Online monitoring and early warning method for abnormal state of power distribution area

Abstract

The invention provides an online monitoring and early warning method for an abnormal state of a power distribution area, which comprises the following steps of: firstly, acquiring voltage and current sequence data of the power distribution area from a historical database, and acquiring a three-phase voltage value and a current value in real time; calculating a Pearson's correlation coefficient between the voltage data and the current data, and judging the connection group of the distribution transformer according to a preset threshold value and a judgment condition; feature extraction is carried out on the collected three-phase voltage and current values to obtain voltage features and current features, and a voltage feature strengthening value is obtained by combining the connection group; and finally, constructing a total scoring function, scoring the voltage features, the current features and the voltage feature strengthening values, comparing with a preset threshold value, judging whether the power distribution area has a fault or not, and realizing online monitoring and early warning of an abnormal state. The method is deeply focused on how to improve the diagnosis accuracy of the high-voltage side open-phase fault, and the false alarm rate and the missing report rate of online detection of the power distribution area are effectively reduced through low-frequency measurement data.

Description

Technical Field

[0001] The present invention belongs to the technical field of power monitoring, and in particular relates to an online monitoring and early warning method for abnormal conditions in a distribution station area. Background Art

[0002] As a key component of the smart grid, the safe and stable operation of the distribution network is crucial. Phase loss faults on the high-voltage side of a distribution station (such as line breaks or switch failures) can cause three-phase voltage asymmetry, leading to equipment damage, widespread power outages, and economic losses. Traditional fault handling relies on manual inspections and user reports, resulting in delayed response and high costs. While distribution automation terminals (such as FTUs and DTUs) are used, the full deployment of high-end terminals is expensive, and the massive amount of existing low-frequency measurement data (such as low-voltage side voltage and current data collected every 15 minutes) is underutilized. Furthermore, traditional records of key parameters, such as the connection group of distribution transformers (e.g., Dyn11 and Yyn0), can be inaccurate, affecting the accuracy of fault diagnosis.

[0003] Existing fault handling modes and monitoring technologies include the following categories: 1. Manual inspection and user repair reporting: The most traditional method relies on manual discovery or user complaints, resulting in delayed response, inability to proactively warn, and large losses from failures.

[0004] 2. Local intelligent terminal monitoring: Some TTUs / FTUs have a phase loss detection function based on simple logic such as low-voltage side voltage / current imbalance and voltage over-limit. This function is costly, lacks the ability to distinguish between fault types (for example, it is difficult to distinguish between high-voltage side phase loss and load imbalance), and is difficult to maintain.

[0005] 3. Direct monitoring on the high-voltage side: Installing voltage transformers, current sensors, or dedicated fault indicators on the 10kV high-voltage side to directly monitor voltage and current status has high installation and maintenance costs, high risks, and is not universally applicable.

[0006] 4. Fault diagnosis based on high-frequency transient data: Utilizes the high-frequency transient signal characteristics generated at the moment of fault occurrence to analyze the fault type and location. Existing systems rely on high-speed sampling equipment, and are unable to obtain high-frequency data.

[0007] Existing fault handling models and monitoring technologies have significant shortcomings in terms of response speed, cost, accuracy, and ease of maintenance. These limitations limit the power system's ability to effectively monitor and quickly diagnose faults in distribution substations, and fail to meet the requirements of smart grid construction for the safe and stable operation of distribution networks.

[0008] For example, Chinese patent publication number CN116008682B proposes a method for determining phase loss faults on the high-voltage side of a distribution transformer using a rule engine. Its core is to first determine the connection group using the voltage correlation coefficient and the voltage difference extreme value; then, based on simple logical rules, determine whether a phase loss has occurred; and finally, by grouping the substations that triggered alarms at the same time according to the lines they belong to, determine whether the fault is on the line side or the distribution transformer side. Furthermore, Chinese patent publication number CN109713671B discloses a distribution substation operation and maintenance method, system, storage medium, and electronic equipment. This patent aims to achieve comprehensive operation and maintenance of distribution substations, with very broad objectives. It utilizes multiple data mining algorithms, such as fuzzy clustering and linear fitting, to identify various operating parameters, including connection groups, operating gear positions, and capacity, and generates corresponding operation and maintenance recommendations based on the identification results of these parameters. However, the Chinese patent, published with patent number CN116008682B, uses a fixed logic rule engine, a rigid, non-adaptive judgment method whose accuracy and generalization capabilities are limited by the simplicity of the rules. The Chinese patent, published with patent number CN109713671B, also employs a fragmented approach, using fuzzy clustering to identify gear positions and linear fitting to identify capacity, but fails to establish a unified, comprehensive diagnostic model for high-voltage phase loss faults. Summary of the Invention

[0009] In response to the technical problems in the existing technology, a method for online monitoring and early warning of abnormal conditions in a distribution station area is provided, including: Step S1: Obtain voltage sequence data and current sequence data of the distribution substation from the historical database; collect three-phase voltage values and three-phase current values of the distribution substation in real time; Step S2: Calculate the Pearson correlation coefficient between the voltage sequence data and the current sequence data in step S1; determine the connection group of the distribution transformer corresponding to the distribution station area according to the preset threshold and judgment conditions using the Pearson correlation coefficient; the connection groups include Yyn0 and Dyn11; Step S3: Obtain the three-phase voltage values and three-phase current values obtained in step S1, perform feature extraction, and obtain voltage features and current features respectively; and obtain a voltage feature enhancement value based on the connection group obtained in step S2; Step S4: Construct a total scoring function, use the total scoring function to score the voltage characteristics, current characteristics and voltage characteristic enhancement values based on the connection group, compare the scores with the preset thresholds, and determine whether a fault occurs in the distribution substation area.

[0010] Furthermore, step S1 is specifically as follows: Step S11: obtaining the measurement data of the three-phase voltage and three-phase current on the low-voltage side of all distribution substations within the target monitoring range from a provincial or regional measurement data center serving as a historical database; Step S12: For each distribution substation in the connection group to be identified, extract voltage sequence data and current sequence data from the historical database; obtain the three-phase voltage value and three-phase current value of each distribution substation at the latest collection time and the previous collection time for real-time fault identification; Step S13: performing validity check on the acquired voltage sequence data, current sequence data, three-phase voltage values, and three-phase current values, and eliminating damaged data and missing data.

[0011] Furthermore, step S2 is specifically as follows: Step S21: Obtain the voltage sequence data and current sequence data in step S1; calculate the correlation between the voltage sequence data, the voltage sequence data includes Phase voltage sequence, Phase voltage sequence and Phase voltage sequence; specifically: calculate Phase voltage sequence With phase b voltage sequence Pearson correlation coefficient ;calculate Phase voltage sequence and Phase voltage sequence Pearson correlation coefficient ;calculate Phase voltage sequence and Phase voltage sequence Pearson correlation coefficient ; Step S22: Calculate the correlation between the voltage variation and the current variation based on the current sequence data, specifically: Calculate the voltage change, specifically: Get the moment Corresponding Phase voltage change , expressed as: ; ; in, For the moment Corresponding Phase voltage, For the moment Corresponding Phase voltage; Calculate the current change, specifically: Get the moment Corresponding Phase current change , expressed as: ; ; in, For the moment Corresponding Phase current, For the moment Corresponding Phase current; Calculate the Pearson correlation coefficient between the voltage change and the current change of the corresponding phase, specifically: calculate Pearson correlation coefficient between phase voltage change and current change ; calculate Pearson correlation coefficient between phase voltage change and current change ; calculate Pearson correlation coefficient between phase voltage change and current change ; Step S23: Based on the calculated Pearson correlation coefficient, set the judgment threshold of the Pearson correlation coefficient between phase voltages The judgment threshold of the Pearson correlation coefficient of the voltage change and the current change ; The specific process of determining the Yyn0 connection group is as follows: If the following conditions are met at the same time: ; ; ; ; Then the connection group of the distribution transformer is determined to be Yyn0; The Dyn11 connection group determination process is as follows: If the judgment conditions of the Yyn0 connection group are not met at the same time, the connection group of the distribution transformer is determined to be Dyn11; Step S24: Associating the identified connection group information with the ID of the corresponding distribution transformer and the distribution substation and storing the associated information.

[0012] Furthermore, step S3 is specifically as follows: Step S31: Obtain voltage variation , current change , voltage change rate and current change rate ; expressed as: ; ; ; ; ; in, For time, is the time interval, To express farewell, To say goodbye The voltage corresponding to time t is To say goodbye The voltage change, To say goodbye The current corresponding to time t, To say goodbye The current change, To say goodbye The voltage change rate, To say goodbye The rate of change of current; To say goodbye exist Voltage corresponding to time; To say goodbye in Current corresponding to time; Step S32: Obtain voltage features through feature extraction, the voltage features include: Negative sequence voltage component , expressed as: ; Zero-sequence voltage component , expressed as: ; Voltage imbalance , expressed as: ; Maximum phase voltage drop rate , expressed as: ; Minimum phase voltage current value , specifically: ; in, represents the phase rotation operator; represents the maximum value function; represents the minimum evaluation function; Indicates the positive sequence voltage component obtained in real time; Indicates taking the absolute value; represents the core rotation operator; Further obtain current characteristics, which include: Negative sequence current component , expressed as: ; Zero-sequence current component , expressed as: ; Current imbalance , expressed as: ; Maximum phase current amplification rate , expressed as: ; in, Expressing separation The current corresponding to time t is, Expressing separation The current corresponding to time t is, Expressing separation The current corresponding to time t; Indicates the positive sequence current component obtained in real time; Step S33: Based on the connection group obtained in step S2, obtain the voltage characteristic enhancement value, specifically: If the connection group is Dyn11, assuming that the voltage drop occurs on two phases and the third phase is stable, calculate the voltage characteristic enhancement value; Voltage signature enhancement values include: The normalized deviation of the two-phase voltage and the voltage of the other phase , expressed as: ; Normalized fluctuation of stable phase voltage , expressed as: ; in, represents the instantaneous value of the voltage of the first phase at time t, represents the instantaneous value of the voltage of the second phase at time t, represents the instantaneous value of the third phase voltage at time t, is the rated phase voltage, which is 220V; Indicates the change in the instantaneous value of the third phase voltage; If the connection group is Yyn0, calculate the corresponding voltage characteristic enhancement value; The corresponding voltage characteristic enhancement values include: Ratio of minimum phase voltage to rated voltage , expressed as: ; in, is the minimum phase voltage amplitude; Average voltage drop rate of non-lowest voltage phase , that is, assuming that phase X is the phase with the lowest voltage, calculate the average voltage drop rate of the other two phases Y and Z, which can be expressed as: .

[0013] Furthermore, the total score function in step S4 Specifically: ; in, is the comprehensive score of voltage characteristics, is the comprehensive score of current characteristics, It is a composite score of the voltage reinforcement value for a specific connection group; pass Function Compute 、 and Rating value of if , it is determined to be a phase failure on the high-voltage side, where is the total score fault judgment threshold.

[0014] Further, The function is expressed as: ; in, express The input of the function, express The first threshold of the function, express The second threshold of the function, 、 and Respectively represent the scores corresponding to different inputs; The comprehensive score of voltage characteristics is expressed as:

[0015] ; in, 、 、 and Respectively represent the scoring weights of voltage unbalance, zero-sequence voltage component, maximum phase voltage drop rate and minimum phase voltage current value; The comprehensive score of the current characteristics is expressed as:

[0016] ; in, 、 and Respectively represent the scoring weights of current unbalance, zero-sequence current component and maximum phase current increase rate; The composite score based on the voltage enhancement value of a specific connection group is as follows: If the connection group is Dyn11, the corresponding comprehensive score is expressed as:

[0017] ; in, is the comprehensive score of the voltage enhancement value of the connection group Dyn11, represents the scoring weights of the two-phase voltages and the normalized deviation from the other phase voltage, The scoring weight represents the normalized fluctuation of the stable phase voltage; If the connection group is Yyn0, the comprehensive score is expressed as:

[0018] ; in, It represents the comprehensive score of the voltage enhancement value of the connection group Dyn11. and They represent the scoring weights of the ratio of the lowest phase voltage to the rated voltage and the average sag rate of the non-lowest voltage phases respectively.

[0019] The technical effects of the present invention include:

[0020] This invention significantly improves the accuracy of high-voltage side phase loss fault diagnosis: Based on the automatic identification of connection groups, differentiated, refined analysis rules based on voltage variation characteristics are designed for two common distribution transformer connection groups, Dyn11 and Yyn0, and current variation characteristics are combined to assist in differentiation. Compared with universal imbalance criteria or simple logic that does not distinguish between connection groups, this method can more accurately identify high-voltage side phase loss faults, effectively reducing false alarm and missed alarm rates. Furthermore, this invention deeply focuses on how to improve the diagnostic accuracy of the specific high-voltage side phase loss fault, using low-frequency measurement data to provide a high-precision, modeled fault diagnosis and early warning method. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a flowchart of the steps of this application. DETAILED DESCRIPTION

[0022] The following describes the embodiments of the present invention through specific examples. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention.

[0023] Reference Figure 1 , a method for online monitoring and early warning of abnormal conditions in a distribution station area, comprising: Step S1: Obtain voltage sequence data and current sequence data of the distribution substation from the historical database; collect three-phase voltage values and three-phase current values of the distribution substation in real time; Step S2: Calculate the Pearson correlation coefficient between the voltage sequence data and the current sequence data in step S1; determine the connection group of the distribution transformer corresponding to the distribution station area according to the preset threshold and judgment conditions using the Pearson correlation coefficient; the connection groups include Yyn0 and Dyn11; Step S3: Obtain the three-phase voltage values and three-phase current values obtained in step S1, perform feature extraction, and obtain voltage features and current features respectively; and obtain a voltage feature enhancement value based on the connection group obtained in step S2; Step S4: Construct a total scoring function, use the total scoring function to score the voltage characteristics, current characteristics and voltage characteristic enhancement values based on the connection group, compare the scores with the preset thresholds, and determine whether a fault occurs in the distribution substation area.

[0024] Furthermore, step S1 is specifically as follows: Step S11: obtaining the measurement data of the three-phase voltage and three-phase current on the low-voltage side of all distribution substations within the target monitoring range from a provincial or regional measurement data center serving as a historical database; Step S12: For each distribution substation in the connection group to be identified, extract voltage sequence data and current sequence data from the historical database; obtain the three-phase voltage value and three-phase current value of each distribution substation at the latest collection time and the previous collection time for real-time fault identification; Step S13: performing validity check on the acquired voltage sequence data, current sequence data, three-phase voltage values, and three-phase current values, and eliminating damaged data and missing data.

[0025] Furthermore, step S2 is specifically as follows: Step S21: Obtain the voltage sequence data and current sequence data in step S1; calculate the correlation between the voltage sequence data, the voltage sequence data includes Phase voltage sequence, Phase voltage sequence and Phase voltage sequence; specifically: calculate Phase voltage sequence With phase b voltage sequence Pearson correlation coefficient ;calculate Phase voltage sequence and Phase voltage sequence Pearson correlation coefficient ;calculate Phase voltage sequence and Phase voltage sequence Pearson correlation coefficient ; Step S22: Calculate the correlation between the voltage variation and the current variation based on the current sequence data, specifically: Calculate the voltage change, specifically: Get the moment Corresponding Phase voltage change , expressed as: ; ; in, For the moment Corresponding Phase voltage, For the moment Corresponding Phase voltage; Calculate the current change, specifically: Get the moment Corresponding Phase current change , expressed as: ; ; in, For the moment Corresponding Phase current, For the moment Corresponding Phase current; Calculate the Pearson correlation coefficient between the voltage change and the current change of the corresponding phase, specifically: calculate Pearson correlation coefficient between phase voltage change and current change ; calculate Pearson correlation coefficient between phase voltage change and current change ; calculate Pearson correlation coefficient between phase voltage change and current change ; Step S23: Based on the calculated Pearson correlation coefficient, set the judgment threshold of the Pearson correlation coefficient between phase voltages The judgment threshold of the Pearson correlation coefficient of the voltage change and the current change ; The specific process of determining the Yyn0 connection group is as follows: If the following conditions are met at the same time: ; ; ; ; Then the connection group of the distribution transformer is determined to be Yyn0; The Dyn11 connection group determination process is as follows: If the judgment conditions of the Yyn0 connection group are not met at the same time, the connection group of the distribution transformer is determined to be Dyn11; Step S24: Associating the identified connection group information with the ID of the corresponding distribution transformer and the distribution substation and storing the associated information.

[0026] Furthermore, step S3 is specifically as follows: Step S31: Obtain voltage variation , current change , voltage change rate and current change rate ; expressed as: ; ; ; ; ; in, For time, is the time interval, To express farewell, To say goodbye The voltage corresponding to time t is To say goodbye The voltage change, To say goodbye The current corresponding to time t, To say goodbye The current change, To say goodbye The voltage change rate, To say goodbye The rate of change of current; To say goodbye exist Voltage corresponding to time; To say goodbye in Current corresponding to time; Step S32: Obtain voltage features through feature extraction, the voltage features include: Negative sequence voltage component , expressed as: ; Zero-sequence voltage component , expressed as: ; Voltage imbalance , expressed as: ; Maximum phase voltage drop rate , expressed as: ; Minimum phase voltage current value , specifically: ; in, represents the phase rotation operator; represents the maximum value function; represents the minimum evaluation function; Indicates the positive sequence voltage component obtained in real time; Indicates taking the absolute value; represents the core rotation operator; Further obtain current characteristics, which include: Negative sequence current component , expressed as: ; Zero-sequence current component , expressed as: ; Current imbalance , expressed as: ; Maximum phase current amplification rate , expressed as: ; in, Expressing separation The current corresponding to time t is, Expressing separation The current corresponding to time t is, Expressing separation The current corresponding to time t; Indicates the positive sequence current component obtained in real time; Step S33: Based on the connection group obtained in step S2, obtain the voltage characteristic enhancement value, specifically: If the connection group is Dyn11, assuming that the voltage drop occurs on two phases and the third phase is stable, calculate the voltage characteristic enhancement value; Voltage signature enhancement values include: The normalized deviation of the two-phase voltage and the voltage of the other phase , expressed as: ; Normalized fluctuation of stable phase voltage , expressed as: ; in, represents the instantaneous value of the voltage of the first phase at time t, represents the instantaneous value of the voltage of the second phase at time t, represents the instantaneous value of the third phase voltage at time t, is the rated phase voltage, which is 220V; Indicates the change in the instantaneous value of the third phase voltage; If the connection group is Yyn0, calculate the corresponding voltage characteristic enhancement value; The corresponding voltage characteristic enhancement values include: Ratio of minimum phase voltage to rated voltage , expressed as: ; in, is the minimum phase voltage amplitude; Average voltage drop rate of non-lowest voltage phase , that is, assuming that phase X is the phase with the lowest voltage, calculate the average voltage drop rate of the other two phases Y and Z, which can be expressed as: .

[0027] Furthermore, the total score function in step S4 Specifically: ; in, is the comprehensive score of voltage characteristics, is the comprehensive score of current characteristics, It is a composite score of the voltage reinforcement value for a specific connection group; pass Function Compute 、 and Rating value of if , it is determined to be a phase failure on the high-voltage side, where is the total score fault judgment threshold.

[0028] Further, The function is expressed as: ; in, express The input of the function, express The first threshold of the function, express The second threshold of the function, 、 and Respectively represent the scores corresponding to different inputs; The comprehensive score of voltage characteristics is expressed as:

[0029] ; in, 、 、 and Respectively represent the scoring weights of voltage unbalance, zero-sequence voltage component, maximum phase voltage drop rate and minimum phase voltage current value; The comprehensive score of the current characteristics is expressed as:

[0030] ; in, 、 and Respectively represent the scoring weights of current unbalance, zero-sequence current component and maximum phase current increase rate; The composite score based on the voltage enhancement value of a specific connection group is as follows: If the connection group is Dyn11, the corresponding comprehensive score is expressed as:

[0031] ; in, is the comprehensive score of the voltage enhancement value of the connection group Dyn11, represents the scoring weights of the two-phase voltages and the normalized deviation from the other phase voltage, The scoring weight represents the normalized fluctuation of the stable phase voltage; If the connection group is Yyn0, the comprehensive score is expressed as:

[0032] ; in, It represents the comprehensive score of the voltage enhancement value of the connection group Dyn11. and They represent the scoring weights of the ratio of the lowest phase voltage to the rated voltage and the average sag rate of the non-lowest voltage phases respectively.

[0033] The present invention has been described in detail above with reference to the embodiments of the accompanying drawings. A person skilled in the art can make various modifications to the present invention based on the above description. Therefore, certain details in the embodiments should not be construed as limiting the present invention. The present invention shall be protected within the scope defined by the appended claims.

Claims

1. A method for online monitoring and early warning of abnormal conditions in a distribution station area, characterized in that: include: Step S1: Obtain voltage sequence data and current sequence data of the distribution station area from the historical database; Real-time collection of three-phase voltage and three-phase current values in the distribution area; Step S2: Calculate the Pearson correlation coefficient between the voltage sequence data and the current sequence data in step S1; determine the connection group of the distribution transformer corresponding to the distribution station area according to the preset threshold and judgment conditions using the Pearson correlation coefficient; the connection groups include Yyn0 and Dyn11; Step S3: Obtain the three-phase voltage values and three-phase current values in step S1, perform feature extraction, and obtain voltage features and current features respectively; and obtaining a voltage characteristic enhancement value based on the connection group obtained in step S2; Step S4: Construct a total scoring function, use the total scoring function to score the voltage characteristics, current characteristics and voltage characteristic enhancement values based on the connection group, compare the scores with the preset thresholds, and determine whether a fault occurs in the distribution substation area.

2. The method for online monitoring and early warning of abnormal conditions in a distribution station area according to claim 1, characterized in that: Step S1 is specifically as follows: Step S11: obtaining the measurement data of the three-phase voltage and three-phase current on the low-voltage side of all distribution substations within the target monitoring range from a provincial or regional measurement data center serving as a historical database; Step S12: For each distribution substation in the connection group to be identified, extract voltage sequence data and current sequence data from the historical database; obtain the three-phase voltage value and three-phase current value of each distribution substation at the latest collection time and the previous collection time for real-time fault identification; Step S13: performing validity check on the acquired voltage sequence data, current sequence data, three-phase voltage values, and three-phase current values, and eliminating damaged data and missing data.

3. The method for online monitoring and early warning of abnormal conditions in a distribution station area according to claim 1, characterized in that: Step S2 is specifically as follows: Step S21: Obtain the voltage sequence data and current sequence data in step S1; calculate the correlation between the voltage sequence data, the voltage sequence data includes Phase voltage sequence, Phase voltage sequence and Phase voltage sequence; specifically: calculate Phase voltage sequence With phase b voltage sequence Pearson correlation coefficient ;calculate Phase voltage sequence and Phase voltage sequence Pearson correlation coefficient ;calculate Phase voltage sequence and Phase voltage sequence Pearson correlation coefficient ; Step S22: Calculate the correlation between the voltage variation and the current variation based on the current sequence data, specifically: Calculate the voltage change, specifically: Get the moment Corresponding Phase voltage change , expressed as: ； ； in, For the moment Corresponding Phase voltage, For the moment Corresponding Phase voltage; Calculate the current change, specifically: Get the moment Corresponding Phase current change , expressed as: ； ； in, For the moment Corresponding Phase current, For the moment Corresponding Phase current; Calculate the Pearson correlation coefficient between the voltage change and the current change of the corresponding phase, specifically: calculate Pearson correlation coefficient between phase voltage change and current change ; calculate Pearson correlation coefficient between phase voltage change and current change ; calculate Pearson correlation coefficient between phase voltage change and current change ; Step S23: Based on the calculated Pearson correlation coefficient, set the judgment threshold of the Pearson correlation coefficient between phase voltages The judgment threshold of the Pearson correlation coefficient of the voltage change and the current change ; The specific process of determining the Yyn0 connection group is as follows: If the following conditions are met at the same time: ； ； ； ； Then the connection group of the distribution transformer is determined to be Yyn0; The Dyn11 connection group determination process is as follows: If the judgment conditions of the Yyn0 connection group are not met at the same time, the connection group of the distribution transformer is determined to be Dyn11; Step S24: Associating the identified connection group information with the ID of the corresponding distribution transformer and the distribution substation and storing the associated information.

4. The method for online monitoring and early warning of abnormal conditions in a distribution station area according to claim 3 is characterized in that: Step S3 is specifically as follows: Step S31: Obtain voltage variation , current change , voltage change rate and current change rate ; expressed as: ； ； ； ； ； in, For time, is the time interval, To express farewell, To say goodbye The voltage corresponding to time t is To say goodbye The voltage change, To say goodbye The current corresponding to time t, To say goodbye The current change, To say goodbye The voltage change rate, To say goodbye The rate of change of current; To say goodbye exist Voltage corresponding to time; To say goodbye in Current corresponding to time; Step S32: Obtain voltage features through feature extraction, the voltage features include: Negative sequence voltage component , expressed as: ； Zero-sequence voltage component , expressed as: ； Voltage imbalance , expressed as: ； Maximum phase voltage drop rate , expressed as: ； Minimum phase voltage current value , specifically: ； in, represents the phase rotation operator; represents the maximum value function; represents the minimum evaluation function; Indicates the positive sequence voltage component obtained in real time; Indicates taking the absolute value; represents the core rotation operator; Further obtain current characteristics, which include: Negative sequence current component , expressed as: ； Zero-sequence current component , expressed as: ； Current imbalance , expressed as: ； Maximum phase current amplification rate , expressed as: ； in, Expressing separation The current corresponding to time t is, Expressing separation The current corresponding to time t is, Expressing separation The current corresponding to time t; Indicates the positive sequence current component obtained in real time; Step S33: Based on the connection group obtained in step S2, obtain the voltage characteristic enhancement value, specifically: If the connection group is Dyn11, assuming that the voltage drop occurs on two phases and the third phase is stable, calculate the voltage characteristic enhancement value; Voltage signature enhancement values include: The normalized deviation of the two-phase voltage and the voltage of the other phase , expressed as: ; Normalized fluctuation of stable phase voltage , expressed as: ； in, represents the instantaneous value of the voltage of the first phase at time t, represents the instantaneous value of the voltage of the second phase at time t, represents the instantaneous value of the third phase voltage at time t, is the rated phase voltage, which is 220V; Indicates the change in the instantaneous value of the third phase voltage; If the connection group is Yyn0, calculate the corresponding voltage characteristic enhancement value; The corresponding voltage characteristic enhancement values include: Ratio of minimum phase voltage to rated voltage , expressed as: ； in, is the minimum phase voltage amplitude; Average voltage drop rate of non-lowest voltage phase , that is, assuming that phase X is the phase with the lowest voltage, calculate the average voltage drop rate of the other two phases Y and Z, which can be expressed as: .

5. The method for online monitoring and early warning of abnormal conditions in a distribution station area according to claim 4, characterized in that: The total score function in step S4 Specifically: ； in, is the comprehensive score of voltage characteristics, is the comprehensive score of current characteristics, It is a composite score of the voltage reinforcement value for a specific connection group; pass Function Compute 、 and Rating value of if , it is determined to be a phase failure on the high-voltage side, where is the total score fault judgment threshold.

6. The method for online monitoring and early warning of abnormal conditions in a distribution station area according to claim 5, characterized in that: The function is expressed as: ； in, express The input of the function, express The first threshold of the function, express The second threshold of the function, 、 and Respectively represent the scores corresponding to different inputs; The comprehensive score of voltage characteristics is expressed as: ； in, 、 、 and Respectively represent the scoring weights of voltage unbalance, zero-sequence voltage component, maximum phase voltage drop rate and minimum phase voltage current value; The comprehensive score of the current characteristics is expressed as: ； in, 、 and Respectively represent the scoring weights of current unbalance, zero-sequence current component and maximum phase current increase rate; The composite score based on the voltage enhancement value of a specific connection group is as follows: If the connection group is Dyn11, the corresponding comprehensive score is expressed as: ； in, is the comprehensive score of the voltage enhancement value of the connection group Dyn11, represents the scoring weights of the two-phase voltages and the normalized deviation from the other phase voltage, The scoring weight represents the normalized fluctuation of the stable phase voltage; If the connection group is Yyn0, the comprehensive score is expressed as: ； in, It represents the comprehensive score of the voltage enhancement value of the connection group Dyn11. and They represent the scoring weights of the ratio of the lowest phase voltage to the rated voltage and the average sag rate of the non-lowest voltage phases respectively.

Images