阅读说明：本技术 一种变压器振动在线异常检测方法 (Transformer vibration online anomaly detection method ) 是由 宁鑫 朱轲 邓元实 罗洋 张睿 李巍巍 范松海 于 2021-07-19 设计创作，主要内容包括：针对目前如何基于振动法对变压器运行状态进行精细化分析及异常检测,本发明实施例公开了一种变压器振动在线异常检测方法,该方法依据负载电流大小划分变压器运行工况区间,使用加速度传感器获取变压器表面振动信号,判定振动信号所属工况区间,基于Lempel-Ziv算法计算当前振动信号的复杂度,通过与当前变压器运行工况设定的复杂度阈值对比,判断变压器的运行状态。由此,本发明根据负载电流划分变压器运行工况,基于不同工况条件下的历史数据分析评判变压器当前运行状态,实现了考虑工况因素下的变压器振动信号量化分析,使得本发明的变压器异常检测方法更具严谨性。(The embodiment of the invention discloses a transformer vibration online abnormity detection method, which is used for dividing transformer operation working condition intervals according to the magnitude of load current, acquiring a transformer surface vibration signal by using an acceleration sensor, judging the working condition interval to which the vibration signal belongs, calculating the complexity of the current vibration signal based on a Lempel-Ziv algorithm, and judging the operation state of a transformer by comparing the complexity threshold set by the current transformer operation working condition. Therefore, the transformer operation conditions are divided according to the load current, the current operation state of the transformer is analyzed and judged based on historical data under different condition conditions, the quantitative analysis of the vibration signal of the transformer under the condition of considering the condition factors is realized, and the transformer abnormity detection method is more rigorous.)

1. A transformer vibration online abnormity detection method is characterized by comprising the following steps:

dividing transformer operation condition intervals according to the load current;

acquiring a vibration signal of the transformer by using an acceleration sensor, and judging a working condition interval to which the vibration signal belongs;

calculating the complexity of vibration signals in different load current intervals based on a Lempel-Ziv algorithm;

and judging the running state of the transformer by comparing with the complexity threshold set in the section to which the transformer belongs.

2. The online abnormality detection method for transformer vibration according to claim 1, characterized in that a transformer operation condition interval is divided according to load current, specifically comprising:

and dividing the working condition interval of the transformer according to the magnitude of the load current. The rated current of the transformer is set as X, the working condition of the transformer is divided into 5 intervals according to the load current value: 0,0.3X), 0.3X,0.5X, 0.8X,1X), and 1X,1.5X), and these 5 intervals are defined as conditions A, B, C, D and E.

3. The online transformer vibration anomaly detection method according to claim 1, wherein the complexity calculation of the vibration signal is completed based on a Lempel-Ziv algorithm, and specifically comprises the following steps:

1) and carrying out binary coarse graining on the vibration signal sampled every time, namely 5k data points to obtain a binary sequence S. Firstly, an average value is obtained for an original vibration signal time sequence, points larger than the average value are assigned to be 1, points smaller than the average value are assigned to be 0, and a binary sequence S is obtained as { S ═ S₁,S₂,...,S₅₀₀₀}；

2)P₀、Q₀For an empty matrix, let i equal to 0, where complexity c (i) is equal to 0;

3) and entering a circulation. i is i +1, let P_i-1＝{P_i-1S_i}，Q_i-1＝{Q_i-1S_iIs judged P thereafter_i-1Whether or not to include Q_i. If the determination result is yes, the complexity C (i) is not increased, i.e., C (i) ═ C (i-1); if the determination result is "no", C (i) ═ C (i-1) +1, Q_i＝{}. Cycling N5000 times until traversing the binary sequence S;

4) complexity C_NAnd (6) normalizing. For the binary sequence S, the final complexity calculation result is normalized by the following method:

4. the method for detecting the online abnormality of the vibration of the transformer according to claim 1, wherein the step of judging the operation state of the transformer by comparing the operation state with the complexity threshold set in the section to which the transformer belongs specifically comprises the steps of:

the threshold setting method is as follows:

the complexity threshold of the vibration signal in the working condition A interval is the average of the complexity of the historical vibration data and is marked as C_AmeanAnd so on, the other 4 interval thresholds are marked as C_Bmean、C_Cmean、C_DmeanAnd C_Emean。

The running state of the transformer is divided into four types of normal, attention, warning and fault, and the running state judgment rule is as follows:

1) when the absolute value of the difference between the calculated value of C and the belonged interval threshold value is greater than or equal to 0.5 times of the belonged interval threshold value for three times (or more than three times), judging that the current operation state is attention;

2) when the absolute value of the difference between the calculated value of C and the belonged interval threshold is more than 1 time of the belonged interval threshold, judging that the current operation state is an alarm;

3) within 24 hours, when the absolute value of the difference between the calculated value of C and the belonged interval threshold value is more than 1 time within three times or more than three times, the current operation state is judged to be a fault;

4) when any 2 or all 2 of the 1), 2) and 3) are met, the current running state is judged to be the state with the highest abnormal degree, and the sequence of the abnormal degree from low to high is normal, attention, warning and fault;

5) and when any one of the above 1), 2) and 3) is not satisfied, judging that the current operation state is normal.

Technical Field

The invention relates to the field of electric power, in particular to a transformer vibration online abnormity detection method.

Background

Power transformers are important devices in power systems, and their operating conditions have a significant impact on the safety and economic efficiency of the power system. The surface vibration of the transformer mainly originates from the vibration of the winding and the iron core under the excitation of current and voltage, and theoretical analysis and practical experience show that the working state of the winding and the iron core can be analyzed through the vibration signal of the transformer. When structural components such as an iron core and a winding in the transformer are loosened, the vibration intensity is increased, the natural frequency component is increased, and the signal complexity is increased. However, in the actual operation process, the load current is often fluctuated greatly, and when the load current is increased, the electrodynamic force borne by the winding is increased by square times, so that higher harmonic components in the vibration signal are increased, the signal complexity is increased, and a plurality of problems are brought to the abnormal detection of the power transformer based on the vibration analysis method.

Disclosure of Invention

In order to solve the technical problem, the invention provides a transformer vibration online abnormity detection method, which comprises the steps of dividing operation working condition intervals according to the magnitude of load current, obtaining a transformer surface vibration signal by using an acceleration sensor, judging the working condition interval to which the vibration signal belongs, calculating the complexity of the vibration signal in different load current intervals based on a Lempel-Ziv algorithm, and judging the operation state of a transformer by comparing with a complexity threshold set in the belonging interval.

The embodiment of the invention provides the following technical scheme:

a transformer vibration online abnormity detection method comprises the following steps:

dividing transformer operation condition intervals according to the load current;

acquiring a vibration signal on the surface of the transformer by using an acceleration sensor, and judging a working condition interval to which the vibration signal belongs;

calculating the complexity of vibration signals in different load current intervals based on a Lempel-Ziv algorithm;

and judging the running state of the transformer by comparing with the complexity threshold set in the section to which the transformer belongs.

Wherein, divide transformer operating condition interval according to load current size, specifically include:

and dividing the working condition interval of the transformer according to the magnitude of the load current. The rated current of the transformer is set as X, the working condition of the transformer is divided into 5 intervals according to the load current value: 0,0.3X), 0.3X,0.5X, 0.8X,1X), and 1X,1.5X), and these 5 intervals are defined as conditions A, B, C, D and E.

The complexity calculation of the vibration signal is completed based on a Lempel-Ziv algorithm, and the method specifically comprises the following steps:

1) and carrying out binary coarse graining on the vibration signal sampled every time, namely 5k data points to obtain a binary sequence S. Firstly, an average value is obtained for an original vibration signal time sequence, points larger than the average value are assigned to be 1, points smaller than the average value are assigned to be 0, and a binary sequence S is obtained as { S ═ S₁,S₂,...,S₅₀₀₀}；

2)P₀、Q₀For an empty matrix, let i equal to 0, where complexity c (i) is equal to 0;

3) and entering a circulation. i is i +1, let P_i-1＝{P_i-1S_i}，Q_i-1＝{Q_i-1S_iIs judged P thereafter_i-1Whether or not to include Q_i. If the determination result is yes, the complexity C (i) is not increased, i.e., C (i) ═ C (i-1); if the determination result is "no", C (i) ═ C (i-1) +1, Q_i{ }. Cycling N5000 times until traversing the binary sequence S;

4) complexity C_NAnd (6) normalizing. For the binary sequence S, the final complexity calculation result is normalized by the following method:

the method for judging the running state of the transformer by comparing the complexity threshold set in the section to which the transformer belongs specifically comprises the following steps:

the threshold setting method is as follows:

the complexity threshold of the vibration signal in the working condition A interval is the average of the complexity of the historical vibration data and is marked as C_AmeanAnd so on, the other 4 interval thresholds are marked as C_Bmean、C_Cmean、C_DmeanAnd C_Emean。

The running state of the transformer is divided into four types of normal, attention, warning and fault, and the running state judgment rule is as follows:

1) when the absolute value of the difference between the calculated value of C and the belonged interval threshold value is greater than or equal to 0.5 times of the belonged interval threshold value for three times (or more than three times), judging that the current operation state is attention;

2) when the absolute value of the difference between the calculated value of C and the belonged interval threshold is more than 1 time of the belonged interval threshold, judging that the current operation state is an alarm;

3) within 24 hours, when the absolute value of the difference between the calculated value of C and the belonged interval threshold value is more than 1 time within three times or more than three times, the current operation state is judged to be a fault;

4) when any 2 or all 2 of the 1), 2) and 3) are met, the current running state is judged to be the state with the highest abnormal degree, and the sequence of the abnormal degree from low to high is normal, attention, warning and fault;

5) and when any one of the above 1), 2) and 3) is not satisfied, judging that the current operation state is normal.

Compared with the prior art, the technical scheme has the following advantages:

the inventor of the application considers that the load current is always large in fluctuation in the running process of the transformer, and when the load current is increased, the electrodynamic force borne by the winding is increased by square times, so that higher harmonic components in vibration signals are increased, and the signal complexity is increased; similarly, when the internal structural components of the transformer, such as the iron core and the winding, are abnormally loosened, higher harmonics in the vibration signal are increased, so that the complexity of the signal is increased. Therefore, the transformer operation conditions are divided according to the load current, the current operation state of the transformer is analyzed and judged based on historical data under different condition conditions, the quantitative analysis of the vibration signal of the transformer under the condition of considering the condition factors is realized, and the transformer abnormity detection method is more rigorous.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

FIG. 1 is a schematic diagram of a transformer vibration online anomaly detection method;

FIG. 2 is a flow chart of complexity calculation for completing a vibration signal based on a Lempel-Ziv algorithm.

Detailed Description

As described in the background art, how to accurately judge the operation state of the transformer based on the vibration method is a technical problem that needs to be solved urgently by those skilled in the art.

In order to solve the technical problem, the invention provides a transformer vibration online abnormity detection method, which comprises the steps of dividing transformer operation working condition intervals according to load current, obtaining a transformer surface vibration signal by using an acceleration sensor, judging the working condition interval to which the vibration signal belongs, calculating the complexity of the vibration signal of different load current intervals based on a Lempel-Ziv algorithm, and judging the operation state of a transformer by comparing with a threshold set in each load current interval. According to the method, the complexity of the vibration signal of the transformer is quantitatively analyzed by using a Lempel-Ziv algorithm, and the running state of the transformer is analyzed and judged by considering the factors of the running working condition of the transformer, so that the method is higher in practicability and higher in detection precision.

Fig. 1 is a schematic diagram of a transformer vibration online anomaly detection method, which includes:

step 1: and dividing the operation condition interval of the transformer according to the load current.

The transformer has large load current, heavy load or overload operation, and the safety reliability and the service life of the transformer are influenced. The working conditions of the transformer are divided according to the magnitude of the load current. The rated current of the transformer is set as X, the working condition of the transformer is divided into 5 intervals according to the load current value: 0,0.3X), 0.3X,0.5X, 0.8X,1X), and 1X,1.5X), and these 5 intervals are defined as conditions A, B, C, D and E.

Step 2: and acquiring a vibration signal of the surface of the transformer by using an acceleration sensor, and judging a working condition interval to which the vibration signal belongs.

The method comprises the steps of placing a vibration acceleration sensor at a preset power transformer surface vibration measuring point, setting the sampling rate of the vibration acceleration sensor to be 5kHz, setting the sampling length of each time to be 5k, namely obtaining 5k sampling values by sampling each time, setting the sampling time length of each time to be 1 second, and recording the load current size while sampling each time.

And step 3: and completing the complexity calculation of the vibration signal based on a Lempel-Ziv algorithm. The calculation flow chart refers to fig. 2.

The complexity calculation of the vibration signal is completed based on a Lempel-Ziv algorithm, and the method specifically comprises the following steps:

1) the vibration signal of each sampling, i.e. 5k data points, is subjected to binary coarse graining. Firstly, an average value is obtained for an original vibration signal time sequence, points larger than the average value are assigned to be 1, points smaller than the average value are assigned to be 0, and a binary sequence S is obtained as { S ═ S₁,S₂,...,S₅₀₀₀}；

2)P₀、Q₀For an empty matrix, let i equal to 0, where complexity c (i) is equal to 0;

3) entering a loop, i is equal to i +1, and enabling P_i-1＝{P_i-1S_i}，Q_i-1＝{Q_i-1S_iIs judged P thereafter_i-1Whether or not to include Q_iIf the determination result is yes, complexity C (i) does not increase, that is, C (i) ═ C (i-1). If the result of the judgment is "no", C (i) ═ C (i-1) +1, Q_i{ }. Cycling N5000 times until traversing the binary sequence S;

4) complexity C_NAnd (6) normalizing. For the binary sequence S, the final complexity calculation result is normalized by the following method:

and 4, step 4: and judging the running state of the transformer by comparing with the complexity threshold set in the section to which the transformer belongs.

The threshold setting method is as follows:

the complexity threshold of the vibration signal in the working condition A interval is the average of the complexity of the historical vibration data and is marked as C_AmeanAnd so on, the other 4 interval thresholds are marked as C_Bmean、C_Cmean、C_DmeanAnd C_Emean。

The running state of the transformer is divided into four types of normal, attention, warning and fault, and the running state judgment rule is as follows:

1) when the absolute value of the difference between the calculated value of C and the belonged interval threshold value is greater than or equal to 0.5 times of the belonged interval threshold value for three times (or more than three times), judging that the current operation state is attention;

2) when the absolute value of the difference between the calculated value of C and the belonged interval threshold is more than 1 time of the belonged interval threshold, judging that the current operation state is an alarm;

3) within 24 hours, when the absolute value of the difference between the calculated value of C and the belonged interval threshold value is more than 1 time within three times or more than three times, the current operation state is judged to be a fault;

4) when any 2 or all 2 of the 1), 2) and 3) are met, the current running state is judged to be the state with the highest abnormal degree, and the sequence of the abnormal degree from low to high is normal, attention, warning and fault;

5) and when any one of the above 1), 2) and 3) is not satisfied, judging that the current operation state is normal.

In conclusion, the inventor of the application finds that the load current of the transformer is often fluctuated greatly in the actual operation process, and when the load current is increased, the electrodynamic force borne by the winding is increased by square times, so that the higher harmonic components in the vibration signals are increased, and the signal complexity is increased; similarly, when the internal structural components of the transformer, such as the iron core and the winding, are abnormally loosened, higher harmonics in the vibration signal are increased, so that the complexity of the signal is increased. Therefore, the inventors considered that it is necessary to consider an important factor of current change when abnormality detection is performed on the vibration signal. Therefore, the transformer operation conditions are divided according to the load current, the current operation state of the transformer is analyzed and judged based on historical data under different condition conditions, the quantitative analysis of the vibration signal of the transformer under the condition of considering the condition factors is realized, and the transformer abnormity detection method is more rigorous. Specifically, the vibration signals in different load current intervals are subjected to complexity calculation by using a Lempel-Ziv algorithm, the influence of overlarge current change on the vibration signals is restrained to a certain extent, the transformer vibration signals can be subjected to quantitative analysis more finely, misjudgment caused by current change is reduced in the process of abnormality detection, and the detection method is more rigorous.

In the description, each part is described in a progressive manner, each part is emphasized to be different from other parts, and the same and similar parts among the parts are referred to each other.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.