System for monitoring operation of multi-source data fusion

Abstract

The present application relates to the technical field of industrial measurement and control, in particular to a system operation monitoring system for multi-source data fusion, which comprises a jump behavior identification module, a confidence degree screening module, a behavior consistency extraction module, a power recovery anomaly detection module and an operation state monitoring output module.In the present application, by extracting the sudden trend in the current and voltage signals of the operating equipment, combining the response misalignment between the signals, the behavior change of the equipment can be accurately described and the potential abnormal segments can be identified, the signal segments are screened by means of multi-dimensional confidence degree factors, the low reliability data content is eliminated, and then by triggering the time interleaving characteristics between behaviors and the behavior type consistency judgment, the representative characteristic time segments are constructed, and then by combining the trend delay and fluctuation amplitude change of the power signal in the voltage recovery process, the abnormal behavior of the equipment power recovery process can be accurately marked.

Description

Technical Field

[0001] This invention relates to the field of industrial measurement and control technology, and in particular to a system operation monitoring system that integrates multi-source data. Background Technology

[0002] Industrial measurement and control technology refers to a class of engineering technology systems that continuously monitor and regulate the operation of key links in industrial production and operation to ensure the safety, reliability and efficiency of the system. It regulates production behavior through specific control logic and makes timely responses and adjustments based on the real-time situation on site.

[0003] Traditional station power system operation monitoring refers to methods used to observe and record the operation status of internal power supply systems such as substations and distribution rooms, thereby determining whether various power units within the station are in normal condition and whether there are abnormal power, energy consumption deviations, or operational deviations.

[0004] Traditional power system operation monitoring typically relies on basic current and voltage status records and manual judgment. This approach cannot effectively distinguish between behavioral misalignments and normal fluctuations in equipment during sudden switching or power disturbances. In complex load switching or voltage recovery processes, it is difficult to identify situations where equipment power has not recovered in a timely manner or where there is a response lag. Especially in scenarios with multiple devices operating collaboratively or high-frequency start-stop, a single data source is prone to misjudgment and omission of information, leading to delayed system status identification or untimely early warning response. For example, when multiple devices operate continuously in a short period of time, signal interference may be superimposed, masking the abnormal performance of a critical device, thereby affecting the overall operational stability of the power supply system and the efficiency of fault diagnosis. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a system operation monitoring system that integrates multi-source data.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a system operation monitoring system for multi-source data fusion includes:

[0007] The jump behavior recognition module acquires current sampling data and voltage on / off signals of the operating equipment in the station power system, identifies the jump trend behavior of each device, and obtains the behavior response misalignment data segment.

[0008] The confidence screening module calculates the direction score and amplitude deviation of the signal segment corresponding to the trigger point of the device's jump trend behavior based on the behavior response misalignment data segment, and filters the data source segments whose confidence needs to be lowered.

[0009] The behavior consistency extraction module removes data source segments whose confidence level needs to be lowered, constructs an interlaced time interval composed of adjacent trigger points based on the retained device jump trend behavior trigger points, analyzes the overlap of trigger points in the interval, and extracts behavior consistency feature segments.

[0010] The power recovery anomaly detection module extracts power sampling data of the station power system load equipment based on the interleaved time interval of the behavioral consistency feature segments, determines the power recovery delay during the voltage recovery process, and marks the power recovery anomaly segment.

[0011] The operation status monitoring output module generates an abnormal entry sequence in chronological order based on the power recovery abnormal segment, and outputs the operation status monitoring results of the station's power system.

[0012] As a further embodiment of the present invention, the behavioral response misalignment data segment includes trigger point time difference, adjacent trigger point misalignment identifier, and current mutation segment; the data source segment whose confidence needs to be lowered includes direction score value, amplitude deviation value, and low confidence label; the behavioral consistency feature segment includes interleaved time index, repeated trigger frequency, and behavioral event type; the power recovery anomaly segment includes power recovery start point, power trend start point, and power slope fluctuation; and the operating status monitoring result includes abnormal equipment identifier, abnormal data entry, and output time series.

[0013] As a further aspect of the present invention, the jump behavior recognition module includes:

[0014] The electrical parameter acquisition structured submodule acquires current sampling data and voltage on / off signals of operating equipment in the station power system. It calculates the rate of change sequence between continuous sampling points of the current sampling data, extracts abrupt change segments, and combines the edge transition time points in the voltage on / off signals to perform unified period normalization processing on the sampling information of the two types of signals to obtain the structured electrical parameter change sequence.

[0015] The behavior trajectory synchronization submodule calls the current change rate segment and voltage jump point sequence in the structured electrical parameter change sequence to synchronize the position of each device behavior trajectory, and pairs each change rate peak with the corresponding voltage change point to obtain the trajectory group corresponding to the device behavior.

[0016] The response offset recognition submodule determines whether the time offset threshold is exceeded based on the time difference information of each pair of signals in the trajectory group corresponding to the device behavior. It extracts the time interval, device number and signal alignment status information of the behavior segments that meet the conditions to obtain the behavior response misalignment data segment.

[0017] As a further aspect of the present invention, the confidence screening module includes:

[0018] The response segment extraction submodule extracts the signal change segment in the current change rate sequence and the direction change segment in the voltage on / off signal for each jump trend behavior trigger point based on the behavior response misalignment data segment, and obtains the jump signal aligned segment group.

[0019] The offset calculation submodule calls each current change rate segment and voltage direction segment in the jump signal alignment segment group to construct a time matching vector group, calculates the cosine of the angle between the time matching vector groups as the direction score, extracts the maximum value and the average value of the reference segment in each current change segment, calculates the ratio of the two as the amplitude deviation, and obtains the combination of direction score and amplitude deviation.

[0020] The confidence value determination submodule determines whether the score and deviation items in the combination of direction score and amplitude deviation exceed the preset offset score threshold and amplitude deviation threshold, filters the signal segments that meet the dual conditions, extracts the device number and time index of the corresponding segment, marks the data content whose confidence does not meet the requirements, and obtains the data source segments whose confidence needs to be lowered.

[0021] As a further aspect of the present invention, the behavior consistency extraction module includes:

[0022] The behavior point integration submodule removes the data source segments whose confidence level needs to be lowered, extracts the time index and device number information of the retained device jump trend behavior trigger points, and obtains the jump behavior interleaved sequence group;

[0023] The interval frequency aggregation submodule calls the alternating sequence group of the jump behavior, matches the event types defined in the behavior label set, counts the frequency of occurrence of the event type in each combination, and filters the segment group in the jump behavior set.

[0024] The coupling feature filtering submodule performs a conditional independence test on the relationship between the behavior frequency and the corresponding device on / off state based on the group of segments with the jump behavior concentration, calculates the chi-square value and filters the statistical correlation degree of the segments, and retains the behavior consistency feature fragments.

[0025] As a further aspect of the present invention, the power recovery anomaly detection module includes:

[0026] The power data extraction submodule calls each interleaved time interval in the behavior consistency feature segment to obtain the power sampling sequence of the load device in the corresponding time period and constructs a load power time series structure set.

[0027] The recovery feature calculation submodule extracts the time index of the voltage recovery start point and the power trend start change point according to the load power time sequence structure set, calculates the response interval between the two, obtains the power slope sequence within a fixed period after recovery, determines the fluctuation amplitude between each slope in the continuous segment, and obtains the power recovery response feature group.

[0028] The delay segment marking submodule, based on the power recovery response feature group, filters segments whose response interval exceeds the recovery offset threshold and whose slope fluctuation amplitude exceeds the stability limit, extracts the corresponding time range and load device identification information, and marks them as abnormal state intervals in the power recovery process to obtain power recovery abnormal segments.

[0029] As a further aspect of the present invention, the operation status monitoring output module includes:

[0030] The abnormal indicator extraction submodule calls the data structure of each segment in the power recovery abnormal segment, extracts the recorded load device identifier, response delay time and power change slope fluctuation information, and combines the three indicators according to the device dimension to generate an abnormal indicator combination set.

[0031] The item order construction submodule arranges all the status information items corresponding to the abnormal devices in chronological order according to the time index of each group of data in the abnormal indicator combination set, and generates an abnormal behavior time sequence item sequence.

[0032] The monitoring result output submodule writes the sorted content in the abnormal behavior time sequence entry into the station power system operation status record table to obtain the station power system operation status monitoring results.

[0033] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0034] In this invention, by extracting the abrupt change trends in the current and voltage signals of operating equipment and combining them with the response misalignment between signals, it is possible to accurately characterize changes in equipment behavior and identify potential abnormal segments. By using multi-dimensional confidence factors to filter signal segments and eliminate low-reliability data, and then by judging the temporal interleaving characteristics and consistency of behavior types between triggered behaviors, representative characteristic time segments are constructed. Combined with the trend delay and fluctuation amplitude changes of power signals during voltage recovery, the abnormal behavior of equipment power recovery process is accurately marked, ultimately forming an abnormal state sequence arranged in chronological order. This enables highly timely monitoring of operating status. This processing logic significantly improves the robustness and identification accuracy of data analysis, and strengthens the early warning capability and fault response efficiency of station power systems in practical application scenarios for potential risks. Attached Figure Description

[0035] Figure 1 This is a system flowchart of the present invention;

[0036] Figure 2 This is a flowchart of the jump behavior recognition module of the present invention;

[0037] Figure 3 This is a flowchart of the confidence screening module of the present invention;

[0038] Figure 4 This is a flowchart of the behavior consistency extraction module of the present invention;

[0039] Figure 5 This is a flowchart of the power recovery anomaly detection module of the present invention;

[0040] Figure 6 This is a flowchart of the operation status monitoring and output module of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0042] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0043] Please see Figure 1 The system operation monitoring system for multi-source data fusion includes:

[0044] The jump behavior recognition module acquires current sampling data and voltage on / off signals of the operating equipment in the station power system, identifies the jump trend behavior of each device, and obtains the behavior response misalignment data segment.

[0045] The confidence screening module calculates the direction score and amplitude deviation of the signal segment corresponding to the trigger point of the device's jump trend behavior based on the misaligned behavior data segment, and filters the data source segments whose confidence needs to be lowered.

[0046] The behavior consistency extraction module removes data source segments whose confidence level needs to be lowered, constructs staggered time intervals composed of adjacent trigger points based on the retained device jump trend behavior trigger points, analyzes the overlap of trigger points within the intervals, and extracts behavior consistency feature segments.

[0047] The power recovery anomaly detection module extracts power sampling data of the station power system load equipment based on the interlaced time intervals of behavioral consistency feature segments, determines the power recovery delay during the voltage recovery process, and marks the power recovery anomaly segment.

[0048] The operation status monitoring output module generates a sequence of abnormal entries in chronological order based on the power recovery abnormal segment, and outputs the operation status monitoring results of the station's power system.

[0049] The behavioral response misalignment data segment includes trigger point time difference, adjacent trigger point misalignment identifier, and current change segment. The data source segment whose confidence needs to be lowered includes direction score value, amplitude deviation value, and low confidence label. The behavioral consistency feature segment includes interleaved time index, repeated trigger frequency, and behavioral event type. The power recovery anomaly segment includes power recovery start point, power trend start point, and power slope fluctuation. The operation status monitoring results include abnormal equipment identifier, abnormal data entry, and output time series.

[0050] Please see Figure 2 The jump behavior recognition module includes:

[0051] The electrical parameter acquisition structured submodule acquires current sampling data and voltage on / off signals of operating equipment in the station power system. It calculates the rate of change sequence between continuous sampling points of the current sampling data, extracts abrupt change segments, and combines the edge transition time points in the voltage on / off signals to perform unified period normalization processing on the sampling information of the two types of signals to obtain the structured electrical parameter change sequence.

[0052] To acquire current sampling data and voltage on / off signals from operating equipment in the station's power supply system, the rate of change between consecutive sampling points of the current sampling data is first calculated. Assuming five current data points are collected every second, and the current of a certain device is 10A, 10.2A, 11A, 9.5A, and 8A respectively, then its rate of change is: (Point 2 is...) Point 3 is Point 4 is Point 5 is These rates of change are used to construct a rate of change sequence. Then, abrupt changes are identified within this sequence, and a change threshold can be set. When the rate of change of two consecutive points exceeds the threshold, it is marked as a sudden change segment. Therefore, points 3 to 5 are marked as sudden changes. After the sudden change is extracted, the edge transition time point in the voltage on / off signal is extracted. Assuming that the voltage state changes from "on" to "off" at the 3rd second, the edge transition time is identified as the 3rd second. The current sudden change segment and the voltage transition time are uniformly normalized. Assuming the sampling period is 1 second and the normalization period length is 5 seconds, the current change rate sequence and the voltage transition point information are synchronized to this period. For example, if the transition occurs in the 3rd second and coincides with the current change sudden change segment, then the structured electrical parameter change sequence constructed in this period is: the change rate peak position is 3, and the voltage transition time point is 3. After normalization, a complete structured sample is formed.

[0053] The behavior trajectory synchronization submodule calls the current change rate segment and voltage jump point sequence in the structured electrical parameter change sequence to synchronize the position of each device behavior trajectory, and pairs each change rate peak with the corresponding voltage change point to obtain the trajectory group corresponding to the device behavior.

[0054] The current rate of change segment and voltage jump point sequence from the structured electrical parameter change sequence are invoked to synchronize the behavioral trajectory position of each device. First, based on the time point where the rate of change peak occurs, such as the 3rd second mentioned above, this time is designated as the key point in the current change trajectory of that device. Then, it is checked whether there are any overlapping or close time points in the voltage jump time points, and a time pairing threshold is set as follows. The time interval is from 2.8 seconds to 3.2 seconds. If the voltage jump occurs at 3.1 seconds, it is successfully paired with the current change peak. The pair of peaks and jump points are matched as a set of trajectory points. Then, according to the sampling device number and time point, a behavior trajectory group is constructed. For example, if the device ID is D001, the trajectory group is recorded as [D001, t=3.0, current change, voltage jump]. The above process is performed for each device to obtain the trajectory group corresponding to the device behavior. Each trajectory group corresponds to a synchronous behavior sample composed of the electrical parameter changes of a device within a specific time period.

[0055] The response offset recognition submodule determines whether the time offset threshold is exceeded based on the time difference information of each pair of signals in the trajectory group corresponding to the device behavior. It extracts the time interval, device number and signal alignment status information of the behavior segments that meet the conditions to obtain the behavior response misalignment data segment.

[0056] Based on the time difference of each pair of signals in the trajectory group corresponding to the device behavior, determine whether it exceeds the set time offset threshold. The set offset threshold is 0.5 seconds. If the current change time in a certain trajectory group is 3 seconds and the voltage change time is 3.7 seconds, then its time difference is... Seconds, because The segment was identified as a misaligned response and extracted. The time interval for recording this segment was 0.7 seconds, the device number was D001, and the alignment status was "misaligned". If a device experiences a current surge at the 4th second and a voltage jump at the 4.3rd second, the difference is... If the threshold condition is met, the record is ignored. This process is repeated to extract all response offset segments. Each record includes the time interval (in seconds), device number (e.g., D002), and signal alignment status (e.g., not aligned).

[0057] Please see Figure 3 The confidence level filtering module includes:

[0058] The response segment extraction submodule extracts the signal change segment in the current change rate sequence and the direction change segment in the voltage on / off signal for each jump trend behavior trigger point based on the behavior response misalignment data segment, thus obtaining the jump signal aligned segment group.

[0059] Based on the behavioral response misalignment data segment, the signal change segment of each jump trend trigger point in the current change rate sequence and the direction change segment in the voltage on / off signal are extracted. First, the trigger point of the voltage jump trend is located. Suppose that the on / off jump occurs at the 5th second. According to the offset record, the current change rate of the device changes abruptly at the 4.6th second. The current change rate data within a 1-second range before and after the jump point are extracted to form the signal change segment of the trigger point. If the sampling frequency in this period is 5Hz, the current change rate segment consists of 10 points. Then, the corresponding direction change is extracted from the voltage on / off signal. The voltage is determined to be high (e.g., 220V) before the transition and low (e.g., 0V) after the transition. The direction change can be defined as changing from 1 to 0. The direction change segment is then constructed as a set of direction vectors [1, 0]. The current change rate segment and the voltage direction change segment are combined to form a transition signal alignment segment group. If the device number is D015, this segment group is recorded as [D015, 5th second, current change rate segment, 4.0–5.0 seconds; voltage direction segment, 5.0–6.0 seconds]. This method is used to execute all behavioral response misaligned data segments and extract the aligned transition segment group one by one.

[0060] The offset calculation submodule calls the current change rate segment and voltage direction segment of each group in the jump signal alignment segment group to construct a time matching vector group. It calculates the cosine of the angle between the time matching vector groups as the direction score, extracts the maximum value and the mean value of the reference segment in each group of current change segments, calculates the ratio of the two as the amplitude deviation, and obtains the combination of direction score and amplitude deviation.

[0061] The time-matching vector group is constructed by calling the current rate of change segment and voltage direction segment in each segment group of the jump signal alignment segment. First, the sampling sequence of the current rate of change segment is extracted from each jump event. Suppose the current rate of change segment of a certain device is 0.02, 0.04, 0.08, -0.01, -0.03, corresponding to the voltage direction change from "on" to "off", which can form a voltage direction vector of 1, 0, 0, 0, 0, that is, the direction jump is only reflected in the first and second positions. Then the time-matching vector group is:

[0062] Current rate of change vector: ;

[0063] Voltage direction change vector: ;

[0064] Next, we will consider the current rate of change vector. After performing maximum absolute value normalization, the maximum value is 0.08, and the normalization result is: .

[0065] Then, the orientation score is calculated, and the similarity between the two vectors is determined by the cosine angle between them, as shown in the following formula:

[0066] ;

[0067] in, The angle between two vectors; Directional score, representing the degree of consistency between the directions of two vectors, is an indicator used to measure the consistency (similarity) between a current rate of change signal and a voltage direction jump. This index identifies whether the current rate of change and voltage direction exhibit consistent physical behavior at the jump moment, thus helping to determine whether the event is a valid or typical power system jump behavior. A score close to 1 indicates that the direction of the current change and the voltage jump are highly consistent; a score close to 0 indicates almost no correlation; a negative score indicates that the two directions are opposite. In power system event analysis, a high directional score means that the event is highly likely to be a voltage-current consistent physical jump behavior. : The normalized vector of the rate of change of current; : Voltage direction change vector; The dot product of vectors represents the projective product of two vectors in the same direction. :vector The modulus length indicates its size; :vector The modulus indicates its size.

[0068] The specific numerical calculations are as follows:

[0069] Dot product: ;

[0070] Length of the module: ;

[0071] Length of the module: ;

[0072] Final Direction Score: .

[0073] Next, the amplitude deviation is calculated, and the maximum value in this current change rate segment is extracted. Then extract the average rate of change of the reference segment (e.g., the previous second). The ratio between the two is calculated as follows:

[0074] ;

[0075] in, Amplitude deviation indicates the amplification factor of the signal abrupt change relative to the background change; : The maximum value in the current rate of change segment; : The average rate of change of the reference segment (the period without mutations).

[0076] Ultimately, the result for this segment is a direction score of 0.209 and an amplitude deviation of 4.0. The combined record is [direction score = 0.209, amplitude deviation = 4.0]. The above processing is performed on all the jump signal aligned segment groups to generate a complete combination data set of direction score and amplitude deviation.

[0077] The confidence value determination submodule determines whether the score and deviation items in the combination of direction score and amplitude deviation exceed the preset offset score threshold and amplitude deviation threshold, filters the signal segments that meet the dual conditions, extracts the device number and time index of the corresponding segment, marks the data content whose confidence does not meet the requirements, and obtains the data source segments whose confidence needs to be lowered.

[0078] Based on the score and deviation items in the combination of direction score and amplitude deviation, it is determined whether the preset offset score threshold and amplitude deviation threshold are exceeded. Initially, the offset score threshold is set to 0.3 and the amplitude deviation threshold to 2.5. When a data point has a direction score of 0.209 (less than 0.3) and an amplitude deviation of 4.0 (greater than 2.5), it is identified as an abnormal signal segment because it simultaneously meets the two conditions of insufficient direction score and excessive amplitude deviation. Further, the device number and time index of the segment are extracted. Assuming the segment belongs to device D037 and the time index is 12 seconds, it is marked as data content with insufficient confidence and recorded as [D037, 12 seconds, direction score = 0.209, amplitude deviation = 4.0, insufficient confidence]. This process is repeated for each combination of data, with conditions being filtered only when the direction score... And amplitude deviation Only when the data segment is in a certain condition is it recorded as a data segment whose confidence level needs to be lowered; otherwise, it is not processed, eventually forming a set of abnormal data segments marked with insufficient confidence.

[0079] Please see Figure 4 The behavior consistency extraction module includes:

[0080] The behavior point integration submodule removes data source fragments whose confidence level needs to be lowered, extracts the time index and device number information of the retained device jump trend behavior trigger points, and obtains the jump behavior interleaved sequence group;

[0081] Equipment abrupt change behavior refers to sudden and significant changes in current sampling data or voltage on / off signals during equipment operation, reflecting the instantaneous behavioral characteristics of the equipment suddenly transitioning from one stable operating state to another (such as startup, shutdown, or switching). These changes are typically manifested as signal patterns such as sudden increases or decreases in the rate of change of current and abrupt changes in voltage state, exhibiting identifiable abrupt trends and serving as key indicators of equipment behavioral state transitions. Data source segments whose confidence levels need to be lowered are removed. The time index and device number information of the device jump trend behavior trigger points are extracted and retained. First, data records marked as insufficient confidence are excluded from all identified jump behavior segments, such as the previously identified segment [D037, 12 seconds], which will be removed. The remaining parts are retained as valid behavior points. Then, the retained segments are traversed, and the time index and device number of each one are extracted, such as [D012, 10 seconds], [D015, 15 seconds], [D026, 18 seconds]. These valid behavior trigger information are integrated and sorted by time to form a jump behavior interleaved sequence group. If multiple devices have jump behavior in close time, they are merged into multiple device interleaved behavior sequence nodes on the time axis, such as [D012, 10 seconds], [D015, 10.2 seconds], [D012, 11 seconds], [D026, 11.3 seconds], etc., forming a continuous device behavior event stream.

[0082] The interval frequency aggregation submodule calls the alternating sequence group of jump behavior, matches the event types defined in the behavior label set, counts the frequency of occurrence of event types in each combination, and filters the segment groups in the jump behavior set;

[0083] The system calls the alternating sequence group of transition behaviors, matches the event types defined in the behavior label set, and counts the frequency of occurrence of event types in each combination. First, it sets behavior label sets such as "switching operation", "power switching", "device abnormality" and other event labels. It matches the trigger characteristics of the transition sequence according to the behavior pattern. For example, if a device has three transitions within 5 seconds, it matches the "switching operation" behavior pattern. Then, it marks the behavior as a switching event and writes the matching type into the behavior sequence node. After matching all transition behaviors, it counts the number of occurrences of each type of event in each fixed time interval, such as every 10 seconds. For example, "switching operation" occurs twice in [0-10 seconds], "power switching" occurs once, and "device abnormality" occurs three times in [10-20 seconds]. The frequency statistics results are organized into an interval event frequency table. The system filters out the cases where a certain type of event occurs densely in a certain time period and extracts that time period as the segment group in the transition behavior set.

[0084] The coupling feature filtering submodule performs conditional independence tests on the relationship between behavior frequency and corresponding device on / off status based on the concentrated segment group of jump behavior, calculates the chi-square value and filters the statistical correlation degree of the segments, and retains behavior consistency feature fragments.

[0085] Based on the concentrated segments of jump behavior, a conditional independence test is performed on the relationship between the frequency of the behavior and the on / off state of the corresponding equipment. The chi-square value is calculated to screen the statistical correlation of the segments and retain the behavioral consistency feature fragments. First, for each concentrated segment of jump behavior, the joint distribution of the equipment jump frequency and on / off state is statistically analyzed. For example, in a 10-second segment, equipment D012 jumps 3 times, is in the power-on state 4 times and is in the power-off state 1 time; equipment D015 jumps 1 time, is in the power-on state 2 times and is in the power-off state 3 times. The observed frequency table is presented as follows: under high frequency jump behavior (≥3), there are 4 times of power-on and 1 time of power-off; under low frequency jump behavior (≤1), there are 2 times of power-on and 3 times of power-off, with a total of 10 records.

[0086] The formula used to calculate the chi-square value is as follows:

[0087] ;

[0088] in, Chi-square statistic, used to assess the degree of deviation between observed data and expected data; : No. Class behavior frequency in The frequency of observations under various on / off states; : No. The expected frequency of an item represents the theoretical frequency when the frequency of the behavior is independent of the on / off state. The calculation formula is as follows: , : No. Total frequency of rows (i.e., the total number of times a certain frequency category of a row appears); : No. The total frequency of the column (i.e., the total number of times the power-on or power-off state occurs); Total sample size, which is the sum of the frequencies of all observed data.

[0089] Assume the high-frequency behavior (3 times) is in the powered-on state: Power off state: Low-frequency behavior (1 time) in the powered-on state: Power off state: Total of rows: , ; Total: , Total number of samples: .

[0090] Calculate the expected frequency: , , , .

[0091] Substitute into the formula: .

[0092] The judgment threshold was set at 3.84 (degrees of freedom 1, confidence level 0.05), because Therefore, there is no significant statistical relationship between the frequency of behavior and the on / off state within this segment, and it does not constitute a consistent coupling characteristic, so it needs to be removed. If the observation results within another segment make If the frequency of a behavior is statistically correlated with its on / off state, it is considered a consistent segment and retained. This calculation process is repeated for all segments. A behaviorally consistent segment is a signal segment that has a significant statistical correlation between the frequency of the switching behavior and the on / off state of the equipment. Based on the image content, to determine whether a segment is a behaviorally consistent segment, a joint frequency distribution table of the switching frequency and the on / off state of the equipment in that segment needs to be compiled, and an independence test is performed using the chi-square test. If the calculated chi-square value χ² is greater than the set threshold of 3.84 (degrees of freedom 1, confidence level 0.05), it indicates a significant statistical correlation between the switching behavior and the on / off state of the equipment, and the segment can be considered a behaviorally consistent segment and retained. Conversely, if χ² is less than 3.84, it indicates no significant relationship between the two, and the segment does not constitute a consistent coupling feature and should be removed. Finally, all segments that meet the statistical significance are identified as behaviorally consistent segments, and a set of statistically significant feature segments that meet the behavioral consistency criteria is ultimately selected.

[0093] Please see Figure 5 The power recovery anomaly detection module includes:

[0094] The power data extraction submodule calls each staggered time interval in the behavior consistency feature segment to obtain the power sampling sequence of the load device in the corresponding time period and constructs a load power time series structure set.

[0095] The system retrieves the power sampling sequence of the load devices within each interleaved time interval of the behavior consistency feature segment, and constructs a load power time series structure set. First, it performs data extraction operations on each retained transition behavior segment. Assuming that a certain interleaved time interval is from the 30th to the 40th second, the devices involved include D011 and D014. The system samples power at a frequency of once per second, so each device will extract 10 sampling points in this segment. For example, the power values ​​of D011 are 120W, 123W, 128W, etc., indicating the load state change trend in this segment. At the same time, the power values ​​collected by D014 in the same interval are 98W, 101W, 105W, etc., forming a time series set of power changing with time. The power sampling sequence corresponding to each device is matched with the time index to form a complete load power time series structure set. The record format includes device number, start and end time index, and sampling value sequence.

[0096] The recovery characteristic calculation submodule extracts the time index of the voltage recovery start point and the power trend start change point based on the load power time sequence structure set, calculates the response interval between the two, obtains the power slope sequence within a fixed period after recovery, determines the fluctuation amplitude between each slope in the continuous segment, and obtains the power recovery response characteristic group.

[0097] Based on the load power time sequence structure set, the time indices of the voltage recovery start point and the power trend start point are extracted. The response interval between the two is calculated, and the power slope sequence within a fixed period after the recovery is obtained. The fluctuation amplitude between the slopes in the continuous segment is determined, and the power recovery response feature group is obtained. First, the recovery point is identified. For example, the voltage of device D011 recovers from 0V to 220V in the 35th second, which is the voltage recovery start point. Observing its power sampling data, the power rises from 120W to 135W starting from the 36th second. The 36th second is defined as the power trend start point, and the response interval is 1 second. Subsequently, the fixed period is set to 5 seconds. The system extracts power data from 36 to 40 seconds, calculates the power change per second, and obtains four slope values. For example, the slope is 6W / s from 120W to 126W, 6W / s from 126W to 132W, and 3W / s from 135W. The difference between adjacent slopes is calculated one by one. For example, the difference between 6W / s and 6W / s is 0, and the difference between 6W / s and 3W / s is 3, forming a slope fluctuation sequence. The fluctuation amplitude is further analyzed, and the maximum fluctuation value is recorded as 3W / s. Finally, the system forms the response characteristic group of the device within a specified period, including the response interval of 1 second and the maximum slope fluctuation of 3W / s.

[0098] The delay segment marking submodule, based on the power recovery response feature group, filters segments whose response interval exceeds the recovery offset threshold and whose slope fluctuation exceeds the stability limit, extracts the corresponding time range and load device identification information, and marks them as abnormal state intervals in the power recovery process to obtain power recovery abnormal segments;

[0099] Based on the power recovery response feature group, segments with response intervals exceeding the recovery offset threshold and slope fluctuations exceeding the stability limit are filtered out. The corresponding time range and load device identification information are extracted and marked as abnormal state intervals in the power recovery process, resulting in abnormal power recovery segments. First, the recovery offset threshold is set to 2 seconds, and the stability limit is that the slope fluctuation must not exceed 2.5W / s. When a device's response interval is 3 seconds, exceeding the 2-second threshold, and the power slope fluctuation amplitude is 3W / s, also exceeding the fluctuation limit, since both conditions meet the abnormal criteria, the power recovery segment of this device is marked as an abnormal segment. Its device number, such as D011, and the time range of 35 to 40 seconds are recorded and written into the abnormal data structure. For example, another device, D018, has a response interval of 1 second, but the slope fluctuation reaches 4W / s. Although the response interval is normal, the fluctuation exceeds the limit, so it is also judged as an abnormal recovery segment. Its time range, device number, and other key information are extracted to finally form a complete set of abnormal power recovery segments.

[0100] Please see Figure 6 The operation status monitoring output module includes:

[0101] The abnormal indicator extraction submodule calls the data structure of each segment in the power recovery abnormal segment, extracts the recorded load device identifier, response delay time and power change slope fluctuation information, and combines the three indicators according to the device dimension to generate an abnormal indicator combination set.

[0102] The system calls upon the data structure of each segment in the power recovery anomaly segment to extract the recorded load device identifier, response delay time, and power change slope fluctuation information. These three indicators are combined by device dimension to generate an anomaly indicator combination set. First, the corresponding field content is parsed from each recovery segment marked as an anomaly. For example, device D021, in the 50-55 second segment, has a response delay of 3 seconds and a maximum power slope fluctuation of 3.8W / s. These two indicators are combined with the device number to form an anomaly data entry. Then, device D034, in the 60-65 second segment, records a response delay of 2.5 seconds and a slope fluctuation of 4.1W / s, also generating a data entry. All entries are organized by device number dimension. The device identifier, response delay, and slope fluctuation are packaged into an anomaly indicator combination set, with structures such as [D021, 3 seconds, 3.8W / s] and [D034, 2.5 seconds, 4.1W / s]. The combination results for all devices are used for sorting and record construction processing.

[0103] The item order construction submodule arranges all status information items corresponding to abnormal devices in chronological order according to the time index of each group of data in the abnormal indicator combination set, and generates an abnormal behavior time sequence item sequence.

[0104] Based on the time index of each data set in the abnormal indicator combination set, all status information entries corresponding to abnormal devices are arranged in chronological order to generate an abnormal behavior time sequence. First, the time index range contained in each combination entry is extracted. For example, the abnormal segment of D021 is 50 to 55 seconds, D034 is 60 to 65 seconds, and D011 is 52 to 58 seconds. All entries are sorted according to their start time, i.e., 50 seconds, 52 seconds, and 60 seconds in sequence. Then, according to the sorting result, the corresponding device identifier, response delay, and slope fluctuation information are output in sequence to form an abnormal behavior time sequence arranged by time. For example, the first entry is [D021, 3 seconds, 3.8W / s], the second entry is [D011, 2 seconds, 3.2W / s], and the third entry is [D034, 2.5 seconds, 4.1W / s]. The sorting method ensures the consistency of the time axis of abnormal information and ensures that the system has the logical order of event evolution when reading abnormalities later.

[0105] The monitoring result output submodule, based on the sorted content in the abnormal behavior time sequence item sequence, writes the structure of each item into the station power system operation status record table to obtain the station power system operation status monitoring results;

[0106] Based on the sorted content in the sequence of abnormal behavior time-series entries, the structure of each entry is written into the station power system operation status record table to obtain the operation status monitoring results. First, a record row is generated for each time-series entry, with fields including equipment number, abnormal start time, response delay time, maximum power slope fluctuation value, etc. For example, the first entry [D021, 50 seconds, 3 seconds, 3.8W / s] is written into the first row of the record table, and then the second entry [D011, 52 seconds, 2 seconds, 3.2W / s] is written into the second row. Each record is bound to a time index and equipment number. At the same time, the system will mark the status of each record as "power recovery abnormal" and add a number for querying. Finally, all entries are written into the station power system monitoring data file according to the time sequence to generate a complete operation status monitoring result table for the operation and maintenance system to read, compare and analyze.

[0107] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A system operation monitoring system that integrates multi-source data, characterized in that, The system includes: The jump behavior recognition module acquires current sampling data and voltage on / off signals of the operating equipment in the station power system, identifies the jump trend behavior of each device, and obtains the behavior response misalignment data segment. The confidence screening module calculates the direction score and amplitude deviation of the signal segment corresponding to the trigger point of the device's jump trend behavior based on the behavior response misalignment data segment, and filters the data source segments whose confidence needs to be lowered. The behavior consistency extraction module removes data source segments whose confidence level needs to be lowered, constructs an interlaced time interval composed of adjacent trigger points based on the retained device jump trend behavior trigger points, analyzes the overlap of trigger points in the interval, and extracts behavior consistency feature segments. The behavior consistency extraction module includes: The behavior point integration submodule removes the data source segments whose confidence level needs to be lowered, extracts the time index and device number information of the retained device jump trend behavior trigger points, and obtains the jump behavior interleaved sequence group; The interval frequency aggregation submodule calls the alternating sequence group of the jump behavior, matches the event types defined in the behavior label set, counts the frequency of occurrence of the event type in each combination, and filters the segment group in the jump behavior set. The coupling feature filtering submodule performs a conditional independence test on the relationship between the behavior frequency and the on / off state of the corresponding device based on the group of segments with concentrated jump behaviors, calculates the chi-square value and filters the statistical correlation degree of the segments, and retains the behavior consistency feature fragments. The power recovery anomaly detection module extracts power sampling data of the station power system load equipment based on the interleaved time interval of the behavioral consistency feature segments, determines the power recovery delay during the voltage recovery process, and marks the power recovery anomaly segment. The power recovery anomaly detection module includes: The power data extraction submodule calls each interleaved time interval in the behavior consistency feature segment to obtain the power sampling sequence of the load device in the corresponding time period and constructs a load power time series structure set. The recovery feature calculation submodule extracts the time index of the voltage recovery start point and the power trend start change point according to the load power time sequence structure set, calculates the response interval between the two, obtains the power slope sequence within a fixed period after recovery, determines the fluctuation amplitude between each slope in the continuous segment, and obtains the power recovery response feature group. The delay segment marking submodule, based on the power recovery response feature group, filters segments whose response interval exceeds the recovery offset threshold and whose slope fluctuation amplitude exceeds the stability limit, extracts the corresponding time range and load device identification information, and marks them as abnormal state intervals in the power recovery process to obtain power recovery abnormal segments; The operation status monitoring output module generates an abnormal entry sequence in chronological order based on the power recovery abnormal segment, and outputs the operation status monitoring results of the station's power system.

2. The system operation monitoring system for multi-source data fusion according to claim 1, characterized in that, The behavioral response misalignment data segment includes trigger point time difference, adjacent trigger point misalignment identifier, and current mutation segment. The data source segment whose confidence needs to be lowered includes direction score value, amplitude deviation value, and low confidence label. The behavioral consistency feature segment includes interleaved time index, repeated trigger frequency, and behavioral event type. The power recovery anomaly segment includes power recovery start point, power trend start point, and power slope fluctuation. The operating status monitoring results include abnormal equipment identifier, abnormal data entry, and output time series.

3. The system operation monitoring system for multi-source data fusion according to claim 1, characterized in that, The jump behavior recognition module includes: The electrical parameter acquisition structured submodule acquires current sampling data and voltage on / off signals of operating equipment in the station power system. It calculates the rate of change sequence between continuous sampling points of the current sampling data, extracts abrupt change segments, and combines the edge transition time points in the voltage on / off signals to perform unified period normalization processing on the sampling information of the two types of signals to obtain the structured electrical parameter change sequence. The behavior trajectory synchronization submodule calls the current change rate segment and voltage jump point sequence in the structured electrical parameter change sequence to synchronize the position of each device behavior trajectory, and pairs each change rate peak with the corresponding voltage change point to obtain the trajectory group corresponding to the device behavior. The response offset recognition submodule determines whether the time offset threshold is exceeded based on the time difference information of each pair of signals in the trajectory group corresponding to the device behavior. It extracts the time interval, device number and signal alignment status information of the behavior segments that meet the conditions to obtain the behavior response misalignment data segment.

4. The system operation monitoring system for multi-source data fusion according to claim 3, characterized in that, The confidence level filtering module includes: The response segment extraction submodule extracts the signal change segment in the current change rate sequence and the direction change segment in the voltage on / off signal for each jump trend behavior trigger point based on the behavior response misalignment data segment, and obtains the jump signal aligned segment group. The offset calculation submodule calls each current change rate segment and voltage direction segment in the jump signal alignment segment group to construct a time matching vector group, calculates the cosine of the angle between the time matching vector groups as the direction score, extracts the maximum value and the average value of the reference segment in each current change segment, calculates the ratio of the two as the amplitude deviation, and obtains the combination of direction score and amplitude deviation. The confidence value determination submodule determines whether the score and deviation items in the combination of direction score and amplitude deviation exceed the preset offset score threshold and amplitude deviation threshold, filters the signal segments that meet the dual conditions, extracts the device number and time index of the corresponding segment, marks the data content whose confidence does not meet the requirements, and obtains the data source segments whose confidence needs to be lowered.

5. The system operation monitoring system for multi-source data fusion according to claim 1, characterized in that, The operation status monitoring output module includes: The abnormal indicator extraction submodule calls the data structure of each segment in the power recovery abnormal segment, extracts the recorded load device identifier, response delay time and power change slope fluctuation information, and combines the three indicators according to the device dimension to generate an abnormal indicator combination set. The item order construction submodule arranges all the status information items corresponding to the abnormal devices in chronological order according to the time index of each group of data in the abnormal indicator combination set, and generates an abnormal behavior time sequence item sequence. The monitoring result output submodule writes the sorted content in the abnormal behavior time sequence entry into the station power system operation status record table to obtain the station power system operation status monitoring results.

Images