A substation circuit breaker fault identification method and sound and light alarm system

Abstract

The present invention relates to the field of high-voltage power grid operation fault alarm safety technology, and specifically to a substation circuit breaker fault identification method and an acoustic and visual alarm system, wherein the fault identification method comprises the following steps: obtaining a vibration signal of the circuit breaker; decomposing the vibration signal to obtain a time-frequency data set; using the time-frequency data set to obtain an energy response sequence of vibration signals of different frequency bands in the time domain; using the energy response sequence to obtain the energy distribution characteristics of the vibration signal in the time domain under the corresponding frequency band; selecting at least one target energy response sequence from the energy response sequences of multiple frequency bands according to the energy distribution characteristics; reconstructing at least one group of time domain signals using the target energy response sequence; respectively comparing the similarity of at least one group of time domain signals with a standard time domain signal, and identifying the circuit breaker fault state according to the similarity comparison result. The substation circuit breaker fault identification method provided by the present invention can realize the comprehensive identification of possible fault states of vibration signals.

Description

Technical Field

[0001] The present invention relates to the technical field of high-voltage power grid operation fault alarm safety, and in particular to a transformer substation circuit breaker fault identification method and an audible and visual alarm system. Background Art

[0002] Substation circuit breakers (CBs) are an indispensable core component of the power system. Their key function is to accurately control and protect the circuit. Under normal operating conditions and in the face of various faults, circuit breakers can quickly and effectively open or close the circuit to ensure the stable and safe operation of the power system. However, once a circuit breaker fails, it will pose a serious threat to the safety and stability of the entire power system.

[0003] Through in-depth research on the reliability of high-voltage circuit breakers, many research results at home and abroad have shown that mechanical failure is the main cause of circuit breaker failure. These mechanical failures include but are not limited to deformation of the transmission mechanism, electromagnet jamming, crank jamming, lock failure, screw loosening, abnormal contact position, poor lubrication, buffer failure, and component damage. These failures not only affect the normal operation of the circuit breaker, but may also cause power system interruption, resulting in huge economic losses and social impact.

[0004] Therefore, it is of great significance to carry out research on the monitoring and diagnosis methods of mechanical faults of high-voltage circuit breakers and formulate reasonable maintenance and overhaul strategies based on them to improve the reliability of power systems. However, the traditional regular maintenance method can no longer meet the needs of the intelligent development of modern high-voltage circuit breakers due to its low efficiency and the possibility of over-maintenance or under-maintenance. Therefore, there is an urgent need for an intelligent and efficient substation circuit breaker fault identification technology and a supporting sound and light alarm system to monitor the status of the circuit breaker in real time, quickly and accurately identify faults, provide timely warnings, and formulate scientific maintenance and overhaul plans. Summary of the invention

[0005] In response to the needs of practical applications, the present invention provides a substation circuit breaker fault identification method and an sound and light alarm system, which aim to achieve real-time monitoring, accurate identification and rapid early warning of circuit breaker mechanical failures, thereby reducing circuit breaker maintenance costs and improving the safety and intelligent maintenance level of the power system.

[0006] In a first aspect, the present invention provides a method for identifying a substation circuit breaker fault, comprising the following steps: obtaining a vibration signal of a circuit breaker executing a target operation instruction; decomposing the vibration signal to obtain a time-frequency data set of the vibration signal; using the time-frequency data set to obtain an energy response sequence of vibration signals of different frequency bands in a time domain range; using the energy response sequence to obtain an energy distribution feature of the vibration signal in the time domain range under the corresponding frequency band; based on the energy distribution feature, selecting at least one target energy response sequence from energy response sequences of multiple frequency bands; using the target energy response sequence, reconstructing at least one group of time domain signals; respectively comparing the similarity of at least one group of time domain signals with a standard time domain signal, and identifying the fault state of the circuit breaker when executing the target operation instruction based on the similarity comparison result. The substation circuit breaker fault identification method provided by the present invention avoids the influence of noise signals on subsequent fault state identification by decomposing and screening the target energy response sequence of the vibration signal of the circuit breaker executing the target operation instruction; and realizes comprehensive identification of possible fault states of the vibration signal through reconstruction of the target energy response sequence or combination and reconstruction of multiple target energy response sequences, thereby ensuring the accuracy and comprehensiveness of the circuit breaker fault identification, which not only helps to provide more reliable data support for the maintenance and repair of the power system, but also helps to improve the fault monitoring and diagnosis capabilities of the power system.

[0007] Optionally, the step of obtaining the vibration signal of the circuit breaker executing the target operation instruction comprises the following steps: installing a vibration sensor on the operating mechanism housing of the circuit breaker, the vibration sensor being used to obtain the vibration signal of the operating mechanism when the circuit breaker executes the target operation instruction; connecting the vibration sensor to the control system signal of the circuit breaker, and using the control system to control the vibration sensor so that the sampling state of the vibration sensor is synchronized with the execution state of the operating mechanism. The present invention can accurately obtain the vibration signal of the circuit breaker when executing the target operation instruction in real time by installing a vibration sensor on the operating mechanism housing of the circuit breaker and connecting it to the control system signal so that the sampling state of the vibration sensor is synchronized with the execution state of the operating mechanism, thereby improving the accuracy and timeliness of fault identification.

[0008] Optionally, the decomposition of the vibration signal to obtain a time-frequency data set of the vibration signal includes the following steps: setting a number of continuous time domain windows within the time domain range of the vibration signal, the time domain range being from the sampling start time to the sampling end time; setting a wavelet packet basis function and the number of decomposition layers, using the wavelet packet basis function to decompose the vibration signal in any time domain window layer by layer until the number of decomposition layers is reached, and obtaining the wavelet packet decomposition coefficients of the vibration signal in different frequency bands in the corresponding time domain window; summarizing the wavelet packet decomposition coefficients of the vibration signal in different frequency bands in the continuous time domain windows, and obtaining the time-frequency data set of the vibration signal according to the summary results, the frequency domain range of the time-frequency data set being from zero to the Nyquist frequency. The present invention decomposes complex vibration signals into time-frequency data sets that are easier to analyze through time-frequency decomposition technology, which helps to improve the accuracy and efficiency of vibration signal processing.

[0009] Optionally, the use of the time-frequency data set to obtain the energy response sequence of vibration signals of different frequency bands within the time domain includes the following steps: obtaining an energy matrix corresponding to the time-frequency data set, the number of rows of the energy matrix being consistent with the number of time domain windows, and the number of columns of the energy matrix being consistent with the number of decomposed frequency bands; decomposing the energy matrix by columns to obtain the energy response sequence of vibration signals of different frequency bands within the time domain. The present invention can obtain the energy response sequence of vibration signals of different frequency bands within the time domain by obtaining the energy matrix corresponding to the time-frequency data set and decomposing the energy matrix by columns, which enables the energy changes of the signal in different frequency bands to be accurately captured and analyzed, thereby improving the accuracy and effectiveness of fault identification.

[0010] Optionally, the use of the energy response sequence to obtain the energy distribution characteristics of the vibration signal in the time domain range under the corresponding frequency band includes the following steps: using an energy distribution characteristic calculation model, combined with energy response sequences under different frequency bands, to obtain the energy distribution characteristics of the vibration signal in the time domain range under the corresponding frequency band. By extracting energy distribution characteristics, the present invention can effectively capture key fault information in the signal, thereby reducing the interference of noise and non-fault characteristics, and helping to improve the accuracy of recognition.

[0011] Optionally, the energy distribution characteristic calculation model is: Among them, H(f j ) represents the energy response sequence The energy distribution characteristics of , i is a positive integer, i represents the time domain window number, m represents the number of time domain windows, It represents the energy response value corresponding to the jth frequency band in the ith time domain window. The energy distribution characteristic calculation model provided by the present invention can accurately obtain the distribution uniformity of signal energy in different time domain windows and frequency bands.

[0012] Optionally, the step of selecting at least one target energy response sequence from energy response sequences of multiple frequency bands according to the energy distribution characteristics comprises the following steps: obtaining an energy distribution characteristic threshold; comparing the energy distribution characteristics of different energy response sequences with the energy distribution characteristic threshold respectively; and selecting at least one target energy response sequence from energy response sequences of multiple frequency bands according to the comparison result, wherein the energy distribution characteristic of any target energy response sequence exceeds the energy distribution characteristic threshold. The present invention obtains an energy distribution characteristic threshold and compares the energy distribution characteristics of different energy response sequences with the threshold to select a target energy response sequence whose energy distribution characteristic exceeds the threshold. This screening method can effectively highlight important fault signals, reduce interference from noise and irrelevant signals, and improve the accuracy and efficiency of fault identification, thereby more reliably detecting the fault state of the circuit breaker.

[0013] Optionally, the step of reconstructing at least one set of time domain signals by using the target energy response sequence comprises the following steps: using at least one target energy response sequence to reconstruct the corresponding time domain signal by inverse wavelet packet transform. The present invention reconstructs the corresponding time domain signal by inverse wavelet packet transform using a single or multiple target energy response sequences. This method can accurately restore the fault characteristics of the vibration signal, improve the accuracy and reliability of signal reconstruction, thereby more effectively identifying and diagnosing the fault state of the circuit breaker and enhancing the fault detection capability of the system.

[0014] Optionally, the method of comparing the similarity of at least one group of time domain signals with the standard time domain signal and identifying the fault state of the circuit breaker when executing the target operation instruction according to the similarity comparison result comprises the following steps: using a similarity comparison model to obtain the similarity between any reconstructed time domain signal and any standard time domain signal; judging the fault state of the circuit breaker when executing the target operation instruction according to the similarity threshold and the similarity. The present invention uses the reconstructed time domain signal to perform a similarity comparison with the standard signal to comprehensively identify various fault states, including single faults and complex multiple fault states, thereby reducing false positives and false negatives.

[0015] In the second aspect, the substation circuit breaker fault sound and light alarm system provided by the present invention includes a data input module, a data storage module, a data processing module and a sound and light output module; the data input module, the data storage module, the data processing module and the sound and light output module are connected to each other; wherein the data input module is used to input the vibration signal of the circuit breaker to execute the target operation instruction; the data storage module is used to store a computer program, and the computer program includes program instructions; the data processing module is configured to call the program instructions, execute the substation circuit breaker fault identification method, and generate a sound and light alarm instruction based on the execution result of the substation circuit breaker fault identification method; the sound and light output module is used to receive and execute the sound and light alarm instruction. The substation circuit breaker fault sound and light alarm system provided by the present invention can stably and efficiently execute the substation circuit breaker fault identification method, and then generate a sound and light alarm instruction to achieve real-time early warning, and promptly remind maintenance personnel to take measures to prevent the expansion of faults. The substation circuit breaker fault sound and light alarm system provided by the present invention not only reduces manual intervention and improves the intelligent level of fault detection, but also reduces false alarms and missed alarms through accurate fault identification and rapid early warning, thereby improving the operational reliability and safety of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of a method for identifying a substation circuit breaker fault provided by the present invention;

[0017] Figure 2 This is a structural diagram of the substation circuit breaker fault sound and light alarm system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0019] In one embodiment, see Figure 1 , Figure 1 This is a flow chart of the substation circuit breaker fault identification method provided by the present invention. Figure 1 As shown, the substation circuit breaker fault identification method provided by the embodiment of the present invention includes the following steps:

[0020] S01. Obtain a vibration signal of a circuit breaker executing a target operation instruction.

[0021] In this embodiment, the vibration signal is generated by the movement of the operating mechanism when the circuit breaker is running, and reflects the physical quantity of the mechanical movement state of the operating mechanism, such as displacement, velocity or acceleration.

[0022] Furthermore, the step of obtaining the vibration signal of the circuit breaker executing the target operation instruction in step S01 includes the following steps:

[0023] S011. Install a vibration sensor on the operating mechanism housing of the circuit breaker, wherein the vibration sensor is used to obtain a vibration signal of the operating mechanism when the circuit breaker executes a target operating instruction.

[0024] In this embodiment, the vibration sensor is any one of piezoelectric, piezoresistive, and capacitive accelerometers.

[0025] S012: Connect the vibration sensor to the control system signal of the circuit breaker, and use the control system to control the vibration sensor so that the sampling state of the vibration sensor is synchronized with the execution state of the operating mechanism.

[0026] In this embodiment, the control system is used to manage and control the operation of the circuit breaker, including closing, opening and other functions. Generally, the control system mainly includes a control unit and a human-computer interaction unit.

[0027] Among them, the control unit is responsible for interpreting operating instructions, processing sensor inputs (such as current, voltage, temperature and vibration data), and issuing control commands based on preset logic and algorithms; the human-computer interaction unit includes an interface for interacting with the control system, through which relevant technicians can monitor the status of the circuit breaker, manually input operating commands, or modify control parameters.

[0028] Furthermore, the sampling frequency of the vibration sensor refers to the number of times the vibration signal is collected per unit time, usually expressed in Hz (times per second). Usually, the sampling frequency is preset in advance by relevant technicians based on industry standards or relevant experience, such as 1kHz, 2kHz, etc. It can be known that the sampling frequency determines the frequency resolution of subsequent signal processing and the highest frequency that can be observed (Nyquist frequency).

[0029] S02. Decompose the vibration signal to obtain a time-frequency data set of the vibration signal.

[0030] In this embodiment, the time domain range of the time-frequency data set is from the sampling start time to the sampling end time, and the frequency domain range of the time-frequency data set is from zero to the Nyquist frequency; any time-frequency data in the time-frequency data set represents the signal energy at a specific time and a specific frequency, for example, a time-frequency data can represent the signal energy E(10ms,100Hz) at a frequency of 100Hz at the 10th millisecond from the sampling start time.

[0031] Furthermore, the decomposing of the vibration signal in step S02 to obtain a time-frequency data set of the vibration signal includes the following steps:

[0032] S021. Setting a plurality of continuous time domain windows within the time domain range of the vibration signal, wherein the time domain range is from the sampling start time to the sampling end time.

[0033] In this embodiment, the long-term signal from the sampling start time to the sampling end time is divided into a plurality of continuous short time periods, and each short time period is a time domain window.

[0034] In each time domain window, the signal can be considered to be approximately stationary, so the signal in each window can be analyzed, such as wavelet decomposition. It can be understood that the role of the time domain window is to be able to capture the time-varying characteristics of the signal in time-frequency analysis.

[0035] S022. Set a wavelet packet basis function and a decomposition layer number, and use the wavelet packet basis function to decompose the vibration signal in any time domain window layer by layer until the decomposition layer number is reached, and obtain the wavelet packet decomposition coefficients of the vibration signal in the corresponding time domain window in different frequency bands.

[0036] The wavelet packet basis function is a basic function for signal decomposition and signal reconstruction. Common wavelet packet basis functions can be extended functions based on wavelet basis functions such as Daubechies wavelet, Coiflets wavelet, Haar wavelet and Symlets wavelet. It can be understood that different wavelet packet basis functions have different characteristics and are suitable for different types of signal processing.

[0037] The number of decomposition levels refers to the number of levels at which the signal is decomposed in the wavelet transform. The number of decomposition levels determines the degree of refinement of the decomposition. The more levels there are, the higher the frequency resolution of the decomposition. Generally speaking, the number of decomposition levels cannot exceed the logarithm of the signal length. Too many decomposition levels will cause the division of frequency bands to exceed the frequency resolution of the signal itself, thus affecting the accuracy of the analysis and the computational efficiency.

[0038] Furthermore, the number of decomposition layers satisfies the following constraints: m∈N * , 1≤m <log 2 ((T e -T s)·f s ), where m represents the number of decomposition levels, N* represents a positive integer, and T e Indicates the sampling start time value, T s Indicates the end time of sampling, f s Indicates the sampling frequency.

[0039] S023. Summarize the wavelet packet decomposition coefficients of the vibration signal in different frequency bands in the continuous time domain window, and obtain a time-frequency data set of the vibration signal according to the summary result.

[0040] It is understandable that based on the wavelet packet transform from step S021 to step S022, step S023 can obtain the wavelet packet decomposition coefficients of different frequency bands in each time domain window. Further, step S023 uses the square of the wavelet packet decomposition coefficient to characterize the signal energy of the corresponding frequency band.

[0041] Therefore, the time-frequency data set of the vibration signal in this embodiment is represented by the energy matrix E T×F Represents, where the number of rows of the energy matrix is ​​consistent with the number of time domain windows, and the number of columns of the energy matrix is ​​consistent with the number of decomposed frequency bands.

[0042] S03. Using the time-frequency data set, obtain energy response sequences of vibration signals of different frequency bands within the time domain.

[0043] In this embodiment, the energy response sequence refers to a sequence in which the energy of vibration signals in different frequency bands changes over time in the entire time domain.

[0044] Furthermore, the step S03 uses the time-frequency data set to obtain energy response sequences of vibration signals of different frequency bands in the time domain, including the following steps: obtaining an energy matrix E corresponding to the time-frequency data set T×F ; Decompose the energy matrix E by columns T×F , obtain the energy response sequence of vibration signals of different frequency bands in the time domain

[0045] Specifically, any frequency band f j The energy response sequence of the vibration signal in the time domain is Includes multiple energy response values: Wherein, j is a positive integer and represents the frequency band number.

[0046] Furthermore, the energy response sequence Any energy response value It represents the energy response value corresponding to the jth frequency band in the i-th time domain window, where i is a positive integer and represents the sequence number of the time domain window.

[0047] S04. Using the energy response sequence, obtain energy distribution characteristics of the vibration signal in the time domain at a corresponding frequency band.

[0048] In this embodiment, the energy distribution characteristics refer to statistical characteristics of the vibration signal energy in the corresponding frequency band within the entire time domain, such as average energy value, maximum energy value, energy standard deviation, energy entropy, etc.

[0049] Furthermore, the use of the energy response sequence in step S04 to obtain the energy distribution characteristics of the vibration signal in the time domain range under the corresponding frequency band specifically includes the following steps: using an energy distribution characteristic calculation model, combined with energy response sequences under different frequency bands, to obtain the energy distribution characteristics of the vibration signal in the time domain range under the corresponding frequency band.

[0050] In this embodiment, any frequency band f j Lower energy response sequence The energy distribution characteristic calculation model is as follows: Among them, H(f j ) represents the energy response sequence The energy distribution characteristics of , i is a positive integer, i represents the time domain window number, m represents the number of time domain windows, Represents the energy response value corresponding to the jth frequency band in the i-th time domain window.

[0051] In this embodiment, the energy distribution characteristic H(f j ) represents the frequency band f j The uniformity of energy distribution in the time domain, energy distribution characteristics H(f j ) is larger, the more uniform the signal energy distribution is; the energy distribution characteristic H(f j ) is smaller, the more discrete the signal energy distribution is.

[0052] S05. Filter out at least one target energy response sequence from energy response sequences of multiple frequency bands according to the energy distribution characteristics.

[0053] Furthermore, the step S05 of selecting at least one energy response sequence from energy response sequences of multiple frequency bands according to the energy distribution characteristics comprises the following steps:

[0054] An energy distribution characteristic threshold is obtained, and energy distribution characteristics of different energy response sequences are compared with the energy distribution characteristic threshold respectively, and at least one energy response sequence whose energy distribution characteristic exceeds the energy distribution characteristic threshold is screened out.

[0055] In this embodiment, the energy distribution characteristic threshold refers to an energy entropy threshold used to screen and determine whether there is abnormal energy concentration in the vibration signal during the fault diagnosis process.

[0056] S06. Reconstruct at least one set of time domain signals using the target energy response sequence.

[0057] Furthermore, when only one set of target energy response sequences is screened out in step S05, step S06 utilizes this set of target energy response sequences to reconstruct corresponding time domain signals.

[0058] Specifically, using a single target energy response sequence The corresponding time domain signal is reconstructed by inverse wavelet packet transform. The reconstructed time domain signal X:{x(t 1 ), x(t 2 ), x(t 3 ), ..., x(t m )}, where any signal x(t i ) satisfies the following calculation model: in, represents the energy response value corresponding to the jth frequency band in the i-th time domain window, φ(t i ) represents the wavelet packet basis function of the i-th time domain window.

[0059] Furthermore, when only two or more target energy response sequences are screened out in step S05, step S06 utilizes different numbers of combinations of different target energy response sequences to reconstruct corresponding time domain signals.

[0060] For example, when three groups of target energy response sequences a, b, and c are screened out in step S05, step S06 not only uses the three groups of target energy response sequences a, b, and c for separate reconstruction, but also uses two groups of a and b, two groups of b and c, c, and three groups of a, b, and c for combined reconstruction.

[0061] Furthermore, using a set of target energy response sequences e (T,fa) and b group target energy response sequence combination e (T,fb) , reconstruct the corresponding time domain signal X:{x(t 1 ), x(t 2 ), x(t 3 ), ..., x(t m )}, where any signal x(t i ) satisfies the following calculation model: Wherein, a and b are both positive integers and a≠b, a and b represent the frequency band numbers. represents the energy response value corresponding to the ath frequency band in the i-th time domain window, represents the energy response value corresponding to the bth frequency band in the i-th time domain window, φ(t i ) represents the wavelet packet basis function of the i-th time domain window.

[0062] Furthermore, using a set of target energy response sequences b group target energy response sequence combination And group c target energy response sequence combination Reconstruct the corresponding time domain signal X: {x(t 1 ), x(t 2 ), x(t 3 ), ..., x(t m )}, where any signal x(t i ) satisfies the following calculation model: Wherein, a, b, c are all positive integers and a≠b, a≠c, c≠b, a, b and c all represent frequency band numbers. represents the energy response value corresponding to the ath frequency band in the i-th time domain window, represents the energy response value corresponding to the bth frequency band in the i-th time domain window, represents the energy response value corresponding to the cth frequency band in the ith time domain window, φ(t i ) represents the wavelet packet basis function of the i-th time domain window.

[0063] S07. Compare the similarities of at least one group of time domain signals with the standard time domain signal respectively, and identify the fault state of the circuit breaker when executing the target operation instruction according to the similarity comparison result.

[0064] It is understandable that the standard time domain signal includes the vibration signal of the circuit breaker executing the target operation instruction under normal working state, the vibration signal of the circuit breaker executing the target operation instruction under a single fault state, and the vibration signal of the circuit breaker executing the target operation instruction under multiple fault states. Furthermore, the fault state includes fault states such as contact wear, loose insulating rod, inability to maintain the closed state when closing, broken opening, loose soft connection, etc.

[0065] Furthermore, the step S07 of comparing the similarity of at least one group of time domain signals with the standard time domain signal, and identifying the fault state of the circuit breaker when executing the target operation instruction according to the similarity comparison result, includes the following steps:

[0066] S071. Using a similarity comparison model, obtain the similarity between any reconstructed time domain signal and any standard time domain signal.

[0067] Furthermore, the similarity comparison between any group of reconstructed time domain signals and any standard time domain signal in step S071 can be achieved through data similarity calculation models such as Euclidean distance, Pearson correlation coefficient, cosine similarity, or by constructing a corresponding classification or recognition model through support vector machine (SVM), random forest, convolutional neural network (CNN), etc. to obtain the corresponding similarity.

[0068] Specifically, when the Euclidean distance is used to compare the similarity between the reconstructed time domain signal and the standard time domain signal, the similarity between any group of reconstructed time domain signals and a standard time domain signal satisfies the following calculation formula: Among them, d(x(t), s(t)) represents the similarity between the reconstructed time domain signal x(t) and the standard time domain signal s(t), x(t i ) represents the reconstructed time domain signal corresponding to the i-th time domain window, s(t i ) represents the standard time domain signal corresponding to the i-th time domain window.

[0069] S072. Determine the fault state of the circuit breaker when executing the target operation instruction according to the similarity threshold and the similarity.

[0070] It can be understood that the similarity threshold is a critical value used to determine whether the reconstructed time domain signal is sufficiently similar to the standard time domain signal; further, the similarity threshold can be set empirically or determined by statistical analysis of historical data.

[0071] Specifically, if the similarity between a group of reconstructed time domain signals and a standard time domain signal is greater than or equal to a similarity threshold, it is considered that the fault state represented by the reconstructed time domain signal is similar to that represented by the corresponding standard time domain signal, and then the circuit breaker is determined to be in the corresponding normal state or fault state.

[0072] In one embodiment, see Figure 2 , Figure 2 This is a structural diagram of the substation circuit breaker fault sound and light alarm system provided by an embodiment of the present invention.

[0073] like Figure 2 As shown, the substation circuit breaker fault sound and light alarm system provided in the embodiment of the present invention includes a data input module, a data storage module, a data processing module and a sound and light output module, and the data input module, the data storage module, the data processing module and the sound and light output module are interconnected.

[0074] Furthermore, the data input module is used to input a vibration signal of the circuit breaker executing a target operation instruction; the data storage module is used to store a computer program, the computer program includes program instructions, and the data processing module is configured to call the program instructions, execute the substation circuit breaker fault identification method, and generate an audible and visual alarm instruction based on the execution result of the substation circuit breaker fault identification method; the audible and visual output module is used to receive and execute the audible and visual alarm instruction.

[0075] When the circuit breaker executes the target operation instruction, the vibration sensor of the data input module obtains the vibration signal and transmits the signal to the data processing module; the data processing module executes the substation circuit breaker fault identification method according to the vibration signal: if the data processing module detects that the circuit breaker has a fault, it generates an audible and visual alarm instruction according to the fault identification result; the audible and visual output module receives the audible and visual alarm instruction, starts the speaker and the light alarm, and sends out an audible and visual alarm to remind the maintenance personnel that the circuit breaker has a fault.

[0076] 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, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents.

Claims

1. A method for identifying a substation circuit breaker fault, characterized in that: The steps include: Acquire a vibration signal of the circuit breaker executing a target operation instruction; Decomposing the vibration signal to obtain a time-frequency data set of the vibration signal; Using the time-frequency data set, obtaining energy response sequences of vibration signals of different frequency bands within the time domain; Using the energy response sequence, obtaining energy distribution characteristics of the vibration signal in the time domain range under the corresponding frequency band; According to the energy distribution characteristics, screening out at least one target energy response sequence from energy response sequences of multiple frequency bands; Reconstructing at least one set of time domain signals using the target energy response sequence; Comparing the similarity of at least one group of time domain signals with the standard time domain signal respectively, and identifying the fault state of the circuit breaker when executing the target operation instruction according to the similarity comparison result; Decomposing the vibration signal to obtain a time-frequency data set of the vibration signal comprises the following steps: A plurality of continuous time domain windows are set within the time domain range of the vibration signal, wherein the time domain range is from the sampling start time to the sampling end time; Setting a wavelet packet basis function and a decomposition layer number, using the wavelet packet basis function to decompose the vibration signal in any time domain window layer by layer until the decomposition layer number is reached, and obtaining the wavelet packet decomposition coefficients of the vibration signal in the corresponding time domain window in different frequency bands; Summarize the wavelet packet decomposition coefficients of the vibration signal in different frequency bands in the continuous time domain window, and obtain a time-frequency data set of the vibration signal according to the summary result, wherein the frequency domain range of the time-frequency data set is zero to Nyquist frequency; The method of using the time-frequency data set to obtain energy response sequences of vibration signals of different frequency bands in the time domain includes the following steps: Obtain an energy matrix corresponding to the time-frequency data set, wherein the number of rows of the energy matrix is ​​consistent with the number of time domain windows, and the number of columns of the energy matrix is ​​consistent with the number of decomposed frequency bands; Decomposing the energy matrix by columns to obtain energy response sequences of vibration signals of different frequency bands within the time domain; The step of reconstructing at least one set of time domain signals by using the target energy response sequence comprises the following steps: At least one target energy response sequence is used to reconstruct a corresponding time domain signal through inverse wavelet packet transform.

2. The method for identifying a substation circuit breaker fault according to claim 1, characterized in that: The step of obtaining a vibration signal of the circuit breaker executing a target operation instruction comprises the following steps: A vibration sensor is installed on the operating mechanism housing of the circuit breaker, wherein the vibration sensor is used to obtain a vibration signal of the operating mechanism when the circuit breaker executes a target operating instruction; The vibration sensor is connected to the control system signal of the circuit breaker, and the vibration sensor is controlled by the control system so that the sampling state of the vibration sensor is synchronized with the execution state of the operating mechanism.

3. The method for identifying a substation circuit breaker fault according to claim 2, characterized in that: The method of using the energy response sequence to obtain the energy distribution characteristics of the vibration signal in the time domain in the corresponding frequency band comprises the following steps: The energy distribution characteristic calculation model is used to combine the energy response sequences under different frequency bands to obtain the energy distribution characteristics of the vibration signal in the time domain range under the corresponding frequency band.

4. The method for identifying a substation circuit breaker fault according to claim 3, characterized in that: The energy distribution characteristic calculation model is: ,in, Represents the energy response sequence The energy distribution characteristics of , i is a positive integer, i represents the time domain window number, represents the number of time domain windows, Represents the energy response value corresponding to the jth frequency band in the i-th time domain window.

5. The method for identifying a substation circuit breaker fault according to claim 4, characterized in that: The step of selecting at least one target energy response sequence from energy response sequences of multiple frequency bands according to the energy distribution characteristics comprises the following steps: Obtaining energy distribution characteristic threshold; comparing the energy distribution characteristics of different energy response sequences with the energy distribution characteristic thresholds respectively; According to the comparison result, at least one target energy response sequence is screened out from the energy response sequences of the multiple frequency bands, and the energy distribution feature of any target energy response sequence exceeds the energy distribution feature threshold.

6. The method for identifying a substation circuit breaker fault according to claim 5, characterized in that: The method of comparing the similarity of at least one group of time domain signals with the standard time domain signal and identifying the fault state of the circuit breaker when executing the target operation instruction according to the similarity comparison result comprises the following steps: Using a similarity comparison model, the similarity between any reconstructed time domain signal and any standard time domain signal is obtained; According to the similarity threshold and the similarity, a fault state of the circuit breaker when executing the target operation instruction is determined.

7. A substation circuit breaker fault sound and light alarm system, characterized in that: It includes a data input module, a data storage module, a data processing module and an acoustic and optical output module; the data input module, the data storage module, the data processing module and the acoustic and optical output module are connected to each other; Among them, the data input module is used to input the vibration signal of the circuit breaker to execute the target operation instruction; the data storage module is used to store the computer program, and the computer program includes program instructions; the data processing module is configured to call the program instructions, execute the substation circuit breaker fault identification method described in any one of claims 1-6, and generate an audible and visual alarm instruction based on the execution result of the substation circuit breaker fault identification method; the audible and visual output module is used to receive and execute the audible and visual alarm instruction.

Images