A method and system for automatic intelligent testing of low voltage transformers
Through automated intelligent testing methods, the performance differences of transformer are analyzed using energy loss rate and temperature data, and the volatility of sliding windows is analyzed, which solves the problem of insufficient accuracy of the existing test methods and realizes the efficiency and timeliness of transformer performance detection.
Patent Information
- Application Number
- CN202410987205.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-07-23
AI Technical Summary
The existing transformer testing methods are affected by temperature and output power changes, resulting in poor accuracy of performance and quality testing and difficulty in detecting faults in a timely manner.
Using the intelligent automation test method of low-voltage transformer, the voltage, current and temperature data are collected by laying new coils, the energy loss rate is calculated, the performance difference sequence is obtained, the volatility is analyzed using sliding windows, and the reference fragment is selected to determine the possibility of performance problems. .
It improves the accuracy and timeliness of transformer performance detection, realizes automatic intelligent testing of transformers, promptly detects performance problems, extends equipment life, and reduces maintenance costs.
Smart Images

Figure CN118518974B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of measuring electric variables, and in particular to an automatic intelligent testing method and system for low-voltage transformers. Background Art
[0002] A low-voltage transformer is a device that converts high-voltage electricity into low-voltage electricity. It is commonly used in urban and rural power distribution systems. In the process of transformer voltage conversion, in order to ensure its normal operation and good performance, it is necessary to collect and analyze various parameters during its working process to verify whether its various functions and characteristics meet the requirements, monitor potential faults or abnormalities, and take measures in advance to repair or prevent them. By testing and evaluating the working status, electrical performance and safety of the transformer, the stable operation and reliability of the power supply system can be guaranteed, thereby improving energy utilization, extending equipment life, and reducing operating costs and maintenance costs.
[0003] Existing transformer testing methods may be affected by factors such as the temperature of the transformer during use and changes in the transformer output power, thereby having a certain impact on the accuracy of the transformer performance and quality test, resulting in poor accuracy in the performance and quality test of the transformer during use, which is not conducive to timely and accurate detection of transformer faults. Summary of the invention
[0004] The present invention provides a low-voltage transformer automatic intelligent testing method and system to solve the existing problems.
[0005] The present invention provides a low-voltage transformer automatic intelligent testing method and system using the following technical solutions:
[0006] An embodiment of the present invention provides a low-voltage transformer automatic intelligent testing method, the method comprising the following steps:
[0007] Lay out additional coils and collect the voltage, current and temperature of the transformer primary coil, secondary coil and additional coil within a preset time;
[0008] According to the voltage and current of the coil, the coil power is obtained, and the energy loss rate of the transformer at any time is obtained according to the coil power. According to the energy loss rate of the transformer, the transformer temperature, and the power of the transformer coil, the transformer performance difference sequence is obtained;
[0009] A sliding window is set to traverse the transformer performance difference sequence to obtain sequence fragments. According to the number of data points in the sequence fragments and the values of the data points, the volatility of the sequence fragments when the sliding window is at any position is determined. According to the volatility of the sequence fragments when the sliding window is at any position, the values of all data points in the sequence fragments, and the energy loss rate of the transformer at the moment corresponding to all data points in the sequence fragments, a reference fragment is selected to obtain the latest sequence fragment and the volatility of the latest sequence fragment. The difference in volatility between the latest sequence fragment and the reference fragment is analyzed to determine the possibility of performance problems in the transformer at the latest moment.
[0010] Furthermore, the transformer performance difference sequence is obtained according to the energy loss rate of the transformer, the transformer temperature, and the power of the transformer coil, and the specific method includes:
[0011] According to the temperature of the primary coil and the secondary coil, the temperature of the transformer is determined. Moment and The temperature and energy loss rate of the transformer at the moment, the power of the new coil and the secondary coil, determine Moment and The difference between the transformer performance at the same time;
[0012] Within a preset time, the difference values of the transformer performance at all two adjacent moments before the latest moment are calculated to obtain several difference values, which are recorded as data points. A time axis is drawn, and the sequence formed by all data points on the time axis is obtained in chronological order, which is recorded as the transformer performance difference sequence.
[0013] Furthermore, the Moment and The specific calculation method of the difference between the transformer performance at the moment is:
[0014]
[0015] In the formula, express Moment and The difference between the transformer performance at the same time; express The temperature of the transformer at the moment; express The temperature of the transformer at the moment; express Energy loss rate of the transformer at the moment; express Energy loss rate of the transformer at this moment; express The power of the newly added coil at this moment; express The power of the newly added coil at this moment; express The power of the secondary coil at each moment; express The power of the secondary coil at each moment; represents the first preset hyperparameter; Indicates taking the absolute value.
[0016] Furthermore, the obtaining of the sequence fragments, and determining the volatility of the sequence fragments when the sliding window is at any position according to the number of data points in the sequence fragments and the values of the data points, includes the following specific methods:
[0017] The transformer performance difference sequence is traversed with a sliding window of preset size. During the traversal process, the time range corresponding to all data points contained in the sliding window at each position is a sequence segment. The sliding window is located at the first position according to the value of the data point in the sliding window and the mean value of all data points in the sliding window. The volatility of the series segment when there are 1 data point.
[0018] Furthermore, the reference segment is selected according to the volatility of the sequence segment when the sliding window is at any position, the values of all data points in the sequence segment, and the energy loss rate of the transformer at the moment corresponding to all data points in the sequence segment, including the specific method of:
[0019] According to the sliding window located at The volatility of the sequence fragment, the mean of all data point values in the sliding window, and the mean of the energy loss rate of the transformer at the time corresponding to all data points in the sliding window are used to determine the sliding window is located at the The reference credibility of the sequence fragment when there are data points;
[0020] The entire transformer performance difference sequence is traversed from left to right using a sliding window to obtain reference credibility of several sequence segments, and the sequence segment with the largest reference credibility is selected as the reference segment.
[0021] Furthermore, the sliding window is located at When there are data points, the specific calculation method of the reference credibility of the sequence fragment is:
[0022]
[0023] In the formula, Indicates that the sliding window is located at The reference credibility of the sequence fragment when there are data points; Indicates that the sliding window is located at The volatility of the sequence fragment when there are 1 data point; Indicates that the sliding window is located at When there are data points, it is the mean of all data points in the sequence segment; Indicates that the sliding window is located at When there are data points, the mean value of the energy loss rate of the transformer at the time corresponding to all data points in the sequence segment; Represents an exponential function with a natural constant as base.
[0024] Furthermore, the method of obtaining the latest sequence segment and the volatility of the latest sequence segment, analyzing the volatility difference between the latest sequence segment and the reference segment, and determining the possibility of the transformer having a performance problem at the latest moment includes the following specific methods:
[0025] A sliding window of preset size is traversed on the time axis. When the sliding window is located at the latest moment on the time axis, the range contained in the position of the sliding window is recorded as the latest sequence segment, and the volatility of the latest sequence segment is obtained. According to the volatility of the latest sequence segment, the volatility of the reference segment, the mean of the data points in the latest sequence segment, the mean of the data points in the reference segment, and the range of the data points in the latest sequence segment, the possibility of performance problems in the transformer at the latest moment is determined.
[0026] Furthermore, the specific calculation method of the possibility that the transformer has a performance problem at the latest moment is:
[0027]
[0028] In the formula, Indicates the possibility of transformer performance problems at the latest moment; Indicates the volatility of the latest sequence segment; represents the volatility of the reference segment; Represents the mean of the data points in the latest sequence segment; represents the mean of the data points within the reference segment; Indicates the range of the data points within the latest sequence segment; represents the first preset hyperparameter; Represents the normalization function.
[0029] Furthermore, the specific calculation method of the energy loss rate of the transformer at any time is:
[0030]
[0031] In the formula, express Energy loss rate of the transformer at this moment; express The power of the secondary coil at each moment; express The power of the primary coil at this moment.
[0032] A low-voltage transformer automation intelligent testing system comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the processor implements any one of the steps of a low-voltage transformer automation intelligent testing method.
[0033] The beneficial effects of the technical solution of the present invention are as follows: the coil power is obtained according to the voltage and current of the coil, the energy loss rate of the transformer at any time is obtained according to the coil power, the transformer performance difference sequence is obtained according to the energy loss rate of the transformer, the transformer temperature, and the power of the transformer coil, representing the performance changes of the transformer at different times, a sliding window is set to traverse the transformer performance difference sequence, and sequence fragments are obtained, and the volatility of the sequence fragment when the sliding window is at any position is determined according to the number of data points and the values of the data points in the sequence fragment, and a reference fragment is selected according to the volatility of the sequence fragment when the sliding window is at any position, the values of all data points in the sequence fragment, and the energy loss rate of the transformer at the time corresponding to all data points in the sequence fragment. The reference fragment can represent the performance of the transformer at the most stable time, the volatility of the sequence fragment corresponding to the latest time is obtained, the volatility difference between the sequence fragment corresponding to the latest time and the reference fragment is analyzed, the possibility of the transformer having performance problems at the latest time is determined, the transformer performance problems are discovered in time, the accuracy and timeliness of transformer performance detection are improved, and the automatic intelligent testing of the transformer is realized. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0035] Figure 1 The present invention is a flowchart of the steps of an automated intelligent testing method for low-voltage transformers. DETAILED DESCRIPTION
[0036] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a low-voltage transformer automatic intelligent testing method and system proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.
[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
[0038] The specific scheme of the automatic intelligent testing method for low-voltage transformer in the area provided by the present invention is described in detail below with reference to the accompanying drawings.
[0039] See also Figure 1 , which shows a flowchart of a method for automatic intelligent testing of low-voltage transformers provided by an embodiment of the present invention, the method comprising the following steps:
[0040] Step S001: Arrange a newly added coil, and collect the voltage, current and temperature of the primary coil, secondary coil and newly added coil of the transformer within a preset time.
[0041] It should be noted that in order to monitor the use of the transformer and analyze various parameters during the use of the transformer, it is first necessary to collect various data during the use of the transformer.
[0042] Specifically, in order to implement the low-voltage transformer automatic intelligent testing method proposed in this embodiment, it is first necessary to collect various parameters of the transformer during use. The specific process is as follows:
[0043] A new coil is arranged at the magnetic steel position of the secondary coil of the transformer, which is recorded as the newly added coil. A voltmeter and an ammeter with strong anti-interference ability are used to collect the voltage and current values of the primary coil, the secondary coil and the newly added coil of the transformer within a preset time. The collection frequency is once every 0.2 seconds. At the same time, temperature sensors are arranged in the primary coil and the secondary coil of the transformer to collect the temperature data of the primary coil and the secondary coil.
[0044] It should be noted that the preset time refers to the first 24 hours of use of the transformer.
[0045] So far, the voltage, current and temperature of the primary coil, secondary coil and newly added coil of the transformer within the preset time are obtained through the above method.
[0046] Step S002: Obtain coil power according to coil voltage and current, obtain energy loss rate of transformer at any time according to coil power, and obtain transformer performance difference sequence according to transformer energy loss rate, transformer temperature and transformer coil power.
[0047] It should be noted that during the operation of the transformer, the magnetic energy generated by the primary coil connected to the high voltage is transmitted to the iron core of the secondary coil, and the secondary coil generates a certain induced voltage and current by utilizing the principle of magnetoelectricity. During the energy conversion process of the transformer, due to the resistance of the coil itself and the eddy current phenomenon in the magnetoelectricity process, a certain amount of heat will be generated, causing the temperature of the transformer to rise and the magnetic conductivity to be affected. The increase in the temperature of the transformer will also increase the resistance of the coil inside the transformer, thereby affecting the energy conversion efficiency of the transformer. When the coil inside the transformer ages and the magnetic conductivity decreases, the working efficiency of the transformer will also decrease, thereby losing more energy. Therefore, in order to test the performance and quality of the transformer during use, it is necessary to first obtain the energy loss rate of the transformer at any time, and obtain the transformer performance difference sequence based on the energy loss rate of the transformer at any time, the temperature of the transformer, and the power of the transformer coil.
[0048] Step (2.1), obtain the coil power according to the coil voltage and current, and obtain the energy loss rate of the transformer at any time according to the coil power.
[0049] Specifically, according to the voltage and current of the primary coil and the secondary coil, the power of the primary coil and the secondary coil at any time is obtained through a power calculation formula, and according to the power of the primary coil and the secondary coil, the energy loss rate of the transformer at any time is determined.
[0050] As an embodiment, a specific method for calculating the energy loss rate of the transformer at any time is as follows:
[0051]
[0052] In the formula, express Energy loss rate of the transformer at this moment; express The power of the secondary coil at each moment; express The power of the primary coil at this moment.
[0053] It should be noted that the primary coil is the input end of the transformer, and the secondary coil is the output end of the transformer. express The ratio between the transformer output power and input power at the moment is: The energy conversion efficiency of the transformer at this moment, That is The energy loss rate of the transformer at this moment.
[0054] It should be noted that the power calculation formula is a known technology and will not be described in detail here.
[0055] Step (2.2), obtaining the transformer performance difference sequence according to the transformer energy loss rate, transformer temperature, and transformer coil power.
[0056] It should be noted that the energy loss rate of the transformer is related to the transformer temperature, which means that when the transformer temperature is the same and the transformer performance is stable, the energy conversion efficiency of the transformer will be relatively similar at any time, that is, the energy loss rate of the transformer at any time is also relatively similar. However, if the transformer has performance problems during operation, such as quality problems or coil short circuits, the energy conversion efficiency of the transformer will be affected. Therefore, in order to determine whether the transformer has performance problems, it is necessary to calculate the difference between the transformer performance at any two adjacent moments.
[0057] Specifically, according to Moment and The temperature and energy loss rate of the transformer at the moment, the power of the new coil and the secondary coil, determine Moment and The difference between the transformer performance at the same time.
[0058] As an example, Moment and The specific calculation method of the difference between the transformer performance at the moment is:
[0059]
[0060] In the formula, express Moment and The difference between the transformer performance at the same time; express The temperature of the transformer at the moment; express The temperature of the transformer at the moment; express Energy loss rate of the transformer at this moment; express Energy loss rate of the transformer at this moment; express The power of the newly added coil at this moment; express The power of the newly added coil at this moment; express The power of the secondary coil at each moment; express The power of the secondary coil at each moment; represents the first preset hyperparameter; Indicates taking the absolute value.
[0061] It should be noted that express Moment and The temperature difference of the transformer at two moments. Because the energy loss rate of the transformer is positively correlated with the temperature of the transformer, when the temperature of the transformer at two moments is similar, the energy loss rate of the transformer is also similar. express Moment and The ratio of the transformer temperature difference at each moment to the transformer energy loss rate difference, The less the value approaches 1, the Moment and The greater the performance difference of the transformer at each moment, The larger the value; express Moment and The power difference of the newly added coil at each moment. When there is no performance problem with the transformer, the power of the transformer secondary coil at adjacent moments is close, and the power of the newly added coil is also close. express Moment and The ratio of the power difference of the newly added coil to the power difference of the secondary coil at any given moment, The less the value approaches 1, the Moment and The greater the performance difference of the transformer at each moment, The larger the value.
[0062] It should be noted that the temperature of the transformer is the average of the temperatures of the primary coil and the secondary coil.
[0063] It should be noted that based on experience The value of is 0.001, which can be adjusted according to actual conditions and is not specifically limited in this embodiment.
[0064] Within a preset time, the difference values of the transformer performance at all two adjacent moments before the latest moment are calculated to obtain several difference values, which are recorded as data points. A time axis is drawn, and the sequence formed by all data points on the time axis is obtained in chronological order, which is recorded as the transformer performance difference sequence.
[0065] So far, the transformer performance difference sequence is obtained through the above method.
[0066] Step S003, set a sliding window to traverse the transformer performance difference sequence, obtain sequence fragments, determine the volatility of the sequence fragments when the sliding window is at any position according to the number of data points in the sequence fragments and the values of the data points, select a reference fragment according to the volatility of the sequence fragments when the sliding window is at any position, the values of all data points in the sequence fragments, and the energy loss rate of the transformer at the moment corresponding to all data points in the sequence fragments, obtain the latest sequence fragment and the volatility of the latest sequence fragment, analyze the difference in volatility between the latest sequence fragment and the reference fragment, and determine the possibility of performance problems of the transformer at the latest moment.
[0067] It should be noted that, under normal circumstances, the performance differences of the transformers at any two adjacent moments should be nearly consistent. However, during use, the transformer may have performance problems such as coil aging and coil short circuit due to the use environment and use time, resulting in changes in the transformer performance difference sequence. In order to determine the possibility of performance problems of the transformer at the latest moment, it is necessary to first select a reference segment in the transformer performance difference sequence. The reference segment can best represent the performance of the transformer when it is stable. By comparing and analyzing the sequence segment corresponding to the latest moment with the reference segment, the possibility of performance problems of the transformer at the latest moment can be determined.
[0068] Step (3.1) sets a sliding window to traverse the transformer performance difference sequence to obtain sequence fragments. According to the number of data points in the sequence fragments and the values of the data points, the volatility of the sequence fragments when the sliding window is at any position is determined. According to the volatility of the sequence fragments when the sliding window is at any position, the values of all data points in the sequence fragments, and the energy loss rate of the transformer at the moments corresponding to all data points in the sequence fragments, a reference fragment is selected.
[0069] First, a sliding window of preset size is used to traverse the transformer performance difference sequence. During the traversal process, the time range corresponding to all data points contained in the sliding window at each location is a sequence segment. The sliding window is located at the first segment according to the value of the data point in the sequence segment and the mean value of all data points in the sequence segment. The volatility of the series segment when there are 1 data point.
[0070] As an embodiment, the sliding window is located at The specific calculation method of the volatility of the series fragment when there are data points is:
[0071]
[0072] In the formula, Indicates that the sliding window is located at The volatility of the series segment at each data point; Indicates the size of the sliding window, that is, the number of data points contained in the sequence segment; Indicates that the sliding window is located at data points, the The value of the data point; Indicates that the sliding window is located at When there are data points in the sequence, it is the mean of all data point values in the sequence segment.
[0073] It should be noted that Indicates that the sliding window is located at When there are data points, the mean of all data points in the sequence segment is The smaller the value, the more stable the transformer performance is within the sequence segment. Indicates that the sliding window is located at When there are data points, the sum of the absolute values of the differences between the values of all data points in the sequence segment and the mean of all data points in the sequence segment is The smaller the value, the smaller the difference of data points in the sequence segment, that is, the more stable the transformer performance in the sequence segment.
[0074] It should be noted that the size of the sliding window is preset based on experience. , which can be adjusted according to actual conditions and is not specifically limited in this embodiment.
[0075] It should be noted that in this embodiment, the sliding window is located at data point, refers to the The data point is taken as the center point, and the sliding window includes the two data points before and after the center point. In addition, when the center point of the sliding window is located at the beginning or the end of the transformer performance difference sequence, that is, the two data points before or after the center point may not exist. Therefore, in this embodiment, the non-existent data points are calculated as 0.
[0076] Then, according to the sliding window at the When there are data points, the volatility of the sequence segment, the mean of all data point values in the sequence segment, and the mean of the energy loss rate of the transformer at the moment corresponding to all data points in the sequence segment are used to determine the sliding window at the first The reference confidence level of the sequence fragment when there are 1 data point.
[0077] As an embodiment, the sliding window is located at When there are data points, the specific calculation method of the reference credibility of the sequence fragment is:
[0078]
[0079] In the formula, Indicates that the sliding window is located at The reference credibility of the sequence fragment when there are data points; Indicates that the sliding window is located at The volatility of the sequence fragment when there are 1 data point; Indicates that the sliding window is located at When there are data points, it is the mean of all data points in the sequence segment; Indicates that the sliding window is located at When there are data points, the mean value of the energy loss rate of the transformer at the time corresponding to all data points in the sequence segment; Represents an exponential function with a natural constant as base.
[0080] It should be noted that The smaller the value is, the higher the energy conversion efficiency of the transformer is within the preset time contained in the sequence segment, and the better the transformer performance is. The larger the value, the closer the sliding window is to the The greater the number of data points, the greater the reference credibility of the sequence fragment.
[0081] The entire transformer performance difference sequence is traversed from left to right using a sliding window to obtain reference credibility of several sequence segments, and the sequence segment with the largest reference credibility is selected as the reference segment.
[0082] In step (3.2), the latest sequence segment and the volatility of the latest sequence segment are obtained, and the volatility difference between the latest sequence segment and the reference segment is analyzed to determine the possibility of performance problems of the transformer at the latest moment.
[0083] It should be noted that the volatility of the latest sequence segment obtained is analyzed with the volatility of the reference segment. When the difference is large, it means that the transformer has performance problems at the latest moment.
[0084] Specifically, a sliding window of a preset size is traversed on the time axis. When the sliding window is located at the latest moment on the time axis, the range contained in the position of the sliding window is recorded as the latest sequence segment, and the volatility of the latest sequence segment is obtained. According to the volatility of the latest sequence segment, the volatility of the reference segment, the mean of the data points in the latest sequence segment, the mean of the data points in the reference segment, and the range of the data points in the latest sequence segment, the possibility of a performance problem of the transformer at the latest moment is determined. When the possibility of a performance problem of the transformer at the latest moment is greater than a preset first threshold, it can be considered that the transformer has a performance problem at that moment.
[0085] As an embodiment, a specific method for calculating the possibility that the transformer has a performance problem at the latest moment is:
[0086]
[0087] In the formula, Indicates the possibility of transformer performance problems at the latest moment; Indicates the volatility of the latest sequence segment; represents the volatility of the reference segment; Represents the mean of the data points in the latest sequence segment; represents the mean of the data points within the reference segment; Indicates the range of the data points within the latest sequence segment; represents the first preset hyperparameter; Represents the normalization function.
[0088] It should be noted that represents the volatility ratio of the latest sequence segment to the reference segment, The larger the value, the greater the fluctuation of the latest sequence segment, and the greater the possibility that the transformer has performance problems; It represents the ratio of the mean value of the data points in the latest sequence segment to the mean value of the data points in the reference segment. The larger the value, the greater the difference in transformer performance at the latest moment, and the greater the possibility of performance problems.
[0089] It should be noted that the first threshold is preset to 0.4 based on experience, and can be adjusted according to actual conditions, and is not specifically limited in this embodiment.
[0090] At this point, the possibility of the transformer having performance problems at the latest moment is obtained through the above method.
[0091] Through the above steps, the automated intelligent testing of low-voltage transformers is completed.
[0092] An embodiment of the present invention further provides a low-voltage transformer automatic intelligent testing system, the system comprising the following modules:
[0093] A data acquisition module is used to arrange the newly added coils and collect the voltage, current and temperature of the primary coil, secondary coil and newly added coil of the transformer within a preset time;
[0094] The performance difference sequence acquisition module is used to obtain the coil power according to the coil voltage and current, obtain the energy loss rate of the transformer at any time according to the coil power, and obtain the transformer performance difference sequence according to the transformer energy loss rate, transformer temperature, and transformer coil power;
[0095] The transformer performance judgment module is used to set a sliding window to traverse the transformer performance difference sequence, obtain sequence fragments, determine the volatility of the sequence fragments when the sliding window is at any position according to the number of data points and the values of the data points in the sequence fragments, select the reference fragment according to the volatility of the sequence fragments when the sliding window is at any position, the values of all data points in the sequence fragments, and the energy loss rate of the transformer at the moment corresponding to all data points in the sequence fragments, obtain the latest sequence fragment and the volatility of the latest sequence fragment, analyze the difference in volatility between the latest sequence fragment and the reference fragment, and determine the possibility of performance problems of the transformer at the latest moment.
[0096] The present invention also proposes an automated intelligent testing device for low-voltage transformer substations, which includes automated intelligent testing equipment for low-voltage transformer substations and a memory, a processor, and a computer program stored in the memory and executable on the processor contained in the automated intelligent testing equipment for low-voltage transformer substations. When the processor executes the computer program, the steps of the method described in steps S001 to S003 are implemented.
[0097] In this embodiment, the coil power is obtained according to the voltage and current of the coil, and the energy loss rate of the transformer at any time is obtained according to the coil power. According to the energy loss rate of the transformer, the transformer temperature, and the power of the transformer coil, a transformer performance difference sequence is obtained to represent the performance change of the transformer at different times. A sliding window is set to traverse the transformer performance difference sequence to obtain sequence fragments. According to the number of data points in the sequence fragments and the values of the data points, the volatility of the sequence fragments when the sliding window is at any position is determined. According to the volatility of the sequence fragments when the sliding window is at any position, the values of all data points in the sequence fragments, and the energy loss rate of the transformer at the time corresponding to all data points in the sequence fragments, a reference fragment is selected, and the reference fragment can represent the performance of the transformer at the most stable time. The volatility of the sequence fragment corresponding to the latest time is obtained, and the difference in volatility between the sequence fragment corresponding to the latest time and the reference fragment is analyzed to determine the possibility of performance problems of the transformer at the latest time, timely discover transformer performance problems, improve the accuracy and timeliness of transformer performance detection, and realize automatic intelligent testing of transformers.
[0098] It should be noted that the The model is only used to represent negative correlation and constrain the output of the model to be in In the specific implementation, it can be replaced by other models with the same purpose. This embodiment is only based on The model is described as an example without any specific limitation. is the input to the model.
[0099] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A low voltage transformer automatic intelligent testing method, characterized in that: The method comprises the following steps: Lay out additional coils and collect the voltage, current and temperature of the transformer primary coil, secondary coil and additional coil within a preset time; According to the voltage and current of the coil, the coil power is obtained, and the energy loss rate of the transformer at any time is obtained according to the coil power. According to the energy loss rate of the transformer, the transformer temperature, and the power of the transformer coil, the transformer performance difference sequence is obtained; A sliding window is set to traverse the transformer performance difference sequence to obtain sequence fragments. According to the number of data points in the sequence fragments and the values of the data points, the volatility of the sequence fragments when the sliding window is at any position is determined. According to the volatility of the sequence fragments when the sliding window is at any position, the values of all data points in the sequence fragments, and the energy loss rate of the transformer at the moment corresponding to all data points in the sequence fragments, a reference fragment is selected to obtain the latest sequence fragment and the volatility of the latest sequence fragment. The difference in volatility between the latest sequence fragment and the reference fragment is analyzed to determine the possibility of performance problems of the transformer at the latest moment. The specific method of obtaining the transformer performance difference sequence according to the transformer energy loss rate, transformer temperature, and transformer coil power includes: According to the temperature of the primary coil and the secondary coil, the temperature of the transformer is determined. Moment and The temperature and energy loss rate of the transformer at the moment, the power of the new coil and the secondary coil, determine Moment and The difference between the transformer performance at the same time; Within a preset time, the difference values of the transformer performance at all two adjacent moments before the latest moment are calculated to obtain a number of difference values, which are recorded as data points, and a time axis is drawn to obtain a sequence formed by all data points on the time axis in chronological order, which is recorded as a transformer performance difference sequence; Said Moment and The specific calculation method of the difference between the transformer performance at the moment is: In the formula, express Moment and The difference between the transformer performance at the same time; express The temperature of the transformer at the moment; express The temperature of the transformer at the moment; express Energy loss rate of the transformer at the moment; express Energy loss rate of the transformer at the moment; express The power of the newly added coil at this moment; express The power of the newly added coil at this moment; express The power of the secondary coil at each moment; express The power of the secondary coil at each moment; represents the preset first hyperparameter; Indicates taking the absolute value.
2. According to claim 1, a low-voltage transformer automatic intelligent testing method is characterized in that: The obtaining of the sequence fragments and determining the volatility of the sequence fragments when the sliding window is at any position according to the number of data points in the sequence fragments and the values of the data points include the following specific methods: The transformer performance difference sequence is traversed with a sliding window of preset size. During the traversal process, the time range corresponding to all data points contained in the position of the sliding window at each time is a sequence fragment. The sliding window is located at the first position according to the value of the data point in the sliding window and the mean value of all data points in the sliding window. The volatility of the series segment when there are 1 data point.
3. According to claim 1, a low-voltage transformer automatic intelligent testing method is characterized in that: The method of selecting a reference segment according to the volatility of the sequence segment when the sliding window is at any position, the values of all data points in the sequence segment, and the energy loss rate of the transformer at the moment corresponding to all data points in the sequence segment includes: According to the sliding window located at The volatility of the sequence fragment, the mean of all data point values in the sliding window, and the mean of the energy loss rate of the transformer at the time corresponding to all data points in the sliding window are used to determine the sliding window is located at the The reference credibility of the sequence fragment when there are data points; The entire transformer performance difference sequence is traversed from left to right using a sliding window to obtain reference credibility of several sequence segments, and the sequence segment with the largest reference credibility is selected as the reference segment.
4. According to claim 3, a low-voltage transformer automatic intelligent testing method is characterized in that: The sliding window is located at When there are data points, the specific calculation method of the reference credibility of the sequence fragment is: In the formula, Indicates that the sliding window is located at The reference credibility of the sequence fragment when there are data points; Indicates that the sliding window is located at The volatility of the sequence fragment when there are data points; Indicates that the sliding window is located at When there are data points, it is the mean of all data points in the sequence segment; Indicates that the sliding window is located at When there are data points, the mean value of the energy loss rate of the transformer at the time corresponding to all data points in the sequence segment; Represents an exponential function with a natural constant as base.
5. According to claim 1, a low-voltage transformer automatic intelligent testing method is characterized in that: The method of obtaining the latest sequence segment and the volatility of the latest sequence segment, analyzing the volatility difference between the latest sequence segment and the reference segment, and determining the possibility of the transformer having a performance problem at the latest moment includes the following specific methods: A sliding window of preset size is traversed on the time axis. When the sliding window is located at the latest moment on the time axis, the range contained in the position of the sliding window is recorded as the latest sequence segment, and the volatility of the latest sequence segment is obtained. According to the volatility of the latest sequence segment, the volatility of the reference segment, the mean of the data points in the latest sequence segment, the mean of the data points in the reference segment, and the range of the data points in the latest sequence segment, the possibility of performance problems in the transformer at the latest moment is determined.
6. According to claim 5, a low-voltage transformer automatic intelligent testing method is characterized in that: The specific calculation method of the possibility that the transformer has performance problems at the latest moment is: In the formula, Indicates the possibility of transformer performance problems at the latest moment; Indicates the volatility of the latest sequence segment; represents the volatility of the reference segment; Represents the mean of the data points in the latest sequence segment; represents the mean of the data points within the reference segment; Represents the range of data points within the latest series segment; represents the preset first hyperparameter; Represents the normalization function.
7. According to claim 1, a low-voltage transformer automatic intelligent testing method is characterized in that: The specific calculation method of the energy loss rate of the transformer at any time is: In the formula, express Energy loss rate of the transformer at the moment; express The power of the secondary coil at each moment; express The power of the primary coil at this moment.
8. An automatic intelligent test system for low-voltage transformers, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.
