Power grid data fault diagnosis method based on intelligent power monitoring instrument

Abstract

The invention relates to the technical field of power system monitoring, and provides a power grid data fault diagnosis method and device based on an intelligent power monitoring instrument, and the method comprises the steps: collecting the operation data of a power grid through the intelligent power monitoring instrument, and generating a fault feature vector containing the multi-dimensional feature parameters of voltage, current and power factor; matching and comparing the feature vector with a preset fault criterion; determining a specific fault type based on a comparison result of the characteristic parameters and a threshold value, including identifying voltage abnormity through a voltage deviation rate, identifying harmonic exceeding through a current harmonic total distortion rate, and identifying a three-phase imbalance fault through a three-phase imbalance degree; for judgment of the three-phase unbalance degree, mean square error statistic calculation is adopted, and a safety coefficient is introduced for threshold comparison; and the fault influence priority is evaluated by quantifying the degree of deviation of each characteristic parameter from the threshold. The fault diagnosis efficiency and reliability can be improved, and intelligent decision support is provided for safe operation of a power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of power system monitoring, and more specifically, the present invention relates to a power grid data fault diagnosis method based on intelligent power monitoring instruments. Background Art

[0002] With the continuous development of the power system, the complexity and reliability requirements of the power grid are getting higher and higher. As an important part of the modern power grid, intelligent power monitoring instruments can collect real-time power grid operation data, such as key parameters like voltage, current, power factor, etc., providing a rich data basis for the operation monitoring and fault diagnosis of the power grid. Traditional power grid fault diagnosis methods mainly rely on manual experience or simple threshold judgment, and these methods have many limitations when facing complex power grid faults. For example, manual experience is difficult to handle sudden complex faults, and simple threshold judgment is easily affected by changes in the power grid operation environment, resulting in misjudgment or missed judgment. In addition, although some existing fault diagnosis technologies can identify single fault types, they often cannot accurately judge the weights and mutual relationships of various faults when facing compound faults, thus affecting the timeliness and accuracy of fault handling.

[0003] In the process of implementing the embodiments of the present invention, the inventors found that there are at least the following problems or defects in the prior art: the existing fault diagnosis methods do not extract the features of power grid operation data comprehensively enough and cannot make full use of the multi-dimensional data collected by intelligent power monitoring instruments; when determining the fault type, there is a lack of an effective identification and weight calculation mechanism for compound faults, resulting in insufficient diagnostic accuracy in complex fault scenarios; in addition, the methods for generating and processing fault feature vectors in the prior art are relatively single and cannot adapt to dynamic changes such as power grid load fluctuations, thus affecting the real-time performance and reliability of fault diagnosis. Summary of the Invention

[0004] The present invention provides a power grid data fault diagnosis method, device and equipment based on intelligent power monitoring instruments.

[0005] In the first aspect of the present invention, a power grid data fault diagnosis method based on intelligent power monitoring instruments is provided, including:

[0006] Generating a fault feature vector according to the power grid operation data collected by the intelligent power monitoring instrument, where the fault feature vector includes multi-dimensional feature parameters of voltage, current and power factor;

[0007] Determining the power grid fault type according to the matching relationship between the fault feature vector and the preset fault criterion, where the preset fault criterion includes the determination thresholds of short-circuit fault criterion, grounding fault criterion and overload fault criterion.

[0008] Further, determining the power grid fault type includes:

[0009] Step S100: Determine whether the voltage deviation rate in the fault feature vector exceeds a first threshold. If it exceeds, it is determined as a voltage abnormal fault; otherwise, proceed to step S200;

[0010] Step S200: Determine whether the total harmonic distortion rate of the current in the fault feature vector exceeds a second threshold. If it exceeds, it is determined as a harmonic over-standard fault; otherwise, proceed to step S300;

[0011] Step S300: Determine whether the three-phase unbalance degree in the fault feature vector exceeds a third threshold. If it exceeds, it is determined as a three-phase unbalance fault; otherwise, it is determined that the power grid is operating normally.

[0012] Further, determining whether the three-phase unbalance degree exceeds the third threshold includes:

[0013] Calculate the mean square deviation statistic of each phase current in the fault feature vector. When the mean square deviation statistic satisfies the following formula, it is determined to exceed the third threshold:

[0014] σ abc >K·σ baSe

[0015] where σ abc represents the mean square deviation statistic of the three-phase current, σ base represents the reference unbalance degree threshold, and K represents a preset safety factor.

[0016] Further, calculating the mean square deviation statistic of each phase current in the fault feature vector includes:

[0017] Step S10: Segment the current data collected by the intelligent power monitoring instrument according to a set time window;

[0018] Step S20: Calculate the root mean square values of phase A, phase B, and phase C for each segment of current data respectively;

[0019] Step S30: Calculate the mean square deviation statistic according to the dispersion degree of the three-phase root mean square values.

[0020] Further, the length of the set time window is determined according to the power grid load fluctuation period, specifically satisfying:

[0021] T w =α·T load

[0022] where T w represents the time window length, T load represents the power grid load characteristic period, and α represents an adjustment coefficient with a value range of 0.5 - 2.0.

[0023] Further, generating the fault feature vector includes:

[0024] Step B100: Obtain the voltage instantaneous value sequence V from the intelligent power monitoring instrument t , the current instantaneous value sequence I t and the power factor sequence PF t ;

[0025] Step B200: Perform moving average filtering on the voltage instantaneous value sequence to obtain the voltage characteristic parameter V avg ;

[0026] Step B300: Perform fast Fourier transform on the current instantaneous value sequence and extract the harmonic characteristic parameter H thd ;

[0027] Step B400: Calculate the statistical variance of the power factor sequence to obtain the power factor fluctuation parameter σ pf ;

[0028] Step B500: Combine V avg , H thd and σ pf to generate the fault feature vector.

[0029] Further, the moving average filtering satisfies:

[0030]

[0031] where k represents the sampling point sequence number variable, V t-kΔt represents the voltage instantaneous value of the kth sampling point before the tth moment, V avg represents the filtered average voltage value, N represents the number of sampling points in the moving window, and Δt represents the sampling interval time.

[0032] Further, it also includes:

[0033] When it is determined that there is a compound fault, calculate the fault weight according to the correlation degree of each parameter in the fault feature vector, specifically satisfying:

[0034]

[0035] where W i represents the weight coefficient of the ith type of fault, P i represents the actual measured value of the ith type of characteristic parameter, Th i represents the determination threshold of the ith type of fault, j represents the fault type sequence number variable, Th j represents the determination threshold of the jth type of fault, and n represents the total number of preset fault types.

[0036] In a second aspect of the present invention, there is provided a power grid data fault diagnosis device based on an intelligent power monitoring instrument, including:

[0037] A feature extraction module configured to generate a fault feature vector based on the power grid operation data collected by the intelligent power monitoring instrument;

[0038] A fault determination module configured to determine the type of power grid fault according to the matching relationship between the fault feature vector and a preset fault criterion.

[0039] In a third aspect of the present invention, there is provided a power monitoring terminal device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the power grid data fault diagnosis method according to any one of the first aspect.

[0040] According to the above embodiments of the present invention, there are at least the following beneficial effects: The power grid data fault diagnosis method of the present invention can make full use of the multi-dimensional data collected by the intelligent power monitoring instrument to generate a fault feature vector. Through comprehensive analysis of multi-dimensional characteristic parameters such as voltage, current, and power factor, it can comprehensively reflect the operation state of the power grid and improve the accuracy of fault diagnosis. At the same time, this method can quickly determine the type of power grid fault according to the matching relationship between the fault feature vector and the preset fault criterion, including various fault types such as short-circuit faults, grounding faults, overload faults, voltage abnormalities, harmonic over-limit, and three-phase imbalance, effectively improving the efficiency and reliability of fault diagnosis. In addition, for complex fault scenarios, the present invention can also calculate the fault weight according to the correlation degree of each parameter in the fault feature vector, provide a more accurate basis for fault handling, further optimize the fault handling process, and reduce power outage time and economic losses.

[0041] The power grid data fault diagnosis device and the power monitoring terminal device of the present invention can realize the automation and intelligence of fault diagnosis. Through the collaborative work of the feature extraction module and the fault determination module, a fault feature vector can be quickly generated and the fault type can be determined without manual intervention, which can improve the efficiency and timeliness of fault diagnosis. At the same time, this device and equipment can dynamically adjust the time window length according to the power grid load fluctuation period to ensure that the generation of the fault feature vector can adapt to the change of the power grid operation state, further improving the real-time performance and reliability of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown in an exemplary rather than restrictive manner, wherein:

[0043] Figure 1 A schematic flowchart of a power grid data fault diagnosis method based on an intelligent power monitoring instrument provided by an embodiment of the present invention;

[0044] Figure 2 A schematic structural diagram of a power grid data fault diagnosis device based on an intelligent power monitoring instrument provided by an embodiment of the present invention;

[0045] Figure 3 A schematic structural diagram of an electronic device according to an embodiment of the present invention is schematically shown. Specific embodiments

[0046] Hereinafter, the principles and spirit of the present invention will be described with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and then implement the present invention, rather than limiting the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0047] Those skilled in the art know that the embodiments of the present invention can be implemented as a system, a device, a device, a method, or a computer program product. Therefore, the present invention can be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0048] It should be noted that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction and does not have any limiting meaning.

[0049] Hereinafter, reference is made to Figure 1 , Figure 1 A schematic flowchart of a power grid data fault diagnosis method based on an intelligent power monitoring instrument provided by an embodiment of the present invention. As Figure 1 shown, a power grid data fault diagnosis method based on an intelligent power monitoring instrument includes:

[0050] Generating a fault feature vector according to the power grid operation data collected by the intelligent power monitoring instrument, where the fault feature vector includes multi-dimensional feature parameters of voltage, current, and power factor;

[0051] Determining the type of power grid fault according to the matching relationship between the fault feature vector and a preset fault criterion, where the preset fault criterion includes determination thresholds of a short-circuit fault criterion, a ground fault criterion, and an overload fault criterion.

[0052] It should be noted that the present invention proposes a method for diagnosing grid data faults based on intelligent power monitoring instruments. This method collects grid operation data through intelligent power monitoring instruments and generates a fault feature vector containing multi-dimensional characteristic parameters such as voltage, current, and power factor. Among them, voltage refers to the potential difference in power transmission in the grid, current is the flow of charge in the circuit, and power factor is the ratio of the actual power to the apparent power in the grid, reflecting the power consumption efficiency of the grid. By matching the fault feature vector with the preset fault criteria, the type of grid fault can be determined. The preset fault criteria include the determination thresholds of short-circuit fault criteria, grounding fault criteria, and overload fault criteria, and these thresholds are set according to the safety standards and empirical data of grid operation and are used to judge whether a grid fault occurs and the type of the fault.

[0053] Specifically, the fault feature vector mentioned in the present invention is generated by processing the grid operation data collected by the intelligent power monitoring instrument. Among them, the first threshold is the determination threshold of the voltage deviation rate and is used to judge whether the voltage is abnormal; the second threshold is the determination threshold of the total harmonic distortion rate of the current and is used to judge whether the harmonics exceed the standard; the third threshold is the determination threshold of the three-phase unbalance degree and is used to judge whether the three-phase currents are balanced. These thresholds can be set according to the specific operation conditions and safety standards of the grid. For example, the first threshold of the voltage deviation rate can be set to ±10% of the rated voltage of the grid, the second threshold of the total harmonic distortion rate of the current can be set to 5%, and the third threshold of the three-phase unbalance degree can be set to 2%. In addition, the preset safety factor K is a parameter used to adjust the judgment strictness, and its value range is 0.5 - 2.0, which can be adjusted according to the operation environment and reliability requirements of the grid.

[0054] Preferably, the process of generating the fault feature vector can be further refined. First, obtain the voltage instantaneous value sequence, current instantaneous value sequence, and power factor sequence from the intelligent power monitoring instrument. Then, perform a moving average filtering process on the voltage instantaneous value sequence to smooth the voltage data and extract the voltage characteristic parameters. Moving average filtering is a commonly used signal processing method that eliminates noise and mutations by calculating the average value within a certain time window. Next, perform a fast Fourier transform on the current instantaneous value sequence to extract the harmonic characteristic parameters in the current. The fast Fourier transform is an efficient signal analysis method that can convert the time-domain signal into the frequency-domain signal, thereby extracting the harmonic components in the current. Finally, calculate the statistical variance of the power factor sequence to obtain the power factor fluctuation parameter. The statistical variance is used to measure the fluctuation degree of the power factor and reflects the stability of the grid operation state. By combining these parameters, a complete fault feature vector is generated, providing basic data for subsequent fault diagnosis.

[0055] In some embodiments, determining the type of grid fault includes:

[0056] Step S100: Determine whether the voltage deviation rate in the fault feature vector exceeds a first threshold. If it exceeds, it is determined as a voltage abnormal fault; otherwise, proceed to step S200;

[0057] Step S200: Determine whether the total harmonic distortion rate of the current in the fault feature vector exceeds a second threshold. If it exceeds, it is determined as a harmonic over - standard fault; otherwise, proceed to step S300;

[0058] Step S300: Determine whether the three - phase unbalance degree in the fault feature vector exceeds a third threshold. If it exceeds, it is determined as a three - phase unbalance fault; otherwise, it is determined that the power grid is operating normally.

[0059] It should be noted that when determining the power grid fault type, the power grid data fault diagnosis method of the present invention adopts a set of step - by - step logical processes. First, determine whether the voltage deviation rate in the fault feature vector exceeds the first threshold. If it exceeds, it is determined as a voltage abnormal fault. The voltage deviation rate refers to the deviation degree between the actual voltage of the power grid and the rated voltage, reflecting the stability of the power grid voltage. If the voltage deviation rate does not exceed the first threshold, then further determine whether the total harmonic distortion rate of the current exceeds the second threshold to determine whether there is a harmonic over - standard fault. The total harmonic distortion rate of the current refers to the ratio of the harmonic component to the fundamental component in the current, reflecting the distortion degree of the current waveform. If the total harmonic distortion rate of the current also does not exceed the second threshold, then continue to determine whether the three - phase unbalance degree exceeds the third threshold to determine whether there is a three - phase unbalance fault. The three - phase unbalance degree refers to the unbalance degree between the three - phase currents or voltages, reflecting the symmetry of the three - phase system of the power grid. If the three - phase unbalance degree does not exceed the third threshold, it is determined that the power grid is operating normally. This step - by - step judgment logic can gradually screen out the fault types and improve the accuracy and efficiency of fault diagnosis.

[0060] Specifically, the first threshold is the determination threshold of the voltage deviation rate, used to judge whether the voltage is abnormal. For example, the first threshold can be set to ±10% of the rated voltage of the power grid. When the voltage deviation rate exceeds this value, it is considered that the power grid has a voltage abnormal fault. The second threshold is the determination threshold of the total harmonic distortion rate of the current, used to judge whether the harmonics exceed the standard. For example, the second threshold can be set to 5%. When the total harmonic distortion rate of the current exceeds this value, it is considered that the power grid has a harmonic over - standard fault. The third threshold is the determination threshold of the three - phase unbalance degree, used to judge whether the three - phase currents are balanced. For example, the third threshold can be set to 2%. When the three - phase unbalance degree exceeds this value, it is considered that the power grid has a three - phase unbalance fault. The setting of these thresholds can be adjusted according to the specific operating conditions and safety standards of the power grid to adapt to different power grid environments and fault diagnosis requirements.

[0061] Preferably, when determining whether the three-phase unbalance degree exceeds the third threshold, the calculation process can be further refined. First, calculate the mean square deviation statistic of the current in each phase in the fault feature vector. The mean square deviation statistic is an index to measure the degree of data dispersion. By calculating the degree of dispersion of the root mean square values of the three-phase currents, the specific value of the three-phase unbalance degree can be obtained. When specifically calculating, the current data collected by the intelligent power monitoring instrument can be segmented according to a set time window first, and then the root mean square values of phase A, phase B, and phase C are calculated for each segment of current data respectively. The root mean square value reflects the effective value of the current and is an important parameter to measure the magnitude of the current. Then, calculate the mean square deviation statistic according to the degree of dispersion of the three-phase root mean square values. When the mean square deviation statistic exceeds the reference unbalance degree threshold multiplied by a preset safety factor, it is determined as a three-phase unbalance fault. The reference unbalance degree threshold is a standard value set according to the power grid operation experience and is used to judge whether the three-phase unbalance degree is within the normal range. The preset safety factor is a parameter used to adjust the judgment strictness and can be adjusted according to the operation environment and reliability requirements of the power grid. For example, the value range is 0.5 - 2.0. Through this detailed calculation process, the three-phase unbalance fault can be judged more accurately, and the accuracy of fault diagnosis can be improved.

[0062] In some embodiments, determining whether the three-phase unbalance degree exceeds the third threshold includes:

[0063] Calculating the mean square deviation statistic of the current in each phase in the fault feature vector, and determining that it exceeds the third threshold when the mean square deviation statistic satisfies the following formula:

[0064] σ abc >K·σ base

[0065] Where σ abc represents the mean square deviation statistic of the three-phase current, σ base represents the reference unbalance degree threshold, and K represents the preset safety factor.

[0066] It should be noted that the method for determining whether the three-phase unbalance degree exceeds the third threshold in the present invention is realized by calculating the mean square deviation statistic of the current in each phase in the fault feature vector. The three-phase unbalance degree refers to the degree of imbalance between the three-phase currents and is an important index to measure the symmetry of the three-phase system of the power grid. When the degree of imbalance of the three-phase currents exceeds a certain threshold, it may cause abnormal operation or faults of the power grid equipment. The mean square deviation statistic is a commonly used statistical method to measure the degree of data dispersion. By calculating the mean square deviation statistic of each phase current and comparing it with the product of the reference unbalance degree threshold and the preset safety factor, it can be accurately judged whether the three-phase unbalance degree exceeds the allowable range.

[0067] Specifically, the mean square error statistic is obtained by calculating the degree of dispersion of the root mean square values of each phase current. The reference unbalance threshold is a reference value set based on grid operation experience and safety standards, and is used to determine whether the three-phase unbalance degree is within the normal range. The preset safety factor is an adjustment parameter used to adjust the strictness of the judgment according to the specific operating environment and reliability requirements of the power grid. For example, the value range of the preset safety factor can be from 0.5 to 2.0. During the calculation process, first, the current data collected by the intelligent power monitoring instrument needs to be segmented, and then the root mean square values of phase A, phase B, and phase C in each segment of data are calculated separately. The root mean square value reflects the effective value of the current and is an important parameter for measuring the magnitude of the current. The mean square error statistic obtained by calculating the degree of dispersion of these root mean square values can intuitively reflect the unbalance degree of the three-phase current.

[0068] Preferably, in order to calculate the three-phase unbalance degree more accurately, the segmentation process of the current data can be optimized. The length of the set time window is determined according to the grid load fluctuation period, and specifically can be calculated by multiplying the grid load characteristic period by an adjustment factor. The value range of the adjustment factor is from 0.5 to 2.0, and can be adjusted according to the fluctuation characteristics of the grid load. For example, if the grid load fluctuates greatly, the value of the adjustment factor can be appropriately increased to ensure that the time window can cover enough data points, thereby improving the calculation accuracy. When calculating the root mean square value of each segment of current data, the standard root mean square calculation method can be used, that is, first square the current data, then calculate the average value, and finally take the square root. The root mean square value obtained in this way can accurately reflect the magnitude of each phase current. Then, according to the degree of dispersion of these root mean square values, the mean square error statistic is calculated, specifically by calculating the sum of the squares of the differences between the root mean square values of each phase and the average value, and then taking the square root. When the calculated mean square error statistic exceeds the reference unbalance threshold multiplied by the preset safety factor, it can be determined that the three-phase unbalance degree exceeds the third threshold, thereby confirming that there is a three-phase unbalance fault in the power grid.

[0069] In some embodiments, calculating the mean square error statistic of each phase current in the fault feature vector includes:

[0070] Step S10: Segment the current data collected by the intelligent power monitoring instrument according to the set time window;

[0071] Step S20: Calculate the root mean square values of phase A, phase B, and phase C for each segment of current data respectively;

[0072] Step S30: Calculate the mean square error statistic according to the degree of dispersion of the three-phase root mean square values.

[0073] It should be noted that the process of calculating the mean square deviation statistic of each phase current in the fault feature vector of the present invention includes three main steps: segmenting the current data, calculating the root mean square value of each segment of current data, and calculating the mean square deviation statistic based on the degree of dispersion of the root mean square values. This process aims to accurately evaluate the unbalance degree of the three-phase current by analyzing the fluctuation characteristics of the current data. The mean square deviation statistic is a statistical method used to measure the degree of dispersion of data, and the root mean square value of the current is a quantitative representation of the effective value of the current, reflecting the actual power transmission capacity of the current. Through these steps, key data support can be provided for subsequent fault diagnosis.

[0074] Specifically, for the segmentation of the current data, the length of the set time window is determined according to the grid load fluctuation period, and its purpose is to ensure that the characteristics of the grid load change can be captured during the analysis process. The calculation formula for the time window length is the grid load characteristic period multiplied by an adjustment coefficient, and the value range of the adjustment coefficient is from 0.5 to 2.0, and the specific value can be adjusted according to the actual operation conditions of the grid. For example, if the grid load changes rapidly, the adjustment coefficient can be appropriately reduced to improve the sensitivity of the data. For each segment of current data, the root mean square values of phase A, phase B, and phase C are calculated separately. The calculation method of the root mean square value is to first square the current data, then calculate the average value, and finally take the square root. This process can effectively eliminate the noise and transient interference in the current data, so as to obtain a more accurate effective value of the current. The calculation of the mean square deviation statistic is based on the degree of dispersion of the three-phase root mean square values, and is realized by calculating the sum of the squares of the differences between each phase root mean square value and the average value, and then taking the square root. This statistic can intuitively reflect the unbalance degree between the three-phase currents.

[0075] Preferably, in order to further improve the accuracy and reliability of the current data processing, a moving average filtering process can be introduced when calculating the root mean square value. Moving average filtering is a commonly used signal processing method, which smooths the data by calculating the average value within a certain time window, thereby reducing the influence of noise. In the present invention, the moving average filtering process can be performed on each segment of current data separately, and the window size can be adjusted according to the sampling frequency of the current data and the grid load fluctuation characteristics. For example, for current data with a higher sampling frequency, the window size can be appropriately reduced to improve the real-time performance of the data. When calculating the mean square deviation statistic, the calculation process can be further refined. For example, first calculate the average value of the three-phase current root mean square values, then calculate the square of the difference between each phase root mean square value and the average value, and finally add these squares of the differences and take the square root to obtain the final mean square deviation statistic. Through this detailed calculation process, the unbalance degree of the three-phase current can be more accurately reflected, providing a more reliable basis for grid fault diagnosis.

[0076] In some embodiments, the length of the set time window is determined according to the power grid load fluctuation period, specifically satisfying:

[0077] T w = α·T load

[0078] Where T w represents the time window length, T load represents the power grid load characteristic period, and α represents the adjustment coefficient with a value range of 0.5 - 2.0.

[0079] It should be noted that the set time window length mentioned in the present invention is determined according to the power grid load fluctuation period, and its purpose is to better adapt to the changes in the power grid operation state and ensure that the fault feature vector can accurately reflect the real-time operation of the power grid. The calculation formula for the time window length is the power grid load characteristic period multiplied by an adjustment coefficient. This dynamic adjustment mechanism enables the fault diagnosis method to flexibly respond to the load changes under different power grid environments, thereby improving the accuracy and real-time performance of fault diagnosis.

[0080] Specifically, the power grid load characteristic period refers to the periodic change law presented by the power grid load within a certain period of time, and this parameter can be obtained through the analysis of the power grid historical load data. The adjustment coefficient is a parameter used to optimize the time window length, and its value range is from 0.5 to 2.0. The specific value of the adjustment coefficient can be selected according to the load fluctuation characteristics of the power grid. For example, for a power grid with large load fluctuations, the adjustment coefficient can be appropriately increased to ensure that the time window can cover enough data points, thereby improving the representativeness of the fault feature vector. The calculation formula for the time window length is the power grid load characteristic period multiplied by the adjustment coefficient, and this formula ensures that the time window length can be dynamically adjusted according to the actual operation of the power grid.

[0081] Preferably, when determining the time window length, the calculation process can be further refined. First, analyze the power grid historical load data, extract its periodic change characteristics, and obtain the power grid load characteristic period. Then, select a suitable adjustment coefficient according to the actual operation of the power grid. For example, the optimal adjustment coefficient can be selected by evaluating the accuracy of the fault diagnosis results under different adjustment coefficients. Finally, multiply the power grid load characteristic period by the adjustment coefficient to obtain the set time window length. In practical applications, the calculation of the time window length can be dynamically updated according to the real-time load data of the power grid to ensure that the fault diagnosis method can reflect the operation state of the power grid in real time. Through this dynamic adjustment mechanism, the accuracy and reliability of fault diagnosis can be effectively improved, providing a strong guarantee for the safe operation of the power grid.

[0082] In some embodiments, generating the fault feature vector includes:

[0083] Step B100: Obtain the instantaneous voltage value sequence V from the intelligent power monitoring instrument t , the instantaneous current value sequence I t and the power factor sequence PF t ;

[0084] Step B200: Perform moving average filtering on the instantaneous voltage value sequence to obtain the voltage characteristic parameter V avg ;

[0085] Step B300: Perform fast Fourier transform on the instantaneous current value sequence to extract the harmonic characteristic parameter H thd ;

[0086] Step B400: Calculate the statistical variance of the power factor sequence to obtain the power factor fluctuation parameter σ pf ;

[0087] Step B500: Combine V avg , H thd and σ pf to generate the fault feature vector.

[0088] It should be noted that in the present invention, the process of generating the fault feature vector is realized by performing a series of processes on the grid operation data collected by the intelligent power monitoring instrument. These data include the instantaneous voltage value sequence, the instantaneous current value sequence, and the power factor sequence. Through the processing of these data, key characteristic parameters that can reflect the grid operation state can be extracted, thereby generating the fault feature vector. The fault feature vector is the basis for fault diagnosis. It includes the voltage characteristic parameter, the current harmonic characteristic parameter, and the power factor fluctuation parameter. These parameters can comprehensively reflect the grid operation state and provide an important basis for subsequent fault diagnosis.

[0089] Specifically, the voltage instantaneous value sequence refers to the voltage data continuously collected by the intelligent power monitoring instrument within a certain period of time, and these data reflect the real-time change of the grid voltage. The current instantaneous value sequence refers to the current data collected within the same period of time, reflecting the real-time change of the grid current. The power factor sequence refers to the power factor data collected within the same period of time, reflecting the proportional relationship between the actual power and the apparent power in the grid. In the process of generating the fault feature vector, first, perform a moving average filtering process on the voltage instantaneous value sequence to smooth the voltage data and extract voltage characteristic parameters. Moving average filtering is a commonly used signal processing method that eliminates noise and mutations by calculating the average value within a certain time window. Then, perform a fast Fourier transform on the current instantaneous value sequence to extract the harmonic characteristic parameters in the current. The fast Fourier transform is an efficient signal analysis method that can convert the time-domain signal into a frequency-domain signal, thereby extracting the harmonic components in the current. Finally, calculate the statistical variance of the power factor sequence to obtain the power factor fluctuation parameter. The statistical variance is used to measure the fluctuation degree of the power factor and reflects the stability of the grid operation state. By combining these parameters, a complete fault feature vector is generated.

[0090] Preferably, in the process of generating the fault feature vector, the processing methods of each step can be further refined. For the moving average filtering process of the voltage instantaneous value sequence, a suitable moving window size can be set, for example, determined according to the sampling frequency of the grid and the load fluctuation characteristics. The selection of the window size should be able to smooth the data without losing important voltage change information. When performing a fast Fourier transform on the current instantaneous value sequence, specific frequency band harmonic characteristic parameters can be extracted, such as extracting the 2nd to 50th harmonic components, because these harmonic components usually have a greater impact on the grid operation state. For the calculation of the statistical variance of the power factor sequence, the standard variance calculation method can be used, that is, first calculate the average value of the power factor, then calculate the square of the difference between each power factor data point and the average value, and finally calculate the average value of these squared differences. Through these detailed processing steps, each parameter in the fault feature vector can be extracted more accurately, thereby improving the accuracy and reliability of fault diagnosis.

[0091] In some embodiments, the moving average filtering process satisfies:

[0092]

[0093] where k represents the sampling point sequence number variable, V t-kΔt represents the voltage instantaneous value of the kth sampling point before the t moment, V avg represents the filtered voltage average value, N represents the number of sampling points of the moving window, and Δt represents the sampling interval time.

[0094] It should be noted that the moving average filtering process mentioned in the present invention is used to smooth the instantaneous voltage value sequence to extract voltage characteristic parameters. Moving average filtering is a commonly used data processing method that reduces noise and mutations in data by calculating the average value within a certain time window, thereby obtaining more stable voltage characteristic parameters. This method can effectively reduce the impact of transient interference in power grid operation on fault diagnosis and improve the accuracy and reliability of fault feature vectors.

[0095] Specifically, the process of moving average filtering involves several key parameters. The sampling point serial number variable represents the position in the data sequence and is used to identify each sampling point. The average filtered voltage is obtained by calculating the average value of the instantaneous voltage values of all sampling points within the window, which reflects the average voltage level within the window. The number of sampling points in the moving window determines the size of the window, that is, the number of sampling points included each time the average value is calculated. The sampling interval time represents the time interval between adjacent sampling points, which is related to the sampling frequency and determines the density of the data. In practical applications, the size of the moving window can be adjusted according to the sampling frequency of the power grid and the load fluctuation characteristics. For example, for a power grid with a high sampling frequency, the window size can be appropriately reduced to improve the real-time performance of the data; while for a power grid with large load fluctuations, the window size can be appropriately increased to smooth the data.

[0096] Preferably, when implementing the moving average filtering process, the operation steps can be further refined. First, determine the size of the moving window according to the sampling frequency of the power grid and the load fluctuation characteristics. For example, if the sampling frequency of the power grid is 100Hz, that is, 100 data points are collected per second, a suitable window size, such as 10 sampling points, can be selected according to the periodic characteristics of the load fluctuation. Then, starting from the first sampling point, calculate the average value of the instantaneous voltage values of all sampling points within the window to obtain the first average filtered voltage. Next, move the window backward by one sampling point and repeat the calculation of the average value until all sampling points are processed. In this way, a smooth voltage characteristic parameter sequence can be obtained. This processing method can effectively reduce noise and mutations in voltage data, enabling the fault feature vector to more accurately reflect the actual operating state of the power grid, thereby improving the accuracy and reliability of fault diagnosis.

[0097] In some embodiments, it further includes:

[0098] When it is determined that there is a composite fault, calculate the fault weight according to the correlation degree of each parameter in the fault feature vector, specifically satisfying:

[0099]

[0100] Where W i represents the weight coefficient of the i-th type of fault, and P iRepresents the actual measured value of the i-th type of characteristic parameter, Th i Represents the determination threshold of the i-th type of fault, j represents the variable of the fault type serial number, Th j Represents the determination threshold of the j-th type of fault, and n represents the total number of preset fault types.

[0101] It should be noted that the fault weight calculation method mentioned in the present invention is based on the correlation degree of each parameter in the fault feature vector. This method assigns a weight coefficient to each fault type by quantifying the difference between each fault parameter and the preset fault threshold. The weight coefficient reflects the relative importance of each fault type in the composite fault scenario, thus providing a more accurate basis for fault diagnosis and treatment. This method is particularly applicable to the composite fault scenarios that may occur in the power grid, and can help quickly identify and handle the main faults, improving the reliability and safety of the power grid operation.

[0102] Specifically, the calculation of fault weights involves several key parameters. The variable of the fault type serial number is used to identify different fault types, such as short-circuit faults, grounding faults, overload faults, etc. The actual measured value of the characteristic parameter refers to the real-time values of parameters such as voltage, current, and power factor extracted from the power grid operation data. The determination threshold is set according to the power grid operation standards and experience, and is the critical value for judging whether a fault occurs. The weight coefficient is obtained by calculating the difference between the actual measured value of each characteristic parameter and the corresponding determination threshold, and reflects the significance of each fault type in the current power grid operation state. For example, if the difference between the actual measured value of a certain characteristic parameter and the determination threshold is large, it indicates that the significance of this fault type in the current power grid state is high, so its weight coefficient will also be large.

[0103] Preferably, when calculating the fault weight, the operation steps can be further refined. First, determine the determination threshold of each fault type, and these thresholds can be set according to the power grid operation standards and historical fault data. Then, extract the actual measured values of each characteristic parameter from the fault feature vector. Next, calculate the absolute difference between the actual measured value of each characteristic parameter and the corresponding determination threshold. Finally, divide the absolute difference of each characteristic parameter by the sum of the absolute differences of all characteristic parameters to obtain the weight coefficient of each fault type. This calculation method can ensure that the sum of the weight coefficients is 1, thus providing a standardized weight distribution. In this way, the importance of different fault types can be more intuitively compared, providing more accurate guidance for the diagnosis and treatment of composite faults.

[0104] The above embodiments of the present invention have the following beneficial effects: The present invention can automatically generate a fault feature vector including multi-dimensional features such as voltage, current, and power factor based on the power grid operation data collected by intelligent power monitoring instruments, and quickly determine typical fault types such as short circuit, grounding, and overload through the matching relationship with preset fault criteria. The step-by-step threshold comparison method can identify fault states such as voltage anomaly, harmonic over-standard, and three-phase imbalance in sequence. Among them, the determination of the three-phase imbalance degree can improve the accuracy of determination by calculating the mean square deviation statistic of each phase current and introducing a safety factor. By setting a time window related to the load fluctuation period, the data sampling efficiency can be optimized, and data processing methods such as moving average filtering and fast Fourier transform can ensure the reliability of the characteristic parameters.

[0105] This method can calculate the weight coefficients of each fault type for compound fault situations, and realize the priority assessment of fault impacts by quantifying the deviation degree between the characteristic parameters and the thresholds. Based on the collaborative work of the feature extraction module and the fault determination module, a complete fault diagnosis process can be constructed. Finally, by the processor executing corresponding instructions, efficient operation on the power monitoring terminal device can be achieved. This systematic diagnosis method can significantly improve the real-time performance and accuracy of power grid fault identification, and at the same time reduce the workload of manual judgment, providing intelligent support for the safe operation of the power grid.

[0106] As Figure 2 shown, a power grid data fault diagnosis device based on an intelligent power monitoring instrument in some embodiments, the device includes:

[0107] A feature extraction module 201, configured to generate a fault feature vector according to the power grid operation data collected by the intelligent power monitoring instrument;

[0108] A fault determination module 202, configured to determine the power grid fault type according to the matching relationship between the fault feature vector and the preset fault criterion.

[0109] It can be understood that the various modules described in the power grid data fault diagnosis device based on the intelligent power monitoring instrument correspond to the respective steps in the power grid data fault diagnosis method described with reference to Figure 1 Therefore, the operations, features, and beneficial effects described above for the power grid data fault diagnosis method based on the intelligent power monitoring instrument also apply to the power grid data fault diagnosis device based on the intelligent power monitoring instrument and the modules included therein, and will not be elaborated here.

[0110] Next, refer to Figure 3, which shows a schematic structural diagram of an electronic device 300 suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3 The terminal device shown is merely an example and should not impose any limitations on the functions and usage scopes of the embodiments of the present invention.

[0111] As Figure 3 shown, the electronic device 300 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 301, which may perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage device 308 into the random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0112] Generally, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or wiredly to exchange data. Although Figure 3 the electronic device 300 with various devices is shown, it should be understood that it is not required to implement or include all the shown devices. More or fewer devices may be alternatively implemented or included. Figure 3 Each block shown in

[0113] Further, the storage medium of the embodiments of the present application stores program instructions capable of implementing all of the above methods. Among them, the program instructions can be stored in the above storage medium in the form of a software product, including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, or terminal devices such as computers, servers, mobile phones, and tablets.

[0114] The above description is only some preferred embodiments of the present invention and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present invention is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features having similar functions disclosed in the embodiments of the present invention.

Claims

1. A power grid data fault diagnosis method based on an intelligent power monitoring instrument, characterized in that, Including the following steps: Generating a fault feature vector based on the grid operation data collected by the intelligent power monitoring instrument, wherein the fault feature vector includes multi-dimensional feature parameters of voltage, current and power factor; Determining the grid fault type according to the matching relationship between the fault feature vector and the preset fault criterion, wherein the preset fault criterion includes the determination thresholds of short-circuit fault criterion, grounding fault criterion and overload fault criterion.

2. The power grid data fault diagnosis method according to claim 1, wherein Determining the grid fault type includes: Step S100: Judging whether the voltage deviation rate in the fault feature vector exceeds the first threshold. If it exceeds, it is determined as a voltage abnormal fault; otherwise, go to step S200; Step S200: Judging whether the total harmonic distortion rate of the current in the fault feature vector exceeds the second threshold. If it exceeds, it is determined as a harmonic over-standard fault; otherwise, go to step S300; Step S300: Judging whether the three-phase unbalance degree in the fault feature vector exceeds the third threshold. If it exceeds, it is determined as a three-phase unbalance fault; otherwise, it is determined that the grid operation is normal.

3. The power grid data fault diagnosis method according to claim 2, wherein Judging whether the three-phase unbalance degree exceeds the third threshold includes: Calculating the mean square deviation statistic of each phase current in the fault feature vector, and determining that it exceeds the third threshold when the mean square deviation statistic satisfies the following formula: σ abc >K·σ base Among them, σ abc represents the standard deviation statistic of three-phase current, and σ base represents the reference unbalance threshold, and K represents the preset safety factor.

4. The power grid data fault diagnosis method according to claim 3, wherein Calculating the mean square deviation statistic of each phase current in the fault feature vector includes: Step S10: Segmenting the current data collected by the intelligent power monitoring instrument according to the set time window; Step S20: Calculating the root mean square values of phase A, phase B and phase C for each segment of current data respectively; Step S30: Calculating the mean square deviation statistic according to the dispersion degree of the three-phase root mean square values.

5. The power grid data fault diagnosis method according to claim 4, characterized in that The length of the set time window is determined according to the grid load fluctuation period, specifically satisfying: T w = α · T load where T w represents the time window length, and T load represents the power grid load characteristic period. α represents the adjustment coefficient and its value range is 0.5 - 2.

0.

6. The power grid data fault diagnosis method according to claim 1, wherein Generating a fault feature vector includes: Step B100: Obtain the instantaneous voltage value sequence V from the intelligent power monitoring instrument t , the instantaneous current value sequence I t and the power factor sequence PF t ; Step B200: Perform a moving average filtering process on the instantaneous voltage value sequence to obtain a voltage characteristic parameter V avg ; Step B300: Perform a fast Fourier transform on the sequence of instantaneous current values to extract the harmonic feature parameter H thd ; Step B400: Calculate the statistical variance of the power factor sequence to obtain the power factor fluctuation parameter σ pf ; Step B500: Combine V avg , H thd and σ pf to generate the fault feature vector.

7. The power grid data fault diagnosis method according to claim 6, characterized in that The moving average filtering process satisfies: where k represents the sampling point sequence number variable, V t-kΔt represents the instantaneous voltage value of the k-th sampling point before time t, V avg represents the average voltage value after filtering, N represents the number of sampling points in the sliding window, and Δt represents the sampling interval time.

8. The power grid data fault diagnosis method according to claim 1, wherein, It also includes: When it is determined that there is a composite fault, calculating the fault weight according to the correlation degree of each parameter in the fault feature vector, specifically satisfying: Among which W i represents the weight coefficient of the i-th type of fault, P i represents the actual measured value of the i-th type of characteristic parameter, Th i represents the determination threshold of the i-th type of fault, j represents the fault type serial number variable, Th j represents the determination threshold of the j-th type of fault, and n represents the total number of preset fault types.

9. A power grid data fault diagnosis device based on an intelligent power monitoring instrument, characterized in that, Including: A feature extraction module configured to generate a fault feature vector based on the grid operation data collected by the intelligent power monitoring instrument; A fault determination module configured to determine the grid fault type according to the matching relationship between the fault feature vector and the preset fault criterion.

10. A power monitoring terminal device, characterized in that, Including: At least one processor; A memory communicatively connected to the at least one processor; Wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the grid data fault diagnosis method according to any one of claims 1-8.

Images