A real-time operating status assessment method for distribution transformers

Abstract

The invention discloses a method for evaluating the real-time operation state of a distribution transformer, which comprises: step 1, using the least square method based on orthogonal polynomials to perform fitting processing on the collected real-time electrical state quantity data, and to complete the missing electrical state quantity data; step 2, calculating the real-time load rate and voltage variation of the distribution transformer according to the completed data, and using the two as indicators for evaluating the operation state of the distribution transformer; step 3, establishing a deduction model corresponding to each indicator according to the calculation results of the real-time load rate and voltage deviation of the distribution transformer, and constructing a distribution transformer state evaluation model based on the above, obtaining the real-time state score of the distribution transformer, and realizing quantitative evaluation of the operation state of the distribution transformer; the technical problems such as the use of breakpoint data in the prior art to evaluate the real-time operation state of the distribution transformer, which affects the accuracy and reliability of the state evaluation due to incomplete data information, are solved.

Description

Technical Field

[0001] The invention belongs to transformer operation status monitoring technology, and in particular relates to a real-time operation status evaluation method for a distribution transformer. Background Art

[0002] There are many methods for evaluating the real-time operating status of distribution transformers. However, the existing technology for evaluating distribution transformers generally collects the transformer breakpoint status data information, performs comprehensive processing, and then comes up with the evaluation result; and the use of breakpoint data affects the accuracy and reliability of status evaluation because of incomplete data information. Summary of the invention

[0003] The technical problem to be solved by the present invention is: to provide a method for evaluating the real-time operating status of a distribution transformer, so as to solve the technical problems that the existing technology uses breakpoint data to evaluate the real-time operating status of a distribution transformer, which affects the accuracy and reliability of the status evaluation due to incomplete data information.

[0004] Technical solution of the present invention:

[0005] A method for evaluating the real-time operating status of a distribution transformer, comprising:

[0006] Step 1: Using the least square method based on orthogonal polynomials, the collected real-time electrical state quantity data is fitted and the missing electrical state quantity data is supplemented;

[0007] Step 2: Calculate the real-time load rate and voltage variation of the distribution transformer according to the completed data and use them as indicators for evaluating the operating status of the distribution transformer;

[0008] Step 3: According to the real-time load rate of the distribution transformer and the calculation results of the voltage deviation, a deduction model corresponding to each indicator is established, and based on this, a distribution transformer status assessment model is constructed to obtain the real-time status score of the distribution transformer and realize the quantitative assessment of the operating status of the distribution transformer.

[0009] Step 4: Based on the calculation result of step 3, a graph showing the relationship between the real-time status score of the distribution transformer and time is drawn, with time as the horizontal axis and the status score of the distribution transformer as the vertical axis.

[0010] The method of fitting the collected real-time electrical state quantity data by using the least square method based on orthogonal polynomials and completing the missing electrical state quantity data is as follows:

[0011] Assume (x i ,y i )(i=0,1,…,m) is a set of data of variables x,y, ω i >0 (i=0,1,…,m) is the weight coefficient of each point. The least square method can be expressed as In the function

[0012]

[0013] Minimize the sum of squares of the errors between all data and the fitted data, that is,

[0014]

[0015] Where: S is the n+1-dimensional subspace of the continuous function space C[a,b], is a set of basis functions of S;

[0016] The matrix form of the least squares fitting method equation system is as follows:

[0017]

[0018] Where:

[0019]

[0020]

[0021] By solving equation (3), we can find the function s * The coefficients of the basis functions of (x) are a0, a1, …, a n ;

[0022] Assume a set of points And the weight coefficient of each point If the polynomial family satisfy:

[0023]

[0024] Then it is called For a point set The belt weight Orthogonal families of polynomials;

[0025] Substituting equation (6) into equation (3), we can get the matrix expression of the least squares fitting equation system based on orthogonal polynomials, as shown below:

[0026]

[0027] Through the following recursive relationship:

[0028]

[0029]

[0030] Construct an orthogonal polynomial. After determining the fitting order, substitute the constructed orthogonal polynomial into equation (7) to obtain the coefficients of the corresponding basis function.

[0031] When there are missing items in the electrical state quantity data, the 6 adjacent data records of the data record are selected as the basic data set and the number of fitting times is set to 2; at the same time, time is used as the independent variable x, and the missing corresponding electrical state quantity is the dependent variable y; the set of independent variable x {x1, x2, x3, x4, x5, x6} and the set of dependent variable y {y1, y2, y3, y4, y5, y6} are respectively constructed, and the values ​​in the set x and the set y correspond one to one; after the set x and the set y are determined, they are substituted into the least squares equation group based on orthogonal polynomials mentioned above, the corresponding odd basis function coefficients are obtained, and the data time corresponding to the missing data is substituted to obtain the electrical state quantity.

[0032] The method for calculating the real-time load rate and voltage variation of the distribution transformer is:

[0033] The load factor calculation formula is as follows:

[0034]

[0035] Where: A, B, C represent the three-phase currents of distribution transformer A, B, C respectively, and capacity represents the operating capacity of distribution transformer;

[0036] Voltage deviation refers to the degree of deviation between the actual voltage and the nominal voltage. The calculation formula is as follows:

[0037]

[0038] Where: U T is the real-time voltage of the distribution transformer, U R is the system nominal phase voltage.

[0039] According to the calculation results of the real-time load rate and voltage deviation of the distribution transformer, the deduction model corresponding to each indicator is established, and the distribution transformer status evaluation model is constructed based on this, and the real-time status score of the distribution transformer is obtained. The method for realizing the quantitative evaluation of the operating status of the distribution transformer specifically includes:

[0040] Step 3.1. According to the distribution transformer operation specification guidelines, the real-time state quantity characterizing the transformer operation status is compared with the corresponding standard value; according to the distribution transformer real-time state quantity value, it is divided into four states: normal, caution, abnormal and emergency; the real-time state quantity value range and scoring interval corresponding to each indicator item are shown in Table 2 below:

[0041] Table 2 The value range and scoring interval of each indicator corresponding to the state quantity

[0042] Indicator name normal Notice abnormal urgent Load factor [0,0.8) [0.8,1) [1,2) [2,+∞) Voltage deviation [0,5％) [5％,7％) [7％,10％) [10％,+∞) Deduction interval [0,15) [15,25) [25,40) [40,100] Score range (85，100] (75，85] (60，75] [0,60]

[0043] Step 3.2: Establish the corresponding relationship between equipment evaluation indicators and health status, and use piecewise function to establish the equipment health status deduction model as shown below:

[0044]

[0045] In the formula, i = 1, 2, respectively represents the load rate and voltage deviation index; From Table 2, it can be seen that for the load rate index, S Li , S Hi , S Si 、a i 、b i 、c i The values ​​of are 15, 25, 40, 0.8, 1, and 2 respectively. The voltage deviation index is 15, 25, 40, 5%, 7%, and 10% respectively. cof is the continuous influence coefficient considering the influence of continuous overload on the health status of the distribution transformer. The values ​​are as follows:

[0046]

[0047] Where n is the number of consecutive overloads. When n is 0, the cof value is 1. For the voltage deviation indicator, the cof value is 1.

[0048] Step 3.3, calculate the comprehensive deduction value of the distribution transformer; after calculating the load rate of the distribution transformer and the voltage deviation deduction value, consider the real-time weight ratio of each indicator value and calculate the comprehensive deduction value of the distribution transformer as shown below:

[0049]

[0050] When the score calculated by the above formula is greater than 100, the comprehensive deduction value is taken as 100;

[0051] Step 3.4: After obtaining the comprehensive deduction values ​​of the real-time indicators of the distribution transformer, calculate the comprehensive score of the distribution transformer as follows:

[0052] S=100-S D

[0053] (15).

[0054] Beneficial effects of the present invention:

[0055] Based on the real-time electrical state data of distribution transformers, the present invention proposes a distribution transformer state assessment method that only needs to consider the load rate and voltage deviation. This method can effectively assess the transformer operating state every 15 minutes, while taking into account the impact of continuous overload on the operating state of the distribution transformer. It can provide effective support for early diagnosis and early warning of distribution transformers in a timely manner, and facilitate the formulation of differentiated operation and maintenance strategies for distribution transformers. It solves the technical problems of the existing technology of using breakpoint data to assess the real-time operating state of distribution transformers, which affects the accuracy and reliability of state assessment due to incomplete data information. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a schematic diagram of the real-time status score of distribution transformers. DETAILED DESCRIPTION

[0057] A method for evaluating the real-time operating status of a distribution transformer comprises the following steps:

[0058] Step 1: Use the least square method based on orthogonal polynomials to fit the real-time three-phase current and voltage data derived from the application metering automation system to fill in the missing data of the three-phase current and voltage caused by the acquisition device.

[0059] The monitoring device usually installed on the distribution transformer will collect the real-time electrical status data of the transformer, record it every 15 minutes, and store the collected data in the metering automation system. The specific data format is shown in Table 1.

[0060] Table 1 Real-time electrical status data of distribution transformer

[0061]

[0062]

[0063] However, sometimes due to device reasons, the electrical state quantity data may be missing. Simply treating its value as 0 or other values ​​will have a certain impact on the evaluation of the real-time operating status of the distribution transformer, and may easily lead to misjudgment of the results. Therefore, the present invention adopts the least squares method based on orthogonal polynomials to fit the real-time electrical state quantity data of the application metering automation system, and to fill in the missing electrical state quantity data to reduce the impact of missing data on the evaluation of the operating status. The basic principle of the least squares method based on orthogonal polynomials is as follows:

[0064] (1) Least squares fitting principle

[0065] Assume (x i ,y i )(i=0,1,…,m) is a set of data of variables x,y, ωi >0 (i=0,1,…,m) is the weight coefficient of each point. The least square method can be expressed as In the function

[0066]

[0067] Minimize the sum of squares of the errors between all data and the fitted data, that is,

[0068]

[0069] Where: S is the n+1-dimensional subspace of the continuous function space C[a,b], is a set of basis functions of S.

[0070] The matrix form of the least squares fitting method equation system is as follows:

[0071]

[0072] Where:

[0073]

[0074]

[0075] By solving equation (3), we can find the function s * The coefficients of the basis functions of (x) are a0, a1, …, a n .

[0076] (2) Least squares fitting principle based on orthogonal polynomials

[0077] In practical applications, when the least squares method is directly used for data fitting, the coefficient matrix of formula (3) is often ill-conditioned, which will make the calculation result unstable. However, using orthogonal polynomials as basis functions for data fitting can effectively avoid the instability of the calculation result caused by the ill-conditioned coefficient matrix. Therefore, the present invention adopts the least squares method based on orthogonal polynomials for data fitting, and its basic principle is as follows:

[0078] Assume a set of points And the weight coefficient of each point If the polynomial family satisfy:

[0079]

[0080] Then it is called For a point set The belt weight Orthogonal families of polynomials.

[0081] Substituting equation (6) into equation (3), we can get the matrix expression of the least squares fitting equation system based on orthogonal polynomials, as shown below:

[0082]

[0083] The present invention is through the following recursive relationship:

[0084]

[0085]

[0086] Construct an orthogonal polynomial. After determining the fitting order, substitute the constructed orthogonal polynomial into equation (7) to obtain the coefficients of the corresponding basis function.

[0087] When there are missing items in the electrical state quantity data, select the adjacent 6 data records of the data record as the basic data set (the first three and last three records of the data record, when the data record is less than three data records in front of the data record, fill in from the back. Similarly, when the data records in the back are insufficient, fill in from the front), and set the number of fitting times to 2. At the same time, take time as the independent variable x, and the missing corresponding electrical state quantity as the dependent variable y. For example, when the current of sequence number 5 in Table 1 is missing, select the 3 data time and current data before and after the record data to form the set of independent variable x {x1, x2, x3, x4, x5, x6} and the set of dependent variable y {y1, y2, y3, y4, y5, y6}, respectively, and the values ​​in set x and set y correspond one to one.

[0088] After the set x and the set y are determined, they are brought into the least squares equations based on orthogonal polynomials described above to obtain the corresponding odd basis function coefficients, and the data time corresponding to the missing data is brought in to obtain the electrical state quantity. It should be noted that since the data time of the application metering automation system is in the format of "year-month-day hour-minute", the data time needs to be converted during calculation, such as 2020-7-1 00:00 is 0, 2020-7-1 00:15 is 1, 2020-7-1 00:30 is 2, and so on.

[0089] Step 2: Utilize the real-time data of the application metering automation system processed in step 1 to calculate the real-time load rate and voltage variation of the distribution transformer and use them as indicators for evaluating the operating status of the distribution transformer.

[0090] Load factor is an important indicator to measure the state of distribution transformer, and its calculation formula is as follows:

[0091]

[0092] Where: A, B, C represent the three-phase currents of distribution transformer A, B, C respectively, and capacity represents the operating capacity of distribution transformer.

[0093] Voltage deviation refers to the degree of deviation between the actual voltage and the nominal voltage. It is the main parameter reflecting the voltage quality. Its calculation formula is as follows:

[0094]

[0095] Where: U T is the real-time voltage of the distribution transformer, U R is the system nominal phase voltage.

[0096] By substituting the three-phase currents A, B, and C and the operating capacity of the distribution transformer in each period in Table 1 into formula (10) and substituting the real-time voltage and the nominal voltage into formula (11), the distribution transformer load rate and voltage deviation at the corresponding moment can be obtained.

[0097] After the load factor and voltage deviation of the distribution transformer are calculated, the load condition of the distribution transformer is judged according to the obtained load factor value. As long as the load factor value is greater than or equal to 1, it means that the distribution transformer is overloaded at the corresponding moment. If the time difference between two overloads of the distribution transformer is 15 minutes, it is defined as continuous overload.

[0098] Step 3: According to the real-time load rate of the distribution transformer and the calculation results of the voltage deviation, a deduction model corresponding to each indicator is established, and based on this, a distribution transformer status assessment model is constructed to obtain the real-time status score of the distribution transformer and realize the quantitative assessment of the operating status of the distribution transformer.

[0099] First, according to the relevant operating specifications and guidelines for distribution transformers, the real-time state quantity that characterizes the operating status of the transformer is compared and analyzed with its corresponding standard value. According to the real-time state value of the distribution transformer, it is divided into four states: normal, attention, abnormal, and emergency. The real-time state value range and scoring interval corresponding to each indicator item are shown in the following table:

[0100] Table 2 The value range and scoring interval of each indicator corresponding to the state quantity

[0101] Indicator name normal Notice abnormal urgent Load factor [0,0.8) [0.8,1) [1,2) [2,+∞) Voltage deviation [0,5％) [5％,7％) [7％,10％) [10％,+∞) Deduction interval [0,15) [15,25) [25,40) [40,100] Score range (85，100] (75，85] (60，75] [0,60]

[0102] Secondly, in order to establish the corresponding relationship between equipment evaluation indicators and health status, a piecewise function is used to establish the equipment health status deduction model, as shown below:

[0103]

[0104] In the formula, i = 1, 2, respectively represents the load rate and voltage deviation index. From Table 2, we can see that for the load rate index, S Li , SHi , S Si 、a i 、b i 、c i The values ​​of are 15, 25, 40, 0.8, 1, and 2, respectively. The voltage deviation index is 15, 25, 40, 5%, 7%, and 10%, respectively. cof is the continuous influence coefficient that considers the impact of continuous overload on the health status of the distribution transformer. Its values ​​are as follows:

[0105]

[0106] Where n is the number of consecutive overloads. When n is 0, the cof value is 1. For the voltage deviation indicator, the cof value is 1.

[0107] Next, calculate the comprehensive deduction value of the distribution transformer. After calculating the load rate of the distribution transformer and the voltage deviation deduction value, consider the real-time weight ratio of each indicator value and calculate the comprehensive deduction value of the distribution transformer, as shown below:

[0108]

[0109] Note: When the score calculated by the above formula is greater than 100, the comprehensive deduction value is 100.

[0110] Finally, after obtaining the comprehensive deduction values ​​of each real-time indicator of the distribution transformer, the comprehensive score of the distribution transformer is calculated as follows:

[0111] S=100-S D (15)

[0113] Step 4: Based on the calculation result of step 3, a graph showing the relationship between the real-time status score of the distribution transformer and time is drawn, with time as the horizontal axis and the status score of the distribution transformer as the vertical axis.

[0114] After obtaining the real-time status score of the distribution transformer, the real-time operating status of the distribution transformer can be clearly reflected. At the same time, in order to more intuitively and clearly reflect the changes in the status of the distribution transformer over time and facilitate the formulation of the distribution transformer operation and maintenance strategy, a graph showing the relationship between the real-time status score of the distribution transformer and time is drawn with data time as the horizontal axis and the status score of the distribution transformer as the vertical axis. At the same time, the critical lines of normal and attention, attention and abnormal, abnormal and emergency states are drawn in the graph, so that the real-time operating status of the distribution transformer can be more intuitively displayed as shown in the attached figure. Figure 1 shown.

Claims

1. A method for evaluating the real-time operating status of a distribution transformer, comprising: Step 1: Using the least square method based on orthogonal polynomials, the collected real-time electrical state quantity data is fitted and the missing electrical state quantity data is supplemented; The method for completing the missing electrical state quantity data is: Assume (x i ,y i )(i=0,1,…,m) is a set of data of variables x,y, ω i >0 (i=0,1,…,m) is the weight coefficient of each point. The least square method can be expressed as In the function Minimize the sum of squares of the errors between all data and the fitted data, that is, Where: S is the n+1-dimensional subspace of the continuous function space C[a,b], is a set of basis functions of S; The matrix form of the least squares fitting method equation system is as follows: Where: By solving equation (3), we can find the function s * The coefficients of the basis functions of (x) are a0, a1, …, a n ; Assume a set of points And the weight coefficient of each point If the polynomial family satisfy: Then it is called For a point set The belt weight Orthogonal families of polynomials; Substituting equation (6) into equation (3), we can get the matrix expression of the least squares fitting equation system based on orthogonal polynomials, as shown below: Through the following recursive relationship: Construct an orthogonal polynomial. After determining the fitting order, substitute the constructed orthogonal polynomial into equation (7) to obtain the coefficients of the corresponding basis function. When there are missing items in the electrical state quantity data, select the 6 adjacent data records of the data record as the basic data set and set the number of fitting times to 2; at the same time, take time as the independent variable x, and the missing corresponding electrical state quantity as the dependent variable y; respectively construct the set of independent variable x {x1, x2, x3, x4, x5, x6} and the set of dependent variable y {y1, y2, y3, y4, y5, y6}, and the values ​​in the set x and the set y correspond one to one; after the set x and the set y are determined, bring them into the least squares equations based on orthogonal polynomials mentioned above, obtain the corresponding odd basis function coefficients, and bring the data time corresponding to the missing data into the electrical state quantity; Step 2: Calculate the real-time load rate and voltage deviation of the distribution transformer according to the completed data and use them as indicators for evaluating the operating status of the distribution transformer; Step 3: According to the real-time load rate of the distribution transformer and the calculation results of the voltage deviation, a deduction model corresponding to each indicator is established, and based on this, a distribution transformer status assessment model is constructed to obtain the real-time status score of the distribution transformer and realize the quantitative assessment of the operating status of the distribution transformer.

2. A method for evaluating the real-time operating status of a distribution transformer according to claim 1, characterized in that: Step 4: Based on the calculation result of step 3, a graph showing the relationship between the real-time status score of the distribution transformer and time is drawn, with time as the horizontal axis and the status score of the distribution transformer as the vertical axis.

3. A method for evaluating the real-time operating status of a distribution transformer according to claim 1, characterized in that: The method for calculating the real-time load rate and voltage deviation of the distribution transformer is: The load factor calculation formula is as follows: Where: A, B, C represent the three-phase currents of distribution transformer A, B, C respectively, and capacity represents the operating capacity of distribution transformer; Voltage deviation refers to the degree of deviation between the actual voltage and the nominal voltage. The calculation formula is as follows: Where: U T is the real-time voltage of the distribution transformer, U R is the system nominal phase voltage.

Images