy of fault diagnosis improved, but the false alarm rate is also reduced, providing a scientific basis for preventive maintenance and safety management of substations.

[0036] Furthermore, the abnormal temperature region of the target substation is identified, and an abnormal temperature function is constructed to detect the set of time points of abrupt temperature changes in the target substation, thereby determining the potential time of fault occurrence. The method used is as follows: Within the abnormal time window, infrared thermal imaging is performed on the relevant equipment in the target substation using the same sampling interval as the electrical parameters. The signal intensity corresponding to each pixel in the infrared image represents the infrared radiation intensity of that pixel, and the temperature of that pixel is obtained based on Planck's radiation law. Data was collected during the normal operation of the relevant equipment to statistically analyze the average temperature and standard deviation of the surface area of ​​the equipment. A hotspot threshold was then set based on the three-standard-deviation rule, using the following formula: in, For the set hotspot threshold, This indicates the average temperature of the surface area of ​​the equipment. This represents the standard deviation of the temperature across a surface area of ​​the equipment. This is an empirical coefficient, with a value of 3; The temperature of each pixel in the infrared image is compared with a set hotspot threshold to filter out pixels that exceed the threshold and mark them as hotspot pixels. A two-pass connected region labeling algorithm is used to select temperature anomaly regions in the infrared image, and characteristic indicators of these regions are statistically analyzed. These characteristic indicators include the hotspot area ratio, the average temperature of the hotspots, and the maximum temperature of the hotspots. The method used is as follows: in, , , They represent in The hotspot area ratio, average hotspot temperature, and maximum hotspot temperature at any given time. This represents the total number of pixels in the infrared image. Indicates in The number of all hotspot pixels detected at any given time. Indicates the location is hotspot pixels at Temperature value at time, For at any time The set of coordinates of pixels identified as hotspots; The abnormal temperature function is constructed based on these characteristic indicators, and the formula used is as follows: in, Indicates in Anomaly index of the temperature anomaly region at any given time; Based on the constructed abnormal temperature function, the hot spot area of ​​the target substation is detected within the abnormal time window according to the pre-set sampling interval. The abnormal index corresponding to each sampling time is obtained, and all abnormal indices are arranged in descending order. The top 5% of abnormal indices are selected as monitoring objects, and their corresponding sampling times are recorded as the set of abnormal temperature change time points of the substation. The intersection of the set of abnormal change times of electrical parameters and the set of abnormal change times of substation temperature is used as a potential time for monitoring substation faults.

[0037] Please see Figure 2 The present invention also provides a substation monitoring system, which is used to execute the above-described substation monitoring method, including: The power modeling module is used to collect the waveform data of current and voltage of the target substation bus in real time, perform time synchronization processing on the collected current and voltage waveform data, perform power analysis based on the processed current and voltage waveform data, and construct a discrete expression of the power of the target substation bus. The abnormal time window extraction module calculates the change in electrical energy within adjacent sampling intervals based on the trapezoidal integral method, sets a threshold for electrical energy change, filters out the electrical energy jump interval based on the change in electrical energy, and detects the start and end points of the electrical energy jump based on the statistical test method of adaptive sliding window to determine the abnormal time window of the substation. The electrical parameter mutation analysis module is used to collect electrical parameters of the target substation within an abnormal time window. It constructs a mutation degree index by performing first-order and second-order difference calculations on the electrical parameters of the target substation within the abnormal time window, so as to detect the set of abnormal mutation time points of electrical parameters. The potential fault timing determination module is used to acquire thermal images of the surfaces of relevant equipment in the target substation, identify temperature anomaly areas in the target substation, and construct an abnormal temperature function by statistically analyzing the characteristic indicators of the temperature anomaly areas to detect the set of abnormal temperature change time points in the target substation. Based on the set of abnormal change time points of electrical parameters and the set of abnormal change time points of substation temperature, the potential timing of the fault occurrence is determined.

[0038] A substation monitoring device, wherein the monitoring device uses the above-mentioned substation monitoring method for monitoring.

[0039] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0040] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0041] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0042] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method of monitoring a substation, characterized by, The specific steps include: Step 1: Real-time acquisition of current and voltage waveform data of the target substation bus, time synchronization processing of the acquired current and voltage waveform data, and energy analysis based on the processed current and voltage waveform data to construct a discrete expression of the energy of the target substation bus. Step 2: Calculate the change in electrical energy within adjacent sampling intervals in the discrete expression of electrical energy based on the trapezoidal integral method, set a threshold for electrical energy change, filter out the electrical energy jump interval based on the change in electrical energy, and detect the start and end points of the electrical energy jump based on the statistical test method of adaptive sliding window in order to determine the abnormal time window of the substation. Step 3: Collect electrical parameters of the target substation within the abnormal time window, and construct a mutation degree index by performing first-order and second-order difference calculations on the electrical parameters within the abnormal time window of the target substation to detect the set of abnormal mutation time points of electrical parameters. Step 4: Obtain thermal images of the surfaces of relevant equipment in the target substation, determine the temperature anomaly areas of the target substation, and statistically analyze the characteristic indicators of the temperature anomaly areas to construct an abnormal temperature function, so as to detect the set of abnormal temperature change time points of the target substation. Based on the set of abnormal change time points of electrical parameters and the set of abnormal change time points of substation temperature, the potential time of fault occurrence can be determined.

2. The substation monitoring method of claim 1, wherein, The collected current and voltage waveform data are processed for time synchronization to construct a discrete expression for the bus power of the target substation. The method used is as follows: First, a unified time axis is determined, and the sampling time points of the voltage signal are used as the reference time series. For data points of the current signal that do not exist at the reference time, the linear interpolation method is used to calculate the current value corresponding to these moments. Finally, a set of synchronous current and voltage data at exactly the same time points is formed, and the same sampling interval is set. The sampling interval is the reciprocal of the sampling frequency. The instantaneous power of the busbar in the three-phase AC system of the substation is calculated at each moment. Based on the instantaneous power of the busbar in the three-phase AC system of the substation at each moment, and combined with the preset sampling time interval, the cumulative electrical energy of the substation during the total sampling time is calculated.

3. A substation monitoring method according to claim 2, characterized in that, The method used to calculate the change in electrical energy within adjacent sampling intervals to screen out electrical energy jump intervals is as follows: The change in electrical energy within adjacent sampling intervals is calculated using the trapezoidal integral method. For the th For each sampling point, take the sampling points on both sides. and The total instantaneous power at a given point is estimated using two trapezoidal approximations. The energy approximation on the left is equal to the energy approximation in the interval [missing value]. The trapezoidal energy estimate is calculated by multiplying half the sum of the powers at both ends by the sampling time interval. The energy approximation on the right side is equal to the energy value within the interval. The trapezoidal energy estimation is that half of the sum of the power at both ends is multiplied by the sampling time interval, and the change in electrical energy between two adjacent intervals is taken as the absolute value of the difference between these two approximate energies. The change in power consumption between two adjacent sampling intervals is compared with a set power consumption change threshold. If the change in power consumption exceeds the power consumption change threshold, the two adjacent sampling intervals are marked as power consumption jump intervals.

4. The substation monitoring method according to claim 3, characterized in that, The statistical test method based on adaptive sliding window is used to detect the start and end points of power jumps in order to determine the abnormal time window of the substation. The method is as follows: Within the selected energy jump interval, two adjacent sliding windows of equal length are chosen. and These are used to detect the first and second sampling intervals within the energy jump interval, respectively, to calculate the mean and variance of the energy data in the two windows, and then to perform statistical hypothesis testing to determine whether the energy data in the two windows come from the same distribution. The significance level was set based on hypothesis testing. And calculate the degrees of freedom between two adjacent sliding windows. In the standard distribution critical value table, based on the degrees of freedom and Find the corresponding critical value ; At this point, starting from the left end of the energy jump interval, move the sliding window point by point to the right. Select the first one that satisfies: The point in time is taken as the starting point of the energy jump; Starting from the right end of the energy jump interval, move the sliding window point by point to the left. Select the first one that satisfies: The time point is used as the termination point of the power jump, and the abnormal time window of the substation is determined based on the start and end points of the power jump.

5. A substation monitoring method according to claim 4, characterized in that, The method used to perform first-order and second-order difference calculations on the electrical parameters within the abnormal time window of the substation is as follows: Abnormal time window of the target substation Electrical parameters and their corresponding timestamps are collected, and the first-order and second-order differences of the electrical parameter sequences are calculated respectively. First, within the abnormal time window of the substation, the electrical parameter sequences are collected. And its corresponding timestamp, the first-order difference value is the electrical parameter at... The first difference at time t is equal to the parameter value at that time. Subtract the parameter value from the previous moment The first-order difference represents the change in the parameter between two adjacent sampling points, i.e., the instantaneous rate of change; the second-order difference value is the electrical parameter sequence at... The second difference value at time t is equal to the parameter value at that time. Subtract twice the parameter value from the previous moment In addition to the parameter values ​​from the previous two time points The resulting second-order difference represents the change in the first-order difference, i.e., the acceleration of the parameter change, where... This represents the index of a time step in the first-order difference sequence. Let be the index of time in the second-order difference sequence, and , .

6. A substation monitoring method according to claim 5, characterized in that, The method used to construct a mutation rate index to detect a set of time points with abnormal mutations in electrical parameters is as follows: A mutation rate index is constructed based on the first and second difference values ​​of the electrical parameter sequence within the abnormal time window. A detection time window is established, and its length is defined as follows: The local standard deviation of the difference is calculated to measure the degree of abrupt change. Statistical historical normal data Based on the distribution, the 95th percentile was selected as the mutation threshold. By detecting time windows, each moment within a different time window is examined. If a problem exists... If a time point is detected as an abnormal change in electrical parameters, then that time point is determined as an abnormal change in electrical parameters. All abnormal change time points detected within the abnormal time window are grouped into a set and defined as the set of abnormal change time points for electrical parameters.

7. A substation monitoring method according to claim 6, characterized in that, The method used to identify the temperature anomaly zone of the target substation, construct an abnormal temperature function to detect the set of time points of abrupt temperature changes in the target substation, and determine the potential time of fault occurrence is as follows: Within the abnormal time window, infrared thermal imaging is performed on the relevant equipment in the target substation using the same sampling interval as the electrical parameters. The signal intensity corresponding to each pixel in the infrared image represents the infrared radiation intensity of that pixel, and the temperature of that pixel is obtained based on Planck's radiation law. Data was collected during the normal operation of the relevant equipment to statistically analyze the average temperature and standard deviation of the surface area of ​​the relevant equipment, and hot spot thresholds were set based on the three-standard-deviation rule. The temperature of each pixel in the infrared image is compared with a set hotspot threshold to filter out pixels that exceed the hotspot threshold and mark them as hotspot pixels. A two-pass connected region labeling algorithm is used to select temperature anomaly areas in the infrared image.

8. A substation monitoring method according to claim 7, characterized in that, The method used to identify the temperature anomaly zone of the target substation, construct an abnormal temperature function to detect the set of time points of abrupt temperature changes in the target substation, and determine the potential time of fault occurrence is as follows: The system collects characteristic indicators of abnormal temperature areas, including hot spot area ratio, average hot spot temperature, and maximum hot spot temperature. Based on these indicators, an abnormal temperature function is constructed. The system then detects the hot spot areas of the target substation within an abnormal time window according to a pre-set sampling interval. The abnormal index corresponding to each sampling time is obtained, and all abnormal indices are arranged in descending order. The top 5% of abnormal indices are selected as monitoring objects, and their corresponding sampling times are recorded as the set of abnormal temperature change time points in the substation. The intersection of the set of abnormal change times of electrical parameters and the set of abnormal change times of substation temperature is used as a potential time for monitoring substation faults.

9. A substation monitoring system, characterized in that, The monitoring system is used to execute the substation monitoring method according to any one of claims 1-8, including: The power modeling module is used to collect the waveform data of current and voltage of the target substation bus in real time, perform time synchronization processing on the collected current and voltage waveform data, perform power analysis based on the processed current and voltage waveform data, and construct a discrete expression of the power of the target substation bus. The abnormal time window extraction module calculates the change in electrical energy within adjacent sampling intervals based on the trapezoidal integral method, sets a threshold for electrical energy change, filters out the electrical energy jump interval based on the change in electrical energy, and detects the start and end points of the electrical energy jump based on the statistical test method of adaptive sliding window to determine the abnormal time window of the substation. The electrical parameter mutation analysis module is used to collect electrical parameters of the target substation within an abnormal time window. It constructs a mutation degree index by performing first-order and second-order difference calculations on the electrical parameters of the target substation within the abnormal time window, so as to detect the set of abnormal mutation time points of electrical parameters. The potential fault timing determination module is used to acquire thermal images of the surfaces of relevant equipment in the target substation, identify temperature anomaly areas in the target substation, and construct an abnormal temperature function by statistically analyzing the characteristic indicators of the temperature anomaly areas to detect the set of abnormal temperature change time points in the target substation. Based on the set of abnormal change time points of electrical parameters and the set of abnormal change time points of substation temperature, the potential timing of the fault occurrence is determined.

10. A substation monitoring device, characterized in that, The monitoring device uses the substation monitoring method described in claim 1 for monitoring.

Images