Power equipment operation safety state prediction method and system based on monitoring data

Abstract

The invention relates to the technical field of data processing, and discloses a power equipment operation safety state prediction method and system based on monitoring data, and the method comprises the steps: obtaining two continuous monitoring periods before a current moment, taking the two continuous monitoring periods as a first monitoring period and a second monitoring period, obtaining various state data, and obtaining a to-be-predicted time period; obtaining a first data sequence, obtaining a first fluctuation trend index, and obtaining a first correlation coefficient; obtaining a second data sequence, obtaining a second fluctuation trend index, and obtaining a second correlation coefficient between any two kinds of state data; and screening out two corresponding state data of which the difference value between the first correlation coefficient and the second correlation coefficient is smaller than a preset change threshold, taking the two state data as to-be-predicted data, and obtaining data change of the to-be-predicted data in the prediction time period according to a second data sequence of the to-be-predicted data. The method can improve the prediction accuracy, enhance the interpretability of the prediction result and improve the adaptability.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and specifically to a method and system for predicting the operational safety status of power equipment based on monitoring data. Background Technology

[0002] In the field of power equipment operation and maintenance, it is common to predict the operational safety status data of power equipment in order to understand the operating status of power equipment in advance.

[0003] Currently, power equipment status data is mainly used to predict the operational safety status of power equipment. However, some technologies rely solely on the temporal changes of a single status data point, neglecting the physical correlations and co-evolutionary patterns between multiple data sources. This fragmented prediction model contradicts the actual operating mechanisms of the equipment. For example, a sudden rise in ambient temperature can cause a short-term increase in the equipment's casing temperature, which may be misjudged as an internal fault (when actual current and vibration data show no abnormalities). Furthermore, some existing technologies lack a quantitative mechanism for the correlation stability between multiple data sources, making it impossible to distinguish between spurious fluctuations caused by random disturbances and genuine fault precursors. Simultaneously, random fluctuations in the detected status data within a monitoring period (such as sudden load changes or transient electromagnetic interference) can lead to spurious correlations between data. For instance, temperature and vibration data may show a high correlation in one monitoring period due to random disturbances, but this correlation may drastically decrease in the next period. If temperature and vibration are predicted in the near future, the predicted values ​​will deviate from the actual trajectories.

[0004] In summary, the current prediction accuracy of power equipment operation safety status data is poor and has many shortcomings. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for predicting the operational safety status of power equipment based on monitoring data.

[0006] According to one aspect of the present invention, a method for predicting the operational safety status of power equipment based on monitoring data is provided, comprising: S1: acquiring two consecutive periods prior to the current moment as a first monitoring period and a second monitoring period, acquiring various status data of the power equipment during operation within the monitoring period, and acquiring a time period to be predicted; S2: acquiring a first data sequence of various status data within the first monitoring period, acquiring a first fluctuation trend index corresponding to various status data based on the first data sequence of various status data, and acquiring a first correlation coefficient between any two types of status data based on the first fluctuation trend index of various status data; S3: acquiring a second data sequence of various status data within the second monitoring period, acquiring a second fluctuation trend index corresponding to various status data based on the second data sequence of various status data, and acquiring a second correlation coefficient between any two types of status data based on the second fluctuation trend index of various status data; S4: selecting two types of status data whose difference between the first correlation coefficient and the second correlation coefficient is less than a preset change threshold and using them as data to be predicted, and acquiring the data change of the data to be predicted within the prediction time period based on the second data sequence of the data to be predicted.

[0007] This scheme establishes a cross-cycle physical correlation stabilization mechanism. It acquires various state data from two consecutive monitoring cycles, then obtains the fluctuation trend indicators of corresponding state parameters within each cycle, as well as the relationship coefficient between any two parameters. Combinations of any two state parameters are filtered, and combinations with a cross-cycle correlation coefficient difference less than a preset threshold are selected. The stability of the correlation between state data and at least one other data point is used as a prediction parameter, improving the prediction confidence. Simultaneously, this correlation coefficient filtering effectively distinguishes between real pre-fault conditions and short-term noise fluctuations, effectively avoiding false correlations caused by random disturbances such as load changes, thus improving prediction accuracy. Finally, based on recent behavioral patterns, linear extrapolation prediction is used. For stable data, the direction and rate of change of the sequence at the end of the second monitoring cycle are directly combined with the current value to generate predicted values ​​for future time points.

[0008] Optionally, obtaining the first fluctuation trend index corresponding to various state data based on the first data sequence of various state data includes: sequentially obtaining the difference between two adjacent monitoring values ​​in the first data sequence of state data and using it as the first relative change; obtaining the theoretical maximum and theoretical minimum values ​​corresponding to the state data; and obtaining the first absolute change corresponding to the first relative change based on the first relative change, the theoretical maximum value, and the theoretical minimum value; and summing the absolute changes corresponding to each first absolute change in the first data sequence of state data and using it as the first fluctuation trend index. In this scheme, the first absolute change is the ratio of the first relative change to the difference between the theoretical maximum value and the theoretical minimum value.

[0009] Optionally, the first correlation coefficient between any two sets of state data can be obtained based on the first fluctuation trend index of various state data, and is expressed as follows: ;in, Let be the first correlation coefficient between the i-th and j-th state data. This is the first fluctuation trend indicator for the i-th state data. This is the first fluctuation trend indicator for the j-th state data. This is the scaling factor. When the first fluctuation trend indicators of two state data are highly consistent in both value and direction, such as when the temperature of A increases and the current of B increases synchronously, it indicates that there is a strong physical coupling relationship between the two. At this time, the correlation coefficient approaches its maximum value. Conversely, the correlation coefficient decreases significantly, or even returns to zero.

[0010] Optionally, obtaining the second fluctuation trend index corresponding to various state data based on the second data sequence of various state data includes: sequentially obtaining the difference between two adjacent monitoring values ​​in the second data sequence of state data and using it as the second relative change, obtaining the theoretical maximum and theoretical minimum values ​​corresponding to the state data, and obtaining the second absolute change corresponding to the second relative change based on the second relative change, the theoretical maximum value, and the theoretical minimum value; and accumulating the absolute changes corresponding to each second absolute change in the second data sequence of state data and using it as the second fluctuation trend index. In this scheme, the second absolute change is the ratio of the second relative change to the difference between the theoretical maximum value and the theoretical minimum value.

[0011] Optionally, the second correlation coefficient between any two sets of state data can be obtained based on the second fluctuation trend index of various state data, and is expressed as follows: ;in, The second correlation coefficient is the relationship between the i-th and j-th state data. This is the second fluctuation trend indicator for the i-th state data. This is the second fluctuation trend indicator for the j-th state data. This is the scaling factor. When the second fluctuation trend indicators of the two state data are highly consistent in both value and direction, such as when the temperature of A increases and the current of B increases synchronously, it indicates that there is a strong physical coupling relationship between the two. At this time, the correlation coefficient approaches its maximum value. Conversely, the correlation coefficient decreases significantly or even returns to zero.

[0012] Optionally, obtaining the data changes of the data to be predicted within the prediction period based on the second data sequence of the data to be predicted includes: obtaining the rate of change of the data to be predicted within the prediction period based on the second data sequence of the data to be predicted; and generating predicted values ​​of the data to be predicted at multiple time points within the prediction period based on the monitored value of the data to be predicted at the current moment and the rate of change within the prediction period.

[0013] According to another aspect of the present invention, a power equipment operation safety status prediction system based on monitoring data is provided. The system includes: an acquisition module, configured to acquire two consecutive monitoring periods for the power equipment before the current moment, which are designated as a first monitoring period and a second monitoring period, and to acquire various status data during the operation of the power equipment, and to acquire a time period to be predicted; a first data processing module, configured to acquire a first data sequence of various status data within the first monitoring period, and to acquire a first fluctuation trend index corresponding to various status data based on the first data sequence of various status data, and to acquire a first correlation coefficient between any two types of status data based on the first fluctuation trend index of various status data; a second data processing module, configured to acquire a second data sequence of various status data within the second monitoring period, and to acquire a second fluctuation trend index corresponding to various status data based on the second data sequence of various status data, and to acquire a second correlation coefficient between any two types of status data based on the second fluctuation trend index of various status data; and a data prediction module, configured to filter out two types of status data whose difference between the first correlation coefficient and the second correlation coefficient is less than a preset change threshold, and to acquire the data change of the data to be predicted within the prediction time period based on the second data sequence of the data to be predicted.

[0014] Optionally, the first data processing module is further configured to: sequentially obtain the difference between two adjacent monitoring values ​​in the first data sequence based on the first data sequence of the state data and use it as the first relative change, obtain the theoretical maximum and theoretical minimum values ​​corresponding to the state data, and obtain the first absolute change corresponding to the first relative change based on the first relative change, the theoretical maximum and the theoretical minimum values; and accumulate the absolute changes corresponding to each first absolute change in the first data sequence of the state data and use it as the first fluctuation trend indicator.

[0015] According to another aspect of the present invention, a power equipment operation safety status prediction device based on monitoring data is provided, characterized in that it includes: a memory storing a computer program thereon; and a processor for executing the computer program in the memory to implement the above-described power equipment operation safety status prediction method based on monitoring data.

[0016] According to another aspect of the present invention, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the above-described method for predicting the safe operating status of power equipment based on monitoring data.

[0017] The beneficial effects of this invention are: 1. To address the shortcomings of existing technologies that neglect the collaborative evolution of multi-source data, a cross-cycle physical correlation stability mechanism is established. This mechanism proposes to screen stable data combinations by comparing correlation coefficients across two monitoring cycles. By utilizing the fluctuation trend indicators of the first monitoring cycle (medium-term history) and the second monitoring cycle (recent dynamics), the correlation strength between any two states of data is quantified (the stronger the trend and the closer the amplitude, the higher the correlation coefficient). Combinations with a cross-cycle correlation coefficient difference less than a preset threshold are then selected. This mechanism ensures that the data included in the prediction has long-term stability supported by physical mechanisms, effectively avoids false correlations caused by random disturbances such as load changes, and fundamentally distinguishes between real fault precursors and noise fluctuations, resulting in more accurate predictions.

[0018] 2. Design hierarchical data prediction admission rules to avoid misjudgment in single data prediction. If a state data has a stable correlation with at least one other data, it is listed as a subject to be predicted. When a state data satisfies stability with multiple data at the same time, its prediction confidence is further improved, while state data that is completely disconnected is completely excluded. This rule retains the simplicity of independent prediction of single data, while strengthening data reliability through cross-validation of correlation networks, and avoids interference such as sudden rise in environmental temperature being misjudged as internal failure.

[0019] Linear extrapolation prediction based on recent behavioral patterns, for stable data, directly uses the direction and rate of change of the sequence at the end of the second monitoring period, combined with the current value to generate predicted values ​​for future time points, which can improve the accuracy of prediction results, enhance the interpretability of prediction results, and improve adaptability. Attached Figure Description

[0020] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0021] Figure 1 This is a partial flowchart illustrating an embodiment of the power equipment operation safety status prediction method based on monitoring data provided by the present invention. Figure 2 This is another part of the flowchart illustrating an embodiment of the power equipment operation safety status prediction method based on monitoring data provided by the present invention; Figure 3 A schematic diagram illustrating the steps of an embodiment of the power equipment operation safety status prediction method based on monitoring data provided by the present invention; Figure 4 This is a schematic diagram of an embodiment of the power equipment operation safety status prediction device based on monitoring data provided by the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0024] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0025] Figure 1 , Figure 2 and Figure 3 The diagram illustrates a flowchart and steps of an embodiment of the power equipment operation safety status prediction method based on monitoring data according to the present invention. This method is executed by a power equipment operation safety status prediction system based on monitoring data. Figures 1 to 3 As shown, the method includes the following steps: S1: Obtain two consecutive monitoring cycles for the power equipment before the current time and use them as the first and second monitoring cycles, and obtain various status data of the power equipment during operation, and obtain the time period to be predicted; S2: Obtain the first data sequence of various state data within the first monitoring period, and obtain the first fluctuation trend index corresponding to various state data based on the first data sequence of various state data, and obtain the first correlation coefficient between any two state data based on the first fluctuation trend index of various state data.

[0026] This step specifically includes: S21: Based on the first data sequence of the state data, the difference between two adjacent monitoring values ​​in the first data sequence is obtained sequentially and used as the first relative change. The theoretical maximum value and theoretical minimum value corresponding to the state data are obtained, and the first absolute change corresponding to the first relative change is obtained based on the first relative change, the theoretical maximum value and the theoretical minimum value.

[0027] This step involves constructing a normalized index reflecting the micro-fluctuations of the data. For the time series of data for each state within the first monitoring period, such as the temperature data of the switchgear, the difference between two adjacent monitoring points is calculated sequentially, i.e., the value at the later time minus the value at the previous time. The sign must be retained to reflect the direction of change, generating a signed relative change sequence. For example, the change between two consecutive points in the temperature sequence: +0.3℃ and -0.2℃. Subsequently, combined with the theoretical extreme range of this state under the current operating environment, such as the safe temperature threshold of the switchgear casing of 10~85℃, the relative change is normalized, i.e., (first relative change) / (theoretical maximum value - theoretical minimum value) = first absolute change. Using the difference between the theoretical maximum and minimum values ​​as a benchmark, each relative change is converted into a unitless absolute change. For example, for the change of +0.3℃, the normalization formula is 0.3 / (85-10) = 0.004. This operation eliminates the incomparability between data of different dimensions (such as temperature ℃ and current A), while preserving the original fluctuation direction (positive value increases / negative value decreases) and relative amplitude, forming a standardized sequence of absolute changes.

[0028] S22: Accumulate the absolute changes corresponding to each first absolute change in the first data sequence of state data and use it as the first fluctuation trend indicator.

[0029] In this step, based on the absolute change sequence generated in S21, a comprehensive trend index is constructed by accumulating the absolute change values ​​(including positive and negative signs) at each time point. For example, the absolute change sequence of switchgear temperature in a certain monitoring period is [+0.004, -0.003, +0.005], and the accumulated result is +0.006, which is the first fluctuation trend index. This index quantifies the net change direction and intensity of the state data throughout the entire monitoring period: if the accumulated value is significantly positive (e.g., +0.15), it indicates that the data shows an overall upward trend with a large fluctuation range; if the accumulated value is significantly negative (e.g., -0.12), it reflects an overall downward trend; values ​​close to zero (e.g., +0.002) indicate weak fluctuations within the period with no significant directionality. Through this index, the multidimensional state characteristics of the equipment are condensed into a trend scalar that can be compared horizontally, providing a unified input for subsequent cross-data correlation analysis.

[0030] Furthermore, let's take transformer winding temperature data as an example. In S21, the relative changes in the time series within the monitoring period are obtained by differentiating adjacent points [-1.5℃, +2.2℃, -0.8℃]; the theoretical temperature range is 0~120℃, and the normalized reference is 120-0=120℃; the absolute changes are calculated as: [-1.5 / 120≈-0.0125, +2.2 / 120≈+0.0183, -0.8 / 120≈-0.0067]. In S22, the cumulative absolute changes are: -0.0125+0.0183-0.0067=-0.0009, and the first fluctuation trend index is -0.0009, reflecting that the winding temperature fluctuates very little and has no significant upward or downward trend within this period, which is consistent with the steady-state operation characteristics of the equipment.

[0031] S23: Obtain the first correlation coefficient between any two sets of state data based on the first fluctuation trend index of various state data.

[0032] The first correlation coefficient in this step is represented as: ;in, Let be the first correlation coefficient between the i-th and j-th state data. This is the first fluctuation trend indicator for the i-th state data. This is the first fluctuation trend indicator for the j-th state data. This is the scaling factor.

[0033] In this embodiment, it should be noted that throughout the entire expression, As a directional term, when the fluctuation trend index of two state data (such as temperature T and current C) is... , Output 1 when the signs are the same (both rising / falling), and output -1 when the signs are opposite (one rising and one falling); output 0 if either indicator is zero. The correlation must conform to the directional coordination of the physical mechanism; for example, if the ambient temperature rises suddenly, and the equipment casing temperature rises (…). However, the internal current is stable. ),but =0 sets the correlation coefficient to zero, directly avoiding the problem of misjudging internal operational failures by a single data point caused by non-operational anomalies or environmental factors. At the same time, it ensures that a correlation is established only when the two data points evolve in the same direction within the monitoring period (such as when the fan current and temperature increase synchronously when the cooling system is working), strictly following the physical laws of equipment operation.

[0034] Furthermore, To obtain the associated items, calculate the absolute difference between the fluctuation trend indicators of the two states. And multiplied by Then, a non-linear decay is achieved through an exponential function. If the two data trends are significantly different (e.g., a sharp increase in temperature I=0.3, a slight increase in current I=0.05), then... hour This results in e^{-1.25}≈0.28, relatively lowering the correlation coefficient and reflecting a lack of strong physical coupling between the two. When random operating conditions such as instantaneous electromagnetic interference cause the trend values ​​of the two data to accidentally approach each other ( Although e^{0}=1 will output a high correlation, subsequent cross-period stability verification of S4 (e.g., the TV combination correlation coefficient drops sharply from 0.9 to 0.2) can filter out such spurious correlations. Meanwhile, the sensitivity of the exponential function (small...) (Changes cause drastic changes in the correlation coefficient) Ensure that a high correlation value is output only when the fluctuation trends are highly matched (e.g., when the first fluctuation trend index of temperature is 0.15 and the first fluctuation trend index of current is 0.18). , When e^{-0.15}≈0.86), the physical proportional relationship between current load and temperature rise is captured.

[0035] Furthermore, The strong filtering mechanism truncates the product of the sign function and the exponential term to the non-negative range. If the result is negative, it forces an output of 0. This completely eliminates spurious correlations caused by trend divergence. For example, when equipment experiences abnormal operating conditions (such as mechanical jamming leading to temperature rise but vibration decrease), even... Extremely small, because This makes the overall result negative. By resetting it to zero, this mechanism addresses the pain point of spurious correlations caused by random disturbances to a certain extent. It directly filters out abnormal combinations through physical law inversion verification (temperature rise is often accompanied by increased vibration in real faults), thus avoiding missed fault detection.

[0036] Furthermore, scaling factor Dynamic adaptability, As an adjustable parameter to control the decay rate, the greater the thermal inertia of the equipment (such as a transformer), the... The smaller (e.g.) Sensitive devices (such as surge arresters) The larger (e.g.) This overcomes the generalization limitations of traditional models in diverse equipment scenarios. For example, transformer oil temperature and current naturally exhibit similar trends and amplitudes due to their large heat capacity. value( The tolerance can be relaxed; however, in GIS equipment, pressure and current need to be strictly matched in proportion, which is large. value( 0) Enhanced amplitude sensitivity. By adapting to device characteristics, it ensures the accurate representation of the physical mechanisms of different power devices, solving the problem of one-size-fits-all correlation models failing in complex scenarios.

[0037] S3: Obtain the second data sequence of various state data within the second monitoring period, and obtain the second fluctuation trend index corresponding to various state data based on the second data sequence of various state data, and obtain the second correlation coefficient between any two state data based on the second fluctuation trend index of various state data.

[0038] This step specifically includes: S31: Based on the second data sequence of the state data, the difference between two adjacent monitoring values ​​in the second data sequence is obtained sequentially and used as the second relative change. The theoretical maximum and theoretical minimum values ​​corresponding to the state data are obtained, and the second absolute change corresponding to the second relative change is obtained based on the second relative change, the theoretical maximum and the theoretical minimum values.

[0039] This step focuses on quantifying the micro-fluctuations of state data within the second monitoring cycle (the most recent cycle). For each time series of state data, such as switchgear current data, the signed difference between adjacent monitoring points is calculated sequentially, i.e., the value at the later moment minus the value at the previous moment. The sign must be retained to indicate the direction of change, generating a second relative change sequence, i.e., (second relative change) / (theoretical maximum value - theoretical minimum value) = second absolute change; for example, current series change: +0.5A, -0.3A. Subsequently, combined with the theoretical extreme range of this state under the current operating environment, such as the maximum allowable current of the switchgear being 100A, the relative change is normalized using the difference between the theoretical maximum and minimum values ​​(e.g., 100A - 0A = 100A) (e.g., +0.5A is normalized to +0.005). This operation eliminates dimensional differences while strictly preserving the original fluctuation direction (positive values ​​indicate an upward trend, negative values ​​indicate a downward trend) and relative amplitude, forming a standardized second absolute change sequence that can be compared across data types.

[0040] S32: Accumulate the absolute changes corresponding to each second absolute change in the second data sequence of the state data and use it as the second fluctuation trend indicator.

[0041] In this step, based on the absolute change sequence generated in S31, the signed absolute changes at each time point are summed (e.g., the sequence [+0.004, -0.002, +0.006] sums to +0.008) to generate a second volatility trend indicator. This indicator comprehensively reflects the overall intensity and direction of the data's changes in the most recent period: a significantly positive value (e.g., +0.25) indicates a sharp rise in data; a significantly negative value (e.g., -0.18) indicates a continuous decline; and values ​​approaching zero (e.g., +0.003) reflect stability without fluctuation. This indicator provides standardized input for recent dynamics in subsequent correlation analysis.

[0042] For example, taking transformer current data in the second monitoring cycle as an example. In S31, the relative changes in the time series difference are [-2.1A, +1.8A, -0.5A]; the theoretical current range is 0-500A, with a normalized reference of 500A; the absolute changes are calculated as: [-2.1 / 500=-0.0042, +1.8 / 500=+0.0036, -0.5 / 500=-0.001]. In S32, the cumulative absolute changes are: -0.0042+0.0036-0.001=-0.0016, and the second fluctuation trend index is -0.0016, indicating a slight overall decrease in current, consistent with the recent light-load operation of the equipment.

[0043] S33: Obtain the second correlation coefficient between any two state data based on the second fluctuation trend index of various state data.

[0044] The second correlation coefficient in this step is expressed as: ;in, The second correlation coefficient is the relationship between the i-th and j-th state data. This is the second fluctuation trend indicator for the i-th state data. This is the second fluctuation trend indicator for the j-th state data. This is the scaling factor.

[0045] In this embodiment, it should be noted that the calculation logic is the same as that of S2. Specifically, firstly, and A dynamic verification mechanism was implemented, mandating that the fluctuation trends of the two state data evolve in the same direction within the current cycle (second monitoring cycle), ensuring that it only responds to immediate physical coordination changes (such as the synchronous rise of current and temperature when the cooling system starts). Furthermore, Similarly, for items retrieved through correlation, the difference in fluctuation trends... It decays exponentially. The acquisition of values ​​needs to be achieved through a multi-level adaptation mechanism driven by device characteristics, and its core logic is directly linked to the physical characteristics of power equipment and the level of data noise. In the specific process of obtaining the value, a baseline setting is first made based on the physical response characteristics of the equipment. For equipment with high thermal inertia (such as transformer oil temperature), due to the slow change in state and high tolerance for fluctuation trend index differences, a small value needs to be set. Value (e.g.) Given [2,5]), to prevent oversensitivity from causing missed correlations; for highly responsive devices (such as GIS agency current), because the data is easily affected by instantaneous disturbances, a large [missing information] setting is required. Value (e.g.) Given [8,15]), the amplitude matching requirement is strengthened. Furthermore, The value can be corrected using equipment nameplate parameters (such as rated current and heat capacity). First, calculate the theoretical fluctuation correlation strength S: S = {theoretical change threshold of state variable A} / {theoretical change threshold of state variable B}. If S > 1 (e.g., the temperature change threshold is much greater than the vibration threshold), then... Reduce proportionally, for example, S>1. Reduced by 10%.

[0046] Furthermore, the second correlation coefficient output by the expression in this embodiment... The first correlation coefficient in the aforementioned embodiments In S4, cross-period stability comparisons are performed to form the final prediction scheme: First, false correlations are screened out empirical evidence, where a lightning strike causes a strong correlation between temperature and vibration in the second period. However, there was no such correlation in the first cycle. If the preset threshold is 0.2, it is determined to be a transient interference combination and the prediction is excluded; secondly, physical correlation stability capture is realized. In the early stage of cooling pump aging, the correlation between current and temperature is 0.82 in period 1 and 0.79 in period 2. The difference is 0.03 < 0.2, which triggers prediction. The short-term data correlation (S3 expression output) is placed under the long-term stability verification (S4) framework to solve the defect of false correlation caused by random operating condition fluctuations within the monitoring period.

[0047] S4: Select the two types of state data where the difference between the first correlation coefficient and the second correlation coefficient is less than the preset change threshold and use them as the data to be predicted. Then, obtain the data changes of the data to be predicted within the prediction period based on the second data sequence of the data to be predicted.

[0048] This step specifically includes: S41: Obtain the rate of change of the data to be predicted within the prediction time period based on the second data sequence of the data to be predicted; S42: Generate predicted values ​​for the data to be predicted at multiple time points within the prediction period based on the monitored value of the data to be predicted at the current moment and the rate of change within the prediction period.

[0049] In this embodiment, it should be noted that the core of S41 lies in extrapolating future trends based on the latest data behavior. For each data point to be predicted selected in S4, such as the switch cabinet temperature T, its complete time series in the second monitoring period (the most recent period) is extracted, i.e., the second data series. By analyzing the changing patterns of continuous monitoring points at the end of the series, such as the rise and fall patterns of the last few monitoring values, existing algorithms are used to extrapolate the rate of change of the data within the prediction period. Specifically: if the end of the series shows a continuous upward trend, such as an increase in three consecutive monitoring points, the rate of change is determined to be positive; if it continues to decrease, the rate is negative; if the fluctuation is stable, the rate approaches zero. This rate quantifies the evolutionary inertia of the data in the current operating state, while avoiding the introduction of noise into complex models—because S4 has already selected data with stable physical mechanisms, its short-term evolution patterns have high reliability.

[0050] In S42, linear extrapolation-based multi-time-point prediction is implemented. Driven by the rate of change calculated in S41, it overlays real-time monitoring values, such as the measured temperature T at 14:00, to generate predicted values ​​for multiple time points within the predicted time period at equal time intervals. The simplified calculation formula is: Future time-point predicted value = Current monitoring value + Rate of change × Time span.

[0051] In this embodiment, it should be noted that the main purpose of S1 is to establish a time frame and basic data source for subsequent data analysis. Specifically, when implementing prediction, it is first necessary to obtain historical data time series from the monitoring of power equipment (such as sensor networks): select two consecutive monitoring periods before the current moment, such as periodic data based on fixed sampling intervals. Both monitoring periods are close to the current moment, with the earlier period defined as the first monitoring period, i.e., the second period closest to the current moment, representing the prior historical trend of the equipment status; and the later period defined as the second monitoring period, i.e., the period closest to the current moment, representing the subsequent historical trend of the equipment status. The length of these monitoring periods is set according to the equipment characteristics, such as minute-level or hour-level windows. At the same time, S1 will acquire various status data of the equipment during operation, such as the equipment casing temperature, internal current, or vibration amplitude. These data originate from the equipment's operating mechanism, ensuring comprehensive prediction coverage. Finally, S1 determines the time period to be predicted, such as the next few minutes or hours. It defines the prediction target range, enabling subsequent steps to output predicted values ​​of status changes at specific time points, thereby supporting operation and maintenance decisions.

[0052] Furthermore, to illustrate with an example, suppose we assume a switchgear monitoring scenario in a 110kV substation, where the maintenance team sets the monitoring cycle to 2 hours. The first monitoring cycle contains data from 4 to 2 hours prior to the current moment, such as temperature, vibration, and current sequences, reflecting the equipment's operating status under the previous 4 to 2 hours' conditions. The second monitoring cycle contains data from 2 hours prior to the current moment, such as temperature sequences from yesterday to this morning, capturing changes in the equipment under the latest conditions. The forecast period is set to the next hour, indicating that maintenance personnel need to be alerted to potential equipment problems that evening. In this case, S1 automatically extracts these periodic data and status types, providing structured input for subsequent correlation and stability analysis.

[0053] In S2, in-depth analysis of multi-source state data is performed within the first monitoring cycle (i.e., the second closest historical cycle to the current moment). The core objective is to reveal the physical correlation between data by quantifying the fluctuation trends of individual data. Specifically, for each state data, such as temperature, current, and vibration, its time series within that cycle is extracted, i.e., the first data series. The relative change series is obtained by calculating the difference between adjacent monitoring values ​​point by point (preserving the positive and negative signs of the change direction). Subsequently, combined with the theoretical extreme value range of the state under the current operating environment, such as the temperature safety threshold in the equipment design specifications, the relative change is normalized into an absolute change that can be compared across data (eliminating the influence of different dimensions while retaining trend direction information). Finally, the absolute changes are summed to obtain the "first fluctuation trend index." The first fluctuation trend index captures both the overall amplitude and direction of the data change; positive values ​​indicate an upward trend, and negative values ​​indicate a downward trend. Based on this, by obtaining the first correlation coefficient between any two states of data: when the first fluctuation trend indicators of the two states are highly consistent in value and direction, such as when the temperature of A rises and the current of B increases synchronously, it indicates that there is a strong coupling relationship in physical mechanism, and the correlation coefficient approaches the maximum value; otherwise, the correlation coefficient will decrease significantly or even return to zero.

[0054] Furthermore, taking transformer monitoring as an example, oil temperature and cooling fan current data are analyzed during the first monitoring cycle (12:00-14:00). In S21, the oil temperature sequence is differentially analyzed point by point to obtain relative changes, such as +3℃, -2℃, etc., which are then normalized to an absolute change sequence based on the theoretical range of 10 to 85. , The cumulative oil temperature fluctuation trend index is 0.013, showing a slight overall increase. Simultaneously, fan current data is processed: relative change sequences, such as +0.1A and -0.05A; normalized to an absolute change sequence within the theoretical range of 0 to 10A, such as +0.01 and -0.005; the cumulative fluctuation trend index is +0.015, showing a simultaneous slight increase. Since both trends are in the same direction and their amplitudes are similar, assuming the first correlation coefficient calculated in step S22 is 0.89, representing a strong correlation and reflecting the normal response mechanism of the cooling system to temperature rise; while the fluctuation trend index of the winding vibration data during the same period is -0.32, showing a downward trend, and its correlation coefficient with oil temperature is only 0.03, indicating no significant correlation between the two.

[0055] In S3, the focus is on the analysis of state data within the second monitoring cycle (i.e., the cycle closest to the current moment). Its logical framework is completely symmetrical to S2, but its target is different: by real-time evaluation of the fluctuation trend and correlation of the latest data, the latest dynamic coordination law of equipment status is captured. Specifically: First, the time series sequence (second data sequence) of each state data within this cycle is extracted. The second relative change sequence is obtained by calculating the signed difference between adjacent monitoring values ​​point by point (positive values ​​indicate an increase, and negative values ​​indicate a decrease). Then, based on the theoretical extreme value range of the state under the current operating conditions (such as the maximum allowable vibration amplitude or current value of the equipment), the relative change is normalized into a standardized second absolute change (eliminating dimensional differences and retaining the trend direction). Finally, all absolute changes in the sequence are summed to obtain the second fluctuation trend index—this index simultaneously quantifies the overall change intensity and direction of the data in the latest cycle. Based on this, the second correlation coefficient between any two state data is calculated: if the fluctuation trends of the two are highly synchronized in the current cycle (e.g., when the temperature of A rises rapidly, the current of B increases synchronously), the correlation coefficient approaches the upper limit value significantly, reflecting that there is a strong physical coupling between the two; if the trends diverge or the amplitudes differ significantly (e.g., the temperature rises but the vibration remains unchanged), the correlation coefficient decreases sharply or even returns to zero, revealing the risk of correlation failure.

[0056] In S4, the core decision-making stage of the technical solution is to verify the cross-cycle correlation stability of multi-source state data and select reliable data combinations with physical mechanism support for final prediction. The specific operation consists of two key steps: First, compare the correlation coefficient of each data combination in the first monitoring period (medium-term history) and the second monitoring period (recent dynamics) (e.g., the coefficient of temperature-current in period 1 is 0.85, and in period 2 it is 0.82). If the difference is less than a preset threshold (e.g., the threshold is set to 0.1), the correlation of the data combination is determined to be stable across time (not a spurious correlation caused by random disturbances), and it is listed as data to be predicted; conversely, if the difference exceeds the threshold (e.g., the vibration-temperature coefficient is 0.9 in a certain period and drops sharply to 0.2 in the next period), it indicates that the correlation is caused by transient disturbances, and the two are not included in the prediction range for the time being (without affecting the judgment of other data combinations). Subsequently, for the selected stable data combination, based on its time series sequence (i.e., the second data sequence) in the latest cycle (the second monitoring cycle), the rate of data change is deduced by analyzing the numerical change characteristics at the end of the sequence (such as recent continuous rise or oscillating convergence). Then, combined with the real-time monitoring value at the current moment, the predicted values ​​for multiple time points within the predicted time period are generated by linear extrapolation.

[0057] It should also be noted that in S4, the data to be predicted is screened through multi-level correlation stability verification. The mechanism includes three progressive layers of logic: First, if the difference in the correlation coefficient between a certain state data (such as temperature T) and any other state data (such as current C) in the first monitoring period and the second monitoring period is less than a preset threshold (such as 0.1), then T and C are listed as data to be predicted. This mechanism ensures that as long as there is at least one stable physical correlation, the data combination is considered reliable. Second, when the difference in the correlation coefficient between a certain data (such as current C) and multiple data (such as temperature T and voltage U) is less than the threshold at the same time, it means that the data maintains a cooperative evolution law in multiple physical dimensions (such as the thermal effect correlation between C and T, and the power balance correlation between C and U). Its cross-cycle behavior is more consistent, and the prediction confidence is significantly improved. Third, a complete exclusion mechanism: if the difference in the correlation coefficient between a certain state data (such as vibration V) and all other data exceeds the threshold (such as 0.1), it indicates that its fluctuation is completely out of the framework of the physical mechanism of the equipment (such as random electromagnetic interference). This state data is directly excluded as a prediction object to avoid noise interference.

[0058] Furthermore, continuing with the switchgear monitoring scenario (three state data: temperature T, current C, and vibration V), the preset correlation stability threshold is set to 0.1. The temperature T-current C combination has a first correlation coefficient of 0.18 in period 1 and a second correlation coefficient of 0.15 in period 2 (difference 0.03 < 0.1), and is therefore judged as a stable correlation combination. At this point, the sequences of T and C in the second period (the most recent 2 hours) are extracted: the T sequence shows an upward trend of x units / hour at the end, and the C sequence shows a downward trend of y units / hour at the end. Based on the current T and C values, the system predicts the T and C values ​​every 10 minutes in the next hour according to their respective rates of change. The temperature T-vibration V combination has a first correlation coefficient of 0.75 in period 1, but the second correlation coefficient in period 2 drops sharply to 0.15 (difference 0.6 > 0.1), and is judged as a false correlation. Although vibration data is monitored, it is excluded from the prediction range due to unstable correlation, avoiding misjudgments caused by a single instantaneous interference (such as vibration from switch operation). This mechanism accurately distinguishes between real fault precursors and noise fluctuations, improving prediction reliability.

[0059] In summary, the above-described method for predicting the operational safety status of power equipment based on monitoring data first establishes a cross-period physical correlation stability mechanism. Addressing the shortcomings of existing technologies that neglect the collaborative evolution of multi-source data, this method proposes screening stable correlated data combinations by comparing correlation coefficients across two monitoring periods. This involves quantifying the correlation strength between any two sets of state data using fluctuation trend indicators from the first monitoring period (medium-term history) and the second monitoring period (recent dynamics) (the stronger the trend alignment and the closer the amplitude, the higher the correlation coefficient). Furthermore, combinations with a cross-period correlation coefficient difference less than a preset threshold are selected. This mechanism ensures that the data included in the prediction has a physical mechanism supporting it. To ensure long-term stability, the system effectively avoids false correlations caused by random disturbances such as sudden load changes, fundamentally distinguishing between true fault precursors and noise fluctuations. Furthermore, a hierarchical data prediction admission rule is designed to avoid misjudgments from single-data predictions. If a state data point has a stable correlation with at least one other data point, it is listed as a target for prediction. When a state data point simultaneously satisfies stability with multiple data points, its prediction confidence is further improved, while completely disconnected state data is completely excluded. This rule retains the simplicity of independent single-data prediction while strengthening data reliability through cross-validation of correlation networks, preventing interference such as sudden increases in ambient temperature from being misjudged as internal faults. Finally, linear extrapolation prediction based on recent behavioral patterns is used. For stable data, the prediction value for future time points is generated directly based on the direction and rate of change of the sequence at the end of the second monitoring period, combined with the current value. In summary, the entire technical solution tightly couples the physical correlation stability verification and behavioral linear extrapolation processes, achieving a reduction in fault misjudgment rate, enhanced interpretability of prediction results, and improved adaptability to both short-term and long-term operating conditions.

[0060] This solution also provides a system for implementing the above methods, including an acquisition module, a first data processing module, a second data processing module, and a data prediction module. The acquisition module is used to acquire two consecutive monitoring cycles for the power equipment before the current time and use them as the first and second monitoring cycles, acquire various status data of the power equipment during operation, and acquire the time period to be predicted. The first data processing module is used to acquire the first data sequence of various state data within the first monitoring period, acquire the first fluctuation trend index corresponding to various state data based on the first data sequence of various state data, and acquire the first correlation coefficient between any two state data based on the first fluctuation trend index of various state data. The second data processing module is used to acquire the second data sequence of various state data within the second monitoring period, and to acquire the second fluctuation trend index corresponding to various state data based on the second data sequence of various state data, and to acquire the second correlation coefficient between any two state data based on the second fluctuation trend index of various state data. The data prediction module is used to filter out two types of state data where the difference between the first correlation coefficient and the second correlation coefficient is less than a preset change threshold and use them as the data to be predicted. The module also obtains the data changes of the data to be predicted within the prediction period based on the second data sequence of the data to be predicted.

[0061] In one optional approach, the first data processing module is further configured to: sequentially obtain the difference between two adjacent monitoring values ​​in the first data sequence based on the first data sequence of the state data and use it as the first relative change, obtain the theoretical maximum value and theoretical minimum value corresponding to the state data, and obtain the first absolute change corresponding to the first relative change based on the first relative change, the theoretical maximum value and the theoretical minimum value; and accumulate the absolute changes corresponding to each first absolute change in the first data sequence of the state data and use it as the first fluctuation trend indicator.

[0062] In one optional embodiment, the second data processing module is further configured to: sequentially obtain the difference between two adjacent monitoring values ​​in the second data sequence of the state data and use it as the second relative change, obtain the theoretical maximum and theoretical minimum values ​​corresponding to the state data, and obtain the second absolute change corresponding to the second relative change based on the second relative change, the theoretical maximum and the theoretical minimum values; and accumulate the absolute changes corresponding to each second absolute change in the second data sequence of the state data and use it as the second fluctuation trend indicator.

[0063] The acquisition module acquires two consecutive monitoring periods prior to the current moment, designating them as the first and second monitoring periods, and obtains various state data and the time period to be predicted within these two periods. The first data processing module acquires the first data sequence of various state data within the first monitoring period, and obtains the first fluctuation trend index corresponding to each state data based on the first data sequence, and then obtains the first correlation coefficient between any two state data based on the first fluctuation trend index. The second data processing module acquires the second data sequence of various state data within the second monitoring period, and obtains the second fluctuation trend index corresponding to each state data based on the second data sequence, and then obtains the second correlation coefficient between any two state data based on the second fluctuation trend index. The data prediction module filters out two types of state data whose difference between the first and second correlation coefficients is less than a preset change threshold, and uses these as the data to be predicted, obtaining the data changes of the data to be predicted within the prediction time period based on the second data sequence of the data to be predicted. This scheme couples the physical correlation stability verification and behavioral linear extrapolation in a closed loop, achieving a reduction in fault misjudgment rate, enhanced interpretability of prediction results, and improved adaptability to long-term and short-term operating conditions.

[0064] In this embodiment, it should be noted that the specific method of performing the above-mentioned power equipment operation safety status prediction system based on monitoring data has been described in detail in the embodiments of the power equipment operation safety status prediction method based on monitoring data, and will not be elaborated here.

[0065] Figure 4 The diagram shows a structural schematic of an embodiment of the power equipment operation safety status prediction device based on monitoring data according to the present invention. The specific embodiments of the present invention do not limit the specific implementation of the power equipment operation safety status prediction device based on monitoring data.

[0066] like Figure 4 As shown, the power equipment operation safety status prediction device 700 based on monitoring data may include: a processor 701 and a memory 702. The electronic device 700 may also include one or more of the following: a multimedia component 703, an I / O interface 704 (input / output interface), and a communication component 705.

[0067] The processor 701 controls the overall operation of the electronic device 700 to complete all or part of the steps in the power equipment operation safety status prediction method based on monitoring data. The memory 702 stores various types of data to support the operation of the electronic device 700. This data may include, for example, instructions for any application or method operating on the electronic device 700, and application-related data such as contact data, sent and received messages, pictures, audio, video, etc. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. Multimedia component 703 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in memory 702 or transmitted via communication component 705. The audio component also includes at least one speaker for outputting audio signals. I / O interface 704 provides an interface between processor 701 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual or physical buttons. Communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IoT, eMTC, or other 5G technologies, or combinations thereof, is not limited here. Therefore, the corresponding communication component 705 may include: a Wi-Fi module, a Bluetooth module, an NFC module, etc.

[0068] In an exemplary embodiment, the electronic device 700 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method for predicting the safe operating status of power equipment based on monitoring data.

[0069] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided. When executed by a processor, these program instructions implement the steps of the power equipment operation safety status prediction method based on monitoring data described above. For example, the computer-readable storage medium may be the memory 702 including the program instructions described above, which may be executed by the processor 701 of the electronic device 700 to complete the power equipment operation safety status prediction method based on monitoring data described above.

[0070] In another exemplary embodiment, a computer program product is also provided, which includes a computer program executable by a programmable device, the computer program having a code portion for performing the above-described method for predicting the safe operating status of power equipment based on monitoring data when executed by the programmable device.

[0071] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0072] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0073] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for predicting the operational safety status of power equipment based on monitoring data, characterized in that, include: S1: Obtain two consecutive periods before the current time as the first monitoring period and the second monitoring period, and obtain various status data of power equipment operation within the monitoring period, and obtain the time period to be predicted; S2: Obtain the first data sequence of various state data within the first monitoring period, and obtain the first fluctuation trend index corresponding to various state data based on the first data sequence of various state data, and obtain the first correlation coefficient between any two state data based on the first fluctuation trend index of various state data. S3: Obtain the second data sequence of various state data within the second monitoring period, and obtain the second fluctuation trend index corresponding to various state data based on the second data sequence of various state data, and obtain the second correlation coefficient between any two state data based on the second fluctuation trend index of various state data. S4: Select the two types of state data where the difference between the first correlation coefficient and the second correlation coefficient is less than the preset change threshold and use them as the data to be predicted. Then, obtain the data changes of the data to be predicted within the prediction period based on the second data sequence of the data to be predicted.

2. The method for predicting the operational safety status of power equipment based on monitoring data according to claim 1, characterized in that, Specifically, obtaining the first fluctuation trend indicator corresponding to each state data based on the first data sequence of each state data includes: Based on the first data sequence of the state data, the difference between two adjacent monitoring values ​​in the first data sequence is obtained sequentially and used as the first relative change. The theoretical maximum and theoretical minimum values ​​corresponding to the state data are obtained, and the first absolute change corresponding to the first relative change is obtained based on the first relative change, the theoretical maximum and the theoretical minimum values. The absolute changes corresponding to each first absolute change in the first data sequence of each state data are accumulated and used as the first fluctuation trend indicator of the corresponding state data.

3. The method for predicting the operational safety status of power equipment based on monitoring data according to claim 2, characterized in that, The first correlation coefficient is expressed as: ;in, Let be the first correlation coefficient between the i-th and j-th state data. This is the first fluctuation trend indicator for the i-th state data. This is the first fluctuation trend indicator for the j-th state data. This is the scaling factor.

4. The method for predicting the operational safety status of power equipment based on monitoring data according to claim 1, characterized in that, Specifically, obtaining the second fluctuation trend indicator corresponding to each state data based on the second data sequence of each state data includes: Based on the second data sequence of the state data, the difference between two adjacent monitoring values ​​in the second data sequence is obtained sequentially and used as the second relative change. The theoretical maximum and theoretical minimum values ​​corresponding to the state data are obtained, and the second absolute change corresponding to the second relative change is obtained based on the second relative change, the theoretical maximum and the theoretical minimum values. The absolute changes corresponding to each second absolute change in the second data sequence of each state data are accumulated and used as the second fluctuation trend index of the corresponding state data.

5. The method for predicting the operational safety status of power equipment based on monitoring data according to claim 4, characterized in that, The second correlation coefficient is expressed as: ;in, The second correlation coefficient is the relationship between the i-th and j-th state data. This is the second fluctuation trend indicator for the i-th state data. This is the second fluctuation trend indicator for the j-th state data. This is the scaling factor.

6. The method for predicting the operational safety status of power equipment based on monitoring data according to claim 1, characterized in that, Specifically, obtaining the data changes of the data to be predicted within the prediction time period based on the second data sequence of the data to be predicted includes: The rate of change of the data to be predicted within the prediction period is obtained from the second data sequence of the data to be predicted. Based on the monitored value of the data to be predicted at the current moment and the rate of change within the prediction period, predicted values ​​of the data to be predicted are generated for multiple time points within the prediction period.

7. A power equipment operation safety status prediction system based on monitoring data, characterized in that, The system includes: The acquisition module is used to acquire two consecutive periods before the current time as the first monitoring period and the second monitoring period, and to acquire various status data of power equipment operation within the monitoring period, and to acquire the time period to be predicted; The first data processing module is used to acquire the first data sequence of various state data within the first monitoring period, acquire the first fluctuation trend index corresponding to various state data based on the first data sequence of various state data, and acquire the first correlation coefficient between any two state data based on the first fluctuation trend index of various state data. The second data processing module is used to acquire the second data sequence of various state data within the second monitoring period, and to acquire the second fluctuation trend index corresponding to various state data based on the second data sequence of various state data, and to acquire the second correlation coefficient between any two state data based on the second fluctuation trend index of various state data. The data prediction module is used to filter out two types of state data where the difference between the first correlation coefficient and the second correlation coefficient is less than a preset change threshold and use them as the data to be predicted. The module also obtains the data changes of the data to be predicted within the prediction period based on the second data sequence of the data to be predicted.

8. The power equipment operation safety status prediction system based on monitoring data according to claim 7, characterized in that, The first data processing module is used to sequentially obtain the difference between two adjacent monitoring values ​​in the first data sequence according to the first data sequence of the state data and use it as the first relative change amount, and obtain the theoretical maximum value and theoretical minimum value corresponding to the state data, and obtain the first absolute change amount corresponding to the first relative change amount according to the first relative change amount, the theoretical maximum value and the theoretical minimum value; The absolute changes corresponding to each first absolute change in the first data sequence of state data are summed and used as the first fluctuation trend indicator.

9. A power equipment operation safety status prediction device based on monitoring data, characterized in that, include: A memory on which computer programs are stored; A processor is configured to execute the computer program in the memory to implement the power equipment operation safety status prediction method based on monitoring data as described in any one of claims 1 to 6.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the power equipment operation safety status prediction method based on monitoring data as described in any one of claims 1 to 6.

Images