CN114252810A - Transformer sound vibration fault monitoring method, system, equipment and readable storage medium - Google Patents
Transformer sound vibration fault monitoring method, system, equipment and readable storage medium Download PDFInfo
- Publication number
- CN114252810A CN114252810A CN202210033985.2A CN202210033985A CN114252810A CN 114252810 A CN114252810 A CN 114252810A CN 202210033985 A CN202210033985 A CN 202210033985A CN 114252810 A CN114252810 A CN 114252810A
- Authority
- CN
- China
- Prior art keywords
- transformer
- fault
- model
- monitoring
- observer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 80
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000013461 design Methods 0.000 claims abstract description 41
- 238000000605 extraction Methods 0.000 claims abstract description 15
- 239000000654 additive Substances 0.000 claims description 33
- 230000000996 additive effect Effects 0.000 claims description 33
- 230000003044 adaptive effect Effects 0.000 claims description 19
- 238000004590 computer program Methods 0.000 claims description 17
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000007781 pre-processing Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 12
- 238000005259 measurement Methods 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 4
- 238000003745 diagnosis Methods 0.000 abstract description 28
- 238000010586 diagram Methods 0.000 description 11
- 238000012545 processing Methods 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 238000013135 deep learning Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000007635 classification algorithm Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/62—Testing of transformers
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Protection Of Transformers (AREA)
Abstract
The invention discloses a method, a system, equipment and a readable storage medium for monitoring sound vibration faults of a transformer, wherein the method comprises the following steps: acquiring input and output signals of a transformer to be monitored in a normal working state; carrying out feature extraction on the input and output signals to obtain input signal features and output signal features; identifying a model of the transformer to be monitored in a normal working state based on the input signal characteristic and the output signal characteristic; observing by adopting an observer obtained by pre-design based on the identified model to obtain a disturbance observation value; and monitoring the sound vibration fault of the transformer based on the disturbance observation value. The invention provides a transformer sound vibration fault monitoring method, in particular to a transformer sound vibration fault on-line monitoring and early warning method based on a state observer, which can solve the problem that the existing fault diagnosis method based on data (signals) is limited by the number of samples and the types of the samples.
Description
Technical Field
The invention belongs to the technical field of transformer on-line monitoring and fault early warning, and particularly relates to a transformer sound vibration fault monitoring method, a transformer sound vibration fault monitoring system, transformer sound vibration fault monitoring equipment and a readable storage medium.
Background
The safe and stable operation of the transformer is an important factor for improving the quality of electric energy and the reliability of power supply, and the condition monitoring and fault diagnosis of the transformer are important measures for avoiding serious accidents. With the rapid development of modern information technology, sensing data such as temperature, ultrasonic waves, infrared images, vibration signals and the like are applied to transformer fault diagnosis; the vibration monitoring technology is a hotspot researched in recent years, on one hand, when the transformer works, the iron core, the winding and other mechanisms vibrate to generate mechanical waves, and acoustic vibration signals can be generated through propagation and comprise a large amount of equipment running state information; on the other hand, the sound vibration signal is acquired without power failure detection, and the normal operation of the equipment is not influenced, so that the sound vibration signal monitoring has a good application prospect in the field of transformer equipment fault diagnosis.
At present, fault diagnosis can be roughly divided into two categories, namely a data (signal) based fault diagnosis method and a model based fault diagnosis method, wherein the data based fault diagnosis method represented by deep learning can be widely applied to transformer fault diagnosis, but the effectiveness of the method is limited by the number of samples and the types of the samples, and the deep learning method requires a large amount of sample data during training and is only suitable for diagnosis of known fault types. For the transformer, the fault frequency is low, the maintenance is performed regularly, and the sample data in normal operation can be easily obtained, however, the sample data of different fault types of the transformer equipment is often difficult to obtain and needs to be accumulated for a long time, so that the data volume of the fault sample set is small and lacks completeness, the generalization capability of algorithms such as deep learning is greatly reduced, and the accuracy is low.
In the fault diagnosis method based on the model, the system model is used for detecting and diagnosing faults, and the method has the main advantages that unknown fault types can be diagnosed by using the model, and a large amount of data is not needed; through years of research, various model-based fault diagnosis methods such as a Kalman filter, an unknown input observer, a multiple integral observer and the like are formed, and the model-based methods can obtain good application prospects through theoretical and engineering practice verification, but no transformer sound vibration fault online monitoring strategy based on a state observer exists at present.
Disclosure of Invention
The invention aims to provide a transformer sound vibration fault monitoring method, a transformer sound vibration fault monitoring system, transformer sound vibration fault monitoring equipment and a readable storage medium, so as to solve one or more technical problems. The invention provides a transformer sound vibration fault monitoring method, in particular to a transformer sound vibration fault on-line monitoring and early warning method based on a state observer, which can solve the problem that the existing fault diagnosis method based on data (signals) is limited by the number of samples and the types of the samples.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention provides a transformer sound vibration fault monitoring method in a first aspect, which comprises the following steps:
acquiring input and output signals of a transformer to be monitored in a normal working state;
carrying out feature extraction on the input and output signals to obtain input signal features and output signal features;
identifying a model of the transformer to be monitored in a normal working state based on the input signal characteristic and the output signal characteristic;
observing by adopting an observer obtained by pre-design based on the identified model to obtain a disturbance observation value;
and monitoring the sound vibration fault of the transformer based on the disturbance observation value.
The method of the present invention is further improved in that the step of acquiring the input and output signals of the transformer to be monitored in the normal working state specifically comprises:
acquiring original input and output signals of a transformer to be monitored in a normal working state;
preprocessing the original input and output signals to obtain preprocessed input and output signals; and preprocessing is used for removing noise signals and obtaining the structural data with prominent signals of the transformer body.
The method of the present invention is further improved in that, in the feature extraction of the input and output signals to obtain the input signal features and the output signal features, the feature extraction specifically includes: and acquiring the frequency response of the sound vibration signal by a characteristic extraction method.
The method of the present invention is further improved in that the observing based on the identified model by the observer obtained by pre-design to obtain the disturbance observation value specifically includes:
when the identified model is an additive fault model, expressed as,
in the formula, xe＝[x f]TExpanding states for the transformer system, including a transformer system state x and a fault disturbance f, assuming f is bounded and differentiable; d is noise, u is input signal, y is measurement output signal, Ae、Be、Ce、EeR and Q are allIs a transformer system matrix;
the model-aided extended state observer model obtained by pre-design is expressed as,
in the formula, LoFor the observer gain matrix, z ═ zx zf]TIs an estimate of the disturbance caused by system conditions and faults;
wherein by selecting the observer gain LoRealization of zfAnd tracking the fault signal f.
The method of the present invention is further improved in that the observing based on the identified model by the observer obtained by pre-design to obtain the disturbance observation value specifically includes:
when the identified model is a non-additive fault model, it is expressed as,
where u is the input signal, y is the measurement output signal, x1And x2For the transformer system state, G (x)1,x2)，F(x1,x2U) and b are the transformer system matrix, θfAs a fault indicator factor, thetafWhen 0 indicates no failure, θfWhen the value is 1, the fault occurs;
the adaptive observer obtained by the pre-design, denoted,
in the formula I1、l2And l3To observer gain, z1，z2Is an estimate of the state of the system, zfFor fault-induced disturbance estimation, G (z)1,z2)，F(z1,z2,u) And b is an observer system matrix;
lim(zf) Is a clipping function expressed as
Wherein, by adjusting the observer gain l1、l2And l3Obtaining desired tracking characteristics, implementing zfAnd tracking the fault signal f.
The method of the invention is further improved in that the expression for realizing the monitoring of the sound vibration fault of the transformer based on the disturbance observation value is as follows,
The invention provides a transformer sound vibration fault monitoring system in a second aspect, which comprises:
the signal acquisition module is used for acquiring input and output signals of the transformer to be monitored in a normal working state;
the characteristic acquisition module is used for extracting the characteristics of the input signal and the output signal to acquire the characteristics of the input signal and the characteristics of the output signal;
the model identification module is used for identifying a model of the transformer to be monitored in a normal working state based on the input signal characteristics and the output signal characteristics;
the disturbance observation value acquisition module is used for observing by adopting an observer obtained by pre-design based on the identified model to obtain a disturbance observation value;
and the monitoring module is used for realizing the sound vibration fault monitoring of the transformer based on the disturbance observation value.
In a further improvement of the system of the present invention, in the disturbance observation acquisition module,
when the identified model is an additive fault model, expressed as,
in the formula, xe＝[x f]TExpanding states for the transformer system, including a transformer system state x and a fault disturbance f, assuming f is bounded and differentiable; d is noise, u is input signal, y is measurement output signal, Ae、Be、Ce、EeR and Q are all transformer system matrixes;
the model-aided extended state observer model obtained by pre-design is expressed as,
in the formula, LoFor the observer gain matrix, z ═ zx zf]TIs an estimate of the disturbance caused by system conditions and faults;
wherein by selecting the observer gain LoRealization of zfTracking the fault signal f;
when the identified model is a non-additive fault model, it is expressed as,
where u is the input signal, y is the measurement output signal, x1And x2For the transformer system state, G (x)1,x2)，F(x1,x2U) and b are the transformer system matrix, θfAs a fault indicator factor, thetafWhen 0 indicates no failure, θfWhen the value is 1, the fault occurs;
the adaptive observer obtained by the pre-design, denoted,
in the formula I1、l2And l3To observer gain, z1，z2Is an estimate of the state of the system, zfFor fault-induced disturbance estimation, G (z)1,z2)，F(z1,z2,u) And b is an observer system matrix;
lim(zf) Is a clipping function expressed as
Wherein, by adjusting the observer gain l1、l2And l3Obtaining desired tracking characteristics, implementing zfAnd tracking the fault signal f.
A third aspect of the present invention provides a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the transformer ringing fault monitoring method according to any one of the above aspects of the present invention.
A fourth aspect of the present invention provides a computer-readable storage medium, which stores a computer program, and the computer program, when executed by a processor, implements the steps of the transformer sound vibration fault monitoring method according to any one of the above aspects of the present invention.
Compared with the prior art, the invention has the following beneficial effects:
the invention provides a transformer sound vibration fault monitoring method, in particular to a transformer sound vibration fault on-line monitoring and early warning method based on a state observer, which can solve the problem that the existing fault diagnosis method based on data (signals) is limited by the number of samples and the types of the samples. Illustratively explained, the current classification algorithm based on data needs a large amount of training data samples, and when the data samples are insufficient, the generalization capability of the system is weak; in addition, when the training data set does not contain certain type of fault information, the classification algorithm based on data (signals) cannot effectively judge the type of fault, and the method can diagnose unknown faults by using a model. In conclusion, the transformer sound vibration fault on-line monitoring strategy based on the state observer provides a new idea for transformer fault diagnosis.
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 are briefly introduced below; it is obvious that the drawings in the following description are some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
FIG. 1 is a schematic diagram of a transformer acoustic-vibration fault monitoring and early warning based on a state observer according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating a transformer acoustic-vibration fault monitoring and early warning process based on a state observer according to an embodiment of the present invention;
FIG. 3 is a flow chart illustrating model identification according to an embodiment of the present invention;
fig. 4 is a schematic diagram of observer design and fault early warning process in the embodiment of the present invention.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The invention is described in further detail below with reference to the accompanying drawings:
referring to fig. 1, a method for monitoring a transformer acoustic vibration fault according to an embodiment of the present invention includes the following steps:
acquiring input and output signals of a transformer to be monitored in a normal working state;
carrying out feature extraction on the input and output signals to obtain input signal features and output signal features; exemplary, methods that can be employed for feature extraction include fourier transform, and the like;
identifying a model of the transformer to be monitored in a normal working state based on the input signal characteristic and the output signal characteristic; for example, the method adopted by the identification model is specifically a least square method and the like;
observing by adopting an observer obtained by pre-design based on the identified model to obtain a disturbance observation value;
and monitoring the sound vibration fault of the transformer based on the disturbance observation value.
The embodiment of the invention particularly provides a transformer sound vibration fault on-line monitoring and early warning method based on a state observer, which can solve the technical problem that the existing fault diagnosis method based on data (signals) is limited by the number and types of samples.
According to the embodiment of the invention, preferably, after acquiring and acquiring original input and output signals of the transformer to be monitored in a normal working state, preprocessing the original input and output signals to acquire preprocessed input and output signals; wherein the preprocessing is used for realizing the signal structuring processing to remove noise.
Referring to fig. 2 to 4, an embodiment of the method for monitoring and warning a transformer acoustic vibration fault on line based on a state observer includes the following steps:
step 1, preprocessing acoustic vibration signal data, comprising: acquiring a sound vibration signal when the transformer normally works by using a sound vibration sensor, and preprocessing the acquired signal to obtain preprocessed data; as an exemplary alternative, the pre-processing generally includes: pre-emphasis, denoising and sound source separation operations; the data preprocessing can improve the information quantity in the signals through the structured processing of the signals, and improve the quality of state monitoring and fault diagnosis.
Step 2, data feature extraction, comprising: and acquiring the frequency response of the acoustic vibration signal by a feature extraction method.
Step 3. model identification, comprising: and identifying the model of the transformer in the normal working state according to the obtained characteristic data (frequency domain characteristics) of the input and output sound vibration signals of the transformer in the normal state.
Step 4. state observer design, comprising: for the transformer sound vibration monitoring model characteristic, additive and non-additive fault models are distinguished, and further, the transformer state online monitoring is realized respectively based on two technical schemes of a model-assisted extended state observer (MESO for short in the embodiment of the invention) and a non-linear adaptive observer, and the transformer state online monitoring is realized by the following technical schemes from Step 4-1 to Step 4-5.
Step 4-1. additive and non-additive fault model differentiation comprises the following steps: judging whether the transformer sound vibration monitoring model is an additive fault model or not by analyzing the relevant characteristics of the transformer model fault disturbance signal and the input signal, and if the transformer sound vibration monitoring model is the additive fault model, monitoring and early warning the transformer fault through the following steps 4-2-4-3; if the fault model is a non-additive fault model, monitoring and early warning of the transformer fault are carried out through the following steps 4-4-Step 4-5.
Step 4-2. taking the typical state space form as an example, an object model considering fault disturbance f and noise d (which is assumed to be bounded and differentiable) is given by equation (1), where f is taken as the system expansion state:
in the formula, xe＝[x f]TTo expand the state of the system (1), including the transformer system state x and the fault disturbance f, u is the input signal, y is the measurement output signal, Ae、Be、Ce、EeR and Q are system matrices containing object model information.
Step 4-3, designing a model auxiliary extended state observer MESO to carry out state observation on the object of the formula (1); MESO can be expressed as:
in the formula, LoFor the observer gain matrix, z ═ zx zf]TThe estimated system state and disturbance caused by the fault.
The transformer error dynamics obtained from equations (1) and (2) are:
in the formula (3) provided that
Is bounded, then when A ise-LoCeWhen gradually stabilized, z can approach xe. In real industrial process objects
The bounded condition is easy to satisfy, so only a proper observer gain L needs to be selectedoNamely, z can be realizedfAnd tracking the fault signal f.
Illustratively, Step 4-2 to Step 4-3 are fault diagnosis technologies designed based on MESO, and the ESO takes the uncertainty and the external interference inside a transformer model as the total disturbance of the transformer model and estimates the total disturbance in real time through the expansion state. The method is characterized in that the transformer model information is added during ESO design, namely the model assists MESO design, so that the total disturbance estimated by the MESO is ensured to be external disturbance on the premise of reliable model, and the fault is the fault of a transformer model system in fault diagnosis.
Step 4-4. by introducing a fault indication factor theta into the modelf(when taking θ)fWhen 0 indicates no failure, θfWhen 1 indicates a failure), the second-order object model is rewritten into the form of equation (4) where G (x) is1,x2)、F(x1,x2U), b both contain model information:
step 4-5, designing a Luenberger observer to estimate the system state, wherein the specific contents are as follows:
wherein, denotes a fault observation state portion to be designed,/1And l2Is observer gain, z1、z2And zfAnd estimating the system state and fault disturbance. The systematic error dynamics can be expressed as:
constructing a lyapunov function as shown in equation (7):
it is possible to obtain:
only needs to ensure that theta is more than or equal to 0fWhen the content is less than or equal to 1, selecting1、l2Satisfies the condition of H (e)1,e2)<0, then is provided
Can meet the requirements
X cannot be obtained in practical application2When considering the steady state of the system, the method is characterized by
Take e approximately2＝l1e1. Can thus design
Wherein z isfThe variation range should also satisfy 0 ≦ zf≤1。
The Luenberger observer shown in final equation (5) is finally designed as an adaptive observer as shown below:
in the formula I1、l2And l3Is the Luenberger observer gain, lim (z)f) Is the clipping function shown in equation (10).
I.e. by adjusting the observer gain l1、l2And l3Achieving z by obtaining desired tracking characteristicsfAnd tracking the fault signal f.
Illustratively explained in the invention, Step4-4 to Step 4-5 are fault diagnosis technologies based on nonlinear adaptive observer design; according to the transformer fault diagnosis scheme based on the MESO design, all transformer fault disturbances are regarded as additive faults during observer design, and results cannot be visually reflected in non-additive fault MESO design, so that a non-linear adaptive observer is designed to monitor the non-additive faults of the transformer. In the embodiment of the invention, the design idea of the transformer nonlinear adaptive observer is explained in detail by taking a second-order object as an example.
Step 5, judging whether early warning is needed or not, wherein the judgment comprises the following steps: by evaluating the fault condition zfThe system state is monitored, and faults are early warned. I.e. the estimated value z according to ffJudging whether a fault occurs, namely:
in the formula, JthIs the threshold value of the judgment.
In summary, the embodiments of the present invention relate to a transformer sound vibration fault online monitoring and early warning technology based on a state observer, and in particular, to the technical field of transformer online monitoring and fault early warning. A transformer fault diagnosis strategy based on a model is provided aiming at the restriction of a fault diagnosis method based on data (signals) on a transformer data sample and a data type. The technical route of the embodiment of the invention comprises the following steps: firstly, acquiring sound vibration data of a transformer in a normal working state, and preprocessing the acquired signals; then, extracting characteristic data such as data characteristics, frequency characteristics and the like for bearing normal information of the transformer; secondly, identifying a transformer model according to input and output characteristic data under normal work; secondly, designing an MESO or nonlinear adaptive observer to carry out state monitoring; and finally, monitoring whether a fault occurs or not according to an estimated value of the external disturbance, and if so, carrying out early warning. According to the embodiment of the invention, after the transformer object model is added in the design of the extended state observer, the extended state realizes the estimation of the external disturbance of the transformer, namely the estimation of the fault state of the transformer, so that the real-time monitoring of the fault state of the transformer is realized, and the system realizes fault early warning when the output of the estimated state is greater than the threshold value.
In the specific embodiment, taking online monitoring of 500kV transformer faults as an example, the invention realizes the above purpose through the following technical schemes in Step 1 to Step 4:
step 1: and preprocessing the acoustic vibration signal data.
The method includes the steps that input and output sound vibration signals of a transformer in normal working are obtained through a sound vibration sensor, and data preprocessing is performed on the obtained signals, wherein the data preprocessing generally comprises the following steps: pre-emphasis, de-noising, and sound source separation operations.
Step 2: data feature extraction
The transformer has the advantages that no high-power rotating device runs inside the transformer, the vibration amplitude is small, the environmental noise is small, and the like, so that for the audio processing of non-rotating equipment such as the transformer, the existing feature extraction method such as Fourier transform is adopted to obtain the feature data such as the frequency response of the sound vibration signal.
Step 3: model identification
And identifying the model of the transformer in the normal working state according to the obtained characteristic data (frequency domain characteristics) of the input and output sound vibration signals of the transformer in the normal state.
Step 4: state observer design
According to the method, additive and non-additive fault models are distinguished according to the characteristics of the transformer sound vibration monitoring model, and further the online monitoring of the state of the transformer is realized respectively based on two technical schemes of MESO and a non-linear adaptive observer, and the method is realized by the following technical schemes from Step 4-1 to Step 4-5.
Step 4-1: additive and non-additive model differentiation
Judging whether the transformer sound vibration monitoring model is an additive fault model or not by analyzing the relevant characteristics of the model fault disturbance signal and the input signal, and if the model is the additive fault model, carrying out transformer fault monitoring and early warning through the following steps 4-2-4-3; if the fault model is a non-additive fault model, monitoring and early warning of the transformer fault are carried out through the following steps 4-5.
According to the technical scheme, the fault diagnosis technology based on the MESO design comprises the following steps: the ESO treats the uncertainty inside the system and the external disturbances as "total disturbances" of the transformer model system and estimates in real time through the state of the expansions. Model information is considered to be added in the MESO design, so that the total disturbance estimated by the MESO can be ensured to be external disturbance on the premise of reliable transformer model, and the fault is the fault of the transformer model system in fault diagnosis.
The method provided by the embodiment of the invention comprises the following specific steps:
step 4-2: taking a typical state space form as an example, a system model considering fault disturbance f and noise d (which is assumed to be bounded and differentiable) is given as equation (12), where f is taken as the system expansion state:
in the formula, xe＝[x f]TIs the system state, u is the input signal, y is the measurement output signal, Ae、Be、Ce、EeAnd Q is the system matrix.
Step 4-3: MESO is designed to perform state observation for the system of equation (12). MESO can be expressed as:
in the formula, LoFor the observer gain matrix, z ═ zx zf]TThe estimated system state and disturbance caused by the fault. Only the proper observer gain L needs to be selectedoNamely, z can be realizedfAnd tracking the fault signal f.
In the second technical scheme, a fault diagnosis technology based on nonlinear adaptive observer design, for example, in the fault diagnosis scheme based on the MESO design shown in the first technical scheme, all transformer fault disturbances are regarded as additive faults when the observer is designed, and results cannot be visually reflected in the non-additive fault MESO design, so that a nonlinear adaptive observer is designed in the second technical scheme to monitor the non-additive faults of the transformer. The design idea of the nonlinear adaptive observer is explained in detail by taking a second-order object as an example in the invention.
Step 4-4: by introducing a fault indication factor theta in the modelf(when taking θ)fWhen 0 indicates failureBarrier, thetafWhen 1 indicates a failure), the second-order object model is rewritten into the form of equation (14):
step 4-5: designing an adaptive non-linear observer as shown in equation (15):
in the formula I1、l2And l3Is the Luenberger observer gain, lim (z)f) Is the clipping function shown in equation (16).
I.e. by adjusting the observer gain l1、l2And l3Achieving z by obtaining desired tracking characteristicsfAnd tracking the fault signal f.
Step 5: determine whether early warning is needed
By evaluating the fault condition zfThe system state is monitored, and faults are early warned. I.e. the estimated value z according to ffJudging whether a fault occurs, namely:
in the formula, JthIs the threshold value of the judgment.
The specific embodiment of the invention is based on the state monitoring strategy of MESO and the state monitoring strategy of nonlinear adaptive observer, and the points to be protected are as follows:
(1) a state monitoring policy based on MESO. Aiming at the additive fault analysis model of the transformer, the extended state observer realizes real-time estimation of internal disturbance and external disturbance of the transformer system through the extended state, and after the transformer object model is added in the design of the state observer, the extended state realizes estimation of the external disturbance of the transformer, namely the estimation of the fault state of the transformer, so that the real-time monitoring of the fault state of the transformer is realized, and the fault early warning is realized by the transformer system when the output of the estimated state of the transformer is greater than a threshold value.
(2) And (3) a state monitoring strategy based on a nonlinear adaptive observer. Aiming at a non-additive fault analysis model of the transformer, the fault diagnosis scheme based on the MESO design considers all the fault disturbances of the transformer as additive faults when the observer is designed, and the MESO design of the non-additive faults can not visually embody the result, so that a non-linear adaptive observer is invented on the basis of the model-assisted MESO design to monitor and early warn the non-additive faults of the transformer. And monitoring the fault state of the transformer by the extended state of the nonlinear adaptive observer.
The following are embodiments of the apparatus of the present invention that may be used to perform embodiments of the method of the present invention. For details of non-careless mistakes in the embodiment of the apparatus, please refer to the embodiment of the method of the present invention.
The embodiment of the invention provides a transformer sound vibration fault monitoring system, which comprises:
the signal acquisition module is used for acquiring input and output signals of the transformer to be monitored in a normal working state;
the characteristic acquisition module is used for extracting the characteristics of the input signal and the output signal to acquire the characteristics of the input signal and the characteristics of the output signal;
the model identification module is used for identifying a model of the transformer to be monitored in a normal working state based on the input signal characteristics and the output signal characteristics;
the disturbance observation value acquisition module is used for observing by adopting an observer obtained by pre-design based on the identified model to obtain a disturbance observation value;
and the monitoring module is used for realizing the sound vibration fault monitoring of the transformer based on the disturbance observation value.
In yet another embodiment of the present invention, a computer device is provided that includes a processor and a memory for storing a computer program comprising program instructions, the processor for executing the program instructions stored by the computer storage medium. The Processor may be a Central Processing Unit (CPU), or may be other general-purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, etc., which is a computing core and a control core of the terminal, and is specifically adapted to load and execute one or more instructions in a computer storage medium to implement a corresponding method flow or a corresponding function; the processor provided by the embodiment of the invention can be used for the operation of the transformer sound vibration fault monitoring method.
In yet another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory), which is a Memory device in a computer device and is used for storing programs and data. It is understood that the computer readable storage medium herein can include both built-in storage media in the computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space storing an operating system of the terminal. Also, one or more instructions, which may be one or more computer programs (including program code), are stored in the memory space and are adapted to be loaded and executed by the processor. It should be noted that the computer-readable storage medium may be a high-speed RAM memory, or may be a non-volatile memory (non-volatile memory), such as at least one disk memory. One or more instructions stored in the computer-readable storage medium may be loaded and executed by a processor to implement the corresponding steps of the transformer sound vibration fault monitoring method in the above embodiments.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (10)
1. A transformer sound vibration fault monitoring method is characterized by comprising the following steps:
acquiring input and output signals of a transformer to be monitored in a normal working state;
carrying out feature extraction on the input and output signals to obtain input signal features and output signal features;
identifying a model of the transformer to be monitored in a normal working state based on the input signal characteristic and the output signal characteristic;
observing by adopting an observer obtained by pre-design based on the identified model to obtain a disturbance observation value;
and monitoring the sound vibration fault of the transformer based on the disturbance observation value.
2. The method for monitoring the sound vibration fault of the transformer according to claim 1, wherein the step of acquiring the input and output signals of the transformer to be monitored in the normal working state specifically comprises the following steps:
acquiring original input and output signals of a transformer to be monitored in a normal working state;
preprocessing the original input and output signals to obtain preprocessed input and output signals; and preprocessing is used for removing noise signals and obtaining the structural data with prominent signals of the transformer body.
3. The method for monitoring the sound vibration fault of the transformer according to claim 1, wherein the feature extraction is performed on the input and output signals to obtain the input signal features and the output signal features, and the feature extraction specifically comprises: and acquiring the frequency response of the sound vibration signal by a characteristic extraction method.
4. The method for monitoring the sound vibration fault of the transformer according to claim 1, wherein the observation is performed by using an observer obtained by pre-design based on the identified model, and the obtaining of the disturbance observation value specifically comprises:
when the identified model is an additive fault model, expressed as,
in the formula, xe＝[x f]TExpanding states for the transformer system, including a transformer system state x and a fault disturbance f, assuming f is bounded and differentiable; d is noise, u is input signal, y is measurement output signal, Ae、Be、Ce、EeR and Q are all transformer system matrixes;
the model-aided extended state observer model obtained by pre-design is expressed as,
in the formula, LoFor the observer gain matrix, z ═ zx zf]TIs an estimate of the disturbance caused by system conditions and faults;
wherein by selecting the observer gain LoRealization of zfAnd tracking the fault signal f.
5. The method for monitoring the sound vibration fault of the transformer according to claim 1, wherein the observation is performed by using an observer obtained by pre-design based on the identified model, and the obtaining of the disturbance observation value specifically comprises:
when the identified model is a non-additive fault model, it is expressed as,
where u is the input signal, y is the measurement output signal, x1And x2For the transformer system state, G (x)1,x2)，F(x1,x2U) and b are the transformer system matrix, θfAs a fault indicator factor, thetafWhen 0 indicates no failure, θfWhen the value is 1, the fault occurs;
the adaptive observer obtained by the pre-design, denoted,
in the formula I1、l2And l3To observer gain, z1，z2Is an estimate of the state of the system, zfFor fault-induced disturbance estimation, G (z)1,z2)，F(z1,z2,u) And b is an observer system matrix;
lim(zf) Is a clipping function expressed as
Wherein, by adjusting the observer gain l1、l2And l3Obtaining desired tracking characteristics, implementing zfAnd tracking the fault signal f.
6. The method for monitoring the sound vibration fault of the transformer according to claim 4 or 5, wherein the expression for realizing the sound vibration fault monitoring of the transformer based on the disturbance observation value is as follows,
7. A transformer sound vibration fault monitoring system is characterized by comprising:
the signal acquisition module is used for acquiring input and output signals of the transformer to be monitored in a normal working state;
the characteristic acquisition module is used for extracting the characteristics of the input signal and the output signal to acquire the characteristics of the input signal and the characteristics of the output signal;
the model identification module is used for identifying a model of the transformer to be monitored in a normal working state based on the input signal characteristics and the output signal characteristics;
the disturbance observation value acquisition module is used for observing by adopting an observer obtained by pre-design based on the identified model to obtain a disturbance observation value;
and the monitoring module is used for realizing the sound vibration fault monitoring of the transformer based on the disturbance observation value.
8. The system for monitoring the sound vibration fault of the transformer according to claim 7, wherein in the disturbance observation value obtaining module,
when the identified model is an additive fault model, expressed as,
in the formula, xe＝[x f]TExpanding states for the transformer system, including a transformer system state x and a fault disturbance f, assuming bounded and differentiable; d is noise, u is input signal, y is measurement output signal, Ae、Be、Ce、EeR and Q are all transformer system matrixes;
the model-aided extended state observer model obtained by pre-design is expressed as,
in the formula, LoFor the observer gain matrix, z ═ zx zf]TIs an estimate of the disturbance caused by system conditions and faults;
wherein by selecting the observer gain LoRealization of zfTracking the fault signal f;
when the identified model is a non-additive fault model, it is expressed as,
where u is the input signal, y is the measurement output signal, x1And x2 is the transformer system state, G (x)1,x2)，F(x1,x2U) and b are the transformer system matrix, θfAs a fault indicator factor, thetafWhen 0 indicates no failure, θfWhen the value is 1, the fault occurs;
the adaptive observer obtained by the pre-design, denoted,
in the formula I1、l2And l3To observer gain, z1，z2Is an estimate of the state of the system, zfFor fault-induced disturbance estimation, G (z)1,z2)，F(z1,z2,u) And b is an observer system matrix;
lim(zf) Is a clipping function expressed as
Wherein, by adjusting the observer gain l1、l2And l3Obtaining desired tracking characteristics, implementing zfAnd tracking the fault signal f.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor,
the processor, when executing the computer program, implements the steps of the transformer sound vibration fault monitoring method according to any one of claims 1 to 6.
10. A computer-readable storage medium storing a computer program, characterized in that,
the computer program, when being executed by a processor, realizes the steps of the transformer sound vibration fault monitoring method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210033985.2A CN114252810B (en) | 2022-01-12 | 2022-01-12 | Transformer sound vibration fault monitoring method, system, equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210033985.2A CN114252810B (en) | 2022-01-12 | 2022-01-12 | Transformer sound vibration fault monitoring method, system, equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114252810A true CN114252810A (en) | 2022-03-29 |
| CN114252810B CN114252810B (en) | 2024-04-30 |
Family
ID=80796391
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210033985.2A Active CN114252810B (en) | 2022-01-12 | 2022-01-12 | Transformer sound vibration fault monitoring method, system, equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114252810B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110500339A (en) * | 2018-05-18 | 2019-11-26 | 南京理工大学 | A kind of electrohydraulic servo system additivity fault detection method |
| CN111308390A (en) * | 2020-02-21 | 2020-06-19 | 武汉科技大学 | Robust fault detection method for modular multilevel converter based on sliding-mode observer |
| CN112199888A (en) * | 2020-09-30 | 2021-01-08 | 苏州容思恒辉智能科技有限公司 | Rotary equipment fault diagnosis method and system based on deep residual error network and readable storage medium |
| CN112733320A (en) * | 2020-12-09 | 2021-04-30 | 淮阴工学院 | Boost converter actuator fault detection method based on delta operator |
| CN113656938A (en) * | 2021-07-07 | 2021-11-16 | 北京智芯微电子科技有限公司 | Monitoring method, control device and storage medium for distribution network transformer |
-
2022
- 2022-01-12 CN CN202210033985.2A patent/CN114252810B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110500339A (en) * | 2018-05-18 | 2019-11-26 | 南京理工大学 | A kind of electrohydraulic servo system additivity fault detection method |
| CN111308390A (en) * | 2020-02-21 | 2020-06-19 | 武汉科技大学 | Robust fault detection method for modular multilevel converter based on sliding-mode observer |
| CN112199888A (en) * | 2020-09-30 | 2021-01-08 | 苏州容思恒辉智能科技有限公司 | Rotary equipment fault diagnosis method and system based on deep residual error network and readable storage medium |
| CN112733320A (en) * | 2020-12-09 | 2021-04-30 | 淮阴工学院 | Boost converter actuator fault detection method based on delta operator |
| CN113656938A (en) * | 2021-07-07 | 2021-11-16 | 北京智芯微电子科技有限公司 | Monitoring method, control device and storage medium for distribution network transformer |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114252810B (en) | 2024-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109783903B (en) | Industrial water pipeline fault diagnosis method and system based on time sequence | |
| He et al. | A quantitative estimation technique for welding quality using local mean decomposition and support vector machine | |
| CN109469896B (en) | Industrial boiler fault diagnosis method and system based on time series analysis | |
| CN110222765B (en) | Method and system for monitoring health state of permanent magnet synchronous motor | |
| Yu et al. | Nonlinear control performance assessment in the presence of valve stiction | |
| CN116625683A (en) | Wind turbine generator system bearing fault identification method, system and device and electronic equipment | |
| CN113237619B (en) | Fault early warning method, device, equipment and storage medium for variable-speed rotating machinery vibration | |
| CN109142979B (en) | Method and device for detecting abnormal state of power distribution network | |
| Wang et al. | Degradation prediction method by use of autoregressive algorithm | |
| CN114252810B (en) | Transformer sound vibration fault monitoring method, system, equipment and readable storage medium | |
| CN105652795B (en) | 3PTT-2R series-parallel numerical control machine tool servo system fault prediction device and method based on residual observer | |
| CN115269241A (en) | Method, device and storage medium for carrying out anomaly detection on periodic data | |
| CN112525337B (en) | Pretreatment method for vibration monitoring data of mechanical press | |
| CN113516023A (en) | Equipment vibration abnormality diagnosis method and system | |
| CN109357751B (en) | Power transformer winding looseness defect detection system | |
| CN106649204A (en) | Method for identifying exception of dissolved gas in transformer oil | |
| CN112100577A (en) | Long-range correlation-based equipment operation stability online monitoring method and system | |
| CN112729531A (en) | Distribution transformer equipment fault studying and judging method and system | |
| CN111590390B (en) | Cutter wear state real-time assessment method and system, storage medium and terminal | |
| CN111985380B (en) | Bearing degradation process state monitoring method, system, equipment and storage medium | |
| CN113361823B (en) | Fuel cell fault prediction method and system based on prediction data | |
| CN112597425B (en) | Method, device, equipment and medium for detecting abnormal data in flow velocity measurement data | |
| CN113641584B (en) | Monitoring analysis method and device for test case execution | |
| CN117572162A (en) | Partial discharge detection method and system | |
| CN115270888A (en) | Motor fault analysis method, system, terminal equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |