Distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals

Abstract

The present invention discloses a distributed photovoltaic grid-connected safety monitoring system and method based on four fusion terminals, which relate to the technical field of power system monitoring. The system includes: a data acquisition module, a first preprocessing module, a reference vector management module, a second preprocessing module, a matrix management module, a real-time monitoring module and an information feedback module. The data acquisition module is used to classify according to all parameters of the same data acquisition and different acquisitions of the same parameters. The first preprocessing module is used to normalize the parameters. The reference vector management module is used to iteratively output reference vectors for historical data vectors. The second preprocessing module is used to classify and obtain historical data of the same parameters and calculate the statistics of the power grid data. The matrix management module is used to establish a relationship matrix of data items. The real-time monitoring module is used to obtain real-time monitoring data of the distributed power grid and normalize the real-time data. The information feedback module is used to obtain risk assessment value information.

Description

Technical Field

[0001] The present invention relates to the technical field of power system monitoring, and in particular to a distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals. Background Art

[0002] Distributed photovoltaic systems have the advantages of local power generation, local grid connection, local conversion, and local use. They are highly efficient, can alleviate electricity shortages, and are environmentally friendly. They solve the problems of traditional centralized photovoltaic power generation systems, such as long transmission distances, high relay consumption, and complex power dispatching methods that cannot flexibly allocate power.

[0003] Distributed photovoltaic systems are typically built close to users, with common installation locations including residential and commercial rooftops, as well as factory rooftops. These locations are close to people's homes and workplaces, making the operational safety of distributed photovoltaic systems particularly important. Distributed photovoltaic systems operate in a complex environment, susceptible to both human and natural influences. For example, increased contact resistance or loose connectors at the grid-connected interface can hinder the smooth flow of current, resulting in leakage current. Wear, aging, or moisture in insulating materials can also degrade their insulation performance, allowing current to leak through weak points. Therefore, real-time monitoring of the operating status of distributed power grids and timely detection of abnormal conditions are essential to improve system safety and efficiency. Summary of the Invention

[0004] The purpose of the present invention is to provide a distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals, the system comprising:

[0006] Data acquisition module, first preprocessing module, reference vector management module, second preprocessing module, matrix management module, real-time monitoring module and information feedback module;

[0007] The four fusion requirements of photovoltaics are observable, measurable, controllable and adjustable;

[0008] Visible includes: Real-time observation of the power plant's operating status through an intelligent monitoring system, including key information such as equipment performance, environmental conditions, and power output, enabling timely identification and resolution of potential problems;

[0009] Measurable features include: accurate prediction and measurement of power generation and various operating data of power plants through big data models, providing a solid foundation for optimizing power plant performance;

[0010] Controllable includes: remote or automatic control through a digital monitoring platform to ensure safe and reliable operation of the power station;

[0011] Adjustability includes: through intelligent adjustment means, the power station can flexibly respond to changes in grid load and frequency fluctuations;

[0012] The present invention combines observation data to extract the changing rules of photovoltaic power grid operation, extracts the operating characteristics of photovoltaic power grid, further establishes judgment and reference evaluation models, accurately outputs the evaluation results of power grid operation data, and provides a basis for precise regulation of distributed photovoltaic power grid.

[0013] Furthermore, the data acquisition module is used to collect historical data of the load power, terminal temperature, and terminal leakage current of the distributed power grid, and classify them according to all parameters collected in the same time and different collections of the same parameters; the first preprocessing module is used to normalize all parameters collected in the same time and form the normalized data into a historical data vector;

[0014] The reference vector management module is used to obtain several historical data vectors from the first preprocessing module, iterate the historical data vectors and output reference vectors; the second preprocessing module is used to classify and obtain historical data of the same parameters, and calculate the statistics of three data items: the load power of the distributed power grid, the temperature of the terminal and the leakage current of the terminal; the matrix management module is used to obtain the statistics of the three data items, calculate the correlation coefficients between the data items, and establish a relationship matrix of the data items; the real-time monitoring module is used to obtain real-time monitoring data of the distributed power grid and normalize the real-time data; the information feedback module is used to calculate the difference between the real-time monitoring data and the historical records of the distributed power grid, obtain the risk assessment value, and issue an alarm message when the difference exceeds the threshold.

[0015] Furthermore, the data acquisition module is connected to the first preprocessing module and the second preprocessing module respectively, the first preprocessing module is connected to the reference vector management module, the second preprocessing module is connected to the matrix management module, and the information feedback module is connected to the real-time monitoring module, the reference vector management module and the matrix management module respectively.

[0016] Furthermore, the data acquisition module includes: a sensor management unit, a historical data management unit and a data classification unit;

[0017] The sensor management unit is used to manage load power sensors, temperature sensors and leakage current sensors. The load power sensor is used to collect the system load power of a distributed photovoltaic power generation system, the temperature sensor is used to collect the temperature of the wiring terminals at the grid-connected location of a distributed photovoltaic power generation system, and the leakage current sensor is used to collect the leakage current of the wiring terminals at the grid-connected location of a distributed photovoltaic power generation system.

[0018] The historical data management unit is used to collect the system load power, terminal temperature and terminal leakage current at the same time in batches and collect them into data groups. Each data collection corresponds to a data group, and several data groups are collected and recorded in the historical data set.

[0019] The data classification unit is used to classify the data in the historical data set according to the type of data, wherein all system load powers in the historical data set are collected and recorded in the system load power record set, all temperatures of the wiring terminals in the historical data set are collected and recorded in the temperature record set of the wiring terminals, and all leakage currents of the wiring terminals in the historical data set are collected and recorded in the leakage current set of the wiring terminals.

[0020] Furthermore, the first pre-processing module includes: a power processing unit, a temperature processing unit, a leakage current processing unit and a history vector management unit;

[0021] The power processing unit is used to normalize the system load power to obtain normalized power, wherein the rated power P of the preset system load is rated , calculate the normalized power P norm , P norm =P history / P rated , P history Indicates the actual historical value of system load power;

[0022] The temperature processing unit is used to normalize the temperature of the terminal to obtain a normalized temperature, wherein the preset terminal rated temperature T rated , obtain the ambient temperature T at the time of data collection amb , calculate the normalized temperature T norm , T norm =(T history -T amb ) / T rated , T history Indicates the actual historical value of the temperature of the terminal;

[0023] The leakage current processing unit is used to normalize the leakage current of the terminal to obtain the normalized leakage current, wherein the leakage threshold I is preset. hold , calculate the normalized leakage current I leaknorm , I leaknorm =I leak / I hold , I leak Indicates the actual historical value of the leakage current of the terminal;

[0024] The history vector management unit is used to collect the normalized power, normalized temperature and normalized leakage current corresponding to the system load power, terminal temperature and terminal leakage current in each data group, and form the history data vector X corresponding to each data group, X = (P norm , T norm , I leaknorm );

[0025] After data normalization, the evaluation criteria for each data item are unified, the impact of different dimensions between different parameters is reduced, and the accuracy of the data evaluation process is further improved.

[0026] Furthermore, the reference vector management module includes: a data set management unit, an initial vector management unit, a vector iteration unit and a reference vector management unit;

[0027] The data set management unit is used to collect a plurality of historical data vectors, and divide the plurality of historical data vectors into two vector groups, the two vector groups being respectively recorded as a first vector group and a second vector group, wherein the total number of historical data vectors in the second vector group is greater than the total number of historical data vectors in the first vector group;

[0028] The initial vector management unit is used to obtain all historical data vectors in the first vector group, calculate the average value of all normalized power, normalized temperature and normalized leakage current in the historical data vectors, obtain the average normalized power P0, evaluate the normalized temperature T0 and the average normalized leakage current I0, and form the initial vector k0, k0 = (P0, T0, I0);

[0029] The vector iteration unit is used to obtain the initial vector k0 to iterate all the historical data vectors in the second vector group, wherein the iteration formula is: k m =α×q m +(1-α)×k m-1 , k m represents the output result of the mth iteration, q m represents the mth historical data vector in the second vector group, α is the balance coefficient, and satisfies the condition 0<α<1;

[0030] This scheme uses an exponential moving average to improve the accuracy of the reference vector. Considering that the operation of the photovoltaic system may have periodic changes, such as the large difference in power between daytime and nighttime, the direct use of simple average cannot accurately describe the changes in the distributed photovoltaic system, which will introduce more errors in the subsequent comparison process. The iterative method is used to eliminate the calculation errors caused by discrete values ​​as much as possible.

[0031] The reference vector management unit is used to output the result vector when the iterated vector converges, and use the result vector as the reference vector, wherein the total number N1 of historical data vectors in the first vector group is obtained, when m>N1, and ||k m -k m-N1 || / || k m-N1 When ||<δ, output k m , the output vector at this time is recorded as the reference vector μ, where δ represents the vector difference threshold;

[0032] After several iterations, if the iteration result tends to be stable, the iteration result is output as the reference vector. This iteration vector includes the average information of the average value and the change trend brought about by the system change, which is more in line with the actual situation of the power grid system and is conducive to improving the accuracy of the subsequent judgment process.

[0033] Furthermore, the second preprocessing module includes: a mean calculation unit, a variance calculation unit, and a covariance calculation unit;

[0034] The average calculation unit is used to obtain the load power record set, temperature record set and leakage current set from the data classification unit and calculate the average load power P ave , average temperature T ave and the average leakage current I ave ;

[0035] The variance calculation unit is used to respectively calculate the variance Var (P) of the load power of the distributed power grid, the variance Var (T) of the temperature of the terminal, and the variance Var (I) of the leakage current of the terminal;

[0036] The covariance calculation unit is used to calculate the covariance Cov(P, T) of the load power and temperature, the covariance Cov(T, I) of the temperature and leakage, and the covariance Cov(P, I) of the power and leakage.

[0037] Furthermore, the matrix management module includes: a relationship coefficient calculation unit and a relationship matrix management unit;

[0038] The relationship coefficient calculation unit is used to calculate the relationship coefficient ρ xy , , x∈{P, T, I}, y∈{P, T, I}, and x≠y, where {P, T, I} represents a parameter set, in which P represents a corresponding parameter for calculating the load power of the distributed power grid, T represents a corresponding parameter for the temperature of the terminal, and I represents a corresponding parameter for the temperature of the terminal;

[0039] The relationship matrix management unit is used to traverse the parameter set and manage the relationship matrix G, G= .

[0040] Furthermore, the real-time monitoring module includes: a real-time data acquisition unit and a real-time data processing unit;

[0041] The real-time data acquisition unit is used to collect real-time data of a distributed photovoltaic power generation system. The real-time data includes the real-time value P of the system load power of a distributed photovoltaic power generation system. cur , the real-time value of the temperature of the connection terminal at the grid connection point of a distributed photovoltaic power generation system T cur The real-time value of the leakage current I of the terminal at the grid connection point of a distributed photovoltaic power generation system cur , collect the collected data into real-time data group (P cur , T cur , I cur );

[0042] The real-time data processing unit is used to normalize the real-time data to obtain the real-time data vector Y, Y = (P sta , T sta , I sta ), where P sta ==P cur / P rated , T sta =(T cur -T 2 amb ) / T rated , I sta = I cur / I hold , T 2 amb Indicates the ambient temperature of a distributed photovoltaic power generation system when the real-time data group is collected.

[0043] Furthermore, the information feedback module includes: a difference calculation unit, a threshold management unit and an alarm unit;

[0044] The difference calculation unit is used to obtain the real-time data vector, the management relationship matrix and the reference vector, and calculate the difference D. ,

[0045] Wherein, T in the formula for calculating the difference degree D represents the matrix transposition operator, and μ represents the reference vector;

[0046] The threshold management unit is used to manage the alarm threshold γ;

[0047] The alarm unit is used to issue an alarm to the management personnel of a distributed photovoltaic power generation system, wherein the alarm condition is D>γ.

[0048] Compared with existing technologies, the present invention offers the following advantages: by classifying and analyzing historical data from distributed photovoltaic systems, the characteristics of individual parameters and the relationships between them are fully explored. Furthermore, through an iterative approach, reference values ​​are derived from the historical data. By integrating the reference values ​​with the relationships between the parameters, an evaluation model for real-time distributed photovoltaic system data is constructed, eliminating the impact of dimensional differences between the parameters and providing timely, accurate distributed photovoltaic system monitoring information. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a schematic diagram of the first structure of the distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals of the present invention;

[0050] Figure 2 This is a second structural diagram of the distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals of the present invention. DETAILED DESCRIPTION

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0052] Example: Figure 1 and Figure 2 As shown, the present invention provides a technical solution, a distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals;

[0053] The system includes: a data acquisition module, a first preprocessing module, a reference vector management module, a second preprocessing module, a matrix management module, a real-time monitoring module and an information feedback module;

[0054] Among them, the data acquisition module is connected to the first preprocessing module and the second preprocessing module respectively, the first preprocessing module is connected to the reference vector management module, the second preprocessing module is connected to the matrix management module, and the information feedback module is connected to the real-time monitoring module, the reference vector management module and the matrix management module respectively.

[0055] The data acquisition module is used to collect historical data of the load power, terminal temperature and terminal leakage current of the distributed power grid, and classify them according to all parameters collected in the same time and different collections of the same parameters;

[0056] Among them, the data acquisition module includes: sensor management unit, historical data management unit and data classification unit;

[0057] The sensor management unit is used to manage load power sensors, temperature sensors and leakage current sensors. The load power sensor is used to collect the system load power of a distributed photovoltaic power generation system, the temperature sensor is used to collect the temperature of the wiring terminals at the grid-connected location of a distributed photovoltaic power generation system, and the leakage current sensor is used to collect the leakage current of the wiring terminals at the grid-connected location of a distributed photovoltaic power generation system.

[0058] The historical data management unit is used to collect the system load power, terminal temperature and terminal leakage current at the same time in batches and collect them into data groups. Each data collection corresponds to a data group, and several data groups are collected and recorded in the historical data set.

[0059] The data classification unit is used to classify the data in the historical data set according to the type of data, wherein all system load powers in the historical data set are collected and recorded in the system load power record set, all temperatures of the wiring terminals in the historical data set are collected and recorded in the temperature record set of the wiring terminals, and all leakage currents of the wiring terminals in the historical data set are collected and recorded in the leakage current set of the wiring terminals.

[0060] The first pre-processing module is used to normalize all parameters collected at the same time and form the normalized data into a historical data vector;

[0061] The first pre-processing module includes: a power processing unit, a temperature processing unit, a leakage current processing unit and a history vector management unit;

[0062] The power processing unit is used to normalize the system load power to obtain normalized power, wherein the rated power P of the preset system load is rated , calculate the normalized power P norm , P norm =P history / P rated , P history Indicates the actual historical value of system load power;

[0063] The temperature processing unit is used to normalize the temperature of the terminal to obtain a normalized temperature, wherein the preset terminal rated temperature T rated , obtain the ambient temperature T at the time of data collection amb , calculate the normalized temperature T norm , T norm =(T history -T amb ) / T rated , T history Indicates the actual historical value of the temperature of the terminal;

[0064] The leakage current processing unit is used to normalize the leakage current of the terminal to obtain the normalized leakage current, wherein the leakage threshold I is preset. hold , calculate the normalized leakage current I leaknorm , I leaknorm =I leak / I hold , I leak Indicates the actual historical value of the leakage current of the terminal;

[0065] For example, in a distributed photovoltaic system with a rated power of 20kW, the data set measured in a certain measurement is (15.2kW, 48℃, 5mA), and the ambient temperature at this time is 28℃;

[0066] Set T rated =57℃, I hold =30mA;

[0067] P norm =15.2 / 20=0.76, T norm = (48-28) / 57 = 0.35, I leaknorm =5 / 30=0.17;

[0068] The history vector management unit is used to collect the normalized power, normalized temperature and normalized leakage current corresponding to the system load power, terminal temperature and terminal leakage current in each data group, and form the history data vector X corresponding to each data group, X = (P norm , T norm , I leaknorm );

[0069] In the embodiment, one of the historical data vectors is (0.76, 0.35, 0.17).

[0070] The reference vector management module is used to obtain a number of historical data vectors from the first pre-processing module, and iterate the historical data vectors to output reference vectors;

[0071] The reference vector management module includes: a data set management unit, an initial vector management unit, a vector iteration unit and a reference vector management unit;

[0072] The data set management unit is used to collect a plurality of historical data vectors, and divide the plurality of historical data vectors into two vector groups, the two vector groups being respectively recorded as a first vector group and a second vector group, wherein the total number of historical data vectors in the second vector group is greater than the total number of historical data vectors in the first vector group;

[0073] The initial vector management unit is used to obtain all historical data vectors in the first vector group, calculate the average value of all normalized power, normalized temperature and normalized leakage current in the historical data vectors, obtain the average normalized power P0, evaluate the normalized temperature T0 and the average normalized leakage current I0, and form the initial vector k0, k0 = (P0, T0, I0);

[0074] The vector iteration unit is used to obtain the initial vector k0 to iterate all the historical data vectors in the second vector group, wherein the iteration formula is: k m =α×q m +(1-α)×k m-1 , k m represents the output result of the mth iteration, q m represents the mth historical data vector in the second vector group, α is the balance coefficient, and satisfies the condition 0<α<1;

[0075] In the embodiment, the value of α can be selected as 1 / N1. For example, if the first vector group includes 100 historical data vectors, α can be 0.01.

[0076] Calculate the average value of the 100 historical data vectors in the first vector group to obtain the initial vector k0, k0 = (0.72, 0.31, 0.15);

[0077] When the first historical data vector in the second vector group is (0.76, 0.35, 0.17);

[0078] Calculate k1 = 0.01 × (0.76, 0.35, 0.17) + 0.99 × (0.72, 0.32, 0.15) = (0.7204, 0.3203, 0.1502);

[0079] The reference vector management unit is used to output the result vector when the iterated vector converges, and use the result vector as the reference vector, wherein the total number N1 of historical data vectors in the first vector group is obtained, when m>N1, and ||k m -k m-N1 || / || k m-N1 When ||<δ, output k m , the vector output at this time is recorded as the reference vector μ, where δ represents the vector difference threshold. In the embodiment, δ defaults to 5%. After the system runs for a period of time, it can be adjusted according to the operation feedback results.

[0080] The second pre-processing module is used to classify and obtain historical data of the same parameters and calculate the statistics of three data items: load power of the distributed power grid, temperature of the terminal, and leakage current of the terminal.

[0081] The second preprocessing module includes: a mean calculation unit, a variance calculation unit and a covariance calculation unit;

[0082] The average calculation unit is used to obtain the load power record set, temperature record set and leakage current set from the data classification unit and calculate the average load power P ave , average temperature T ave and the average leakage current I ave ;

[0083] The variance calculation unit is used to respectively calculate the variance Var (P) of the load power of the distributed power grid, the variance Var (T) of the temperature of the terminal, and the variance Var (I) of the leakage current of the terminal;

[0084] The covariance calculation unit is used to calculate the covariance Cov(P, T) of the load power and temperature, the covariance Cov(T, I) of the temperature and leakage, and the covariance Cov(P, I) of the power and leakage.

[0085] In the embodiment, n data groups are collected and classified according to the load power of the distributed power grid, the temperature of the terminal and the leakage current of the terminal, and the average power is calculated respectively. , the average temperature and the average value of the leakage current ;

[0086] Covariance of power and temperature: ;

[0087] Covariance of temperature and leakage: ;

[0088] Covariance of power and leakage: ;

[0089] Among them, P i represents the load power of the distributed power grid in the i-th data group, T i Indicates the temperature of the terminal in the i-th data group, I i Indicates the leakage current of the terminal in the i-th data group.

[0090] The matrix management module is used to obtain the statistics of three data items, calculate the correlation coefficients between data items, and establish the relationship matrix of data items;

[0091] The matrix management module includes: a relationship coefficient calculation unit and a relationship matrix management unit;

[0092] The relationship coefficient calculation unit is used to calculate the relationship coefficient ρ xy , , x∈{P, T, I}, y∈{P, T, I}, and x≠y, where {P, T, I} represents a parameter set, in which P represents a corresponding parameter for calculating the load power of the distributed power grid, T represents a corresponding parameter for the temperature of the terminal, and I represents a corresponding parameter for the temperature of the terminal;

[0093] The relationship matrix management unit is used to traverse the parameter set and manage the relationship matrix G, G= ;

[0094] Due to the symmetry of the calculated values, ρ PT =ρ TP ,ρ PI =ρ IP ,ρ IT =ρ TI .

[0095] The real-time monitoring module is used to obtain real-time monitoring data of the distributed power grid and perform normalization processing on the real-time data;

[0096] Among them, the real-time monitoring module includes: a real-time data acquisition unit and a real-time data processing unit;

[0097] The real-time data acquisition unit is used to collect real-time data of a distributed photovoltaic power generation system. The real-time data includes the real-time value P of the system load power of a distributed photovoltaic power generation system. cur , the real-time value of the temperature of the connection terminal at the grid connection point of a distributed photovoltaic power generation system T cur The real-time value of the leakage current I of the terminal at the grid connection point of a distributed photovoltaic power generation system cur , collect the collected data into real-time data group (P cur , T cur , I cur );

[0098] The real-time data processing unit is used to normalize the real-time data to obtain the real-time data vector Y, Y = (P sta , T sta , I sta ), where P sta ==P cur / P rated , T sta =(T cur -T 2 amb ) / T rated , I sta = I cur / I hold , T 2 amb Indicates the ambient temperature of a distributed photovoltaic power generation system when the real-time data group is collected.

[0099] The information feedback module is used to calculate the difference between the real-time monitoring data and the historical records of the distributed power grid, and obtain the risk assessment value. When the difference exceeds the threshold, an alarm message is issued.

[0100] The information feedback module includes: a difference calculation unit, a threshold management unit and an alarm unit;

[0101] The difference calculation unit is used to obtain the real-time data vector, the management relationship matrix and the reference vector, and calculate the difference D. ,

[0102] Wherein, T in the formula for calculating the difference degree D represents the matrix transposition operator, and μ represents the reference vector;

[0103] The threshold management unit is used to manage the alarm threshold γ;

[0104] The alarm unit is used to issue an alarm to the management personnel of a distributed photovoltaic power generation system, where the alarm condition is D>γ;

[0105] In the embodiment, γ is 1, G is , the two real-time data vectors are recorded as Y1 and Y2, when Y1-μ=(0.2,-0.1,0.03), Y2-μ=(0.3,1.2,0.1);

[0106] Calculate the inverse matrix G of G -1 , G -1 = ;

[0107] Difference D1 = 0.246, no alarm is issued at this time;

[0108] The difference D2=1.144, and an alarm is issued at this time.

[0109] Methods include:

[0110] S1. Collect physical parameters in the distributed photovoltaic system, pre-process the physical parameters, and obtain historical data vectors and reference vectors;

[0111] S2. Obtain historical data of physical parameters and calculate statistics of three items of data: load power of the distributed power grid, temperature of the terminal blocks, and leakage current of the terminal blocks;

[0112] S3. Obtain statistics of three items of data: load power, terminal temperature, and terminal leakage current, calculate correlation coefficients between the data items, and establish a relationship matrix for the data items;

[0113] S4. Acquire real-time monitoring data of the distributed power grid and perform normalization processing on the real-time data of the distributed power grid;

[0114] S5. Calculate the difference between the real-time monitoring data and the distributed power grid historical records to obtain a risk assessment value. When the difference exceeds a threshold, an alarm message is issued.

[0115] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals, characterized by: The system includes: Data acquisition module, first preprocessing module, reference vector management module, second preprocessing module, matrix management module, real-time monitoring module and information feedback module; The data acquisition module is used to collect historical data of the load power, terminal temperature and terminal leakage current of the distributed power grid, and classify them according to all parameters collected in the same time and different collections of the same parameters; The first pre-processing module is used to normalize all parameters collected at the same time and form the normalized data into a historical data vector; The reference vector management module is used to obtain a number of historical data vectors from the first pre-processing module, and iterate the historical data vectors to output reference vectors; The reference vector management module includes: a data set management unit, an initial vector management unit, a vector iteration unit and a reference vector management unit; The data set management unit is used to collect a plurality of historical data vectors, and divide the plurality of historical data vectors into two vector groups, the two vector groups being respectively recorded as a first vector group and a second vector group, wherein the total number of historical data vectors in the second vector group is greater than the total number of historical data vectors in the first vector group; The initial vector management unit is used to obtain all historical data vectors in the first vector group, calculate the average value of all normalized power, normalized temperature and normalized leakage current in the historical data vectors, obtain the average normalized power P0, evaluate the normalized temperature T0 and the average normalized leakage current I0, and form the initial vector k0, k0 = (P0, T0, I0); The vector iteration unit is used to obtain the initial vector k0 to iterate all the historical data vectors in the second vector group, wherein the iteration formula is: k m =α×q m +(1-α)×k m-1 , k m represents the output result of the mth iteration, q m represents the mth historical data vector in the second vector group, α is the balance coefficient, and satisfies the condition 0<α<1; The reference vector management unit is used to output the result vector when the iterated vector converges, and use the result vector as the reference vector, wherein the total number N1 of historical data vectors in the first vector group is obtained, when m>N1, and ||k m -k m-N1 || / ||k m-N1 When ||<δ, output k m , the output vector at this time is recorded as the reference vector μ, where δ represents the vector difference threshold; The second pre-processing module is used to classify and obtain historical data of the same parameters and calculate the statistics of three data items: load power of the distributed power grid, temperature of the terminal, and leakage current of the terminal. The matrix management module is used to obtain the statistics of the three data items, calculate the correlation coefficients between the data items, and establish a relationship matrix of the data items; The real-time monitoring module is used to obtain real-time monitoring data of the distributed power grid and perform normalization processing on the real-time data; The information feedback module is used to calculate the difference between the real-time monitoring data and the historical records of the distributed power grid, and obtain the risk assessment value. When the difference exceeds the threshold, an alarm message is issued; The information feedback module includes: a difference calculation unit, a threshold management unit and an alarm unit; The difference calculation unit is used to obtain the real-time data vector, the management relationship matrix and the reference vector, and calculate the difference D. , Wherein, T in the formula for calculating the difference degree D represents the matrix transposition operator, μ represents the reference vector, Y represents the real-time data vector, and G represents the relationship matrix; The threshold management unit is used to manage the alarm threshold γ; The alarm unit is used to issue an alarm to the management personnel of a distributed photovoltaic power generation system, where the alarm condition is D>γ; Among them, the data acquisition module is connected to the first preprocessing module and the second preprocessing module respectively, the first preprocessing module is connected to the reference vector management module, the second preprocessing module is connected to the matrix management module, and the information feedback module is connected to the real-time monitoring module, the reference vector management module and the matrix management module respectively.

2. The distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals according to claim 1 is characterized by: The data acquisition module includes: sensor management unit, historical data management unit and data classification unit; The sensor management unit is used to manage load power sensors, temperature sensors and leakage current sensors. The load power sensor is used to collect the system load power of a distributed photovoltaic power generation system, the temperature sensor is used to collect the temperature of the wiring terminals at the grid-connected location of a distributed photovoltaic power generation system, and the leakage current sensor is used to collect the leakage current of the wiring terminals at the grid-connected location of a distributed photovoltaic power generation system. The historical data management unit is used to collect the system load power, terminal temperature and terminal leakage current at the same time in batches and collect them into data groups. Each data collection corresponds to a data group, and several data groups are collected and recorded in the historical data set. The data classification unit is used to classify the data in the historical data set according to the type of data, wherein all system load powers in the historical data set are collected and recorded in the system load power record set, all temperatures of the wiring terminals in the historical data set are collected and recorded in the temperature record set of the wiring terminals, and all leakage currents of the wiring terminals in the historical data set are collected and recorded in the leakage current set of the wiring terminals.

3. The distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals according to claim 2 is characterized by: The first pre-processing module includes: a power processing unit, a temperature processing unit, a leakage current processing unit and a history vector management unit; The power processing unit is used to normalize the system load power to obtain normalized power, wherein the rated power P of the preset system load is rated , calculate the normalized power P norm , P norm =P history / P rated , P history Indicates the actual historical value of system load power; The temperature processing unit is used to normalize the temperature of the terminal to obtain a normalized temperature, wherein the preset terminal rated temperature T rated , obtain the ambient temperature T at the time of data collection amb , calculate the normalized temperature T norm , T norm =(T history -T amb ) / T rated , T history Indicates the actual historical value of the temperature of the terminal; The leakage current processing unit is used to normalize the leakage current of the terminal to obtain the normalized leakage current, wherein the leakage threshold I is preset. hold , calculate the normalized leakage current I leaknorm , I leaknorm =I leak / I hold , I leak Indicates the actual historical value of the leakage current of the terminal; The history vector management unit is used to collect the normalized power, normalized temperature and normalized leakage current corresponding to the system load power, terminal temperature and terminal leakage current in each data group, and form the history data vector X corresponding to each data group, X = (P norm , T norm , I leaknorm ).

4. The distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals according to claim 3 is characterized by: The second preprocessing module includes: a mean calculation unit, a variance calculation unit and a covariance calculation unit; The average calculation unit is used to obtain the load power record set, temperature record set and leakage current set from the data classification unit and calculate the average load power P ave , average temperature T ave and the average leakage current I ave ; The variance calculation unit is used to respectively calculate the variance Var (P) of the load power of the distributed power grid, the variance Var (T) of the temperature of the terminal, and the variance Var (I) of the leakage current of the terminal; The covariance calculation unit is used to calculate the covariance Cov(P, T) of the load power and temperature, the covariance Cov(T, I) of the temperature and leakage, and the covariance Cov(P, I) of the power and leakage.

5. The distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals according to claim 4 is characterized by: The matrix management module includes: a relationship coefficient calculation unit and a relationship matrix management unit; The relationship coefficient calculation unit is used to calculate the relationship coefficient ρ xy , , x∈{P, T, I}, y∈{P, T, I}, and x≠y, where {P, T, I} represents a parameter set, in which P represents a corresponding parameter for calculating the load power of the distributed power grid, T represents a corresponding parameter for the temperature of the terminal, and I represents a corresponding parameter for the temperature of the terminal; The relationship matrix management unit is used to traverse the parameter set and manage the relationship matrix G, G= .

6. The distributed photovoltaic grid-connected safety monitoring system based on four fusion terminals according to claim 5 is characterized by: The real-time monitoring module includes: a real-time data acquisition unit and a real-time data processing unit; The real-time data acquisition unit is used to collect the real-time data of the distributed photovoltaic power generation system, and the real-time data includes the real-time value P of the system load power of the distributed photovoltaic power generation system. cur , the real-time value of the temperature of the connection terminal at the grid connection point of a distributed photovoltaic power generation system T cur The real-time value of the leakage current I of the terminal at the grid connection point of a distributed photovoltaic power generation system cur , collect the collected data into real-time data group (P cur , T cur , I cur ); The real-time data processing unit is used to normalize the real-time data to obtain the real-time data vector Y, Y = (P sta , T sta , I sta ), where P sta ==P cur / P rated , T sta =(T cur -T 2 amb ) / T rated , I sta = I cur / I hold , T 2 amb Indicates the ambient temperature of a distributed photovoltaic power generation system when the real-time data group is collected.

7. A distributed photovoltaic grid-connected safety monitoring method based on four fusionable terminals, used in a distributed photovoltaic grid-connected safety monitoring system based on four fusionable terminals as claimed in any one of claims 1 to 6, characterized in that: Methods include: S1. Collect physical parameters in the distributed photovoltaic system, pre-process the physical parameters, and obtain historical data vectors and reference vectors; S2. Obtain historical data of physical parameters and calculate statistics of three items of data: load power of the distributed power grid, temperature of the terminal blocks, and leakage current of the terminal blocks; S3. Obtain statistics of three items of data: load power, terminal temperature, and terminal leakage current, calculate correlation coefficients between the data items, and establish a relationship matrix for the data items; S4. Acquire real-time monitoring data of the distributed power grid and perform normalization processing on the real-time data of the distributed power grid; S5. Calculate the difference between the real-time monitoring data and the distributed power grid historical records to obtain a risk assessment value. When the difference exceeds a threshold, an alarm message is issued.

Images