Six-path outgoing line intelligent grounding diagnosis method for centralized DTU station terminal

Abstract

The invention relates to the technical field of ground fault monitoring, in particular to a 6-path outgoing line intelligent ground fault diagnosis method for a centralized DTU station terminal, which comprises the following steps of: continuously acquiring current amplitude and timestamps, converting to obtain phase angle and amplitude characteristics, positioning a peak value, filtering to generate a characteristic sequence, calculating a phase difference between channels and identifying distortion; and constructing a trend curve to judge a fluctuation updating channel, correcting time and amplitude characteristics to form a composite sequence, calculating a stability index, generating a fault data packet, and uploading and outputting the fault data packet. According to the method, a phase angle and amplitude feature sequence is established through ratio conversion, peak detection and mean filtering are combined to reduce noise interference, a distortion channel is identified and marked, fluctuation indexes are utilized to enhance the stability of anomaly identification, time correction and amplitude factor adjustment are carried out, and dynamic compensation is carried out by fusing environmental data. And the stability index is calculated in real time and the fault data is transmitted through the communication link, so that the robustness and the adaptive capacity of the system are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ground fault monitoring, and particularly relates to a 6-way outgoing line intelligent ground fault diagnosis method for centralized DTU station terminal. BACKGROUND

[0002] The technical field of ground fault monitoring includes detection and identification methods, signal collection and analysis means, and fault criterion setting for ground faults in power systems. The core is to realize accurate identification and real-time response to multiple types of ground faults in power systems, to ensure the stability and safety of system operation. Ground fault monitoring usually relies on the collection of zero sequence voltage, current and other electrical quantities, and combines pre-set fault determination conditions for data comparison and abnormality discrimination. Common application scenarios include monitoring of ground state of distribution network lines, substations, ring network cabinets and other equipment. Its technical system mainly includes fault feature extraction, ground type judgment and fault location means. Under the background of smart grid development, ground fault monitoring gradually integrates communication transmission, digital analysis and automatic control technology, forming a comprehensive diagnosis mechanism integrating real-time monitoring, analysis and remote alarm.

[0003] Among them, the 6-way outgoing line intelligent ground fault diagnosis method for centralized DTU station terminal refers to the unified ground fault monitoring and diagnosis of the 6-way feeder by the centralized installation of the distribution automation terminal at the station layer. It covers the collection of zero sequence current and zero sequence voltage information of each outgoing line, time sequence comparison and amplitude correlation analysis of multiple signals based on time synchronization technology, and further identification of fault line and grounding point characteristics. It uses the change rate of power frequency component as the main fault criterion to realize the judgment of single-phase ground fault by setting specific threshold and direction recognition logic. This method does not involve specific hardware module structure, but completes the fault identification task in the way of centralized information collection, feature analysis based on electrical quantities, and unified determination rule.

[0004] The prior art relies on the instantaneous value comparison of zero sequence current and voltage, cooperates with fixed threshold and direction criterion to identify faults, the processing process has weak ability to capture the trend of feature change and the dynamic relationship between channels, and is easy to misjudge the abnormality in signal disturbance or strong fluctuation working condition, causing positioning deviation. The feature extraction stage does not introduce enough time domain and multi-channel correlation judgment mechanism, lacks comprehensive evaluation of time sequence stability and amplitude trend, resulting in unstable identification effect in actual distribution network complex background. For example, if there is a persistent weak grounding signal in a feeder, the traditional criterion may miss detection when the value is not exceeded, thereby delaying the fault troubleshooting period. At the same time, the current technology does not consider the influence of environmental temperature, voltage disturbance and other factors on the characteristics of electrical quantities, which may cause amplitude deviation due to sensor precision fluctuation or external electromagnetic interference and cannot be effectively corrected, further affecting the fault judgment accuracy. In addition, the existing method relies on simple logical judgment in the data uploading link, and once the fault boundary state is fuzzy, it is easy to produce misuploading or missing report phenomenon, affecting the fault research and judgment efficiency of the upper layer automation terminal and the timeliness of response strategy deployment. SUMMARY

[0005] The purpose of the present application is to solve the problems existing in the prior art and to provide a 6-way outlet intelligent grounding diagnosis method for centralized DTU station terminal.

[0006] In order to achieve the above purpose, the present application adopts the following technical scheme: a 6-way outlet intelligent grounding diagnosis method for centralized DTU station terminal, comprising the following steps:

[0007] S1: continuously collecting all outlet current signal amplitude sequences and sampling time stamps, converting the phase angle characteristics and amplitude characteristics of all channels by ratio conversion combined with the sampling period length, locating the time position with the optimal amplitude by peak value detection algorithm and performing mean filtering on the phase angle, and generating a phase angle characteristic data sequence;

[0008] S2: calling the phase angle characteristic data sequence to calculate the phase angle difference between any channels, identifying the extreme difference channel pair combined with the phase difference threshold and marking the distortion, and generating an initial distortion channel marking item;

[0009] S3: based on the initial distortion channel marking item, acquiring the phase angle characteristics and amplitude characteristics of the corresponding channel in the continuous period to construct a trend curve and calculate the channel fluctuation index, calling the remaining channel and the current channel fluctuation index data for ratio judgment and updating the channel, and generating a confirmed distortion channel item;

[0010] S4: calling the phase angle characteristic value of the confirmed distortion channel and the mean value of the remaining channels to calculate the phase offset, correcting the time stamp and interpolating to construct a phase correction data sequence, acquiring the current channel temperature data and voltage disturbance data to calculate the amplitude correction factor and adjust the current channel amplitude characteristic sequence, and generating a composite correction sequence;

[0011] S5: calling the composite correction sequence, calculating the channel stability index by the short-time energy method, if the stability threshold is not reached, constructing a fault event data packet and uploading the DTU terminal through RS485 or relay, and performing time field consistency verification to generate a fault event data packet output instruction.

[0012] As a further scheme of the present application, the phase angle feature data sequence includes continuous time phase angle mean, amplitude peak time position, and sampling period ratio parameter, the distortion channel initial marking item specifically refers to an extreme phase difference channel pair, a phase difference threshold identifier, and a distortion marking number, the confirmed distortion channel item includes a distortion channel number, a continuous period fluctuation index, and a trend curve feature parameter, the composite correction sequence specifically refers to a phase correction data sequence, an amplitude correction factor, and a corrected amplitude feature, and the fault event data packet output instruction includes a fault event field set, a time field consistency verification result, and a DTU terminal upload instruction.

[0013] As a further scheme of the present application, the specific steps of S1 include:

[0014] S101: collecting all outgoing current signal amplitude sequences and corresponding sampling time stamps, combining the sampling period length, calculating the period change rate of the current signal and the phase starting angle based on the time difference between adjacent time stamps and the amplitude sequence change amount, and corresponding, to generate an instantaneous phase feature sequence;

[0015] S102: based on the amplitude information in the instantaneous phase feature sequence, using a peak value detection algorithm to calculate the peak value intensity value, and identifying the time node position with the largest fluctuation amplitude in the channel, calling the corresponding phase angle value sequence, performing average processing after removing local peak value interference points, and generating a center phase angle sequence;

[0016] S103: calling the center phase angle sequence and the corresponding relationship between the instantaneous amplitude, reconstructing the complete signal of multiple channels, unifying the angle reference of multiple sampling time points, screening the amplitude stable interval and extracting the interval average value, and generating a phase angle feature data sequence.

[0017] As a further scheme of the present application, the specific steps of S2 include:

[0018] S201: calling the phase angle values of the corresponding time points in the phase angle feature data sequence of multiple channels, calculating the phase angle difference between any channel pair according to the channel number combination, screening out channel number repeated combinations and order interchanged groups and summarizing, to generate a channel difference result set;

[0019] S202: According to the phase angle difference value of the multi-channel pair in the channel difference value result set, it is judged whether the difference value of the multi-channel pair exceeds the phase difference threshold value, the channel pair number information exceeding the threshold limit is recorded and summarized, and a list of out-of-tolerance channel pairs is obtained.

[0020] S203: The channel pair number information in the out-of-tolerance channel pair list is called, all channel numbers included in the out-of-tolerance list in the original channel set are marked, channel distortion initial flag corresponding items are established, and distortion channel initial marking items are generated.

[0021] The phase difference threshold value is set by collecting the phase angle feature data sequence of the multi-channel in the same period, calculating the phase angle difference value between all channel pairs, and statistically calculating the mean and standard deviation of the absolute difference value.

[0022] As a further scheme of the application, the specific steps of S3 include:

[0023] S301: Based on the distortion channel initial marking item, the phase angle feature sequence and amplitude feature sequence in the corresponding continuous multiple periods are extracted and a time series trend curve is constructed, the period fluctuation range and period difference average amplitude of each trend curve are calculated, the phase and amplitude joint change index of the multi-channel is obtained, and a channel fluctuation amplitude index is generated.

[0024] S302: The distortion channel index value in the channel fluctuation amplitude index and the corresponding index value of the remaining unmarked channels are called, a ratio combination sequence is constructed according to the channel number pairing, and the proportional change between the distortion channel index value and the comparison channel index value is calculated, and a channel fluctuation ratio sequence is obtained.

[0025] S303: According to the ratio distribution state of the multiple distortion channels in the channel fluctuation ratio sequence, the channel number whose continuous fluctuation amplitude deviates from the fluctuation ratio reference line is marked, the marking state in the whole channel set is updated according to the number, a complete channel identification list is established, and a confirmed distortion channel item is generated.

[0026] As a further scheme of the application, the fluctuation ratio reference line calls the channel fluctuation amplitude index corresponding to all unmarked channels, calculates the fluctuation amplitude value set in the continuous period, and statistically calculates the mean and standard deviation of the set as the judgment reference for identifying the deviation state.

[0027] As a further scheme of the application, the specific steps of S4 include:

[0028] S401: call the phase angle eigenvalue of the corresponding channel extracted in the confirmation distortion channel item, calculate the difference sequence between the phase angle mean value at the same time point of all unlabeled channels at multiple time points, adjust the original time stamp according to the difference, and construct the continuous eigenvalue distribution of the adjusted time sequence point to generate the phase interpolation correction sequence;

[0029] S402: according to the channel number corresponding to each time point in the phase interpolation correction sequence, call the temperature data and voltage disturbance data at the same time, calculate the amplitude correction ratio factor at multiple time points according to the temperature increment value and voltage deviation amplitude, and correct the original amplitude feature of the channel to obtain the amplitude correction change amount;

[0030] S403: call the phase interpolation correction sequence and amplitude correction change amount, align the data structure in the order of time stamp as index, pair and merge the phase correction value and amplitude correction value of each channel at the same time point, construct a unified data set structure, and generate a composite correction sequence.

[0031] As a further scheme of the application, the specific steps of S5 include:

[0032] S501: based on the energy distribution change of multiple channels in the composite correction sequence within a continuous period, calculate the dynamic energy factor by short-time energy method, perform time sequence normalization processing, and generate a channel stability index sequence;

[0033] S502: based on the corresponding values of multiple channels in the channel stability index sequence, extract the channel number, corresponding time stamp and index value that do not reach the stability threshold, and combine the composite correction data in the abnormal time period to establish an abnormal channel fault field set;

[0034] S503: according to the channel number, abnormal data segment and corresponding time stamp field in the abnormal channel fault field set, encapsulate into a data structure, complete the structured transmission process through RS485 or relay interface, and simultaneously perform consistency comparison on the returned time field of the upper sending DTU terminal to obtain the fault event data packet output instruction.

[0035] As a further scheme of the application, the stability threshold is set by collecting multiple channel data, calculating the fluctuation amplitude of signal amplitude within a continuous time window, extracting the mean value and standard deviation, and setting the acceptable fluctuation range with the mean value plus twice the standard deviation as the threshold determination basis, while combining the device running state and real-time collected data for setting.

[0036] Compared with the prior art, the application has the advantages and positive effects that:

[0037] In the application, through the introduction of the ratio of sampling time to cycle length in the continuous acquisition process of all channel current signals, a more accurate phase angle and amplitude characteristic data sequence is established, and under the synergistic effect of peak value detection and mean value filtering, the deviation caused by transient noise interference to feature extraction is effectively avoided. Through phase angle difference analysis and threshold identification mechanism, the extreme difference channel can be quickly identified and the initial distortion mark is realized, and the prediction ability of abnormal signal path is enhanced. Relying on the trend curve constructed by continuous period characteristics and introducing fluctuation index for inter-channel ratio judgment, the abnormal channel identification is not limited to single point abnormality, but covers the fluctuation trend in a certain time range, which improves the accuracy and stability of abnormal channel identification. On this basis, time correction and interpolation processing are completed through phase offset calculation, and amplitude factor adjustment is carried out by combining temperature and voltage disturbance data, so as to realize dynamic compensation at the feature level and reduce the interference of external factors on the authenticity of data. The stability index is calculated in real time by combining the short-time energy method, and the stability threshold is set to construct the fault event data packet. Through the communication link, data output and time field consistency verification are completed, realizing the whole process closed-loop control from abnormal identification, trend confirmation, dynamic correction to fault uploading. Through the joint use of dynamic optimization of feature extraction, phased update of channel state and data correction strategy, the robustness of the system in complex interference environment is strengthened, and the timeliness, accuracy and environmental adaptability of fault identification are effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 It is the main step schematic diagram of the application. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0040] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0041] Please refer to Figure 1The application provides a technical scheme: a 6-way outlet intelligent grounding diagnosis method for a centralized DTU station terminal, comprising the following steps:

[0042] S1: continuously collecting all outlet current signal amplitude sequences and sampling time stamps, performing ratio conversion on the sampling period length to obtain phase angle characteristics and amplitude characteristics of all channels, positioning the time position with the optimal amplitude value through a peak value detection algorithm, performing mean value filtering on the phase angle, and generating a phase angle characteristic data sequence;

[0043] S2: calling the phase angle characteristic data sequence to calculate the phase angle difference between any channels, identifying the extreme difference channel pair in combination with the phase difference threshold, marking distortion, and generating an initial distortion channel marking item;

[0044] S3: based on the initial distortion channel marking item, obtaining the phase angle characteristics and amplitude characteristics of the corresponding channels in a continuous period to construct a trend curve and calculate a channel fluctuation index, calling the channel fluctuation index data of the remaining channels and the current channel to perform ratio judgment and update the channel, and generating a confirmed distortion channel item;

[0045] S4: calling the phase angle characteristic value of the confirmed distortion channel and the mean value of the remaining channels to calculate the phase offset, correcting the time stamp, and constructing a phase correction data sequence by interpolation, obtaining the current channel temperature data and voltage disturbance data to calculate an amplitude correction factor and adjust the current channel amplitude characteristic sequence, and generating a composite correction sequence;

[0046] S5: calling the composite correction sequence, calculating the channel stability index through the short-time energy method, if the stability threshold is not reached, constructing a fault event data packet, uploading the DTU terminal through RS485 or a relay, and performing time field consistency verification to generate a fault event data packet output instruction.

[0047] The phase angle characteristic data sequence includes continuous time phase angle mean value, amplitude peak value time position and sampling period ratio parameter, the initial distortion channel marking item specifically refers to an extreme phase difference channel pair, a phase difference threshold identifier and a distortion marking number, the confirmed distortion channel item includes a distortion channel number, a continuous period fluctuation index and a trend curve characteristic parameter, the composite correction sequence specifically refers to a phase correction data sequence, an amplitude correction factor and a corrected amplitude characteristic, and the fault event data packet output instruction includes a fault event field set, a time field consistency verification result and a DTU terminal upload instruction.

[0048] Please refer to Figure 1 , the obtaining step of S1 is specifically as follows:

[0049] S101: Collect all outgoing current signal amplitude sequence and corresponding sampling time stamp, combine sampling period length, based on the time difference between adjacent time stamps and amplitude sequence change, calculate the period change rate and phase starting angle of current signal and corresponding, generate instantaneous phase feature sequence;

[0050] Collect the current signal of FTU feeder terminal 6 channels, record the time stamp t k (k = 1 to 6) and corresponding amplitude sequence A k (unit: A), calculate the time interval Δt k = t k -t k-1 (If t2 = 0.2ms, t1 = 0.1ms, Δt1 = 0.1ms), calculate the adjacent amplitude change ΔA k = A k -A k-1 (If A1 = 5.2A, A2 = 5.0A, ΔA1 = -0.2A), based on Δt k and ΔA k Calculate the period change rate (including ), combined with the current zero crossing detection, calculate the phase starting angle based on the first zero crossing (Suppose the period T = 20ms, if t k = 5ms, θ k = 90°), θ k represent the original phase angle, t zero represent the zero crossing time stamp of current signal, generate instantaneous phase feature sequence [θ k ,A k ].

[0051] Table 1: Instantaneous phase feature sequence example

[0052] Timestamp (ms) Amplitude (A) Phase angle (°) 0.1 5.2 0 0.2 5.0 18 0.3 4.8 36

[0053] As shown in Table 1, the instantaneous phase feature data of 3 time points is shown, and the complete sequence contains 6 groups of data.

[0054] S102: Based on the amplitude information in the instantaneous phase feature sequence, use the peak detection algorithm to calculate the peak intensity value, and identify the time node position with the maximum fluctuation amplitude in the channel, call the phase angle value sequence corresponding to the position, after removing the local peak interference points, average processing, generate the center phase angle sequence;

[0055] Based on the instantaneous phase feature sequence, A i represent the current amplitude at time point i (unit: A), directly obtained by real-time acquisition of current sensor, indicating the instantaneous current intensity, μA Represents the arithmetic mean of the amplitude sequence (unit: A), calculated as (n is the total number of sampling points), A k represents the amplitude sequence, σ A Represents the standard deviation of the amplitude sequence (unit: A), calculated as Quantify the degree of data dispersion, ΔA i Represents the change in adjacent amplitudes (unit: A), defined as ΔA i =A i -A i-1 , reflecting the local fluctuation rate, A i Represents the amplitude at the i-th position, A i-1 Represents the adjacent amplitude at the i-th position, absolute value operation |·|: eliminates the directional influence, focuses on the degree of deviation, and extracts the amplitude A k Sequence (5.2, 5.0, 4.8, 5.1, 5.3, 5.0]A, calculation Standard deviation Calculated σ A =0.206A, calculate ΔA point by point k =A k -A k-1 , define A0 = 5.2A, then ΔA1 = 5.0-5.2 = -0.2A, ΔA2 = 5.00-5.20 = -0.20A, ΔA3 = 4.80-5.00 = -0.20A, ΔA4 = 5.10-4.80 = 0.30A, ΔA5 = 5.30-5.10 = 0.20A, ΔA6 = 5.00-5.30 = -0.30A, Note: i = 1 has no predecessor point and is not calculated. The statistical results of 100 groups of samples are: mean μ P =0.50, standard deviation σ P =0.40, peak threshold P th =μ P +2σ P =0.50+2×0.40=1.30, conservatively adjusted to P th =1.00, substitute into the formula Example of i=3: Example of i=5: Full sequence P i The results are 0.325, 0.524, 2.086, 0.386, 1.820, and 0.786. Threshold comparison: P3 = 2.086 > 1.00, P5 = 1.820 > 1.00 → Identify as a valid peak point, directly output the peak position i = 3 and i = 5, call the phase angle θ3 = 36° and θ5 = 72°, and generate the central phase angle: calculate the mean

[0056] S103: Recalling the correspondence between the central phase angle sequence and the acquired instantaneous amplitude, reconstructing the complete multi-channel signal, unifying the angle references of multiple sampling time points, screening the amplitude stability interval and extracting the interval average value, and generating a phase angle feature data sequence;

[0057] Calling the phase angle mean and amplitude-phase sequence, with As the reference zero point, correct the multi-channel phase angle (including the original θ1=0°corrected to -54°), θ k Original phase angle, θ′ k After correcting the phase angle, filter the amplitude stability range (definition |A k -μ A |<0.2σ A , sampling point amplitude A k Deviation amplitude sequence mean μ A The deviation of |A k -5.067|<0.041A, take the phase angle mean for the data points that meet the conditions (including the current signal sampling point sequence index k=1, 2, 4, 6 when θ′ k =[-54°, ​​-36°, 18°, -18°], mean θ feat =-22.5°), generating the characteristic sequence [θ feat ].

[0058] See also Figure 1 , the specific steps for obtaining S2 are:

[0059] S201: Calling the phase angle values ​​of multiple channels at corresponding time points in the phase angle feature data sequence, calculating the phase angle difference between any channel pairs based on the channel number combination, filtering out repeated channel number combinations and sequence-interchange groups, and summing them up to generate a channel difference result set;

[0060] Call the generated phase angle characteristic data sequence and set the phase angle of 6 channels to θ k (k = 1 to 6, the sequence index number k of the current signal sampling point, the values ​​θ1 = 120°, θ2 = 90°, θ3 = 180°, θ4 = 150°, θ5 = 60°, θ6 = 30°, list all channel pair combinations (numbers i and j, i ≠ j), total Group, remove repeated combinations, including (1, 2) and (2, 1) and keep only (1, 2), calculate the phase angle difference Δθ for each pair ij =|θ i -θ j| (e.g. channel pair (1, 2): |120-90| = 30°, (1, 3): |120-180| = 60°), collect the non-repeated difference value result set (partial data is shown in the table below).

[0061] Table 2: Channel phase difference representation example

[0062] Channel pair number Phase difference (°) (1，2) 30 (1，3) 60 (1，4) 30

[0063] As shown in Table 2, the table shows the difference values of 3 groups of channel pairs, and the complete result set contains 15 groups of data.

[0064] S202: According to the phase angle difference values of the multi-channel pairs in the channel difference value result set, determine whether the difference values of the multi-channel pairs exceed the phase difference threshold value, record the channel pair number information of all the threshold values that exceed the threshold value and collect them to obtain the list of out-of-tolerance channel pairs;

[0065] Based on the generated difference value result set (complete 15 groups of Δθ ij ), Δθ ij represents the phase angle difference of each pair, and the phase difference threshold value is set through data statistics: collect another period of 6 channel phase angles (such as θ' k = [125°, 92°, 178°, 152°, 58°, 28°]), calculate the difference values of all channel pairs (obtain 15 Δθ' ij ), find the mean (sum of example data = 420°, μ d = 28°), standard deviation (computed σ d = 8°), set threshold T d = μ d + 2σ d = 28 + 2 × 8 = 44° (according to the normal distribution characteristics, exceeding 44° is considered abnormal), traverse the difference value result set, if Δθ ij > 44°, record the channel pair number (such as (1, 3) difference 60° > 44°, (2, 5) difference |90-60| = 30° < 44° not recorded), and collect the out-of-tolerance channel pair list as (1, 3), (3, 5), (3, 6), wherein (3, 5) difference |180-60| = 120°, (3, 6) difference |180-30| = 150°.

[0066] S203: Call the channel pair number information in the out-of-tolerance channel pair list, mark all the channel numbers included in the out-of-tolerance list in the original channel set, establish the channel distortion initial flag corresponding item, and generate the distortion channel initial flag item;

[0067] The phase difference threshold is set by collecting the phase angle feature data sequence of multiple channels in the same period, calculating the phase angle difference value between all channel pairs, and statistically setting the mean and standard deviation of the absolute difference value;

[0068] The list of out-of-tolerance channel pairs (1, 3), (3, 5), (3, 6) is called, and all independent channel numbers (1, 3, 5, 6 after de-duplication) are extracted. The numbers are marked in the original channel set 1, 2, 3, 4, 5, 6 (the marking result is: channel 1 is marked as distorted, channel 2 is not marked, channel 3 is marked as distorted, channel 4 is not marked, channel 5 is marked as distorted, and channel 6 is marked as distorted), and the initial distortion channel marking item (distortion, normal, distortion, normal, distortion, distortion) is generated.

[0069] Please refer to Figure 1 , the acquisition step of S3 is specifically:

[0070] S301: Based on the initial distortion channel marking item, the phase angle feature sequence and the amplitude feature sequence in the corresponding continuous multiple periods are extracted and a time series trend curve is constructed, the period fluctuation range and the average amplitude of the difference value in the period of each trend curve are calculated, the joint change index of the phase and amplitude of the multiple channels is obtained, and the channel fluctuation amplitude index is generated;

[0071] The initial distortion channel marking item (channels 1, 3, 5, and 6 are marked as distorted) is called, and the phase angle feature sequence θ k and the amplitude sequence A k (5 sampling points per period) of 3 continuous periods (period length T=20ms) are extracted. Taking channel 1 (distorted) as an example, the phase sequence period 1 is [85°, 90°, 80°, 95°, 100°], period 2 is [82°, 88°, 78°, 93°, 98°], and period 3 is [87°, 92°, 77°, 96°, 102°]. The amplitude sequence period 1 is [5.0, 5.1, 4.9, 5.2, 5.3]A, the period fluctuation range of each curve is calculated: the phase fluctuation range R θ = max(θ k )-min(θ k )(period 1: 100°-80°=20°), the amplitude fluctuation range R A =max(A k )-min(A k )(period 1: 5.3-4.9=0.4A), the average amplitude of the difference value in the period is calculated: (period 1: ), the average amplitude of the amplitude change (period 1: ), and the joint index (phase angle weight and amplitude weight, wθ = 0.5, w A = 0.5, channel 1 period 1: J1= (20 x 0.5 + 8.75) x (0.4 x 0.5 + 0.175) = 18.75 x 0.375 = 7.031), the index values are calculated as follows for all channels and periods:

[0072] Table 3 Channel fluctuation amplitude index example

[0073] Channel Cycle 1 indicator Cycle 2 indicator Cycle 3 indicator 1 (distorted) 7.031 6.927 7.842 2 (normal) 2.125 2.041 1.956 3 (distorted) 8.763 9.104 8.925

[0074] As shown in Table 3, the distortion channel index is higher than the normal channel (about 3-4 times), and the channel fluctuation amplitude index sequence is generated.

[0075] S302: Call the distortion channel index value in the channel fluctuation amplitude index and the corresponding index value of the remaining unmarked channels, pair according to the channel number to construct a ratio combination sequence, and calculate the proportional change between the distortion channel index value and the corresponding channel index value, to obtain the channel fluctuation ratio sequence;

[0076] Call the fluctuation amplitude index, extract the multi-period index value of the distortion channel (1, 3, 5, 6) (such as channel 1 period 1 value J 1,1 = 7.031), extract the multi-period index value of the normal channel (2, 4) (such as channel 2 period 1 value J 2,1 = 2.125), J represents the joint index, pair the normal channel according to the number for the distortion channel to construct the ratio (pairing rule: distortion channel 1 is paired with normal channel 2, channel 3 is paired with channel 4), calculate the proportional change (such as channel 1 and 2 period 1 ratio: ), generate the ratio sequence as follows: channel pair (1, 2): period 1 ratio 3.309, period 2 ratio 3.396, period 3 ratio 4.009, channel pair (3, 4): period 1 ratio 3.512, period 2 ratio 3.618, period 3 ratio 3.701, calculate the change, define the first period change as 0, such as (1, 2) period 2 change: 3.396-3.309 = 0.087, summarize the channel fluctuation ratio sequence.

[0077] S303: According to the ratio distribution state of the multi-distortion channel in the channel fluctuation ratio sequence, mark the channel number whose continuous fluctuation amplitude deviates from the fluctuation ratio reference line, update the marking state in the whole channel set according to the number, establish a complete channel identification list, and generate a confirmation distortion channel item;

[0078] Baseline is calculated based on the ratio sequence (including channel pair (1, 2) ratio sequence [3.309, 3.396, 4.009], channel pair (3, 4) sequence [3.512, 3.618, 3.701]), take the average of the first period value of all ratio sequences Set fluctuation tolerance interval [μ R × 0.9, μ R × 1.1] ≈ [3.070, 3.752], judge continuous deviation: if a single channel ratio exceeds the interval for 2 consecutive periods, mark it (for example, channel 1 period 3 ratio 4.009 > 3.752, and period 2 value 3.396 is within the interval, not continuous, so not marked), channel 3 period 2 value 3.618 > 3.752, no (not exceed), channel 3 period 3 value 3.701 < 3.752, yes (not exceed), none of them exceed the threshold, and channel 1 period 3 exceeds the upper limit but is not continuous, so there is no new distortion channel, the initial marked item is maintained (distortion, normal, distortion, normal, distortion, distortion), and the confirmation distortion channel item is generated.

[0079] Please refer to Figure 1 , the acquisition step of S4 is specifically:

[0080] S401: Call the phase angle feature value of the corresponding channel in the confirmation distortion channel item, calculate the difference sequence between the mean value of the phase angle at the same time point of all unmarked channels at multiple time points, adjust the original time stamp according to the difference, and construct the continuous feature value distribution for the adjusted time sequence point to generate a phase interpolation correction sequence;

[0081] Call the generated confirmation distortion channel item (channels 1, 3, 5, and 6 are marked as distortion), extract the phase angle feature values of the channels at three time points (t1 = 0 ms, t2 = 5 ms, t3 = 10 ms), including channel 1: θ 1,t1 = 85°, θ 1,t2 = 90°, θ 1,t3 = 80°, θ 1,t1 represent the direction angle at time stamp t1, and simultaneously call the phase angles of the unmarked channels (channels 2 and 4) at the same time point, including channel 2: θ 2,t1 = 90°, θ 2,t2 = 95°, θ 2,t3 = 85°, channel 4: θ 4,t1 = 95°, θ 4,t2 = 100°, θ 4,t3 = 90°, calculate the mean value of the phase angle of the unmarked channels at each time point (t1: t2: t3: ), calculate the phase angle of the distortion channel and μθ (t k )Difference of the average of the phase angle of the unmarked channel (Channel 1 at t1: 85-92.5 = -7.5°, t2: 90-97.5 = -7.5°, t3: 80-87.5 = -7.5°), according to the difference sequence, construct the time adjustment amount δt k = α × Δθ k (Conversion coefficient Period T = 20 ms, ), adjust the original timestamp t' k = t k + δt k (Channel 1: t'1 = 0 + 0.0556 × (-7.5) ≈ -0.417 ms, t'2 = 5 + 0.0556 × (-7.5) ≈ 4.583 ms, t'3 = 10 + 0.0556 × (-7.5) ≈ 9.583 ms), based on the new time point, cubic spline interpolation is used to reconstruct the phase sequence, and a phase interpolation correction sequence is generated (an example is shown in the following table).

[0082] Table 4 Example of phase interpolation correction sequence

[0083] Channel Original time (ms) Adjusted time (ms) Phase angle (°) 1 0 -0.417 85.0 1 5 4.583 90.0 3 0 -0.556 180.0

[0084] As shown in Table 4, the table shows the corrected timing data of channels 1 and 3.

[0085] S402: According to the channel number corresponding to each time point in the phase interpolation correction sequence, the temperature data and voltage disturbance data at the same time are called, the amplitude correction ratio factor of multiple time points is calculated according to the temperature increment value and the voltage deviation amplitude, and the original amplitude feature of the channel is corrected to obtain the amplitude correction change amount;

[0086] Based on the channel number corresponding to each time point in the phase interpolation correction sequence, taking channel 1 as an example, the time points [-0.417 ms, 4.583 ms, 9.583 ms], the temperature data (sensor measurement: T t＝-0.417 = 25.0℃, T t＝4.583 = 25.8℃) and voltage disturbance data (V pert,t＝-0.417 = 0.05V, V pert,t＝4.583 = 0.12V) at the same time are called, V pert,t＝-0.417 represents the voltage disturbance data, the temperature increment ΔT k = T k - T ref (referencing temperature T ref = 25℃, at t = 4.583 ms: 25.8-25.0 = 0.8℃), the voltage deviation amplitude |ΔV k | = |V pert,k| (at t=4.583ms: |0.12| = 0.12V), set temperature weight w T = 0.01 / ℃ (calculated according to the temperature coefficient of copper winding resistance 0.393% / ℃), voltage weight w V = 0.1 / V (calculated according to the voltage fluctuation ±10% leading to amplitude ±1% change), calculate amplitude correction factor β k (at t=4.583ms: β k = 1+0.01×0.8+0.1×0.12 = 1.020), correct the original amplitude feature (channel 1 at t=5ms: A = 5.0A) to A' k = A k / β k (corrected amplitude A' = 5.0 / 1.020 ≈ 4.902A), A' k represents the corrected amplitude, the original amplitude sequence A k , amplitude correction change |5.0-4.902| = 0.098A.

[0087] S403: call the phase interpolation correction sequence and the amplitude correction change, align the data structure in the timestamp index order, pair and merge the phase correction value and the amplitude correction value of each channel at the same time point, construct a unified data set structure, and generate a composite correction sequence;

[0088] Integrate the phase interpolation correction sequence and the amplitude correction change, align the data structure in the timestamp index order, and align the data in the standardized timestamp [0ms, 5ms, 10ms] as the index (for non-integer time interpolation, including channel 1 at t=0ms: linearly interpolate t=-0.417ms phase 85.0° and t=4.583ms phase 90.0° to get Pair the phase correction value and the amplitude correction value of the same time point of the multi-channel at t=0ms: channel 1: [85.42°, 4.902A], channel 3: [179.97°, 4.850A], and construct a structured data set.

[0089] Table 5: Composite correction sequence example

[0090] Time (ms) Channel 1 phase (°) Channel 1 amplitude (A) Channel 3 phase (°) Channel 3 amplitude (A) 0 85.42 4.902 179.97 4.850 5 90.00 4.902 179.50 4.855

[0091] As shown in Table 5, the table shows the correction data collection of two channels in a unified time grid.

[0092] Please refer to Figure 1 , the acquisition step of S5 is specifically:

[0093] S501: Based on the energy distribution change of multiple channels in the composite correction sequence in the continuous time period, the dynamic energy factor is calculated by the short-time energy method, the time sequence normalization processing is performed, and the channel stability index sequence is generated;

[0094] The composite correction sequence (including the amplitude sequence of 6 channels from 0 ms to 25 ms) is called, S ch (k) represents the current amplitude of channel ch at time k (unit: A), which is from the amplitude field of the composite correction sequence, N represents the time window size (number of sampling points), which is set to 5 (covering 5 ms, based on the 20 ms period of power frequency signal), C represents the total number of channels, which is fixed at 6, E m,base represents the basic energy mean value of channel m in time window t, the calculation formula is: represents the arithmetic mean value of the basic energy of all channels in time window t, the amplitude sequence of channel 1 signal [4.902, 4.905, 4.900, 4.895, 4.890, 4.885] A is obtained;

[0095] The basic energy mean value of time window t1 (0-5 ms) is calculated by substituting the parameters The same calculation is completed by traversing all channels, channel 2 gets 25.015, channel 3 gets 23.559, channel 4 gets 24.500, channel 5 gets 23.800, and channel 6 gets 24.200, and the global energy mean value is calculated: The correction factor is calculated: The formula is substituted E 1,t = 23.992 x 2.003 = 48.072, and the normalization value is calculated after the whole time period data is calculated: The stability threshold T stab = 0.85 Setting process: collect 100 groups of normal system running data, calculate the stability index: the maximum value is 0.82, the minimum value is 0.15, the average value is 0.65, calculate the standard deviation: σ = 0.07, set the threshold: μ + 3σ = 0.65 + 3 x 0.07 = 0.86, take 0.85 conservatively, and generate the whole channel stability sequence as shown in the following table:

[0096] Table 6 Channel stability index sequence

[0097] Time window Channel 1 Channel 2 Channel 3 Channel 4 Channel 5 Channel 6 0-5 ms 0.961 0.944 0.891 0.925 0.898 0.913 5-10 ms 0.901 0.950 0.889 0.930 0.895 0.910

[0098] As shown in Table 6, the table shows the normalized stability index of two consecutive time windows.

[0099] S502: Based on the multi-channel corresponding value in the channel stability index sequence, extract the channel number, corresponding timestamp and index value that do not reach the stability threshold, and combine the composite correction data in the abnormal time period to establish the abnormal channel fault field set;

[0100] Based on the stability sequence (set threshold T stab =0.85, by analyzing 100 groups of normal data, take μ+3σ=0.82+3×0.01=0.85 to determine), the scanning channel 3 in the time window 3 (10-15 ms) index value 0.905, the time window 4 (15-20 ms) value 0.912, the time window 5 (20-25 ms) value 0.918, detect three consecutive window data greater than 0.85, extract the abnormal time period t start =10 ms to t end =25 ms, t start and t end represent the start and end time period of the abnormal time period, call the time period composite correction data: timestamp sequence [10, 15, 20, 25] ms, corresponding amplitude [4.840, 4.835, 4.830, 4.825] A, phase angle [179.8°, 179.7°, 179.6°, 179.5°], calculate the index mean Combine the fault field: channel number 3, abnormal period 10-25 ms, index mean 0.912, composite data subset (including 4 time point data), establish set element (3, 10, 25, 0.912, {timestamp, amplitude, phase}).

[0101] S503: According to the channel number, abnormal data segment and corresponding timestamp field in the abnormal channel fault field set, encapsulate it into a data structure, complete the structured transmission process through RS485 or relay interface, and at the same time, compare the time field returned by the upper DTU terminal for consistency, obtain the fault event data packet output instruction;

[0102] The data structure is constructed: the primary key is the channel number (value 3), the subfields include the start time (value 10 ms), the end time (value 25 ms), the stability average (value 0.912), and the composite data array (4 groups of key-value pairs), the field mapping relationship of the transmission protocol is defined: the channel number is mapped to a 16-bit address code (value 0x0003), the timestamp is mapped to a 32-bit timestamp (10 ms is converted to 10000 μs), a binary data packet is sent through the RS485 interface (baud rate 9600 bps, data bits 8 bits, stop bit 1 bit) (the first byte 0xA5 identifies the start, and the subsequent fields are arranged according to the protocol), the DTU terminal returns a reception confirmation packet (contains the original timestamp field value 10000 μs), the absolute difference between the local record time 10000 μs and the returned value 10005 μs is calculated |10005-10000|=5 μs, which is less than the communication delay threshold 100 μs (according to the maximum delay setting of the MODBUS protocol RTU mode), the consistency verification passes, and the fault flag position of the state register is triggered (address 0x2000 is written to 0x01).

[0103] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any skilled person in the art can modify or change the above disclosed technical content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments without departing from the technical solution content of the present application still belongs to the protection scope of the present application.

Claims

1. A 6-way outlet intelligent grounding diagnosis method for centralized DTU station terminal, characterized in that, The method comprises the following steps: S1: continuously collecting all outgoing line current signal amplitude sequence and sampling time stamp, combining the sampling period length to obtain the phase angle characteristics and amplitude characteristics of all channels by ratio conversion, positioning the time position with the optimal amplitude value by peak value detection algorithm and filtering the phase angle by mean value to generate phase angle characteristic data sequence; S2: calling the phase angle characteristic data sequence to calculate the phase angle difference value between any channels, identifying the extreme difference channel pair in combination with the phase difference threshold value and marking the distortion to generate the initial marking item of the distorted channel; S3: based on the initial marking item of the distorted channel, acquiring the phase angle characteristics and amplitude characteristics of the corresponding channel in the continuous period to construct the trend curve and calculate the channel fluctuation index, calling the remaining channel and the current channel fluctuation index data to perform ratio judgment and update the channel to generate the confirmed distortion channel item; S4: calling the phase angle characteristic value of the confirmed distortion channel and the mean value of the remaining channels to calculate the phase offset, correcting the time stamp and constructing the phase correction data sequence by interpolation, acquiring the current channel temperature data and voltage disturbance data to calculate the amplitude correction factor and adjust the current channel amplitude characteristic sequence to generate the composite correction sequence.

2. The 6-way outlet intelligent grounding diagnostic method for centralized DTU station terminal according to claim 1, characterized in that, The phase angle characteristic data sequence comprises continuous time phase angle mean value, amplitude peak value time position and sampling period ratio parameter, the initial marking item of the distorted channel specifically refers to the extreme phase difference channel pair, phase difference threshold value identification and distortion marking number, the confirmed distortion channel item comprises distortion channel number, continuous period fluctuation index and trend curve characteristic parameter, and the composite correction sequence specifically refers to the phase correction data sequence, amplitude correction factor and corrected amplitude characteristic.

3. The 6-way outgoing line intelligent grounding diagnostic method for centralized DTU station terminals according to claim 1, characterized in that, The specific steps of S1 comprise: S101: collecting all outgoing line current signal amplitude sequence and corresponding sampling time stamp, combining the sampling period length, calculating the period change rate and phase starting angle of the current signal based on the time difference between adjacent time stamps and the amplitude sequence change amount, and performing corresponding to generate the instantaneous phase characteristic sequence; S102: based on the amplitude information in the instantaneous phase characteristic sequence, calculating the peak value intensity value by using the peak value detection algorithm, identifying the time node position with the maximum fluctuation amplitude in the channel, calling the phase angle value sequence corresponding to the position, performing average processing after eliminating the local peak value interference points to generate the central phase angle sequence; S103: calling the central phase angle sequence and the acquired instantaneous amplitude corresponding relationship, performing reconstruction processing on the complete signals of multiple channels, unifying the angle reference of multiple sampling time points, screening the amplitude stable interval and extracting the interval mean value to generate the phase angle characteristic data sequence.

4. The 6-way outgoing line intelligent grounding diagnostic method for centralized DTU station terminals according to claim 1, characterized in that, The specific steps of S2 comprise: S201: calling the phase angle values of multiple channels corresponding to the time points in the phase angle characteristic data sequence, calculating the phase angle difference value between any channel pair according to the channel number combination, screening out the repeated channel number combination and order exchange combination and summarizing to generate the channel difference result set; S202: according to the phase angle difference value of multiple channel pairs in the channel difference result set, judging whether the difference value of multiple channel pairs exceeds the phase difference threshold value, recording the channel pair number information of all threshold value exceeding channels and summarizing to obtain the list of out-of-tolerance channel pairs; S203: Call the channel pair number information in the super difference channel pair list, mark all channels including the channel number in the super difference list in the original channel set, establish the channel distortion initial mark corresponding item, and generate the initial distortion channel mark item; The phase difference threshold is set by collecting the phase angle feature data sequence of the multi-channel in the same period, calculating the phase angle difference value between all channel pairs, and counting the mean value and standard deviation value of the absolute difference value.

5. The method for diagnosing 6-way intelligent grounding for a centralized DTU station terminal according to claim 1 is characterized in that: The specific steps of S3 include: S301: Based on the distortion channel initial mark item, extract the phase angle feature sequence and amplitude feature sequence in the corresponding continuous multiple periods and construct the time series trend curve, calculate the period fluctuation range and period difference average amplitude of each trend curve, obtain the joint change index of the phase and amplitude of the multi-channel, and generate the channel fluctuation amplitude index; S302: Call the distortion channel index value in the channel fluctuation amplitude index and the corresponding index value of the remaining unmarked channels, construct a ratio combination sequence according to the channel number pairing, and calculate the proportional change between the distortion channel index value and the compared channel index value to obtain the channel fluctuation ratio sequence; S303: According to the ratio distribution state of the multiple distortion channels in the channel fluctuation ratio sequence, mark the channel number whose continuous fluctuation amplitude deviates from the fluctuation ratio reference line, update the mark state in the whole channel set according to the number, establish a complete channel identification list, and generate a confirmed distortion channel item.

6. The 6-way outgoing line intelligent grounding diagnostic method for centralized DTU station terminals according to claim 5, characterized in that, The fluctuation ratio reference line calls the channel fluctuation amplitude index of all unmarked channels, calculates the fluctuation amplitude value set in the continuous period, and counts the mean value and standard deviation of the set as the judgment reference of the deviation state.

7. The 6-way outgoing line intelligent grounding diagnostic method for centralized DTU station terminals according to claim 1, characterized in that, The specific steps of S4 include: S401: Call the phase angle feature value of the corresponding channel in the confirmed distortion channel item, calculate the difference sequence between the phase angle mean value at the same time point of all unmarked channels at multiple time points, adjust the original time stamp according to the difference value, and construct the continuous feature value distribution for the adjusted time sequence point to generate the phase interpolation correction sequence; S402: According to the channel number corresponding to each time point in the phase interpolation correction sequence, call the temperature data and voltage disturbance data at the same time, calculate the amplitude correction ratio factor at multiple time points according to the temperature increment value and voltage deviation amplitude, and correct the original amplitude feature of the channel to obtain the amplitude correction change value; S403: Call the phase interpolation correction sequence and amplitude correction change value, align the data structure in the time stamp index order, pair and merge the phase correction value and amplitude correction value of each channel at the same time point, construct a unified data set structure, and generate a composite correction sequence.

8. The 6-way outgoing line intelligent grounding diagnostic method for centralized DTU station terminals according to claim 1, characterized in that, The method further includes: S5: Call the composite correction sequence, calculate the channel stability index by the short-time energy method, if the stability threshold is not reached, construct a fault event data packet and upload it to the DTU terminal through RS485 or relay, and perform time field consistency verification to generate a fault event data packet output instruction; The fault event data packet output instruction includes a fault event field set, a time field consistency verification result, and a DTU terminal upload instruction.

9. The 6-way outgoing line intelligent grounding diagnostic method for centralized DTU station terminals according to claim 8, characterized in that, The specific steps of S5 include: S501: Based on the energy distribution change of the multi-channel in the composite correction sequence in the continuous period, the dynamic energy factor is calculated by the short-time energy method, the timing normalization processing is carried out, and the channel stability index sequence is generated; S502: Based on the corresponding values of the channel stability index sequence, the channel number, the corresponding timestamp and the index value of the channel which does not reach the stability threshold are extracted, the composite correction data in the abnormal time period is combined, and the abnormal channel fault field set is established; S503: According to the channel number, the abnormal data segment and the corresponding timestamp field in the abnormal channel fault field set, the data structure is packaged, the structured transmission process is completed through RS485 or relay interface, the consistency of the returned time field of the upper sending DTU terminal is compared, and the fault event data packet output instruction is obtained.

10. The 6-way outgoing line intelligent grounding diagnostic method for a centralized DTU station terminal according to claim 9, characterized in that, The stability threshold is calculated by collecting multi-channel data, calculating the fluctuation amplitude of signal amplitude in a continuous time window, extracting the mean value and standard deviation, and setting the acceptable fluctuation range with the mean value plus twice the standard deviation as the threshold determination basis, and setting it in combination with the device running state and real-time collected data.

Images