A smart source tracing method and system for power dispatching systems

Abstract

This invention relates to the field of intelligent source tracing technology for power dispatching systems, and particularly to an intelligent source tracing method and system for power dispatching systems. The method includes: identifying the wind turbine with response deviation requiring tracing; generating the current response time sequence; determining whether the response capability conforms to the historical normal fluctuation range; if not, determining whether there are common factors affecting the wind turbine based on the correlation characteristics between the wind turbine with response deviation and adjacent wind turbines with response deviation; based on the determination of common factors, determining a common response increment sequence according to the response amplitude of adjacent wind turbines with response deviation to reconstruct the common response theoretical sequence; further determining an individual response sequence based on the individual deviation value of the wind turbine with response deviation; identifying whether the wind turbine with response deviation conforms to superimposed individual anomalies based on the individual response sequence to identify abnormal periods, and issuing a deep inspection command. This invention improves the accuracy of identifying individual anomalies when common factors exist in power dispatching source tracing.

Description

Technical Field

[0001] This invention relates to the field of intelligent traceability technology for power dispatching systems, and in particular to an intelligent traceability method and system for power dispatching systems. Background Technology

[0002] As controllable power generation units, wind farms need to participate in grid frequency regulation and power control. The power dispatching system issues commands to wind farms to increase or decrease power generation through automatic generation control, which are then distributed to each wind turbine for execution. In this process, the dispatching system needs to accurately assess the response performance of the wind farms and conduct source analysis on response deviations to clarify responsibility and provide a basis for operation and maintenance decisions.

[0003] However, existing power dispatching tracing methods primarily focus on the overall response deviation of wind farms, lacking refined analysis of the response behavior of individual wind turbines. When the overall power generation increase of a wind farm fails to meet dispatch instructions, current technologies cannot distinguish whether the insufficient response is due to individual turbine malfunctions or a collective decline in response capability caused by common factors. Furthermore, existing methods do not consider the correlation characteristics between wind turbines during the response process, nor do they establish benchmarks for the historical response behavior of individual turbines, leading to a high rate of misjudgment in tracing and failing to provide targeted guidance to maintenance personnel for in-depth inspections of abnormal turbines.

[0004] Chinese Patent Publication No. CN116316586A discloses a method for tracing the source of power jumps in a power system using jump analysis, characterized by the following steps: Step 1: Determine the trend test formula; Step 2: Perform abrupt change test; Step 3: Trace the source of power jump components.

[0005] Therefore, it is evident that the existing technology has the following problems: When the overall power generation of wind turbines does not meet the target increase during power dispatch, the system fails to consider longitudinal comparison of the wind turbines' own response modes and horizontal correlation comparison with adjacent wind turbines. It also fails to consider eliminating the influence of common factors on the wind turbine response process to analyze the individual response process of each wind turbine. As a result, the system is unable to quickly identify individual anomalies and abnormal periods of wind turbines during the problem tracing process due to the influence of common factors on the group. This leads to the problem of low accuracy in tracing the source of abnormal wind turbines and abnormal periods. Summary of the Invention

[0006] To address this, the present invention provides an intelligent tracing method and system for power dispatching systems. This overcomes the problem that existing technologies, when faced with situations where the overall power generation increase of wind turbines does not meet the target increase during power dispatching, fail to consider longitudinal comparison of the wind turbines' own response modes and horizontal correlation comparison with adjacent wind turbines, and fail to consider analyzing the individual response processes of wind turbines by eliminating the influence of common factors affecting the group. As a result, during the problem tracing process, the individual anomalies of wind turbines and abnormal periods cannot be quickly identified due to the influence of common factors affecting the group, leading to low accuracy in tracing abnormal wind turbines and abnormal periods.

[0007] To achieve the above objectives, the present invention provides an intelligent source tracing method for power dispatching systems, comprising: Obtain the target increase value and actual increase value of each wind turbine issued by the power dispatch system in order to identify the wind turbines with response deviations that need to be traced. The response process data of the current cycle of the response deviation fan is obtained to generate the current response time sequence, wherein the response process data includes output data; The deviation of the response deviation fan is evaluated based on the current response time sequence and the historical response time sequence. Based on the comparison of the deviation degree with a preset deviation degree threshold, it is determined whether the response capability of the response deviation fan conforms to the historical normal fluctuation range. In response to the fact that the response capability of the fan with the response deviation does not conform to the historical normal fluctuation range, based on the current correlation characteristics and historical correlation characteristics of the fan with the response deviation and adjacent fans with the response deviation, it is identified whether the fan with the response deviation and adjacent fans with the response deviation have consistency, so as to determine whether there are common factors affecting it. Based on the determination of common factors, the response increment sequence is determined according to the response amplitude of the current response time sequence of adjacent response deviation fans, so as to determine the common response increment sequence. The common response incremental sequence and the rated power of the response deviation fan are used to reconstruct the common response theoretical sequence of the response deviation fan; Based on the individual deviation value between the current response time sequence of the aforementioned response deviation fan and the common response theoretical sequence, a unique response sequence is determined; Based on the absolute extreme values ​​of the individual response sequence and the corresponding duration, identify whether the response deviation fan meets the superimposed individual anomaly to identify the abnormal period, and issue a deep inspection command to the individual abnormal fan based on the abnormal period.

[0008] Furthermore, the process of identifying the source of the response deviation in the fan includes: Based on the difference between the target increase value of each wind turbine issued by the power dispatch system and the actual increase value of each wind turbine, the response deviation value of each wind turbine is determined. Fans with response deviation values ​​greater than a preset response deviation threshold are identified as fans with response deviations that require tracing.

[0009] Furthermore, the process of determining whether the response capability of the fan with the response deviation conforms to the historical normal fluctuation range includes: The response process data of the fan with response deviation in the historical period is obtained to generate several historical response time series sequences; The mean output and standard deviation at each time point are calculated based on several historical response time series. The degree of deviation of the response deviation fan is calculated based on the output data of the current response time sequence and the average output value; In response to the deviation being greater than a preset deviation threshold, it is determined that the response capability of the fan with the deviation does not conform to the historical normal fluctuation range.

[0010] Furthermore, the process of determining whether common factors have an impact includes: Calculate the correlation coefficient between the current response time series of the response deviation fan and the current response time series of adjacent response deviation fans to determine the current correlation characteristics; Calculate the correlation coefficient between the historical response time series of the response deviation fan and the historical response time series of adjacent response deviation fans to determine several historical correlation characteristics; The preset range of association features is determined based on each of the aforementioned historical association features; If the current associated feature falls within the preset range of the associated feature, and it is determined that the response deviation fan and the adjacent response deviation fan are consistent, then it is determined that there is a common factor affecting the fan.

[0011] Furthermore, the process of determining the common response increment sequence includes: The output increment corresponding to each data point is calculated based on the current response time sequence of each adjacent response deviation fan to determine the response amplitude; A response increment sequence is generated based on the ratio of the response amplitude to the rated power of the wind turbine; Based on the median of the response increment sequence at each time point of several adjacent response deviation wind turbines, the common response increment sequence is determined.

[0012] Furthermore, the process of reconstructing the common response theoretical sequence of the aforementioned response deviation fan includes: The theoretical sequence of response increments is determined based on the product of the common response increment sequence and the rated power of the response deviation fan. Based on the sum of the theoretical sequence of the response increment and the corresponding steady-state output, the common theoretical sequence of the response deviation fan is determined.

[0013] Furthermore, the process of determining the individual response sequence includes: Calculate the difference between the current response time sequence of the response deviation fan and the data points corresponding to each data point of the common response theoretical sequence to determine the individual deviation value; The individual response sequence is determined based on the individual deviation value.

[0014] Furthermore, the process of identifying whether the response deviation fan conforms to the superimposed individual anomalies includes: The individual deviation value with the largest absolute value in the individual response sequence is determined as the absolute extreme value. In response to the absolute extreme value being greater than a preset individual deviation threshold, the duration for which the individual deviation value in the individual response sequence is greater than the preset individual deviation threshold is obtained. Based on the determination result that the duration is greater than the preset duration threshold, it is determined that the response deviation fan conforms to the superimposed individual anomaly.

[0015] Furthermore, the process of identifying abnormal time periods includes: In response to the superimposed individual anomalies of the response deviation fan, several time periods in the individual sequence where the individual deviation value is greater than a preset individual deviation threshold are obtained to determine the abnormal time periods.

[0016] Furthermore, an intelligent traceability system for power dispatching systems is also provided, including: The deviation wind turbine identification module is used to obtain the target increase value and the actual increase value of each wind turbine issued by the power dispatch system, so as to identify the wind turbines with response deviations that need to be traced. The longitudinal comparison module for the deviation fan is used to obtain the response process data of the current cycle of the response deviation fan to generate the current response time sequence, and to evaluate the deviation degree of the response deviation fan based on the current response time sequence and the historical response time sequence, so as to determine whether the response capability of the response deviation fan conforms to the historical normal fluctuation range based on the comparison of the deviation degree with the preset deviation degree threshold. The common factor judgment module responds to the fact that the response capability of the response deviation fan does not conform to the historical normal fluctuation range. Based on the current association characteristics and historical association characteristics of the response deviation fan and the adjacent response deviation fans, it identifies whether the response deviation fan and the adjacent response deviation fans have consistency, so as to determine whether there are common factors affecting it. The common template construction module determines the common response increment sequence based on the response amplitude of the current response time sequence of adjacent response deviation fans, based on the judgment result of the existence of common factors, and determines the common response increment sequence. It also restores the common response theoretical sequence of the response deviation fan based on the common response sequence and the rated power of the response deviation fan. The individual sequence analysis module determines the individual response sequence based on the individual deviation value between the current response time sequence of the response deviation fan and the common response theoretical sequence; The individual anomaly identification module identifies whether the response deviation fan conforms to the superimposed individual anomaly based on the absolute extreme value of the individual response sequence and the corresponding duration, thereby identifying the abnormal period. Based on the abnormal period, it issues a deep inspection command to the individual abnormal fan.

[0017] Compared with existing technologies, the beneficial effects of this invention lie in its ability to accurately attribute wind turbine response deviations during power dispatching by constructing a multi-level progressive tracing method that progresses from overall deviation identification to individual anomaly localization. By screening the increased power generation of all wind turbines, the analysis scope is focused on turbines exhibiting response deviations, avoiding waste of computational resources. By longitudinally comparing the current and historical responses of turbines with response deviations, turbines with abnormal response capabilities are initially identified, excluding normal deviations caused by allocation targets exceeding physical capacity. Furthermore, a horizontal correlation comparison is introduced. By analyzing the consistency between turbines with response deviations and adjacent turbines with deviations during the response process, common environmental factors and individual equipment problems are effectively distinguished, solving the problem that traditional methods cannot identify group-wide impacts. By constructing a common template and stripping away common influences, individual anomaly characteristics are further mined within the common context, achieving accurate identification of superimposed anomalies and location of anomaly periods. This invention provides accurate tracing basis for power dispatching systems, improving the accuracy of tracing in situations where common factors and individual anomalies overlap, and enhancing the targeted nature of operation and maintenance work.

[0018] Furthermore, this invention calculates the difference between the target increase value and the actual increase value for each wind turbine as a response deviation value, and compares this deviation value with a preset response deviation threshold, thereby achieving rapid and automated screening of wind turbines requiring traceability. Wind turbines with response deviation values ​​exceeding the threshold are marked as response deviation wind turbines and constitute a set to be analyzed, thus defining a clear scope for all subsequent refined analyses, effectively reducing the complexity of data processing and computational burden, and enabling the entire traceability process to operate efficiently.

[0019] Furthermore, this invention constructs a set of response time series sequences of the wind turbine within a historical period, calculating the mean and standard deviation of the force as a historical benchmark. This historical benchmark fully considers the inherent characteristics of the wind turbine, making subsequent longitudinal comparisons more targeted and accurate. By comparing the current response time series sequence with the historical mean to calculate the overall deviation, the degree of abnormality of the wind turbine relative to its normal performance in the current event can be quantitatively reflected. By comparing the calculated root mean square error with a preset threshold, an objective determination of whether the response capability conforms to the historical normal fluctuation range is achieved, providing a clear trigger condition for subsequent common factor analysis.

[0020] Furthermore, this invention uses the correlation coefficient of output changes between the responding deviation fan and its adjacent deviation fans in current and historical events as a correlation feature. It further determines the impact of common factors by comparing whether the current correlation feature falls within a preset range of historical correlation features. When multiple adjacent fans simultaneously deviate from their respective historical benchmarks and their response relationships maintain a historical synchronization pattern, it can be strongly inferred that an external common factor is causing a decline in the overall response capability, rather than multiple fans experiencing individual failures simultaneously. This judgment logic effectively avoids misjudging external common factors as equipment responsibility and prevents the underestimation of individual problems in the context of group anomalies, significantly improving the rationality and specificity of the tracing process.

[0021] Furthermore, this invention eliminates the incomparability of response amplitudes caused by capacity differences among different wind turbines by dividing the output increment of adjacent turbines by their rated power, enabling statistical averaging of the common behavior of the group. The response increment sequence reflects the relative response amplitude of each wind turbine relative to its rated capacity. Based on this, the common response increment sequence obtained from the median at each time point can represent the typical relative response pattern of this group of wind turbines under the influence of common factors. This provides a basis for subsequently reconstructing the common response theoretical sequence of the target wind turbine, ensuring that the process of stripping away common factors considers both the average behavior of the group and respects the capacity characteristics of the target wind turbine itself.

[0022] Furthermore, this invention restores the group-averaged relative response pattern to the absolute output curve of the target wind turbine when only affected by common factors by multiplying the common response increment sequence by the rated power of the target wind turbine and adding its steady-state output. This restoration process fully considers the target wind turbine's own capacity and initial operating conditions, further obtaining a common response theoretical sequence. This sequence serves as a benchmark for eliminating common influences, providing an accurate reference for subsequent calculations of individual deviations, ensuring that the individual deviation values ​​in subsequent analyses reflect the unique characteristics of the target wind turbine that distinguish it from the group.

[0023] Furthermore, this invention generates a personalized response sequence by calculating the difference between each data point of the current response time sequence and the common response theoretical sequence. This sequence completely eliminates the influence of common factors on the wind turbine's response behavior, retaining only the unique response characteristics of the wind turbine. This processing highlights the individual information that was originally mixed in with the overall response, creating conditions for subsequent identification of superimposed individual anomalies. This sequence is the core data carrier for determining whether individual anomalies exist, and all subsequent determinations regarding individual anomalies are based on this sequence.

[0024] Furthermore, this invention achieves quantitative judgment of superimposed individual anomalies by extracting the absolute extreme values ​​and durations of individual response sequences and comparing these features with preset thresholds. The absolute extreme value reflects the maximum intensity of the anomaly, while the duration reflects the sustained impact of the anomaly; the combination of both can characterize individual anomaly features. By setting dual judgment conditions—an absolute extreme value greater than the individual deviation threshold and a duration greater than the persistence threshold—the accuracy and reliability of anomaly identification are improved.

[0025] Furthermore, this invention, after confirming the superimposed individual anomalies, segments and identifies the individual response sequences, marking intervals that continuously exceed a preset threshold as candidate anomaly periods, thereby achieving precise location of the specific time of anomaly occurrence. From the candidate periods, selection is made based on the maximum absolute deviation value and duration factors, allowing for the screening of the most noteworthy anomaly periods, making the issued in-depth inspection instructions more targeted. The output information includes the anomaly turbine number, common factor determination conclusions, and the period requiring attention, providing maintenance personnel with clear and complete fault clues and improving maintenance efficiency.

[0026] Furthermore, this invention constructs an intelligent traceability system for power dispatching systems. The deviation turbine identification module enables rapid screening, the longitudinal comparison module completes individual historical benchmark comparison, the common factor judgment module performs group correlation analysis, the common template construction module removes common influences, the individual sequence analysis module extracts individual characteristics, and the individual anomaly diagnosis module completes the final judgment and issues instructions. This system provides intelligent traceability auxiliary decision support for dispatching and operation personnel. Attached Figure Description

[0027] Figure 1 This is a flowchart of an intelligent tracing method for power dispatching systems according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the process of identifying fans with response deviations that require tracing, as described in an embodiment of the present invention. Figure 3 This is a flowchart illustrating how to determine whether the response capability of the fan with the response deviation conforms to the historical normal fluctuation range in an embodiment of the present invention. Figure 4 This is a flowchart for determining whether there are common factors affecting the invention. Detailed Implementation

[0028] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0029] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0030] It should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0031] Please see Figure 1 The diagram shows a flowchart of an intelligent tracing method for power dispatching systems according to an embodiment of the present invention. The intelligent tracing method for power dispatching systems according to an embodiment of the present invention includes: Step S1: Obtain the target increase value and actual increase value of each wind turbine issued by the power dispatch system in order to identify the wind turbines with response deviations that need to be traced. Step S2: Obtain the response process data of the current cycle of the response deviation fan to generate the current response time sequence, wherein the response process data includes output data; Step S3: Evaluate the degree of deviation of the response deviation fan based on the current response time sequence and the historical response time sequence, and determine whether the response capability of the response deviation fan conforms to the historical normal fluctuation range by comparing the degree of deviation with a preset deviation threshold. Step S4: In response to the fact that the response capability of the response deviation fan does not conform to the historical normal fluctuation range, based on the current correlation characteristics and historical correlation characteristics of the response deviation fan and the adjacent response deviation fan, identify whether the response deviation fan and the adjacent response deviation fan have consistency, so as to determine whether there are common factors affecting it. Step S5: Based on the determination result of the existence of common factors, determine the response increment sequence according to the response amplitude of the current response time sequence of adjacent response deviation fans, so as to determine the common response increment sequence; Step S6: Based on the common response increment sequence and the rated power of the response deviation fan, reconstruct the common response theoretical sequence of the response deviation fan; Step S7: Determine the individual response sequence based on the individual deviation value between the current response time sequence of the response deviation fan and the common response theoretical sequence; Step S8: Based on the absolute extreme value of the individual response sequence and the corresponding duration, identify whether the response deviation fan meets the superimposed individual anomaly to identify the abnormal period, and issue a deep inspection command to the individual abnormal fan based on the abnormal period.

[0032] It is understood that this invention can be applied to the new energy power grid dispatch automation system. When a wind farm receives an automatic dispatch power generation increase instruction but the actual increase does not meet the instruction requirements, the traceability process is automatically initiated to analyze the response data of each wind turbine in the station, identify wind turbines with response deviations, and further distinguish whether the deviation is caused by a common factor affecting the group of wind turbines. Finally, the specific time period of the abnormality is located and a deep inspection instruction is issued to guide the operation and maintenance personnel to conduct accurate troubleshooting.

[0033] Please see Figure 2 As shown, it is a flowchart for identifying the response deviation fan that needs to be traced in an embodiment of the present invention.

[0034] Specifically, in step S1, the process of identifying the response deviation fan that needs to be traced includes: Step S11: Based on the difference between the target increase value of each wind turbine issued by the power dispatch system and the actual increase value of each wind turbine, determine the response deviation value of each wind turbine. Step S12: Identify the fans whose response deviation values ​​are greater than the preset response deviation threshold as fans with response deviations that need to be traced.

[0035] In this embodiment, the target power increase value issued to each wind turbine in the wind farm during the current dispatch is obtained from the real-time data of the power dispatch system. The unit of the target power increase value is megawatts or kilowatts. The actual power increase value of each wind turbine in this dispatch event is also obtained. The actual power increase value is defined as the change in output of the wind turbine from the steady-state output within five seconds before the command is issued to the end of the dispatch, and the unit is megawatts or kilowatts. The steady-state output is defined as the average output data within five seconds before the command is issued. For each wind turbine, its response deviation value is calculated, which is the target power increase value minus the actual power increase value. The sign of the response deviation value indicates whether the actual power increase of the wind turbine is lower or higher than the target value. The preset response deviation threshold can be set according to the rated power of the wind turbine, for example, one percent of the rated power. Wind turbines with response deviation values ​​greater than the preset threshold are marked as response deviation wind turbines requiring traceability, forming a set of response deviation wind turbines to be analyzed. This step narrows the focus of the investigation from all wind turbines in the field to individual turbines that exhibit significant response deviations, laying the foundation for further detailed analysis.

[0036] In this embodiment, in step S2, for each identified wind turbine with response deviation, a current response time sequence generation step is performed. The response process data includes output data and wind speed data. Output data refers to the active power delivered by the wind turbine to the grid during operation, measured in kilowatts or megawatts. Taking the time when the dispatch instruction for this scheduling event is issued as the starting point and the time when the dispatch ends as the ending point, the active power output data of the wind turbine is extracted, forming a curve with time on the horizontal axis and output value on the vertical axis; this is the current response time sequence. This current response time sequence completely records the entire behavior of the wind turbine during the current dispatch event and serves as the benchmark data for all subsequent comparative analyses.

[0037] Please see Figure 3 As shown, it is a flowchart of an embodiment of the present invention for determining whether the response capability of the response deviation fan conforms to the historical normal fluctuation range.

[0038] Specifically, in step S3, the process of determining whether the response capability of the response deviation fan conforms to the historical normal fluctuation range includes: Step S31: Obtain the response process data of the fan with response deviation in the historical cycle to generate several historical response time series sequences; Step S32: Calculate the mean output and standard deviation at each time point based on several historical response time series. Step S33: Calculate the degree of deviation of the response deviation fan based on the output data of the current response time sequence and the average output value; Step S34: In response to the deviation being greater than a preset deviation threshold, it is determined that the response capability of the fan with the response deviation does not conform to the historical normal fluctuation range. In response to the deviation being less than or equal to a preset deviation threshold, it is determined that the response capability of the fan with the response deviation conforms to the historical normal fluctuation range.

[0039] In this embodiment, for each wind turbine with response deviation, the steps of constructing a historical response time series and evaluating the current deviation are performed. All event records of the wind turbine that received the same scheduling increase command and were operating normally within the past three months are obtained. For each historical event, the same steps as in step S2 are used to form a historical response time series. The response time series of all the above historical events are aligned on the time axis to form a historical response time series set for the wind turbine. For all data points corresponding to each time point in this series set, the average output and standard deviation of all historical sequences at each time point are calculated. The average value represents the typical response sequence that the wind turbine should have under normal conditions, and the standard deviation represents its normal fluctuation range. The current response time series is compared with the data at each time point in the typical response series. The square of the difference between the current output value and the historical average value at each time point is calculated. The sum of the squares is divided by the total number of time points, and the square root of the result is taken to obtain the root mean square error as the overall deviation index. The preset deviation threshold can be set based on the historical standard deviation, such as three times the average historical standard deviation. If the calculated root mean square error is greater than the preset threshold, the current response capability of the wind turbine is determined to be inconsistent with the historical normal fluctuation range, that is, there is a suspicion of its own abnormality, and it needs to proceed to the next step of analysis; otherwise, it is determined to be normal, and its response deviation may be due to the allocation target exceeding the physical capacity, and then an instruction to check the allocation target is issued.

[0040] Please see Figure 4 As shown, it is a flowchart for determining whether there are common factors affecting the invention.

[0041] Specifically, in step S4, the process of determining whether there are common factors influencing the outcome includes: Step S41: Calculate the correlation coefficient between the current response time series of the response deviation fan and the current response time series of the adjacent response deviation fan to determine the current correlation characteristics; Step S42: Calculate the correlation coefficient between the historical response time series of the response deviation fan and the historical response time series of adjacent response deviation fans to determine several historical correlation characteristics; Step S43: Determine the preset range of association features based on each of the historical association features; Step S44: In response to the current associated feature being within the preset range of the associated feature, if it is determined that the response deviation fan and the adjacent response deviation fan are consistent, then it is determined that there is a common factor affecting them. If the current associated feature does not fall within the preset range of the associated features, and it is determined that the response deviation fan and the adjacent response deviation fan are not consistent, then it is determined that there are no common factors affecting the fan.

[0042] In this embodiment, for wind turbines with response deviations determined to be outside the historical normal fluctuation range, a common factor influence determination step is performed. The adjacent wind turbines with response deviations are identified, and a set of wind turbines with response deviations within a geographical distance of less than one kilometer and belonging to the same category identified in step S1 are obtained. For each pair of wind turbines with response deviations and their adjacent wind turbines, the current correlation feature is calculated. The current correlation feature is the output change correlation coefficient, calculated by taking the current response time series of the two wind turbines and calculating the Pearson correlation coefficient between the two series. The coefficient ranges from -1 to 1; the closer it is to 1, the more synchronized the output change trends of the two wind turbines. The response time series of the wind turbine with the same adjacent wind turbine in the past three months are obtained, and the correlation coefficient in each historical scheduling is calculated using the same method. These historical values ​​are statistically analyzed to obtain the historical mean and historical standard deviation of the correlation coefficient. Based on the historical statistical results, a preset range for the correlation feature is set, with the preset range for the correlation coefficient being within the historical mean plus or minus twice the historical standard deviation. If the response deviation fan and most of its adjacent deviation fans (e.g., more than 70% of the adjacent fans) simultaneously meet the conditions that the current correlation coefficient and the current response time difference are within the preset range, then it is determined that there are common factors affecting the fan, such as environmental factors like a sudden drop in regional wind speed or fluctuations in grid voltage; otherwise, it is determined that there are no common factors affecting the fan, and the fan's abnormality is caused by its own reasons, and a deep inspection command can be issued directly.

[0043] Specifically, in step S5, the process of determining the common response increment sequence includes: Step S51: Calculate the output increment corresponding to each data point based on the current response time sequence of each adjacent response deviation fan and determine it as the response amplitude; Step S52: Generate a response increment sequence based on the ratio of the response amplitude to the rated power of the wind turbine; Step S53: Based on the median of the response increment sequence of several adjacent response deviation fans at each time point, determine the common response increment sequence.

[0044] In this embodiment, for each adjacent deviating fan of the response deviation fan, its steady-state output within five seconds before the command is issued is determined. The steady-state output is defined as the average output data within five seconds before the command is issued. For each data point in the response time sequence, the output increment of the fan is calculated. The output increment is the current output minus the steady-state output, reflecting the response amplitude of the fan under the influence of common factors. This output increment is divided by the rated power of the fan, which is the maximum continuous output power indicated on the fan's nameplate, to obtain the normalized increment. The normalized increment is dimensionless and represents the ratio of the fan's current response amplitude to its rated capacity, generating the response increment sequence of the fan. The normalized increments of the response increment sequences of all adjacent deviating fans at each time moment are statistically analyzed, and the median is taken as the common response increment value at that time moment. Based on the common response increment value, a common response increment sequence is determined. This sequence represents the common relative response pattern of this group of adjacent fans under the influence of common factors.

[0045] Specifically, in step S6, the process of restoring the common response theoretical sequence of the response deviation fan includes: Step S61: Determine the theoretical sequence of response increments based on the product of the common response increment sequence and the rated power of the response deviation fan; Step S62: Based on the sum of the theoretical sequence of the response increment and the corresponding steady-state output, determine the common response theoretical sequence of the response deviation fan.

[0046] In this embodiment, the steady-state output of the wind turbine in the five seconds prior to the issuance of the command is obtained. The steady-state output is the average output data of the wind turbine in the five seconds prior to the command issuance. The rated power of the wind turbine is obtained, and each data point in the common response increment sequence is multiplied by the rated power of the wind turbine to obtain the theoretical response increment sequence. This sequence represents the output increment that the wind turbine should have at each moment if it responds according to the average relative amplitude of the adjacent group. The steady-state output of the wind turbine before the command issuance is added to each data point in the theoretical response increment sequence to obtain the common response theoretical sequence. This common response theoretical sequence describes the theoretical output curve of the wind turbine if it is only affected by common factors.

[0047] Specifically, in step S7, the process of determining the individual response sequence includes: Step S71: Calculate the difference between the current response time sequence of the response deviation fan and the data points corresponding to each data point of the common response theoretical sequence to determine the individual deviation value; Step S72: Determine the individual response sequence based on the individual deviation value.

[0048] In this embodiment, the current response time sequence of the wind turbine with response deviation is compared with each data point in the common response theoretical sequence obtained in S6. The current output value is subtracted from the common response theoretical value to obtain the individual deviation value at that moment. A positive individual deviation value indicates that the actual output is higher than the common expectation, and a negative value indicates that the actual output is lower than the common expectation. The individual deviation values ​​at all moments are arranged in chronological order to form the individual response sequence. This sequence completely eliminates the influence of common factors on the wind turbine's response behavior, retaining only the unique individual response characteristics of the wind turbine that distinguish it from the group. It is the core basis for subsequent judgment on whether there is individual anomaly.

[0049] Specifically, in step S8, the process of identifying whether the response deviation fan conforms to the superimposed individual anomalies includes: Step S81: Obtain the individual deviation value with the largest absolute value in the individual response sequence and determine it as the absolute extreme value; Step S82: In response to the absolute extreme value being greater than a preset individual deviation threshold, obtain the duration for which the individual deviation value in the individual response sequence is greater than the preset individual deviation threshold. Step S83: Based on the determination result that the duration is greater than the preset duration threshold, determine that the response deviation fan conforms to the superimposed individual anomaly.

[0050] In this embodiment, the individual response sequence is analyzed to obtain the individual deviation value with the largest absolute value, defined as the absolute extreme value. A preset individual deviation threshold can be set based on the statistical characteristics of the wind turbine's historical individual deviation sequences, for example, five times the standard deviation of historical individual deviations. The absolute extreme value is compared with the preset individual deviation threshold. If the absolute extreme value is greater than the threshold, it is initially suspected to be consistent with an individual anomaly; if the absolute extreme value is less than or equal to the threshold, it is determined not to be consistent with an individual anomaly. Further, the entire individual response sequence is scanned, and the total duration for which the absolute value of the individual deviation value is continuously greater than the preset threshold is counted, i.e., the duration. The preset duration threshold is based on the ninetieths of the historical duration under normal operating conditions of the wind turbine. If the duration is greater than the preset duration threshold, the wind turbine is ultimately determined to be consistent with a superimposed individual anomaly, meaning that in addition to the influence of common factors, the wind turbine also exhibits unique abnormal characteristics exceeding the normal range; if the duration is less than or equal to the preset duration threshold, the wind turbine is ultimately determined not to be consistent with a superimposed individual anomaly, and only the influence of common factors exists.

[0051] Specifically, in step S8, the process of identifying abnormal time periods includes: In response to the superimposed individual anomalies of the response deviation fan, several time periods in the individual sequence where the individual deviation value is greater than a preset individual deviation threshold are obtained to determine the abnormal time periods.

[0052] In this embodiment, after confirming the superimposed individual anomalies, the individual response sequence is further segmented and scanned. Each interval where the absolute value of the individual deviation value continuously exceeds a preset threshold is marked as a candidate anomaly time period. The start time, end time, maximum absolute deviation value within the segment, and duration of each candidate anomaly time period are recorded. From all candidate anomaly time periods, the most noteworthy time period is selected as the focus of in-depth inspection. The selection rule is to prioritize the time period with the largest maximum absolute deviation value within the segment and the time period with the longest duration, thus identifying it as an anomaly time period. Finally, a deep inspection instruction is generated, and the output includes the abnormal fan number, the conclusion of the common factor judgment, and the anomaly time period requiring attention, which is then sent to the operation and maintenance management system.

[0053] Specifically, an embodiment of the present invention provides an intelligent tracing system for power dispatching systems, comprising: The deviation wind turbine identification module is used to obtain the target increase value and the actual increase value of each wind turbine issued by the power dispatch system, so as to identify the wind turbines with response deviations that need to be traced. The longitudinal comparison module for the deviation fan is used to obtain the response process data of the current cycle of the response deviation fan to generate the current response time sequence, and to evaluate the deviation degree of the response deviation fan based on the current response time sequence and the historical response time sequence, so as to determine whether the response capability of the response deviation fan conforms to the historical normal fluctuation range based on the comparison of the deviation degree with the preset deviation degree threshold. The common factor judgment module responds to the fact that the response capability of the response deviation fan does not conform to the historical normal fluctuation range. Based on the current association characteristics and historical association characteristics of the response deviation fan and the adjacent response deviation fans, it identifies whether the response deviation fan and the adjacent response deviation fans have consistency, so as to determine whether there are common factors affecting it. The common template construction module determines the common response increment sequence based on the determination result of the existence of common factors, the response increment sequence is determined according to the response amplitude of the current response time sequence of adjacent response deviation fans, and the common response increment sequence is determined. Based on the common response increment sequence and the rated power of the response deviation fan, the common response theoretical sequence of the response deviation fan is restored. The individual sequence analysis module determines the individual response sequence based on the individual deviation value between the current response time sequence of the response deviation fan and the common response theoretical sequence; The individual anomaly identification module identifies whether the response deviation fan conforms to the superimposed individual anomaly based on the absolute extreme value of the individual response sequence and the corresponding duration, thereby identifying the abnormal period. Based on the abnormal period, it issues a deep inspection command to the individual abnormal fan.

[0054] It is understood that the aforementioned intelligent tracing system for power dispatching systems is used to execute the aforementioned intelligent tracing method for power dispatching systems. The devices and methods described in the embodiments can also be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the connection may be through some interfaces, indirect coupling or communication connection of devices or modules, and may be electrical, mechanical or other forms.

[0055] The modules described as two components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to implement the solution of this embodiment according to actual needs. The functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The units composed of the above modules can be implemented in hardware or in a combination of hardware and software functional units.

[0056] The integrated modules implemented as software functional modules described above can be stored in a computer-readable storage medium. These software functional modules, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods described in the various embodiments of this application.

[0057] It is understood that the module implementing the above functions can be a processor, which can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as being executed by a hardware processor, or being executed by a combination of hardware and software modules within the processor.

[0058] The storage function can be a memory, including high-speed random access memory (RAM), and also non-volatile memory (NVM), such as at least one disk storage device, or a USB flash drive, external hard drive, read-only memory, disk or optical disc, etc.

[0059] The storage medium described above, which provides storage functionality, 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. The storage medium can be any available medium accessible to general-purpose or special-purpose computers.

[0060] In implementation, the storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium. Alternatively, the storage medium can be an integral part of the processor. Both the processor and storage medium can reside within an application-specific integrated circuit (ASIC). Alternatively, the processor and storage medium can exist as discrete components within an electronic device or host device.

[0061] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An intelligent source tracing method for power dispatching systems, characterized in that, include: Obtain the target increase value and actual increase value of each wind turbine issued by the power dispatch system in order to identify the wind turbines with response deviations that need to be traced. The response process data of the current cycle of the response deviation fan is obtained to generate the current response time sequence, wherein the response process data includes output data; The deviation of the response deviation fan is evaluated based on the current response time sequence and the historical response time sequence. Based on the comparison of the deviation degree with a preset deviation degree threshold, it is determined whether the response capability of the response deviation fan conforms to the historical normal fluctuation range. In response to the fact that the response capability of the fan with the response deviation does not conform to the historical normal fluctuation range, based on the current correlation characteristics and historical correlation characteristics of the fan with the response deviation and adjacent fans with the response deviation, it is identified whether the fan with the response deviation and adjacent fans with the response deviation have consistency, so as to determine whether there are common factors affecting it. Based on the determination of common factors, the response increment sequence is determined according to the response amplitude of the current response time sequence of adjacent response deviation fans, so as to determine the common response increment sequence. The common response incremental sequence and the rated power of the response deviation fan are used to reconstruct the common response theoretical sequence of the response deviation fan; Based on the individual deviation value between the current response time sequence of the aforementioned response deviation fan and the common response theoretical sequence, a unique response sequence is determined; Based on the absolute extreme values ​​of the individual response sequence and the corresponding duration, identify whether the response deviation fan meets the superimposed individual anomaly to identify the abnormal period, and issue a deep inspection command to the individual abnormal fan based on the abnormal period.

2. The intelligent tracing method for power dispatching systems according to claim 1, characterized in that, The process of identifying fans with response deviations that require tracing includes: Based on the difference between the target increase value of each wind turbine issued by the power dispatch system and the actual increase value of each wind turbine, the response deviation value of each wind turbine is determined. Fans with response deviation values ​​greater than a preset response deviation threshold are identified as fans with response deviations that require tracing.

3. The intelligent tracing method for power dispatching systems according to claim 2, characterized in that, The process of determining whether the response capability of the fan with the aforementioned response deviation conforms to the historical normal fluctuation range includes: Obtain response process data of the wind turbine with response deviation in historical cycles to generate several historical response time series sequences; The mean output and standard deviation at each time point are calculated based on several historical response time series. The degree of deviation of the response deviation fan is calculated based on the output data of the current response time sequence and the average output value; In response to the deviation being greater than a preset deviation threshold, it is determined that the response capability of the fan with the deviation does not conform to the historical normal fluctuation range.

4. The intelligent tracing method for power dispatching systems according to claim 3, characterized in that, The process of determining whether common factors are at play includes: Calculate the correlation coefficient between the current response time series of the response deviation fan and the current response time series of adjacent response deviation fans to determine the current correlation characteristics; Calculate the correlation coefficient between the historical response time series of the response deviation fan and the historical response time series of adjacent response deviation fans to determine several historical correlation characteristics; The preset range of association features is determined based on each of the aforementioned historical association features; If the current associated feature falls within the preset range of the associated feature, and it is determined that the response deviation fan and the adjacent response deviation fan are consistent, then it is determined that there is a common factor affecting the fan.

5. The intelligent tracing method for power dispatching systems according to claim 4, characterized in that, The process of determining the common response increment sequence includes: The output increment corresponding to each data point is calculated based on the current response time sequence of each adjacent response deviation fan to determine the response amplitude; A response increment sequence is generated based on the ratio of the response amplitude to the rated power of the wind turbine; Based on the median of the response increment sequence at each time point of several adjacent response deviation wind turbines, the common response increment sequence is determined.

6. The intelligent tracing method for power dispatching systems according to claim 5, characterized in that, The process of reconstructing the common response theoretical sequence of the aforementioned response deviation fan includes: The theoretical sequence of response increments is determined based on the product of the common response increment sequence and the rated power of the response deviation fan. Based on the sum of the theoretical sequence of the response increment and the corresponding steady-state output, the common theoretical sequence of the response deviation fan is determined.

7. The intelligent tracing method for power dispatching systems according to claim 6, characterized in that, The process of determining a personality response sequence includes: Calculate the difference between the current response time sequence of the response deviation fan and the data points corresponding to each data point of the common response theoretical sequence to determine the individual deviation value; The individual response sequence is determined based on the individual deviation value.

8. The intelligent tracing method for power dispatching systems according to claim 7, characterized in that, The process of identifying whether the response deviation fan conforms to the superimposed individual anomalies includes: The individual deviation value with the largest absolute value in the individual response sequence is determined as the absolute extreme value. In response to the absolute extreme value being greater than a preset individual deviation threshold, the duration for which the individual deviation value in the individual response sequence is greater than the preset individual deviation threshold is obtained. Based on the determination result that the duration is greater than the preset duration threshold, it is determined that the response deviation fan conforms to the superimposed individual anomaly.

9. The intelligent tracing method for power dispatching systems according to claim 8, characterized in that, The process of identifying abnormal time periods includes: In response to the superimposed individual anomalies of the response deviation fan, several time periods in the individual sequence where the individual deviation value is greater than a preset individual deviation threshold are obtained to determine the abnormal time periods.

10. An intelligent tracing system for power dispatching systems, applied to the intelligent tracing method for power dispatching systems as described in any one of claims 1-9, characterized in that, include, The deviation wind turbine identification module is used to obtain the target increase value and the actual increase value of each wind turbine issued by the power dispatch system, so as to identify the wind turbines with response deviations that need to be traced. The longitudinal comparison module for the deviation fan is used to obtain the response process data of the current cycle of the response deviation fan to generate the current response time sequence, and to evaluate the deviation degree of the response deviation fan based on the current response time sequence and the historical response time sequence, so as to determine whether the response capability of the response deviation fan conforms to the historical normal fluctuation range based on the comparison of the deviation degree with the preset deviation degree threshold. The common factor judgment module responds to the fact that the response capability of the response deviation fan does not conform to the historical normal fluctuation range. Based on the current association characteristics and historical association characteristics of the response deviation fan and the adjacent response deviation fan, it identifies whether the response deviation fan and the adjacent response deviation fan have consistency, so as to determine whether there are common factors affecting it. The common template construction module determines the common response increment sequence based on the response amplitude of the current response time sequence of adjacent response deviation fans, based on the judgment result of the existence of common factors, and determines the common response increment sequence. It also restores the common response theoretical sequence of the response deviation fan based on the common response sequence and the rated power of the response deviation fan. The individual sequence analysis module determines the individual response sequence based on the individual deviation value between the current response time sequence of the response deviation fan and the common response theoretical sequence; The individual anomaly identification module identifies whether the response deviation fan conforms to the superimposed individual anomaly based on the absolute extreme value of the individual response sequence and the corresponding duration, thereby identifying the abnormal period. Based on the abnormal period, it issues a deep inspection command to the individual abnormal fan.

Images