CN112014678A - Three-phase voltage inverter online fault diagnosis method and device and electronic equipment - Google Patents
Three-phase voltage inverter online fault diagnosis method and device and electronic equipment Download PDFInfo
- Publication number
- CN112014678A CN112014678A CN202011121915.XA CN202011121915A CN112014678A CN 112014678 A CN112014678 A CN 112014678A CN 202011121915 A CN202011121915 A CN 202011121915A CN 112014678 A CN112014678 A CN 112014678A
- Authority
- CN
- China
- Prior art keywords
- fault
- training
- data
- input
- model
- 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.)
- Pending
Links
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
-
- 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
- G01R31/26—Testing of individual semiconductor devices
- G01R31/2601—Apparatus or methods therefor
-
- 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
- G01R31/26—Testing of individual semiconductor devices
- G01R31/27—Testing of devices without physical removal from the circuit of which they form part, e.g. compensating for effects surrounding elements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
One or more embodiments of the present disclosure provide an online fault diagnosis method and apparatus for a three-phase voltage inverter, and an electronic device, which can perform accurate diagnosis on a fault type of the inverter in real time and efficiently. The method comprises the following steps: constructing and setting different circuit conditions of the inverter training circuit in different fault types, and acquiring a bus voltage value and a load three-phase current value to determine model training data; training and optimizing the initial neural network model by using model training data to obtain a fault diagnosis model; and determining the fault type of the inverter to be diagnosed by using the fault diagnosis model. The device comprises a training data acquisition module, a model training module and a fault diagnosis module. The electronic device comprises a memory, a processor and a computer program stored on the memory and executable on the processor to implement the fault diagnosis method.
Description
Technical Field
One or more embodiments of the present disclosure relate to the technical field of circuit diagnosis and analysis, and in particular, to an online fault diagnosis method and apparatus for a three-phase voltage inverter, and an electronic device.
Background
The converter system bears core functions of energy conversion, transmission and the like in a locomotive, is an essential key device of the locomotive as an electric energy control core of a traction transmission system of the locomotive, relatively starts to research in the diagnosis field of the locomotive late, and compared with the traditional industrial converter system, the power density required to be responded by the traction converter system applied to the railway transportation field is higher, and the more severe working environment is also borne. One of the major failure causes in a locomotive ac system failure is a three-phase voltage inverter failure.
The existing method for diagnosing the fault of the three-phase voltage inverter mainly comprises three types: the fault diagnosis method comprises a data processing method based on signal sampling (voltage signals and current signals), a fault diagnosis method based on an artificial intelligence algorithm and a fault diagnosis method based on an analytic model. In the fault diagnosis method based on the collected voltage signals, the collected voltage signals are used for constructing fault characteristics by extracting phase voltage signals, and system fault conditions are directly or indirectly reflected by processing the fault characteristics, so that the method is poor in applicability under some complex fault conditions; in the data processing method based on the sampled current signal, a load side phase current signal is generally collected to analyze the signal distortion degree so as to realize fault diagnosis, and the method usually needs to perform more complex operations such as feature extraction, reconstruction or analysis transformation and the like on the basis of collecting the obtained current signal; the fault diagnosis method based on the artificial intelligence algorithm usually needs to extract the characteristics of the voltage and the current value to obtain fault characteristic quantity as a fault diagnosis basis, and the fault diagnosis basis in the method can have the situation of information omission; the fault diagnosis method based on the analytical model mainly realizes the fault diagnosis of the system by modeling the system, analyzing various possible states of each model and comparing the possible states with the actual working state, and the method has disadvantages in the aspects of obtaining system control signals and algorithm complexity.
Disclosure of Invention
In view of this, an object of one or more embodiments of the present disclosure is to provide an online fault diagnosis method for a three-phase voltage inverter, which can diagnose an inverter fault in real time and efficiently, and has high applicability and accurate diagnosis result.
In view of the above, one or more embodiments of the present specification provide an online fault diagnosis method for a three-phase voltage inverter, including:
constructing an inverter training circuit, setting the fault state of a power tube in the inverter training circuit to enable the inverter training circuit to be respectively in different circuit conditions of different fault types, acquiring a bus voltage value and a load three-phase current value under different circuit conditions as training input data, recording the fault type corresponding to the circuit condition as training output data, and forming model training data by the training input data and the training output data;
constructing an initial neural network model, training the initial neural network model by using the model training data, and performing optimization adjustment by adjusting network parameters of the initial neural network model to enable input and output of the initial neural network model to correspond to each other to obtain a fault diagnosis model;
and collecting the bus voltage and the load three-phase current of the inverter to be diagnosed on line as input data, processing the input data by using the fault diagnosis model, and determining the fault type of the inverter to be diagnosed according to the output data of the fault diagnosis model.
Optionally, the fault types include:
the single power tube fault type correspondingly comprises the circuit conditions: a power tube T1 fault, a power tube T2 fault, a power tube T3 fault, a power tube T4 fault, a power tube T5 fault and a power tube T6 fault;
the single-bridge-arm double-tube fault type correspondingly comprises the circuit conditions: power tube T1& T2 failure, power tube T3& T4 failure, and power tube T5& T6 failure;
the type of double-tube fault on the same side of the special bridge arm correspondingly comprises the circuit conditions: a power tube T1& T3 fault, a power tube T1& T5 fault, a power tube T3& T5 fault, a power tube T2& T4 fault, a power tube T2& T6 fault and a power tube T4& T6 fault;
and the type of the fault of the double tubes on the different sides of the different bridge arm correspondingly comprises the circuit conditions: power tube T1& T4 fault, power tube T1& T6 fault, power tube T3& T2 fault, power tube T3& T6 fault, power tube T5& T2 fault and power tube T5& T4 fault.
Optionally, the acquiring the bus voltage value and the load three-phase current value under different circuit conditions as training input data includes:
under different circuit conditions, respectively randomly setting a plurality of bus voltage values
And collecting the load three-phase current value corresponding to the bus voltage value
Value of said bus voltage
Corresponding three-phase current value of the load
Constructing an input vector
；
The training input data comprises a plurality of the input vectors.
Optionally, before the initial neural network model is trained by using the model training data, normalization processing is further performed on the training input data:
wherein, therein
Representing the input vector
To (1)
Dimension data;
the normalized sub-data forms a normalized input vector
。
Optionally, the constructing the initial neural network model includes:
the initial neural network model adopts a back propagation neural network structure and comprises an input layer, a hidden layer and an output layer;
the input network nodes of the input layer correspond to the bus voltage values and the load three-phase current values in the training input data respectively;
the output network nodes of the output layer have the same number with the fault types and respectively correspond to the different fault types.
Optionally, the activation function of the hidden layer is:
wherein the content of the first and second substances,
representing a first input information parameter;
the first input information parameter
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
represents from the first
Each of the input network nodes points to
The input of an implicit network node passes the weights,
representing the number of said input network nodes,
indicating the number of said implicit network nodes,
is shown as
Input values corresponding to the input network nodes;
the output values of the output network nodes are:
wherein the content of the first and second substances,
is shown as
An output value corresponding to each of said output network nodes,
representing the number of said output network nodes,
a parameter representing a second input information is provided,
is shown as
A threshold value for each of the output network nodes;
the second input information parameter
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
represents from the first
Each of the implicit network nodes points to
The output of each of said output network nodes passes a weight,
is shown as
A delivery output value of each of the implicit network nodes;
first, the
The transfer output value calculation formula of each of the implicit network nodes is:
wherein the content of the first and second substances,
is shown as
A threshold value for each of the implicit network nodes.
Optionally, the performing optimization adjustment by adjusting the network parameters of the initial neural network model to make the input and the output of the initial neural network model correspond to each other to obtain a fault diagnosis model includes:
dividing the model training data into training data and a verification data set;
training the initial neural network model using the training input data and the training output data in the training data set;
verifying the trained initial neural network model by using the training input data and the training output data in the verification data set:
adjusting and setting the number of the network layers and the number of the network nodes of the hidden layer;
and adjusting the transmission weight of the initial neural network model and a network node threshold value to ensure that the average error between the output data of the initial neural network model and the training output data is less than a target error threshold value.
In view of the above objects, one or more embodiments of the present specification provide an online fault diagnosis apparatus for a three-phase voltage inverter, including:
the training data acquisition module is configured to construct an inverter training circuit, set the fault state of a power tube in the inverter training circuit to enable the inverter training circuit to be respectively in different circuit conditions of different fault types, acquire a bus voltage value and a load three-phase current value under different circuit conditions as training input data, record the fault type corresponding to the circuit condition as training output data, and form model training data by the training input data and the training output data;
the model training module is configured to construct an initial neural network model, train the initial neural network model by using the model training data, and perform optimization adjustment by adjusting network parameters of the initial neural network model to enable input and output of the initial neural network model to correspond to each other to obtain a fault diagnosis model;
and the fault diagnosis module is configured to acquire the bus voltage and the load three-phase current of the inverter to be diagnosed on line as input data, process the input data by using the fault diagnosis model, and determine the fault type of the inverter to be diagnosed according to the output data of the fault diagnosis model.
In view of the above, one or more embodiments of the present specification provide an electronic device for online fault diagnosis of a three-phase voltage inverter, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the online fault diagnosis method of the three-phase voltage inverter when executing the program.
As can be seen from the above description, the three-phase voltage inverter online fault diagnosis method, apparatus and electronic device provided in one or more embodiments of the present disclosure, through analyzing the three-phase voltage inverter circuit, a plurality of different fault circuit conditions are divided according to specific fault conditions, and the directly acquired load three-phase current value is utilized to carry out optimization training on the neural network model aiming at different fault circuit conditions, the neural network model adopts a Back Propagation (BP) neural network structure, the fault diagnosis model obtained after optimization training can quickly and accurately identify the fault condition of the corresponding inverter circuit only according to the three-phase current loaded in the inverter circuit, in practical application, the data acquisition can be realized by using the current acquisition device in the inverter, and the fault diagnosis can be carried out by acquiring data on line under the condition of not changing the running state of the inverter.
Drawings
In order to more clearly illustrate one or more embodiments or prior art solutions of the present specification, the drawings that are needed in the description of the embodiments or prior art will be briefly described below, and it is obvious that the drawings in the following description are only one or more embodiments of the present specification, and that other drawings may be obtained by those skilled in the art without inventive effort from these drawings.
Fig. 1 is a schematic diagram of an online fault diagnosis method for a three-phase voltage inverter according to one or more embodiments of the present disclosure;
FIG. 2 is a schematic diagram of a three-phase voltage inverter circuit topology;
fig. 3 is a schematic structural diagram of a BP neural network in an online fault diagnosis method for a three-phase voltage inverter according to one or more embodiments of the present disclosure;
fig. 4 is a schematic diagram illustrating a neural network optimization adjustment method in an online fault diagnosis method for a three-phase voltage inverter according to one or more embodiments of the present disclosure;
fig. 5 is a schematic diagram of an online fault diagnosis device for a three-phase voltage inverter according to one or more embodiments of the present disclosure;
fig. 6 is a schematic diagram of an electronic device for online fault diagnosis of a three-wire voltage inverter according to one or more embodiments of the present disclosure.
Detailed Description
For the purpose of promoting a better understanding of the objects, aspects and advantages of the present disclosure, reference is made to the following detailed description taken in conjunction with the accompanying drawings.
It is to be noted that unless otherwise defined, technical or scientific terms used in one or more embodiments of the present specification should have the ordinary meaning as understood by those of ordinary skill in the art to which this disclosure belongs. The use of "first," "second," and similar terms in one or more embodiments of the specification is not intended to indicate any order, quantity, or importance, but rather is used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.
In one aspect, one or more embodiments of the present specification provide an online fault diagnosis method for a three-phase voltage inverter.
As shown in fig. 1, the online fault diagnosis method for a three-phase voltage inverter provided by one or more embodiments of the present disclosure includes:
s1: constructing an inverter training circuit, setting the fault state of a power tube in the inverter training circuit to enable the inverter training circuit to be respectively in different circuit conditions of different fault types, acquiring a bus voltage value and a load three-phase current value under different circuit conditions as training input data, recording the fault type corresponding to the circuit condition as training output data, and forming model training data by the training input data and the training output data;
s2: constructing an initial neural network model, training the initial neural network model by using the model training data, and performing optimization adjustment by adjusting network parameters of the initial neural network model to enable input and output of the initial neural network model to correspond to each other to obtain a fault diagnosis model;
s3: and collecting the bus voltage and the load three-phase current of the inverter to be diagnosed on line as input data, processing the input data by using the fault diagnosis model, and determining the fault type of the inverter to be diagnosed according to the output data of the fault diagnosis model.
According to the three-phase voltage inverter online fault diagnosis method, the three-phase voltage inverter circuit is analyzed, various different fault circuit conditions are divided according to specific fault conditions, the neural network model is optimized and trained by directly acquired load three-phase current values according to different fault circuit conditions, the neural network model adopts a Back Propagation (BP) neural network structure, the fault diagnosis model obtained after optimization training can quickly and accurately identify the fault conditions of the corresponding inverter circuit only according to the load three-phase current in the inverter circuit, data acquisition can be realized by using a current acquisition device carried in the inverter in practical application, and the data can be acquired online for fault diagnosis under the condition that the operation state of the inverter is not changed.
As shown in fig. 2, it is a topology structure diagram of the three-phase voltage inverter circuit. The three-phase voltage inverter adopts a sine wave pulse width modulation control method, as shown in figure 1The inverter circuit is composed of a direct-current bus voltage Udc, filtering and voltage-stabilizing capacitors C1 and C2, six power devices T1-T6, six freewheeling diodes D1-D6 connected in parallel to the semiconductor power devices, load resistors R1-R3 and load inductors L1-L3. The three-phase currents of the loads a, b and c are respectively
The on-off of six semiconductor power tubes is controlled through a sine wave pulse width modulation (SPWM) model, and different voltage values are supplied to a load end.
In the three-phase voltage inverter online fault diagnosis method provided in one or more embodiments of the present specification, it can be determined by analyzing the three-phase voltage inverter circuit, and the types of faults that may occur in the inverter include:
the single power tube fault type correspondingly comprises the circuit conditions: a power tube T1 fault, a power tube T2 fault, a power tube T3 fault, a power tube T4 fault, a power tube T5 fault and a power tube T6 fault;
the single-bridge-arm double-tube fault type correspondingly comprises the circuit conditions: power tube T1& T2 failure, power tube T3& T4 failure, and power tube T5& T6 failure;
the type of double-tube fault on the same side of the special bridge arm correspondingly comprises the circuit conditions: a power tube T1& T3 fault, a power tube T1& T5 fault, a power tube T3& T5 fault, a power tube T2& T4 fault, a power tube T2& T6 fault and a power tube T4& T6 fault;
and the type of the fault of the double tubes on the different sides of the different bridge arm correspondingly comprises the circuit conditions: power tube T1& T4 fault, power tube T1& T6 fault, power tube T3& T2 fault, power tube T3& T6 fault, power tube T5& T2 fault and power tube T5& T4 fault.
In the method for diagnosing the online fault of the three-phase voltage inverter provided in one or more embodiments of the present specification, the acquiring the bus voltage value and the load three-phase current value under different circuit conditions as training input data includes:
in different cases of the circuit, a plurality of circuits are respectively and randomly arrangedThe bus voltage value
And collecting the load three-phase current value corresponding to the bus voltage value
Value of said bus voltage
Corresponding three-phase current value of the load
Constructing an input vector
；
The training input data comprises a plurality of the input vectors.
Under the condition of different circuit faults, the load three-phase current values of the inverter circuit and the change conditions of the current values are different, and under the condition of the same circuit fault, the conditions of the corresponding output load three-phase current values are different when the bus voltage values are different, and the online fault diagnosis method of the three-phase voltage inverter is used for carrying out online fault diagnosis on the bus voltage values
Corresponding three-phase current value of the load
The three-phase voltage inverter online fault diagnosis method comprises the steps that an input vector is formed by a group of training data, a plurality of input vectors corresponding to a plurality of different bus voltages under different circuit conditions form the training input data, circuit fault characteristic information under different fault points, different load conditions and different control method conditions can be comprehensively covered, the input vector capable of comprehensively covering various circuit fault characteristic information is used as training data of an initial neural network model, and the input vector is used as the training data of the initial neural network modelThe initial neural network model can accurately extract and learn characteristic information of various circuit fault conditions, so that the fault diagnosis model obtained after training optimization can accurately diagnose faults of the inverter to be diagnosed.
In the three-phase voltage inverter online fault diagnosis method provided in one or more embodiments of the present specification, before the initial neural network model is trained using the model training data, the training input data is further normalized:
wherein, therein
Representing the input vector
To (1)
Dimension data;
the normalized sub-numberFrom the composition of a normalized input vector
。
As shown in fig. 3, in the three-phase voltage inverter online fault diagnosis method provided in one or more embodiments of the present specification, the constructing an initial neural network model includes:
the initial neural network model adopts a back propagation neural network structure, which includes an input layer, a hidden layer and an output layer, and fig. 3 is a schematic diagram of a Back Propagation (BP) neural network structure.
The input network nodes of the input layer correspond to the bus voltage values and the load three-phase current values in the training input data respectively;
the output network nodes of the output layer have the same number with the fault types and respectively correspond to the different fault types.
The functional characteristics of the back propagation BP neural network are mainly reflected in: model function approximation, training a network to approximate a function by using an input vector and a corresponding output vector; pattern recognition and classification, which is associated with the input vector by a particular output vector, classifies the input vector in a defined suitable manner. In the three-phase voltage inverter online fault diagnosis method, the initial neural network model adopts a BP neural network structure, the input vector corresponds to the input vector, and the bus voltage value
Corresponding three-phase current value of the load
Forming an input vector as a set of training data, the corresponding input network nodes being set to four, respectively
、
Correspondingly, the output vectors of the inverter are corresponding to the determined fault types, the fault types which can possibly occur to the inverter include four fault types, and the number of corresponding output network nodes is also four, and the four corresponding output network nodes respectively correspond to four types, namely a single-power-tube fault type, a single-bridge-arm double-tube fault type, an abnormal-bridge-arm same-side double-tube fault type and an abnormal-bridge-arm abnormal-side double-tube fault type. In some optional embodiments, the output vector of the BP neural network is represented in a four-bit encoding form, where a value of the four-bit encoding is 0 or 1, and when the value is 1, it indicates that the output fault type is the type corresponding to the corresponding encoding bit, for example, when the BP neural network is provided with four output network nodes, output values of the four output network nodes form a four-bit encoding bit 0100, which indicates that the determined fault type is a single-bridge-arm double-pipe fault type corresponding to the second output network node.
In some alternative embodiments, the output network nodes of the BP neural network model may be set to correspond to different fault circuit conditions, and then the output network nodes are set to 21, which correspond to the above 21 different circuit conditions, so that the BP neural network model can be used to directly determine which specific power tube has a fault, besides diagnosing and determining the fault type of the inverter. And the output vector of the BP neural network node is expressed by a 21-bit coding form, and the same coding values are all 0 or 1.
In the three-phase voltage inverter online fault diagnosis method provided in one or more embodiments of the present specification, the activation function of the hidden layer is:
wherein the content of the first and second substances,
representing a first input information parameter;
the above-mentionedFirst input information parameter
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
represents from the first
Each of the input network nodes points to
The input of an implicit network node passes the weights,
representing the number of said input network nodes, which in some alternative embodiments correspond to said input vector dimension by a value of 4,
indicating the number of said implicit network nodes,
is shown as
Input values corresponding to the input network nodes;
the output values of the output network nodes are:
wherein the content of the first and second substances,
is shown as
An output value corresponding to each of said output network nodes,
representing the number of said output network nodes,
a parameter representing a second input information is provided,
is shown as
A threshold value for each of the output network nodes; number of said output network nodes in some alternative embodiments
May be set to 4;
the second input information parameter
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
represents from the first
Each of the implicit network nodes points to
A said output network nodeThe output of the point is passed the weight,
is shown as
A delivery output value of each of the implicit network nodes;
first, the
The transfer output value calculation formula of each of the implicit network nodes is:
wherein the content of the first and second substances,
is shown as
A threshold value for each of the implicit network nodes.
As shown in fig. 4, in the method for diagnosing an online fault of a three-phase voltage inverter according to one or more embodiments of the present disclosure, the performing optimization adjustment by adjusting network parameters of the initial neural network model to make input and output of the initial neural network model correspond to each other to obtain a fault diagnosis model, including:
s401: dividing the model training data into training data and a verification data set;
s402: training the initial neural network model using the training input data and the training output data in the training data set;
s403: verifying the trained initial neural network model by using the training input data and the training output data in the verification data set:
adjusting and setting the number of the network layers and the number of the network nodes of the hidden layer;
and adjusting the transmission weight of the initial neural network model and a network node threshold value to ensure that the average error between the output data of the initial neural network model and the training output data is less than a target error threshold value.
In another aspect, one or more embodiments of the present disclosure provide an online fault diagnosis apparatus for a three-phase voltage inverter.
As shown in fig. 5, one or more embodiments of the present disclosure provide an online fault diagnosis device for a three-phase voltage inverter, including:
a training data acquisition module 501 configured to construct an inverter training circuit, set a fault state of a power tube in the inverter training circuit so that the inverter training circuit is respectively in different circuit conditions of different fault types, acquire a bus voltage value and a load three-phase current value under different circuit conditions as training input data, record the fault type corresponding to the circuit condition as training output data, and form model training data with the training input data and the training output data;
a model training module 502 configured to construct an initial neural network model, train the initial neural network model using the model training data, and perform optimization adjustment by adjusting network parameters of the initial neural network model to make input and output of the initial neural network model correspond to each other, so as to obtain a fault diagnosis model;
the fault diagnosis module 503 is configured to acquire the bus voltage and the load three-phase current of the inverter to be diagnosed on line as input data, process the input data by using the fault diagnosis model, and determine the fault type of the inverter to be diagnosed according to the data output by the fault diagnosis model.
For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, the functionality of the modules may be implemented in the same one or more software and/or hardware implementations in implementing one or more embodiments of the present description.
The apparatus of the foregoing embodiment is used to implement the corresponding method in the foregoing embodiment, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
In another aspect, one or more embodiments of the present description provide a three-phase voltage inverter online fault diagnosis electronic device.
The electronic equipment comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, and when the processor executes the program, the online fault diagnosis method of the three-phase voltage inverter is realized.
Fig. 6 is a schematic diagram illustrating a more specific hardware structure of an electronic device according to this embodiment, where the electronic device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein the processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 are communicatively coupled to each other within the device via bus 1050.
The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits, and is configured to execute a relevant program to implement the three-phase voltage inverter online fault diagnosis method provided in the embodiments of the present disclosure.
The Memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, or the like. The memory 1020 may store an operating system and other application programs, and when the three-phase voltage inverter online fault diagnosis method provided by the embodiments of the present disclosure is implemented by software or firmware, relevant program codes are stored in the memory 1020 and called by the processor 1010 to be executed.
The input/output interface 1030 is used for connecting an input/output module to input and output information. The i/o module may be configured as a component in a device (not shown) or may be external to the device to provide a corresponding function. The input devices may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices may include a display, a speaker, a vibrator, an indicator light, etc.
The communication interface 1040 is used for connecting a communication module (not shown in the drawings) to implement communication interaction between the present apparatus and other apparatuses. The communication module can realize communication in a wired mode (such as USB, network cable and the like) and also can realize communication in a wireless mode (such as mobile network, WIFI, Bluetooth and the like).
It should be noted that although the above-mentioned device only shows the processor 1010, the memory 1020, the input/output interface 1030, the communication interface 1040 and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above-mentioned apparatus may also include only the components necessary for implementing the three-phase voltage inverter online fault diagnosis method according to the embodiment of the present disclosure, and does not necessarily include all the components shown in the drawings.
Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the spirit of the present disclosure, features from the above embodiments or from different embodiments may also be combined, steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present description as described above, which are not provided in detail for the sake of brevity.
In addition, well-known power/ground connections to Integrated Circuit (IC) chips and other components may or may not be shown in the provided figures, for simplicity of illustration and discussion, and so as not to obscure one or more embodiments of the disclosure. Furthermore, devices may be shown in block diagram form in order to avoid obscuring the understanding of one or more embodiments of the present description, and this also takes into account the fact that specifics with respect to implementation of such block diagram devices are highly dependent upon the platform within which the one or more embodiments of the present description are to be implemented (i.e., specifics should be well within purview of one skilled in the art). Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the disclosure, it should be apparent to one skilled in the art that one or more embodiments of the disclosure can be practiced without, or with variation of, these specific details. Accordingly, the description is to be regarded as illustrative instead of restrictive.
While the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic ram (dram)) may use the discussed embodiments.
It is intended that the one or more embodiments of the present specification embrace all such alternatives, modifications and variations as fall within the broad scope of the appended claims. Therefore, any omissions, modifications, substitutions, improvements, and the like that may be made without departing from the spirit and principles of one or more embodiments of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (9)
1. An online fault diagnosis method for a three-phase voltage inverter is characterized by comprising the following steps:
constructing an inverter training circuit, setting the fault state of a power tube in the inverter training circuit to enable the inverter training circuit to be respectively in different circuit conditions of different fault types, acquiring a bus voltage value and a load three-phase current value under different circuit conditions as training input data, recording the fault type corresponding to the circuit condition as training output data, and forming model training data by the training input data and the training output data;
constructing an initial neural network model, training the initial neural network model by using the model training data, and performing optimization adjustment by adjusting network parameters of the initial neural network model to enable input and output of the initial neural network model to correspond to each other to obtain a fault diagnosis model;
and collecting the bus voltage and the load three-phase current of the inverter to be diagnosed on line as input data, processing the input data by using the fault diagnosis model, and determining the fault type of the inverter to be diagnosed according to the output data of the fault diagnosis model.
2. The method of claim 1, wherein the fault type comprises:
the single power tube fault type correspondingly comprises the circuit conditions: a power tube T1 fault, a power tube T2 fault, a power tube T3 fault, a power tube T4 fault, a power tube T5 fault and a power tube T6 fault;
the single-bridge-arm double-tube fault type correspondingly comprises the circuit conditions: power tube T1& T2 failure, power tube T3& T4 failure, and power tube T5& T6 failure;
the type of double-tube fault on the same side of the special bridge arm correspondingly comprises the circuit conditions: a power tube T1& T3 fault, a power tube T1& T5 fault, a power tube T3& T5 fault, a power tube T2& T4 fault, a power tube T2& T6 fault and a power tube T4& T6 fault;
and the type of the fault of the double tubes on the different sides of the different bridge arm correspondingly comprises the circuit conditions: power tube T1& T4 fault, power tube T1& T6 fault, power tube T3& T2 fault, power tube T3& T6 fault, power tube T5& T2 fault and power tube T5& T4 fault.
3. The method of claim 1, wherein the collecting of the bus voltage values and the load three-phase current values under different circuit conditions as training input data comprises:
under different circuit conditions, respectively randomly setting a plurality of bus voltage values
And collecting the load three-phase current value corresponding to the bus voltage value
Value of said bus voltage
Corresponding three-phase current value of the load
Constructing an input vector
；
The training input data comprises a plurality of the input vectors.
4. The method of claim 3, wherein prior to training the initial neural network model with the model training data, the training input data is further normalized by:
wherein, therein
Representing the input vector
To (1)
Dimension data;
the normalized sub-data forms a normalized input vector
。
5. The method of claim 1, wherein constructing the initial neural network model comprises:
the initial neural network model adopts a back propagation neural network structure and comprises an input layer, a hidden layer and an output layer;
the input network nodes of the input layer correspond to the bus voltage values and the load three-phase current values in the training input data respectively;
the output network nodes of the output layer have the same number with the fault types and respectively correspond to the different fault types.
6. The method of claim 5, wherein the activation function of the hidden layer is:
wherein the content of the first and second substances,
representing a first input information parameter;
the first input information parameter
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
represents from the first
Each of the input network nodes points to
The input of an implicit network node passes the weights,
representing the number of said input network nodes,
indicating the number of said implicit network nodes,
is shown as
Input values corresponding to the input network nodes;
the output values of the output network nodes are:
wherein the content of the first and second substances,
is shown as
An output value corresponding to each of said output network nodes,
representing the number of said output network nodes,
a parameter representing a second input information is provided,
is shown as
A threshold value for each of the output network nodes;
the second input information parameter
The calculation formula of (2) is as follows:
wherein the content of the first and second substances,
represents from the first
Each of the implicit network nodes points to
The output of each of said output network nodes passes a weight,
is shown as
A delivery output value of each of the implicit network nodes;
first, the
The transfer output value calculation formula of each of the implicit network nodes is:
wherein the content of the first and second substances,
is shown as
A threshold value for each of the implicit network nodes.
7. The method of claim 6, wherein the optimizing the initial neural network model by adjusting the network parameters of the initial neural network model to correspond the initial neural network model inputs and outputs to obtain a fault diagnosis model, comprises:
dividing the model training data into training data and a verification data set;
training the initial neural network model using the training input data and the training output data in the training data set;
verifying the trained initial neural network model by using the training input data and the training output data in the verification data set:
adjusting and setting the number of the network layers and the number of the network nodes of the hidden layer;
and adjusting the transmission weight of the initial neural network model and a network node threshold value to ensure that the average error between the output data of the initial neural network model and the training output data is less than a target error threshold value.
8. An online fault diagnosis device for a three-phase voltage inverter is characterized by comprising:
the training data acquisition module is configured to construct an inverter training circuit, set the fault state of a power tube in the inverter training circuit to enable the inverter training circuit to be respectively in different circuit conditions of different fault types, acquire a bus voltage value and a load three-phase current value under different circuit conditions as training input data, record the fault type corresponding to the circuit condition as training output data, and form model training data by the training input data and the training output data;
the model training module is configured to construct an initial neural network model, train the initial neural network model by using the model training data, and perform optimization adjustment by adjusting network parameters of the initial neural network model to enable input and output of the initial neural network model to correspond to each other to obtain a fault diagnosis model;
and the fault diagnosis module is configured to acquire the bus voltage and the load three-phase current of the inverter to be diagnosed on line as input data, process the input data by using the fault diagnosis model, and determine the fault type of the inverter to be diagnosed according to the output data of the fault diagnosis model.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011121915.XA CN112014678A (en) | 2020-10-20 | 2020-10-20 | Three-phase voltage inverter online fault diagnosis method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011121915.XA CN112014678A (en) | 2020-10-20 | 2020-10-20 | Three-phase voltage inverter online fault diagnosis method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112014678A true CN112014678A (en) | 2020-12-01 |
Family
ID=73527352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011121915.XA Pending CN112014678A (en) | 2020-10-20 | 2020-10-20 | Three-phase voltage inverter online fault diagnosis method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112014678A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113449771A (en) * | 2021-05-20 | 2021-09-28 | 山东大学 | Four-quadrant converter fault diagnosis method and system based on artificial neural network |
| WO2024087182A1 (en) * | 2022-10-28 | 2024-05-02 | 西门子股份公司 | Fault diagnosis method and apparatus for motor |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000014007A (en) * | 1998-06-22 | 2000-01-14 | Kansai Electric Power Co Inc:The | Control method of active filter |
| CN102053211A (en) * | 2010-12-08 | 2011-05-11 | 沈阳工业大学 | Fault diagnosis method of wind power generation system converter device |
| CN103901882A (en) * | 2014-04-15 | 2014-07-02 | 北京交通大学 | Online monitoring fault diagnosis system and method of train power system |
| CN104459388A (en) * | 2014-11-26 | 2015-03-25 | 国家电网公司 | Permanent magnetic direct-drive wind power generation system integrated fault diagnosis method |
-
2020
- 2020-10-20 CN CN202011121915.XA patent/CN112014678A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000014007A (en) * | 1998-06-22 | 2000-01-14 | Kansai Electric Power Co Inc:The | Control method of active filter |
| CN102053211A (en) * | 2010-12-08 | 2011-05-11 | 沈阳工业大学 | Fault diagnosis method of wind power generation system converter device |
| CN103901882A (en) * | 2014-04-15 | 2014-07-02 | 北京交通大学 | Online monitoring fault diagnosis system and method of train power system |
| CN104459388A (en) * | 2014-11-26 | 2015-03-25 | 国家电网公司 | Permanent magnetic direct-drive wind power generation system integrated fault diagnosis method |
Non-Patent Citations (2)
| Title |
|---|
| 付娟娟: "基于神经网络的有源中点箝位型三电平逆变器故障诊断系统研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
| 韩素敏 等: "基于BP神经网络的逆变器开路故障诊断方法", 《河南理工大学学报（自然科学版）》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113449771A (en) * | 2021-05-20 | 2021-09-28 | 山东大学 | Four-quadrant converter fault diagnosis method and system based on artificial neural network |
| WO2024087182A1 (en) * | 2022-10-28 | 2024-05-02 | 西门子股份公司 | Fault diagnosis method and apparatus for motor |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106443316B (en) | Multi-information detection method and device for deformation state of power transformer winding | |
| CN101907437B (en) | Wavelet difference algorithm-based cable fault localization method | |
| CA2922581C (en) | Method and system for identifying full parameters of element by fault recording and fault locating method | |
| KR102189269B1 (en) | Fault Diagnosis method and system for induction motor using convolutional neural network | |
| CN112014678A (en) | Three-phase voltage inverter online fault diagnosis method and device and electronic equipment | |
| CN112710969A (en) | Open-circuit fault diagnosis method for switching tube of single-phase half-bridge five-level inverter | |
| Manohar et al. | Enhancing resilience of PV-fed microgrid by improved relaying and differentiating between inverter faults and distribution line faults | |
| CN110726898B (en) | Power distribution network fault type identification method | |
| CN113159226B (en) | Inverter fault diagnosis method with integration of depth features and statistical features | |
| CN113358993B (en) | Online fault diagnosis method and system for multi-level converter IGBT | |
| Gargoom et al. | Investigation of effective automatic recognition systems of power-quality events | |
| CN110703151A (en) | Transformer fault diagnosis method based on vibration blind source separation and Bayesian model | |
| CN110901446A (en) | Electric vehicle charging identification method and device based on intelligent electric meter | |
| CN113655348A (en) | Power equipment partial discharge fault diagnosis method based on deep twin network, system terminal and readable storage medium | |
| CN105067958A (en) | Feeder terminal information interaction-based small-current grounding fault location method | |
| Husari et al. | Stator turn fault diagnosis and severity assessment in converter fed induction motor using flat diagnosis structure based on deep learning approach | |
| CN113935237A (en) | Power transmission line fault type distinguishing method and system based on capsule network | |
| CN110907864B (en) | Fault detection method, device and equipment for motor stator winding and storage medium | |
| US20150331062A1 (en) | Failure Detection Method and Detection Device for Inverter | |
| CN113610119B (en) | Method for identifying power transmission line development faults based on convolutional neural network | |
| CN115329816A (en) | Multi-level inverter fault diagnosis method based on simulation data migration learning | |
| CN114330571A (en) | Power system dominant instability mode identification method and system based on transfer learning | |
| Sivapriya et al. | A novel enhanced deep learning-based fault diagnosis approach for cascaded multilevel inverter | |
| CN115935244B (en) | Single-phase rectifier fault diagnosis method based on data driving | |
| CN116449092A (en) | Automobile power detection method, system, computer and readable 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201201 |