Method, device, equipment and medium for evaluating the remaining useful life of enterprise transformers

Abstract

The present invention provides a method, device, equipment and medium for evaluating the remaining effective life of an enterprise transformer, which belongs to the technical field of electrical equipment status monitoring. The method includes the following steps: obtaining the historical load data of the enterprise transformer for one year of operation, and analyzing and processing the historical load data and dividing it into three categories; using a steady-state thermal model and an insulation life loss conversion method to convert the load data of the enterprise transformer into insulation life loss data; performing parameter fitting based on a gamma process on the insulation life loss data of the enterprise transformer to obtain the corresponding parameters of three linear gamma processes; merging the three linear gamma processes, and analyzing to obtain the distribution of the enterprise transformer life, thereby obtaining an estimated value of the remaining effective life of the enterprise transformer. The present invention provides a practical technical solution for evaluating the remaining effective life of a transformer that does not rely on statistical analysis, which is more in line with the actual working conditions of enterprise-level transformers, thereby improving the accuracy of the evaluation results.

Description

Technical Field

[0001] The present invention relates to a method, device, equipment and medium for evaluating the remaining effective life of an enterprise transformer, and specifically to a method, device, equipment and medium for evaluating the remaining effective life of an enterprise transformer based on cluster analysis and a gamma process, and belongs to the technical field of electrical equipment status monitoring. Background Art

[0002] As one of the important devices in the power system, the safe and stable operation of the transformer is of great significance to the reliability of the power system. In order to ensure the safe and stable operation of the transformer, it is necessary to carry out condition monitoring and health management. Among the many indicators for evaluating the health of the transformer, the remaining effective life has become a commonly used technical indicator for transformer health evaluation due to its intuitive and highly operable characteristics.

[0003] At present, the research on the remaining effective life of transformers is mainly focused on transformers in the transmission / distribution network. Since there are a large number of such transformers, the indicators related to health can be obtained by statistics. However, for transformers in enterprise-level power grids, the number is relatively small, and generally does not meet the requirements of statistical analysis, which means that the evaluation of the remaining effective life of transformers in enterprise-level power grids can only be based on the operating conditions of the transformers themselves. In theory, random process is a very good model for analyzing the degradation process of equipment. At present, the random process used to describe the degradation process of transformers is basically concentrated on the Wiener process model. For transformers in the transmission / distribution network, since there are many factors affecting their load, the normal distribution in the Wiener process can well describe the probability distribution of related random variables. However, for transformers in enterprise-level power grids, since the production of enterprises is often highly planned, the factors affecting their transformer load are not as many as those in the transmission / distribution network. In addition, transformer degradation is an irreversible process, that is, the degradation amount in any time period is a value greater than zero. Therefore, the Wiener process with positive or negative increments in any time period is not suitable for describing the degradation process of enterprise-level transformers.

[0004] In addition, due to the development level of technologies such as predictive maintenance, it has not yet reached the point of completely replacing preventive maintenance. Therefore, for enterprises, especially industrial enterprises in heavy asset industries such as petrochemicals and steel, the current equipment maintenance method mainly adopted is preventive maintenance. The implementation of preventive maintenance will have a great impact on the production plan of the enterprise, and then affect the load of the transformer. Due to the attenuation of the transformer, it is closely related to the load of the transformer. How to consider this obvious load change caused by maintenance plans when describing the degradation process of such enterprise-level transformers requires new analysis methods. Therefore, the present invention proposes a method for evaluating the remaining effective life of enterprise transformers. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a method, device, equipment and medium for evaluating the remaining effective life of enterprise transformers, which can solve the problem that statistical analysis cannot be applied to enterprise-level transformer remaining effective life evaluation due to small sample size.

[0006] The technical solution adopted by the present invention to solve the technical problem is:

[0007] In a first aspect, an embodiment of the present invention provides a method for evaluating the remaining effective life of an enterprise transformer, comprising the following steps:

[0008] Step S1, obtaining the historical load data of the enterprise transformer for one year, and analyzing and processing the historical load data based on K-means clustering to divide it into three categories: Case 1, Case 2 and Case 3, wherein Case 1 corresponds to the load of the enterprise transformer during normal operation; Case 2 corresponds to the load of the enterprise transformer during production equipment maintenance; Case 3 corresponds to the load of the enterprise transformer during backup enterprise transformer maintenance;

[0009] Step S2, according to the IEEE C57.91-2011 standard, using a steady-state thermal model and an insulation life loss conversion method to convert the load data of the enterprise transformer into insulation life loss data, wherein the enterprise transformer includes an enterprise transformer without a tap switch and an enterprise transformer with a tap switch;

[0010] Step S3, for the enterprise transformer insulation life loss data obtained in step S2, parameter fitting based on the gamma process is performed to obtain the corresponding three linear gamma processes X 1 ′(t),X 2 ′(t) and X 3 ′(t) parameters;

[0011] Step S4: The three linear gamma processes X corresponding to the insulation life loss of transformers of three types of enterprises, Case 1, Case 2 and Case 3, are transformed into 1 ′(t),X 2 ′(t) and X 3 ′(t) are combined and analyzed to obtain the distribution of the enterprise's transformer life, and then the estimated value of the remaining effective life of the enterprise's transformer is obtained.

[0012] As a possible implementation of this embodiment, step S1 includes the following steps:

[0013] Step S1.1, taking year as the data collection unit and hour as the data collection interval, sampling the voltage and current signals of the enterprise transformer with time stamp, and obtaining the load data of the enterprise transformer every hour in a year by calculation, and obtaining a total of 24×365=8760 load data points;

[0014] Step S1.2, performing K-means cluster analysis on the load data obtained in step S1.1, and dividing it into three categories: Case 1, Case 2 and Case 3;

[0015] The step S1.2 comprises the following steps:

[0016] Step S1.2.1, initialize three cluster centers V1, V2 and V3, where V1 corresponds to the center point of Case 1 load, and its value is the mean of one year's load data; V2 corresponds to the center point of Case 2 load, and its value is the mean of one year's load data multiplied by 0.7; V3 corresponds to the center point of Case 3 load, and its value is the mean of one year's load data multiplied by 1.3; Step S1.2.2, calculate the value of each load point Data i The distance D from the three cluster centers i,j :

[0017] D i,j =|Data i -V j | (1)

[0018] Where i is any integer in [1,8760], j is any integer in [1,3];

[0019] According to Data i The corresponding D i,1 , D i,2 and D i,3 The size of these three values ​​will be i Classified as D i,j The category corresponding to the smallest cluster center is the one that meets the conditions among Case1, Case2 and Case3;

[0020] Step S1.2.3, recalculate the cluster center points of the three categories Case1, Case2 and Case3, and take the mean of all data points contained in the category;

[0021] Step S1.2.4, repeat steps S1.2.2 and S1.2.3 until the change of cluster center points is less than 5%;

[0022] Step S1.2.5, checking the time scales of the data points contained in the two load classes Case 2 and Case 3 obtained by K-means clustering, and determining the month containing the most data points as the month when the enterprise performs equipment maintenance;

[0023] Step S1.2.6, divide the enterprise transformer load data of the equipment maintenance month determined in step S1.2.5 into 3 groups in chronological order, with 10 days as a unit, calculate the average value of each group, and classify the load data corresponding to the group with the highest average value into Case 3; the remaining load data of the month are classified into Case 2; the load data of other months are classified into Case 1.

[0024] As a possible implementation of this embodiment, in step S2, according to the IEEE C57.91-2011 standard, a steady-state thermal model and an insulation life loss conversion method are used to convert the load data of the enterprise transformer without a tap changer into insulation life loss data, and the specific process includes the following steps:

[0025] Step S2.1.1, calculate the temperature rise Δθ of the insulating oil top temperature relative to the ambient temperature TO :

[0026]

[0027] Among them, ΔΘ TO,R It represents the temperature rise of the top of the insulating oil relative to the ambient temperature when the temperature inside the enterprise transformer reaches stability under rated load; K represents the ratio of load to rated load; R represents the ratio between the loss under rated load and the no-load loss; n represents the temperature rise of the top of the insulating oil relative to the ambient temperature, affected by the load;

[0028] Step S2.1.2, calculate the temperature rise Δθ of the hottest point of the winding relative to the top temperature of the insulating oil H for:

[0029] ΔΘ H =ΔΘ H,R K 2m (3)

[0030] Among them, ΔΘ H,R It indicates the temperature rise of the hottest point temperature of the winding relative to the top temperature of the insulating oil under rated load; K indicates the ratio of load to rated load; m indicates the temperature rise of the hottest point temperature of the winding relative to the top temperature of the insulating oil, affected by the load;

[0031] Step S2.1.3, calculate the hottest point temperature Θ of the enterprise transformer winding H :

[0032] Θ H =Θ A +ΔΘ TO +ΔΘ H (4)

[0033] Among them, Θ A represents the ambient temperature (the daily average temperature announced by the local meteorological bureau), ΔΘ TO Indicates the temperature rise of the top temperature of the insulating oil relative to the ambient temperature, ΔΘ H It indicates the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil;

[0034] Step S2.1.4, based on the hottest point temperature θ of the winding H , through the insulation life loss conversion method provided by IEEE C57.91-2011 standard, that is, the insulation life loss T converted to the hottest point temperature of the winding at 110°C:

[0035]

[0036] As a possible implementation of this embodiment, in step S2, according to the IEEE C57.91-2011 standard, a steady-state thermal model and an insulation life loss conversion method are used to convert the load data of the enterprise transformer with a tap changer into insulation life loss data, and the specific process includes the following steps:

[0037] Step S2.2.1, calculate the temperature rise ΔΘ′ of the insulating oil top temperature relative to the ambient temperature when the enterprise transformer is in the corresponding tap changer position, running at full load, and the internal temperature reaches stability TO,R :

[0038]

[0039] Among them, ΔΘ TO,R P is the temperature rise of the top temperature of the insulating oil relative to the ambient temperature when the temperature inside the enterprise transformer reaches a stable state under the nominal rated load; T,R It represents the loss of the enterprise transformer under the nominal rated load; P T ' ,R It represents the loss of the enterprise transformer under the rated load corresponding to the corresponding tapping position; n represents the top temperature of the insulating oil, the temperature rise relative to the ambient temperature, affected by the load;

[0040] Step S2.2.2, calculate the temperature rise ΔΘ′ of the hottest point of the winding relative to the top oil temperature of the insulating oil when the enterprise transformer is in the corresponding tap changer position, running at full load and the internal temperature reaches a stable state. H,R :

[0041]

[0042] Among them, ΔΘ H,R It is the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil when the temperature inside the enterprise transformer reaches stability under the nominal rated load; I R It represents the rated current of the enterprise transformer under the nominal rated load; I R ′ represents the rated current of the enterprise transformer under the rated load corresponding to the corresponding tapping position; m represents the hottest point temperature of the winding, which is the temperature rise relative to the top temperature of the insulating oil, affected by the load;

[0043] Step S2.2.3, calculate the temperature rise ΔΘ′ of the insulating oil top temperature relative to the ambient temperature of the enterprise transformer at the corresponding tap changer position TO :

[0044]

[0045] Among them, K′ represents the ratio between the load and the nominal rated load; R′ represents the ratio between the loss of the enterprise transformer under the nominal rated load and the no-load loss under the corresponding tapping gear; n represents the temperature of the top of the insulating oil, the temperature rise relative to the ambient temperature, affected by the load;

[0046] Step S2.2.4, calculate the temperature rise ΔΘ′ of the hottest point temperature of the winding relative to the top oil temperature of the insulating oil at the corresponding tap changer position of the enterprise transformer H :

[0047] Δθ′ H =Δθ′ H,R K′ 2m (9)

[0048] Among them, K′ represents the ratio between the load and the nominal rated load, m represents the temperature of the hottest point of the winding, which is the temperature rise relative to the top temperature of the insulating oil, affected by the load;

[0049] Step S2.2.5, calculate the hottest point temperature Θ′ of the winding of the enterprise transformer at the corresponding tap changer position H :

[0050] Θ′ H =Θ A +ΔΘ′ TO +ΔΘ′ H (10)

[0051] Among them, Θ A Indicates the ambient temperature, ΔΘ′ TO Indicates the temperature rise of the top temperature of the insulating oil relative to the ambient temperature, ΔΘ′ H It indicates the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil;

[0052] Step S2.2.6, based on the hottest point temperature θ′ of the winding at the corresponding tap changer position of the enterprise transformer H , through the insulation life loss conversion method provided by IEEE, that is, the insulation life loss T' converted to the hottest point temperature of the winding at 110°C:

[0053]

[0054] As a possible implementation of this embodiment, in step S3, the transformer insulation life loss data of three types of enterprises, Case 1, Case 2 and Case 3, and the corresponding linear gamma processes are X 1 (t), X 2 (t) and X 3 (t), whose increments satisfy the corresponding gamma distribution, namely:

[0055] X 1 (τ 1 )-X 1 (t 1 )~Ga(c 1 (τ 1 -t 1 ),u 1 )(12)

[0056] X 2 (τ 2 )-X 2 (t 2 )~Ga(c 2 (τ 2 -t 2 ),u 2 )(13)

[0057] X 3 (τ 3 )-X 3 (t 3 )~Ga(c 3 (τ 3 -t 3 ),u 3 )(14)

[0058] Among them, τ k >t k ≥0, k=1,2,3, parameter c k and the inverse scale parameter u k are greater than 0, and

[0059]

[0060] Among them, I [0,∞) (x) is the indicator function; Γ(a) is the gamma function, expressed as

[0061] In units of years, the maximum likelihood estimation method based on the generalized gamma distribution is used to estimate the parameter c k and the inverse scale parameter u k Parameter fitting is performed to obtain the analytical expression of the parameters after unbiased correction.

[0062] As a possible implementation of this embodiment, step S3 includes the following steps:

[0063] Step S3.1, using the enterprise transformer life loss data of Case1, 1 and u 1 Make an estimate:

[0064]

[0065] Among them, n 1 is the number of enterprise transformer life loss data included in Case 1, For the first 1 Transformer life loss data for each enterprise;

[0066] Step S3.2, using the enterprise transformer life loss data of Case2, 2 and u 2 Make an estimate:

[0067]

[0068] Among them, n 2 is the number of enterprise transformer life loss data included in Case2, For the first 2 Transformer life loss data for each enterprise;

[0069] Step S3.3, using the enterprise transformer life loss data of Case3, 3 and u 3 Make an estimate:

[0070]

[0071] Among them, n 3 is the number of enterprise transformer life loss data included in Case 3, For the first 3 Transformer life loss data for each enterprise;

[0072] Step S3.4, for the linear gamma process X 1 (t), X 2 (t) and X 3The inverse scale parameters of (t) are averaged to obtain the corresponding X 1 (t), X 2 (t) and X 3 (t) Inverse scale parameter of the trilinear gamma process

[0073]

[0074] Step S3.5, gamma process fitting for the loss of insulation life of enterprise transformers over N years:

[0075] Fit the parameters of the gamma process corresponding to the three operating conditions of Case 1, Case 2 and Case 3 in each of the N years and the inverse scale parameter

[0076] Using the average method, calculate the gamma process parameter c′ corresponding to the three operating conditions of Case 1, Case 2 and Case 3 in N years p And the inverse scale parameter u′:

[0077]

[0078] Among them, p=1,2,3.

[0079] As a possible implementation of this embodiment, step S4 includes the following steps:

[0080] Step S4.1, merging three linear gamma processes X 1 ′(t),X 2 ′(t) and X 3 ′(t), and obtain the gamma process X that comprehensively describes the loss of insulation life of the enterprise transformer 1 ' ,2,3 (t);

[0081] Step S4.2, according to the linear gamma process X that comprehensively describes the loss of insulation life of the enterprise transformer 1 ' ,2,3 (t), calculate the distribution of the enterprise transformer life:

[0082]

[0083] Among them, L is the life of the enterprise transformer, the insulation life of the enterprise transformer is 180,000 hours, Γ(a,x) is the incomplete gamma function, and the expression is x≥0,a>0;

[0084] because The value of is usually large, and the distribution of formula (27) is approximated by the inverse Gaussian distribution:

[0085]

[0086] According to the inverse Gaussian distribution shown in formula (28), the estimated value of the enterprise transformer life L is obtained

[0087]

[0088] Step S4.3, estimate the remaining useful life RUL of the enterprise transformer:

[0089]

[0090] Among them, t operation It is the operating time of the enterprise transformer from the time it was put into service to the present moment.

[0091] As a possible implementation of this embodiment, step S4.1 includes the following steps:

[0092] Step S4.1.1, merge the gamma processes of Case 2 and Case 3 corresponding to the equipment maintenance month to obtain the gamma process X that comprehensively describes the loss of insulation life of the enterprise transformer in the equipment maintenance month 2 ' ,3 (t):

[0093]

[0094] in,

[0095]

[0096] From formula (25), we can see that

[0097] Step S4.1.2, merge the gamma process of Case 1 corresponding to the normal operation state of the enterprise transformer and the gamma process X that comprehensively describes the operation state of the enterprise transformer in Case 2 and Case 3 during the equipment maintenance month 2 ' ,3 (t), and obtain the gamma process X that comprehensively describes the annual loss of transformer insulation life in the enterprise 1 ' ,2,3 (t):

[0098] X 1 ' ,2,3 (12t operation2 +t a ′)-X 1 ' ,2,3 (t a ′)=X 1 ′(11t operation2 +ta ′)-X 1 ′(t a ′)+X 2 ' ,3 (t operation2 +t b ′)-X 2 ' ,3 (t b ′) (26) where X 1 ′(11t operation2 +t a ′)-X 1 ′(t a ′)~Ga(11c 1 't operation2 ,u′),

[0099] X 2 ' ,3 (t operation2 +t b ′)-X 2 ' ,3 (t b ′)~Ga(c 2 ' ,3 t operation2 ,u′).

[0100] From formula (26), we can see that

[0101] In a second aspect, an embodiment of the present invention provides a device for evaluating the remaining effective life of an enterprise transformer, comprising:

[0102] The data acquisition module is used to obtain the historical load data of the enterprise transformer during one year of operation, and analyze and process the historical load data based on K-means clustering and divide it into three categories: Case 1, Case 2 and Case 3. Case 1 corresponds to the load of the enterprise transformer during normal operation; Case 2 corresponds to the load of the enterprise transformer during production equipment maintenance; Case 3 corresponds to the load of the enterprise transformer during backup enterprise transformer maintenance;

[0103] A data conversion module, for converting load data of enterprise transformers into insulation life loss data by using a steady-state thermal model and an insulation life loss conversion method according to the IEEE C57.91-2011 standard, wherein the enterprise transformers include enterprise transformers without tap switches and enterprise transformers with tap switches;

[0104] The parameter fitting module is used to perform parameter fitting based on the gamma process for the enterprise transformer insulation life loss data, and obtain the corresponding three linear gamma processes X 1 ′(t),X 2 ′(t) and X3 ′(t) parameters;

[0105] The gamma process merging module is used to merge the three linear gamma processes X corresponding to the insulation life loss of transformers of three types of enterprises, Case 1, Case 2 and Case 3. 1 ′(t),X 2 ′(t) and X 3 ′(t) are combined and analyzed to obtain the distribution of the enterprise's transformer life, and then the estimated value of the remaining effective life of the enterprise's transformer is obtained.

[0106] In a third aspect, an embodiment of the present invention provides a computer device, comprising a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of any of the above-mentioned methods for assessing the remaining effective life of an enterprise transformer.

[0107] In a fourth aspect, an embodiment of the present invention provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for assessing the remaining useful life of any enterprise transformer as described above are executed.

[0108] The beneficial effects of the technical solution of the embodiment of the present invention are as follows:

[0109] The present invention first uses the K-means cluster analysis method to identify the month when the enterprise performs equipment maintenance, and then divides the collected enterprise transformer load data into three categories (i.e., normal working state, production equipment maintenance state, and enterprise transformer maintenance state), and then performs gamma process fitting on the insulation life loss of the enterprise transformer corresponding to the three types of load data, so that it is more in line with the actual operating conditions of the enterprise transformer; after the gamma process fitting is completed, based on the inherent characteristics of the enterprise load, the three linear gamma processes that respectively describe the insulation life loss of the enterprise transformer corresponding to the three types of load data are merged and processed to obtain a gamma process that comprehensively describes the insulation life loss of the enterprise transformer, so that the remaining effective life of the enterprise transformer is analyzed based on the gamma process under the consideration of the actual load characteristics of the enterprise, and then the remaining effective life of the transformer is estimated. The present invention can evaluate the remaining effective life of the enterprise transformer without relying on statistical analysis, which is more in line with the actual working conditions of the enterprise-level transformer, thereby improving the accuracy of the evaluation results.

[0110] The present invention combines cluster analysis of historical load data, application of the IEEE C57.91-2011 standard, and parameter fitting based on the gamma process to more accurately evaluate the insulation life of an enterprise's transformer, and can monitor and predict its insulation condition in real time, providing an effective means for enterprises to save energy and reduce operating costs.

[0111] The present invention not only solves the problem that statistical analysis cannot be applied to the remaining effective life assessment of enterprise-level transformers due to the small number of samples, but also provides a practical technical solution for the remaining effective life assessment of transformers that does not rely on statistical analysis, which is more in line with the actual working conditions of enterprise-level transformers, thereby improving the accuracy of the assessment results. BRIEF DESCRIPTION OF THE DRAWINGS

[0112] Figure 1 is a flow chart of a method for evaluating the remaining effective life of an enterprise transformer according to an exemplary embodiment;

[0113] Figure 2 The present invention is a schematic structural diagram of a device for evaluating the remaining effective life of an enterprise transformer according to an exemplary embodiment. DETAILED DESCRIPTION

[0114] In order to more clearly illustrate the technical features of the solution of the present invention, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings.

[0115] Transformer degradation is an irreversible process, that is, the degradation amount in any time period is a value greater than zero. The Wiener process with positive or negative increments relative to any time period is not suitable for describing the degradation process of enterprise-level transformers, while the gamma process with positive increments in any time period is more suitable. Therefore, the present invention proposes a method for evaluating the remaining effective life of enterprise transformers based on cluster analysis and gamma process. The enterprise transformer described in the present invention refers to the internal power grid transformer of industrial enterprises in heavy asset industries such as petrochemicals and steel.

[0116] like Figure 1 As shown, a method for evaluating the remaining effective life of an enterprise transformer provided by an embodiment of the present invention includes the following steps:

[0117] Step S1, obtaining the historical load data of the enterprise transformer for one year, and analyzing and processing the historical load data based on K-means clustering to divide it into three categories: Case 1, Case 2 and Case 3, wherein Case 1 corresponds to the load of the enterprise transformer during normal operation; Case 2 corresponds to the load of the enterprise transformer during production equipment maintenance; Case 3 corresponds to the load of the enterprise transformer during backup enterprise transformer maintenance;

[0118] Step S2, according to the IEEE C57.91-2011 standard, using a steady-state thermal model and an insulation life loss conversion method to convert the load data of the enterprise transformer into insulation life loss data, wherein the enterprise transformer includes an enterprise transformer without a tap switch and an enterprise transformer with a tap switch;

[0119] Step S3, for the enterprise transformer insulation life loss data obtained in step S2, parameter fitting based on the gamma process is performed to obtain the corresponding three linear gamma processes X 1 ′(t),X 2 ′(t) and X 3 ′(t) parameters;

[0120] Step S4: The three linear gamma processes X corresponding to the insulation life loss of transformers of three types of enterprises, Case 1, Case 2 and Case 3, are transformed into 1 ′(t),X 2 ′(t) and X 3 ′(t) are combined and analyzed to obtain the distribution of the enterprise's transformer life, and then the estimated value of the remaining effective life of the enterprise's transformer is obtained.

[0121] As a possible implementation of this embodiment, step S1 proposes a load clustering method based on K-means clustering analysis, which includes the following steps:

[0122] Step S1.1, taking year as the data collection unit and hour as the data collection interval, sampling the voltage and current signals of the enterprise transformer with time stamp, and obtaining the load data of the enterprise transformer every hour in a year by calculation, and obtaining a total of 24×365=8760 load data points;

[0123] Step S1.2, performing K-means cluster analysis on the load data obtained in step S1.1, and dividing it into three categories: Case 1, Case 2 and Case 3;

[0124] The step S1.2 comprises the following steps:

[0125] Step S1.2.1, initialize three cluster centers V1, V2 and V3, where V1 corresponds to the center point of Case 1 load, and its value is the mean of one year's load data; V2 corresponds to the center point of Case 2 load, and its value is the mean of one year's load data multiplied by 0.7; V3 corresponds to the center point of Case 3 load, and its value is the mean of one year's load data multiplied by 1.3; Step S1.2.2, calculate the value of each load point Data i The distance D from the three cluster centers i,j :

[0126] D i,j=|Data i -V j | (1)

[0127] Where i is any integer in [1,8760], j is any integer in [1,3];

[0128] According to Data i The corresponding D i,1 , D i,2 and D i,3 The size of these three values ​​will be i Classified as D i,j The category corresponding to the smallest cluster center is the one that meets the conditions among Case1, Case2 and Case3;

[0129] Step S1.2.3, recalculate the cluster center points of the three categories Case1, Case2 and Case3, and take the mean of all data points contained in the category;

[0130] Step S1.2.4, repeat steps S1.2.2 and S1.2.3 until the change of cluster center points is less than 5%;

[0131] Step S1.2.5, because the maintenance of production equipment and the maintenance of enterprise transformers are usually carried out together, check the time scales of the data points contained in the two load classes Case 2 and Case 3 obtained by K-means clustering, and determine the month containing the most data points as the month when the enterprise performs equipment maintenance;

[0132] Step S1.2.6, divide the enterprise transformer load data of the equipment maintenance month determined in step S1.2.5 into 3 groups in chronological order, with 10 days as a unit, calculate the average value of each group, and classify the load data corresponding to the group with the highest average value into Case 3; the remaining load data of the month are classified into Case 2; the load data of other months are classified into Case 1.

[0133] Step S2 proposes a steady-state thermal model and insulation life loss conversion method based on the IEEE C57.91-2011 standard, converting the three types of load data, Case 1, Case 2, and Case 3, into the corresponding insulation life loss when the enterprise transformer actually runs for 1 hour. Since the enterprise transformer has two situations, one with a tap changer and one without a tap changer, and the operating states of the enterprise transformer in these two situations are significantly different, it is necessary to perform different processing.

[0134] As a possible implementation of this embodiment, for an enterprise transformer without a tap switch, in step S2, according to the IEEE C57.91-2011 standard, a steady-state thermal model and an insulation life loss conversion method are used to convert the load data of the enterprise transformer without a tap switch into insulation life loss data, and the specific process includes the following steps:

[0135] Step S2.1.1, calculate the temperature rise Δθ of the insulating oil top temperature relative to the ambient temperature TO :

[0136]

[0137] Among them, ΔΘ TO,R It represents the temperature rise of the top of the insulating oil relative to the ambient temperature when the temperature inside the enterprise transformer reaches stability under rated load; K represents the ratio of load to rated load; R represents the ratio between the loss under rated load and the no-load loss; n represents the temperature rise of the top of the insulating oil relative to the ambient temperature, affected by the load;

[0138] Step S2.1.2, calculate the temperature rise Δθ of the hottest point of the winding relative to the top temperature of the insulating oil H for:

[0139] ΔΘ H =ΔΘ H,R K 2m (3)

[0140] Among them, ΔΘ H,R It indicates the temperature rise of the hottest point temperature of the winding relative to the top temperature of the insulating oil under rated load; K indicates the ratio of load to rated load; m indicates the temperature rise of the hottest point temperature of the winding relative to the top temperature of the insulating oil, affected by the load;

[0141] Step S2.1.3, calculate the hottest point temperature Θ of the enterprise transformer winding H :

[0142] Θ H =Θ A +ΔΘ TO +ΔΘ H (4)

[0143] Among them, Θ A represents the ambient temperature (the daily average temperature announced by the local meteorological bureau), ΔΘ TO Indicates the temperature rise of the top temperature of the insulating oil relative to the ambient temperature, ΔΘ H It indicates the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil;

[0144] Step S2.1.4, based on the hottest point temperature θ of the windingH , through the insulation life loss conversion method provided by IEEE, that is, the insulation life loss T converted to the hottest point temperature of the winding at 110°C:

[0145]

[0146] As a possible implementation of this embodiment, for an enterprise transformer with a tap switch, in step S2, according to the IEEE C57.91-2011 standard, a specific process of converting the load data of the enterprise transformer with a tap switch into insulation life loss data by using a steady-state thermal model and an insulation life loss conversion method includes the following steps:

[0147] Step S2.2.1, determine the tap position of the enterprise transformer tap changer by voltage, and calculate the temperature rise ΔΘ′ of the top temperature of the insulating oil relative to the ambient temperature when the enterprise transformer is running at full load at the corresponding tap changer position and the internal temperature reaches a stable state. TO,R :

[0148]

[0149] Among them, ΔΘ TO,R P is the temperature rise of the top temperature of the insulating oil relative to the ambient temperature when the temperature inside the enterprise transformer reaches a stable state under the nominal rated load; T,R It represents the loss of the enterprise transformer under the nominal rated load; P T ' ,R It represents the loss of the enterprise transformer under the rated load corresponding to the corresponding tapping position; n represents the top temperature of the insulating oil, the temperature rise relative to the ambient temperature, affected by the load;

[0150] Step S2.2.2, calculate the temperature rise ΔΘ′ of the hottest point of the winding relative to the top oil temperature of the insulating oil when the enterprise transformer is in the corresponding tap changer position, running at full load and the internal temperature reaches a stable state. H,R :

[0151]

[0152] Among them, ΔΘ H,R It is the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil when the temperature inside the enterprise transformer reaches stability under the nominal rated load; I R It represents the rated current of the enterprise transformer under the nominal rated load; I R ′ represents the rated current of the enterprise transformer under the rated load corresponding to the corresponding tapping position; m represents the hottest point temperature of the winding, which is the temperature rise relative to the top temperature of the insulating oil, affected by the load;

[0153] Step S2.2.3, calculate the temperature rise ΔΘ′ of the insulating oil top temperature relative to the ambient temperature of the enterprise transformer at the corresponding tap changer position TO :

[0154]

[0155] Among them, K′ represents the ratio between the load and the nominal rated load; R′ represents the ratio between the loss of the enterprise transformer under the nominal rated load and the no-load loss under the corresponding tapping gear; n represents the temperature of the top of the insulating oil, the temperature rise relative to the ambient temperature, affected by the load;

[0156] Step S2.2.4, calculate the temperature rise ΔΘ′ of the hottest point temperature of the winding relative to the top oil temperature of the insulating oil at the corresponding tap changer position of the enterprise transformer H :

[0157] Δθ′ H =Δθ′ H,R K′ 2m , (9)

[0158] Among them, K′ represents the ratio between the load and the nominal rated load, m represents the temperature of the hottest point of the winding, which is the temperature rise relative to the top temperature of the insulating oil, affected by the load;

[0159] Step S2.2.5, calculate the hottest point temperature Θ′ of the winding of the enterprise transformer at the corresponding tap changer position H :

[0160] Θ′ H =Θ A +ΔΘ′ TO +ΔΘ′ H (10)

[0161] Among them, Θ A Indicates the ambient temperature, ΔΘ′ TO Indicates the temperature rise of the top temperature of the insulating oil relative to the ambient temperature, ΔΘ′ H It indicates the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil;

[0162] Step S2.2.6, based on the hottest point temperature θ′ of the winding at the corresponding tap changer position of the enterprise transformer H , through the insulation life loss conversion method provided by IEEE, that is, the insulation life loss T' converted to the hottest point temperature of the winding at 110°C:

[0163]

[0164] The values ​​of m and n in equations (2), (3), (6), (7), (8) and (9) are related to the cooling method of the enterprise transformer, as shown in Table 1.

[0165] Table 1 Values ​​of transformer thermal model index m and n

[0166]

[0167] As a possible implementation of this embodiment, in step S3, the transformer insulation life loss data of three types of enterprises, Case 1, Case 2 and Case 3, and the corresponding linear gamma processes are X 1 (t), X 2 (t) and X 3 (t), whose increments satisfy the corresponding gamma distribution, namely:

[0168] X 1 (τ 1 )-X 1 (t 1 )~Ga(c 1 (τ 1 -t 1 ),u 1 ) (12)

[0169] X 2 (τ 2 )-X 2 (t 2 )~Ga(c 2 (τ 2 -t 2 ),u 2 ) (13)

[0170] X 3 (τ 3 )-X 3 (t 3 )~Ga(c 3 (τ 3 -t 3 ),u 3 ) (14)

[0171] Among them, τ k >t k ≥0, k=1,2,3, parameter c k and the inverse scale parameter u k are greater than 0, and

[0172]

[0173] Among them, I [0,∞) (x) is the indicator function; Γ(a) is the gamma function, expressed as Since the data collection time interval is 1 hour, we have τ k -t k =1;

[0174] In units of years, the maximum likelihood estimation method based on the generalized gamma distribution is used to estimate the parameter c k and the inverse scale parameter u k Parameter fitting is performed to obtain the analytical expression of the parameters after unbiased correction.

[0175] As a possible implementation of this embodiment, step S3 includes the following steps:

[0176] Step S3.1, using the enterprise transformer life loss data of Case1, 1 and u 1 Make an estimate:

[0177]

[0178]

[0179] Among them, n 1 is the number of enterprise transformer life loss data included in Case 1, For the first 1 Transformer life loss data for each enterprise;

[0180] Step S3.2, using the enterprise transformer life loss data of Case2, 2 and u 2 Make an estimate:

[0181]

[0182] Among them, n 2 is the number of enterprise transformer life loss data included in Case2, For the first 2 Transformer life loss data for each enterprise;

[0183] Step S3.3, using the enterprise transformer life loss data of Case3, 3 and u 3 Make an estimate:

[0184]

[0185] Among them, n 3 is the number of enterprise transformer life loss data included in Case 3, For the first 3 Transformer life loss data for each enterprise;

[0186] Step S3.4, for the linear gamma process X 1 (t), X 2 (t) and X 3 The inverse scale parameters of (t) are averaged to obtain the corresponding X 1 (t), X 2 (t) and X 3 (t) Inverse scale parameter of the trilinear gamma process

[0187]

[0188] Step S3.5, gamma process fitting for the loss of insulation life of enterprise transformers over N years:

[0189] Fit the parameters of the gamma process corresponding to the three operating conditions of Case 1, Case 2 and Case 3 in each of the N years and the inverse scale parameter

[0190] Using the average method, calculate the gamma process parameter c′ corresponding to the three operating conditions of Case 1, Case 2 and Case 3 in N years p And the inverse scale parameter u′:

[0191]

[0192] Among them, p=1,2,3.

[0193] In years, for X 1 (t), X 2 (t) and X 3 (t) The parameter fitting method for these three linear gamma processes is the maximum likelihood estimation method based on the generalized gamma distribution.

[0194] Because u s (s=1,2,3) is the inverse scale parameter of the gamma distribution, which mainly represents the dispersion of the gamma distribution. When the enterprise performs equipment maintenance, the working status of various equipment does not change significantly, but the number of equipment in working state (i.e., the corresponding total load) changes. Therefore, it can be considered that the dispersion of the gamma distribution represented by formula (15) will not change significantly due to equipment maintenance, that is, it can be considered that X 1 (t), X 2 (t) and X 3 (t) The inverse scale parameters of these three linear gamma processes are equal:

[0195]

[0196] in, Based on the insulation life loss data, the maximum likelihood estimation is used to obtain X s The estimated value of the inverse scale parameter of (t)(s=1,2,3). The processing of formula (22) also simplifies the subsequent analysis of the enterprise transformer life distribution based on the gamma process.

[0197] As a possible implementation of this embodiment, step S4 includes the following steps:

[0198] Step S4.1, merging three linear gamma processes X 1 ′(t),X 2 ′(t) and X 3 ′(t), and obtain the gamma process X that comprehensively describes the loss of insulation life of the enterprise transformer 1 ' ,2,3 (t);

[0199] Step S4.2, according to the linear gamma process X that comprehensively describes the loss of insulation life of the enterprise transformer 1 ' ,2,3 (t), calculate the distribution of the enterprise transformer life:

[0200]

[0201] Among them, L is the life of the enterprise transformer, the insulation life of the enterprise transformer is 180,000 hours, Γ(a,x) is the incomplete gamma function, and the expression is

[0202] because The value of is usually large, and the distribution of formula (27) is approximated by the inverse Gaussian distribution:

[0203]

[0204] According to the inverse Gaussian distribution shown in formula (28), the estimated value of the enterprise transformer life L is obtained

[0205]

[0206] Step S4.3, estimate the remaining useful life RUL of the enterprise transformer:

[0207]

[0208] Among them, t operation It is the operating time of the enterprise transformer from the time it was put into service to the current moment, in hours.

[0209] The IEEE C57.91-2011 standard stipulates that the insulation life of enterprise transformers is: the hottest point temperature of the winding is kept constant at 110°C, and the insulation life of enterprise transformers is 180,000 hours. Therefore, the distribution function of the enterprise transformer life L is shown in formula (27).

[0210] As a possible implementation of this embodiment, step S4.1 includes the following steps:

[0211] Step S4.1.1, merge the gamma processes of Case 2 and Case 3 corresponding to the equipment maintenance month to obtain the gamma process X′ that comprehensively describes the insulation life loss of the enterprise transformer in the equipment maintenance month 2,3 (t):

[0212]

[0213] in,

[0214]

[0215] From formula (25), we can see that

[0216] Step S4.1.2, merge the gamma process of Case 1 corresponding to the normal operation state of the enterprise transformer and the gamma process X that comprehensively describes the operation state of the enterprise transformer in Case 2 and Case 3 during the equipment maintenance month 2 ' ,3 (t), and obtain the gamma process X that comprehensively describes the annual loss of transformer insulation life in the enterprise 1 ' ,2,3 (t):

[0217] X 1 ' ,2,3 (12t operation2 +t a ′)-X 1 ' ,2,3 (t a ′)=X 1 ′(11t operation2 +t a ′)-X 1 ′(t a ′)+X 2 ' ,3 (t operation2 +t b ′)-X 2 ' ,3 (t b ′) (26) where X 1 ′(11t operation2 +t a ′)-X1 ′(t a ′)~Ga(11c 1 't operation2 ,u′),

[0218] X 2 ' ,3 (t operation2 +t b ′)-X 2 ' ,3 (t b ′)~Ga(c 2 ' ,3 t operation2 ,u′).

[0219] From formula (26), we can see that

[0220] like Figure 2 As shown, an embodiment of the present invention provides a device for evaluating the remaining effective life of an enterprise transformer, comprising:

[0221] The data acquisition module is used to obtain the historical load data of the enterprise transformer during one year of operation, and analyze and process the historical load data based on K-means clustering and divide it into three categories: Case 1, Case 2 and Case 3. Case 1 corresponds to the load of the enterprise transformer during normal operation; Case 2 corresponds to the load of the enterprise transformer during production equipment maintenance; Case 3 corresponds to the load of the enterprise transformer during backup enterprise transformer maintenance;

[0222] A data conversion module, for converting load data of enterprise transformers into insulation life loss data by using a steady-state thermal model and an insulation life loss conversion method according to the IEEE C57.91-2011 standard, wherein the enterprise transformers include enterprise transformers without tap switches and enterprise transformers with tap switches;

[0223] The parameter fitting module is used to perform parameter fitting based on the gamma process for the enterprise transformer insulation life loss data, and obtain the corresponding three linear gamma processes X 1 ′(t),X 2 ′(t) and X 3 ′(t) parameters;

[0224] The gamma process merging module is used to merge the three linear gamma processes X corresponding to the insulation life loss of transformers of three types of enterprises, Case 1, Case 2 and Case 3. 1 ′(t),X 2 ′(t) and X 3 ′(t) are combined and analyzed to obtain the distribution of the enterprise's transformer life, and then the estimated value of the remaining effective life of the enterprise's transformer is obtained.

[0225] A computer device provided in an embodiment of the present invention includes a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the method for assessing the remaining effective life of any enterprise transformer as described above.

[0226] Specifically, the above-mentioned memory and processor are general-purpose memories and processors, and when the processor runs the computer program stored in the memory, the above-mentioned method for assessing the remaining effective life of an enterprise transformer can be executed.

[0227] Corresponding to the method for starting the above-mentioned application, an embodiment of the present invention further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for evaluating the remaining effective life of any enterprise transformer are executed.

[0228] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for assessing the remaining effective life of an enterprise transformer, characterized in that: The steps include: Step S1, obtaining the historical load data of the enterprise transformer for one year, and analyzing and processing the historical load data based on K-means clustering, and dividing it into three categories: Case 1, Case 2 and Case 3, wherein Case 1 corresponds to the load of the enterprise transformer during normal operation; Case 2 corresponds to the load of the enterprise transformer during production equipment maintenance; Case 3 corresponds to the load of the enterprise transformer during enterprise backup transformer maintenance; Step S2, according to the IEEE C57.91-2011 standard, using a steady-state thermal model and an insulation life loss conversion method to convert the load data of the enterprise transformer into insulation life loss data, wherein the enterprise transformer includes an enterprise transformer without a tap switch and an enterprise transformer with a tap switch; Step S3, performing parameter fitting based on the gamma process on the enterprise transformer insulation life loss data obtained in step S2 to obtain the parameters of the corresponding three linear gamma processes X1′(t), X2′(t) and X3′(t); In step S4, the three linear gamma processes X1′(t), X2′(t) and X3′(t) corresponding to the insulation life losses of the three types of enterprise transformers, Case 1, Case 2 and Case 3, are merged, and the distribution of the enterprise transformer life is obtained by analysis, thereby obtaining an estimated value of the remaining effective life of the enterprise transformer.

2. The method for assessing the remaining effective life of an enterprise transformer according to claim 1, characterized in that: The step S1 comprises the following steps: Step S1.1, taking year as the data collection unit and hour as the data collection interval, sampling the voltage and current signals of the enterprise transformer with time stamp, and obtaining the load data of the enterprise transformer every hour in a year by calculation, and obtaining a total of 24×365=8760 load data points; Step S1.2, performing K-means cluster analysis on the load data obtained in step S1.1, and dividing it into three categories: Case 1, Case 2 and Case 3; The step S1.2 comprises the following steps: Step S1.2.1, initialize three cluster center points V1, V2 and V3, where V1 corresponds to the center point of Case 1 load, and its value is the mean of one year's load data; V2 corresponds to the center point of Case 2 load, and its value is the mean of one year's load data multiplied by 0.7; V3 corresponds to the center point of Case 3 load, and its value is the mean of one year's load data multiplied by 1.3; Step S1.2.2, calculate each load point Data i The distance D from the three cluster centers i,j : D i,j =|Data i -V j | (1) Where i is any integer in [1,8760], j is any integer in [1,3]; According to Data i The corresponding D i,1 , D i,2 and D i,3 The size of these three values ​​will be i Classified as D i,j The category corresponding to the smallest cluster center is the one that meets the conditions among Case1, Case2 and Case3; Step S1.2.3, recalculate the cluster center points of the three categories Case1, Case2 and Case3, and take the mean of all data points contained in the category; Step S1.2.4, repeat steps S1.2.2 and S1.2.3 until the change of cluster center points is less than 5%; Step S1.2.5, checking the time scales of the data points contained in the two load classes Case 2 and Case 3 obtained by K-means clustering, and determining the month containing the most data points as the month when the enterprise performs equipment maintenance; Step S1.2.6, divide the enterprise transformer load data of the equipment maintenance month determined in step S1.2.5 into 3 groups in chronological order, with 10 days as a unit, calculate the average value of each group, and classify the load data corresponding to the group with the highest average value into Case 3; the remaining load data of the month are classified into Case 2; the load data of other months are classified into Case 1.

3. The method for assessing the remaining effective life of an enterprise transformer according to claim 1, characterized in that: In step S2, according to the IEEE C57.91-2011 standard, the specific process of converting the load data of the enterprise transformer without a tap changer into the insulation life loss data by using the steady-state thermal model and the insulation life loss conversion method includes the following steps: Step S2.1.1, calculate the temperature rise Δθ of the insulating oil top temperature relative to the ambient temperature TO : Among them, ΔΘ TO,R It represents the temperature rise of the top of the insulating oil relative to the ambient temperature when the temperature inside the enterprise transformer reaches stability under rated load; K represents the ratio of load to rated load; R represents the ratio between the loss under rated load and the no-load loss; n represents the temperature rise of the top of the insulating oil relative to the ambient temperature, affected by the load; Step S2.1.2, calculate the temperature rise Δθ of the hottest point of the winding relative to the top temperature of the insulating oil H for: DTH H =DTH H,R K 2m (3) Among them, ΔΘ H,R It indicates the temperature rise of the hottest point temperature of the winding relative to the top temperature of the insulating oil under rated load; K indicates the ratio of load to rated load; m indicates the temperature rise of the hottest point temperature of the winding relative to the top temperature of the insulating oil, affected by the load; Step S2.1.3, calculate the hottest point temperature Θ of the enterprise transformer winding H : I H =Θ A +DTH TO +DTH H (4) Among them, Θ A Indicates the ambient temperature, ΔΘ TO Indicates the temperature rise of the top temperature of the insulating oil relative to the ambient temperature, ΔΘ H It indicates the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil; Step S2.1.4, based on the hottest point temperature θ of the winding H , through the insulation life loss conversion method provided by IEEE, that is, the insulation life loss T converted to the hottest point temperature of the winding at 110°C:

4. The method for assessing the remaining effective life of an enterprise transformer according to claim 1, characterized in that: In step S2, according to the IEEE C57.91-2011 standard, the specific process of converting the load data of the enterprise transformer with a tap changer into the insulation life loss data by using the steady-state thermal model and the insulation life loss conversion method includes the following steps: Step S2.2.1, calculate the temperature rise ΔΘ′ of the insulating oil top temperature relative to the ambient temperature when the enterprise transformer is in the corresponding tap changer position, running at full load, and the internal temperature reaches stability TO,R : Among them, ΔΘ TO,R P is the temperature rise of the top temperature of the insulating oil relative to the ambient temperature when the temperature inside the enterprise transformer reaches a stable state under the nominal rated load; T,R It represents the loss of the enterprise transformer under the nominal rated load; P T ' ,R It represents the loss of the enterprise transformer under the rated load corresponding to the corresponding tapping position; n represents the top temperature of the insulating oil, the temperature rise relative to the ambient temperature, affected by the load; Step S2.2.2, calculate the temperature rise ΔΘ′ of the hottest point of the winding relative to the top oil temperature of the insulating oil when the enterprise transformer is in the corresponding tap changer position, running at full load and the internal temperature reaches a stable state. H,R : Among them, ΔΘ H,R It is the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil when the temperature inside the enterprise transformer reaches stability under the nominal rated load; I R It represents the rated current of the enterprise transformer under the nominal rated load; I R ′ represents the rated current of the enterprise transformer under the rated load corresponding to the corresponding tapping position; m represents the hottest point temperature of the winding, which is the temperature rise relative to the top temperature of the insulating oil, affected by the load; Step S2.2.3, calculate the temperature rise ΔΘ′ of the insulating oil top temperature relative to the ambient temperature of the enterprise transformer at the corresponding tap changer position TO : Among them, K′ represents the ratio between the load and the nominal rated load; R′ represents the ratio between the loss of the enterprise transformer under the nominal rated load and the no-load loss under the corresponding tapping gear; n represents the temperature of the top of the insulating oil, the temperature rise relative to the ambient temperature, affected by the load; Step S2.2.4, calculate the temperature rise ΔΘ′ of the hottest point temperature of the winding relative to the top oil temperature of the insulating oil at the corresponding tap changer position of the enterprise transformer H : ΔΘ′ H =ΔΘ′ H,R K′ 2m , (9) Among them, K′ represents the ratio between the load and the nominal rated load, m represents the temperature of the hottest point of the winding, which is the temperature rise relative to the top temperature of the insulating oil, affected by the load; Step S2.2.5, calculate the hottest point temperature Θ′ of the winding of the enterprise transformer at the corresponding tap changer position H : I' H =Θ A +ΔΘ′ TO +ΔΘ′ H (10) Among them, Θ A Indicates the ambient temperature, ΔΘ′ TO Indicates the temperature rise of the top temperature of the insulating oil relative to the ambient temperature, ΔΘ′ H It indicates the temperature rise of the hottest point of the winding relative to the top temperature of the insulating oil; Step S2.2.6, based on the hottest point temperature θ′ of the winding at the corresponding tap changer position of the enterprise transformer H , through the insulation life loss conversion method provided by IEEE, that is, the insulation life loss T' converted to the hottest point temperature of the winding at 110°C:

5. The method for assessing the remaining effective life of an enterprise transformer according to claim 1, characterized in that: In step S3, the transformer insulation life loss data of three types of enterprises, Case 1, Case 2 and Case 3, correspond to the linear gamma processes X1(t), X2(t) and X3(t), respectively, and their increments satisfy the corresponding gamma distribution, namely: X1(τ1)-X1(t1)~Ga(c1(τ1-t1),u1) (12) X2(τ2)-X2(t2)~Ga(c2(τ2-t2),u2) (13) X3(τ3)-X3(t3)~Ga(c3(τ3-t3),u3) (14) Among them, τ k >t k ≥0,k=1,2,3,τ k and t k Represents any two moments in the gamma process, parameter c k and the inverse scale parameter u k are greater than 0, and Among them, I [0,∞) (x) is the indicator function; Γ(a) is the gamma function, expressed as In units of years, the maximum likelihood estimation method based on the generalized gamma distribution is used to estimate the parameter c k and the inverse scale parameter u k Parameter fitting is performed to obtain the analytical expression of the parameters after unbiased correction.

6. The method for assessing the remaining effective life of an enterprise transformer according to claim 5, characterized in that: The step S3 comprises the following steps: Step S3.1, using the enterprise transformer life loss data of Case1, estimate c1 and u1: Among them, n1 is the number of enterprise transformer life loss data included in Case1, The life loss data of the transformer of the first enterprise in Case 1; Step S3.2, using the enterprise transformer life loss data of Case2, estimate c2 and u2: Among them, n2 is the number of enterprise transformer life loss data included in Case2, The life loss data of the transformer of the 12th enterprise in Case 2; Step S3.3, using the enterprise transformer life loss data of Case3, estimate c3 and u3: Among them, n3 is the number of enterprise transformer life loss data included in Case3, The life loss data of the transformer of the 13th enterprise in Case 3; Step S3.4, average the inverse scale parameters of the linear gamma processes X1(t), X2(t) and X3(t) to obtain the inverse scale parameters of the three linear gamma processes corresponding to X1(t), X2(t) and X3(t) Step S3.5, gamma process fitting for the loss of insulation life of enterprise transformers over N years: Fit the parameters of the gamma process corresponding to the three operating conditions of Case 1, Case 2 and Case 3 in each of the N years and the inverse scale parameter Using the average method, calculate the gamma process parameter c′ corresponding to the three operating conditions of Case 1, Case 2 and Case 3 in N years p And the inverse scale parameter u′: Among them, p=1,2,3.

7. The method for assessing the remaining effective life of an enterprise transformer according to any one of claims 1 to 6, characterized in that: The step S4 comprises the following steps: Step S4.1, merge the three linear gamma processes X1′(t), X2′(t) and X3′(t) to obtain the gamma process X1′ that comprehensively describes the insulation life loss of the enterprise transformer ,2,3 (t); Step S4.2, according to the linear gamma process X1′ that comprehensively describes the loss of insulation life of the enterprise transformer ,2,3 (t), calculate the distribution of the enterprise transformer life: Among them, L is the life of the enterprise transformer, the insulation life of the enterprise transformer is 180,000 hours, Γ(a,x) is the incomplete gamma function, and the expression is x≥0,a>0; because The value of is usually large, and the distribution of formula (27) is approximated by the inverse Gaussian distribution: According to the inverse Gaussian distribution shown in formula (28), the estimated value of the enterprise transformer life L is obtained Step S4.3, estimate the remaining useful life RUL of the enterprise transformer: Among them, t operation The operating time of the enterprise's transformer from the time it was put into service to the current moment.

8. A device for evaluating the remaining effective life of an enterprise transformer, characterized in that: include: The data acquisition module is used to obtain the historical load data of the enterprise transformer during one year of operation, and analyze and process the historical load data based on K-means clustering and divide it into three categories: Case 1, Case 2 and Case 3. Case 1 corresponds to the load of the enterprise transformer during normal operation; Case 2 corresponds to the load of the enterprise transformer during production equipment maintenance; Case 3 corresponds to the load of the enterprise transformer during backup enterprise transformer maintenance; A data conversion module, for converting load data of enterprise transformers into insulation life loss data by using a steady-state thermal model and an insulation life loss conversion method according to the IEEE C57.91-2011 standard, wherein the enterprise transformers include enterprise transformers without tap switches and enterprise transformers with tap switches; The parameter fitting module is used to perform parameter fitting based on the gamma process for the insulation life loss data of the enterprise transformer, and obtain the parameters of the corresponding three linear gamma processes X1′(t), X2′(t) and X3′(t); The gamma process merging module is used to merge the three linear gamma processes X1′(t), X2′(t) and X3′(t) corresponding to the insulation life loss of the three types of enterprise transformers, Case 1, Case 2 and Case 3, and analyze the distribution of the enterprise transformer life, and then obtain the estimated value of the remaining effective life of the enterprise transformer.

9. A computer device, characterized in that: The computer device comprises a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the method for assessing the remaining effective life of an enterprise transformer as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, executes the steps of the method for assessing the remaining useful life of an enterprise transformer as described in any one of claims 1 to 7.

Images