CN115618195A - Sensor circuit fault diagnosis method, system, medium and device - Google Patents

Abstract

The invention belongs to the field of sensor circuit fault diagnosis, and discloses a method, a system, a medium and a device for diagnosing sensor circuit faults, which comprise the steps of acquiring a time domain continuous voltage signal output by a sensor circuit; performing time domain feature extraction and frequency domain feature extraction on the time domain continuous voltage signal to obtain time domain feature data and frequency domain feature data; and calling a preset sensor circuit fault diagnosis model according to the time domain continuous voltage signal, the time domain characteristic data and the frequency domain characteristic data to obtain a sensor circuit fault diagnosis result. The method has the advantages that the combined processing of the feature extraction and the time sequence signal is realized, the correlation between the time domain continuous voltage signal and the sensor circuit fault is considered, the time domain and the frequency domain dual feature extraction is carried out on the time domain continuous voltage signal by using a feature engineering method, the data are jointly input into a sensor circuit fault diagnosis model for fusion, the fault diagnosis result of the sensor circuit is obtained, and the accuracy is greatly improved.

Description

Sensor circuit fault diagnosis method, system, medium and device

technical field

The invention belongs to the field of sensor circuit fault diagnosis, and relates to a sensor circuit fault diagnosis method, system, medium and device.

Background technique

With the rapid development of the Internet of Things technology and the electronic circuit industry, sensor circuits containing a variety of components have been widely used in various fields of industrial production and daily life, and people have higher and higher requirements for the reliability of sensor circuits. Most of the faults in the sensor circuit originate from the failure of the components of the sensor circuit. Process deviation in the actual production process, poor contact in the welding process, and various non-ideal factors in the external environment may cause some components of the sensor circuit to fail, which in turn will cause sensor circuit failure, affect the operation of the equipment, and cause major damage in severe cases. Economic losses, and even derivative dangerous accidents. With the increase in the complexity of sensor circuit components, traditional troubleshooting methods are difficult to meet the existing diagnostic needs. How to quickly locate the location of sensor circuit fault components has gradually become a research hotspot in academia and industry.

The fault diagnosis of hardware circuits such as sensor circuits is to accurately locate the fault point of the circuit by processing and analyzing the output signal of the circuit. At present, there are mainly two methods for fault diagnosis of hardware circuits such as sensor circuits. One method is to extract features from the circuit output signal to obtain a small number of features related to circuit faults, and then use machine learning methods such as SVM to classify faults. One method is to directly use the timing signal or frequency domain signal output by the circuit, input the signal to the neural network for high-dimensional data processing and output classification information, and then complete the diagnosis of circuit faults. However, the diagnostic accuracy of both methods is low.

Contents of the invention

The object of the present invention is to overcome the above-mentioned shortcomings of the prior art, and provide a sensor circuit fault diagnosis method, system, medium and device.

In order to achieve the above object, the present invention adopts the following technical solutions to achieve:

The first aspect of the present invention provides a sensor circuit fault diagnosis method, including:

Obtain the time-domain continuous voltage signal output by the sensor circuit; perform time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal to obtain time-domain feature data and frequency-domain feature data; according to the time-domain continuous voltage signal, time-domain The characteristic data and frequency domain characteristic data are used to call the preset sensor circuit fault diagnosis model to obtain the sensor circuit fault diagnosis result.

Optionally, performing time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal to obtain time-domain feature data and frequency-domain feature data includes: obtaining the maximum value, minimum value, and average value of the time-domain continuous voltage signal One or several of , standard deviation, kurtosis and skewness to obtain time-domain characteristic data; perform time-frequency conversion on the time-domain continuous voltage signal to obtain the frequency-domain continuous voltage signal, and obtain the bandwidth and center of the frequency-domain continuous voltage signal One or several of the frequencies are used to obtain frequency-domain characteristic data.

Optionally, calling a preset sensor circuit fault diagnosis model according to the time domain continuous voltage signal, time domain characteristic data and frequency domain characteristic data to obtain a sensor circuit fault diagnosis result includes: combining time domain characteristic data and frequency domain domain characteristic data to obtain time-frequency mixed data; normalize time-domain continuous voltage signal and time-frequency mixed data to obtain normalized time-frequency voltage data and normalized time-frequency mixed data; The frequency-voltage data and normalized time-frequency mixed data are input into the preset sensor circuit fault diagnosis model to obtain the sensor circuit fault diagnosis result.

Optionally, the preset sensor circuit fault diagnosis model includes DNN neural network, LSTM neural network, fully connected layer network and SoftMax layer network; DNN neural network is used to input normalized time-frequency mixed data, and output DNN neural network Output data to fully connected layer network; LSTM neural network is used to input normalized time-frequency voltage data, output LSTM neural network output data to fully connected layer network; fully connected layer network is used to output data from DNN neural network and LSTM neural network The output data is fully connected, and the output data of the fully connected layer network is obtained and output to the SoftMax layer network; the SoftMax layer network is used to output data according to the input fully connected layer network, and the sensor circuit fault diagnosis result is obtained and output.

Optionally, the SoftMax function of the SoftMax layer network is:

Among them, fc_out(m)[i] is the i-th data of the fully connected layer network output data of the m-th group of time-domain continuous voltage signals, and fc_out(m)[j] is the full connection of the m-th group of time-domain continuous voltage signals The jth data of the layer network output data, m is the number of acquisition groups of time-domain continuous voltage signals, K is the number of fault types of the sensor circuit; s(fc_out(m)[i]) is the i-th fault diagnosis result of the sensor circuit When i>0, it means the probability that the sensor circuit element i fails, and when i=0, it means the probability that the sensor circuit does not fail.

The second aspect of the present invention provides a sensor circuit fault diagnosis system, including:

The data acquisition module is used to acquire the time-domain continuous voltage signal output by the sensor circuit;

The feature extraction module is used to perform time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal to obtain time-domain feature data and frequency-domain feature data;

The diagnosis module is used to invoke a preset sensor circuit fault diagnosis model according to the time-domain continuous voltage signal to obtain a sensor circuit fault diagnosis result.

Optionally, the diagnostic module is specifically used to: combine time-domain feature data and frequency-domain feature data to obtain time-frequency mixed data; normalize time-domain continuous voltage signals and time-frequency mixed data to obtain normalized Normalized time-frequency voltage data and normalized time-frequency mixed data; input the normalized time-frequency voltage data and normalized time-frequency mixed data into the preset sensor circuit fault diagnosis model to obtain the sensor circuit fault diagnosis result.

Optionally, the preset sensor circuit fault diagnosis model includes DNN neural network, LSTM neural network, fully connected layer network and SoftMax layer network; DNN neural network is used to input normalized time-frequency mixed data, and output DNN neural network Output data to fully connected layer network; LSTM neural network is used to input normalized time-frequency voltage data, output LSTM neural network output data to fully connected layer network; fully connected layer network is used to output data from DNN neural network and LSTM neural network The output data is fully connected, and the output data of the fully connected layer network is obtained and output to the SoftMax layer network; the SoftMax layer network is used to output data according to the input fully connected layer network, and the sensor circuit fault diagnosis result is obtained and output.

The third aspect of the present invention provides a sensor circuit fault diagnosis device, including a memory, a processor, and a computer program stored in the memory and operable on the processor, when the processor executes the computer program Steps for realizing the above-mentioned sensor circuit fault diagnosis method.

According to the fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above sensor circuit fault diagnosis method are realized.

Compared with the prior art, the present invention has the following beneficial effects:

The sensor circuit fault diagnosis method of the present invention obtains time-domain feature data and frequency-domain feature data by performing time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal, and then according to the time-domain continuous voltage signal, time-domain feature Data and frequency domain feature data, call the preset sensor circuit fault diagnosis model, and get the sensor circuit fault diagnosis result. When performing sensor circuit fault diagnosis, the joint processing of feature extraction and time series signal is realized, not only considering the time domain continuous voltage The correlation between the signal itself and the fault of the sensor circuit, and the feature engineering method is also used to extract the dual features of the time domain and frequency domain from the time domain continuous voltage signal, and finally these data are jointly input into the preset sensor circuit fault diagnosis model The fusion is carried out to finally obtain the fault diagnosis result of the sensor circuit, so that the accuracy of the fault diagnosis result of the sensor circuit is greatly improved.

Description of drawings

Fig. 1 is the flowchart of the sensor circuit fault diagnosis method of the embodiment of the present invention;

Fig. 2 is a structural block diagram of a sensor circuit fault diagnosis model of an embodiment of the present invention;

Fig. 3 is the principle schematic diagram of the LSTM neural network of the embodiment of the present invention;

Fig. 4 is a structural block diagram of a sensor circuit fault diagnosis system according to an embodiment of the present invention.

detailed description

In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only It is an embodiment of a part of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

It should be noted that the terms "first" and "second" in the description and claims of the present invention and the above drawings are used to distinguish similar objects, but not necessarily used to describe a specific sequence or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances such that the embodiments of the invention described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion, for example, a process, method, system, product or device comprising a sequence of steps or elements is not necessarily limited to the expressly listed instead, may include other steps or elements not explicitly listed or inherent to the process, method, product or apparatus.

As introduced in the background technology, there are currently two methods for fault diagnosis of sensor circuits. One method is to extract features from the circuit output signal to obtain a small number of features related to circuit faults, and then use machine learning methods such as SVM to perform Fault classification; another method is to directly use the timing signal or frequency domain signal output by the circuit, input the signal to the neural network for high data processing and output classification information, and then complete the diagnosis of circuit faults. However, both methods have the problem of low diagnostic accuracy.

In order to improve the above problems, an embodiment of the present invention provides a sensor circuit fault diagnosis method, including: acquiring the time-domain continuous voltage signal output by the sensor circuit; performing time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal to obtain time-domain feature data and frequency-domain feature data; according to the time-domain continuous voltage signal, time-domain feature data and frequency-domain feature data, call a preset sensor circuit fault diagnosis model to obtain a sensor circuit fault diagnosis result. The joint processing of feature extraction and time-series signals is realized, not only considering the correlation between the time-domain continuous voltage signal itself and the sensor circuit fault, but also using the method of feature engineering to perform dual time-domain and frequency-domain analysis of the time-domain continuous voltage signal Feature extraction, and finally these data are jointly input into the sensor circuit fault diagnosis model for fusion, and finally the fault diagnosis result of the sensor circuit is obtained, and the accuracy is greatly improved. The present invention is described in further detail below in conjunction with accompanying drawing:

Referring to Fig. 1, in one embodiment of the present invention, a sensor circuit fault diagnosis method is provided, and the idea of integrating feature extraction and sequential feature processing is proposed, so as to improve the accuracy of sensor circuit fault diagnosis, and finally realize the detection of sensor circuit faults. Accurately classify and locate fault types.

Specifically, the sensor circuit fault diagnosis method includes the following steps:

S1: Obtain the time-domain continuous voltage signal output by the sensor circuit.

S2: Perform time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal to obtain time-domain feature data and frequency-domain feature data.

S3: Calling a preset sensor circuit fault diagnosis model according to the time domain continuous voltage signal, time domain characteristic data and frequency domain characteristic data to obtain a sensor circuit fault diagnosis result.

Optionally, in the S1, when obtaining the time-domain continuous voltage signal output by the sensor circuit, the target sensor circuit can be simulated to obtain the target analog circuit and the output terminal of the target analog circuit can be used as a test point (i.e., data acquisition points), to obtain continuous voltage signals in the time domain, and to sample at time intervals ΔT, to collect N voltage value points in total, expressed as V(m)=[v ₁ ,v ₂ ,...,v _N ], where, m represents the number of acquisition groups of time-domain continuous voltage signals, and the total number of acquisition groups of time-domain continuous voltage signals is M. For each group of time-domain continuous voltage signals, its fault type can be expressed as {F ₀ , F ₁ ,. ..,F _K }, where, F _i ,i∈{1,2,...,K} means that component i of the sensor circuit is faulty, and F ₀ means that the sensor circuit is not faulty.

Optionally, the preset sensor circuit fault diagnosis model can be constructed through a neural network, and high-dimensional features are extracted, processed and fused according to the different characteristics of time-domain continuous voltage signals, time-domain feature data and frequency-domain feature data, Finally, the location and classification prediction of sensor circuit faults are obtained as the diagnosis results.

In summary, the sensor circuit fault diagnosis method of the present invention obtains time-domain feature data and frequency-domain feature data by performing time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal, and according to the time-domain continuous voltage signal, time-domain feature domain characteristic data and frequency domain characteristic data, call the preset sensor circuit fault diagnosis model, and obtain the sensor circuit fault diagnosis result. The correlation between the continuous voltage signal itself and the fault of the sensor circuit also uses the method of feature engineering to perform dual feature extraction in the time domain and frequency domain on the continuous voltage signal in the time domain, and finally these data are jointly input into the preset sensor circuit fault The diagnosis model is fused, and finally the fault diagnosis result of the sensor circuit is obtained, so that the accuracy of the sensor circuit fault diagnosis result is greatly improved.

In a possible implementation manner, performing time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal, and obtaining time-domain feature data and frequency-domain feature data includes: obtaining the maximum value of the time-domain continuous voltage signal, One or more of the minimum value, average value, standard deviation, kurtosis, and skewness to obtain time-domain characteristic data; perform time-frequency conversion on time-domain continuous voltage signals to obtain frequency-domain continuous voltage signals, and obtain frequency-domain continuous voltage One or several of the bandwidth and center frequency of the signal are used to obtain frequency domain characteristic data.

Specifically, regarding the time-domain feature extraction of time-domain continuous voltage signals, in this embodiment, the maximum value, minimum value, average value, standard deviation, kurtosis and skewness are used as time-domain feature indicators, and the time-domain continuous voltage signal To perform time-domain feature extraction is to obtain the maximum value, minimum value, average value, standard deviation, kurtosis and skewness of the time-domain continuous voltage signal, and the time-domain feature data can be expressed as td(m):

td(m)＝[v _max (m), v _min (m), v _avg (m), v _std (m), v _peak (m), v _ske (m)]

为时域连续电压信号的平均值，

为时域连续电压信号的标准差，

为时域连续电压信号的峰度，用于表征概率密度分布曲线在平均值处峰值高低，

为时域连续电压信号的偏度，用于表征概率分布密度曲线相对于平均值不对称程度。Among them, v _max (m)=max(V(m)) is the maximum value of the time-domain continuous voltage signal, v _min (m)=min(V(m)) is the minimum value of the time-domain continuous voltage signal,

is the average value of the continuous voltage signal in the time domain,

is the standard deviation of the time-domain continuous voltage signal,

is the kurtosis of the continuous voltage signal in the time domain, which is used to characterize the peak value of the probability density distribution curve at the average value,

is the skewness of the time-domain continuous voltage signal, which is used to characterize the degree of asymmetry of the probability distribution density curve relative to the average value.

Regarding the frequency-domain feature extraction of the time-domain continuous voltage signal, it is first necessary to perform time-frequency conversion to obtain the frequency-domain continuous voltage signal, and then perform feature extraction on the frequency-domain continuous voltage signal. In this embodiment, the bandwidth and center frequency are used as the frequency domain. The domain characteristic index obtains the bandwidth and center frequency of the continuous voltage signal in the frequency domain, and obtains the characteristic data in the frequency domain. The characteristic data in the frequency domain can be expressed as fd(m):

fd(m)=[band(m), freq(m)]

Wherein, band(m) is the bandwidth of the continuous voltage signal in the frequency domain, and freq(m) is the center frequency of the continuous voltage signal in the frequency domain.

In a possible implementation manner, the calling a preset sensor circuit fault diagnosis model according to the time-domain continuous voltage signal, time-domain characteristic data and frequency-domain characteristic data to obtain a sensor circuit fault diagnosis result includes: domain characteristic data and frequency domain characteristic data to obtain time-frequency mixed data; normalize time domain continuous voltage signal and time-frequency mixed data to obtain normalized time-frequency voltage data and normalized time-frequency mixed data; The normalized time-frequency voltage data and the normalized time-frequency mixed data are input into the preset sensor circuit fault diagnosis model to obtain the sensor circuit fault diagnosis result.

Specifically, the time-domain characteristic data and the frequency-domain characteristic data are firstly combined to form time-frequency mixed data, expressed as tf(m): tf(m)=[fd(m),td(m)].

Then, normalize each group of time-frequency mixed data to obtain the corresponding normalized time-frequency mixed data norm_tf(m), and normalize each group of time-domain continuous voltage signals to obtain the corresponding Normalized time-frequency voltage data norm_V(m).

Optionally, when performing normalization processing on each set of time-frequency mixed data, the min-max normalization method is used to realize the normalization of each set of time-frequency mixed data through the following formula:

Among them, tf _i (m) represents the i-th data in the m-th group of time-frequency mixed data, tf _i ^* (m) represents the normalized value of tf _i (m), and tf _min (m) represents tf( The minimum value of m), tf _max (m) represents the maximum value of tf(m).

Optionally, normalize each group of time-domain continuous voltage signals, or use the min-max standardization method to realize the normalization of each group of time-domain continuous voltage signals through the following formula:

表示v_i(m) 归一化后的数值大小。Among them, v _i (m) represents the i-th data in the m-th group of time-domain continuous voltage signals,

Indicates the normalized numerical value of v _i (m).

In a possible implementation manner, referring to FIG. 2 , the sensor circuit fault diagnosis model includes a DNN neural network, an LSTM neural network, a fully connected layer network, and a SoftMax (classification network) layer network.

Fully connected layers (FC) play the role of "classifier" in the entire convolutional neural network. If operations such as the convolutional layer, pooling layer, and activation function layer map the original data to the hidden layer feature space, the fully connected layer plays the role of mapping the learned distributed feature representation to the sample label space. In practice, fully connected layers can be implemented by convolution operations. The Softmax layer maps several (-∞, +∞) real numbers to the same number of (0,1) real numbers (which can represent probabilities), while ensuring that their sum is 1.

Among them, the DNN neural network is used to input the normalized time-frequency mixed data, and output the output data of the DNN neural network to the fully connected layer network; the LSTM neural network is used to input the normalized time-frequency voltage data, and output the output data of the LSTM neural network to the full network Connection layer network; the fully connected layer network is used to perform full connection processing on the output data of the DNN neural network and the output data of the LSTM neural network to obtain the output data of the fully connected layer network and output it to the SoftMax layer network; Connect layer network to output data, get sensor circuit fault diagnosis result and output.

Among them, the normalized time-frequency voltage data and normalized time-frequency mixed data are input into the preset sensor circuit fault diagnosis model, and the specific process of obtaining the sensor circuit fault diagnosis result is as follows:

Step 1: Input the normalized time-frequency mixed data norm_tf(m) into the DNN neural network to obtain the output data d_out(m) of the DNN neural network.

Step 2: Input the normalized time-frequency voltage data norm_V(m) into the LSTM neural network to obtain the output data l_out(m) of the LSTM neural network.

Step 3: Combine the output data d_out(m) of the DNN neural network in step 1 and the output data l_out(m) of the LSTM neural network in step 2, input them into the fully connected layer network, and obtain the output data fc_out(m) of the fully connected layer network ), and fc_out(m) is a K+1-dimensional vector, where K is the number of fault types of the sensor circuit, the number of fault types of the sensor circuit is related to the number of components of the sensor circuit, and the sensor circuit has K components in total, then the fault type is {Component 1 failed, component 2 failed, ..., component K failed}.

Step 4: Take the output data fc_out(m) of the fully connected layer network as the input of the SoftMax layer network, and obtain a vector with a dimension of K+1 after calculation by the SoftMax layer network, which is the result of the fault diagnosis of the sensor circuit.

Specifically, in this embodiment, according to the different characteristics of time-frequency mixed data and time-frequency voltage data, DNN (Deep Neural Networks, deep neural network) decision-making neural network and LSTM (Long Short Term Memory, long-term short-term memory network) neural network are used respectively. The network performs feature extraction, and finally combines the two network data to obtain the final predicted value, so as to improve the predictive ability of the sensor circuit fault diagnosis model.

Optionally, the DNN neural network is a fully connected neural network including an input layer, an output layer and two hidden layers. Among them, the forward propagation function of the hidden layer is:

Among them, y _i is the i-th output of the hidden layer, x _j is the j-th input of the hidden layer, w _i,j is the weight of the j-th input corresponding to the i-th output, b _i corresponds to the i-th input bias. It should be noted that the forward propagation formula of the fully connected layer network involved in this embodiment is consistent with the above formula.

Optionally, referring to FIG. 3 , each cell of the LSTM neural network implements forward propagation through an input gate, a forget gate and an output gate.

Among them, the update of the forget gate can be expressed as:

f _t = σ·(W _f h _t-1 +U _f x _t +b _f )

Among them, σ is the sigmoid activation function, W _f , U _f and b _f are the coefficients and biases of the forgetting gate, all of which are trainable parameters, h _t-1 is the hidden output state of the t-1th cell, x _{t is} the The t-th input of the sequence corresponds to the value of the t-th element of the normalized time-frequency voltage data norm_V(m) input in this round, and f _t is the updated state of the t-th forget gate.

The update of the input gate can be expressed as:

i _t = σ·(W _i h _t-1 +U _i x _t +b _i )

C _t ＝f _t ·C _t-1 +i _t ·C _t

均为输入门的中间状态量，

表示当前的记忆状态，i_t表示对

的遗忘能力。Among them, W _i , U _i , b _i , W _c , U _c and b _c are the coefficients and biases of the input gates, all of which are trainable parameters, C _t-1 is the long-term state at the previous moment, and C _t is the t-th The updated state of the input gate represents the long-term state at the current _moment , it and

Both are the intermediate state quantities of the input gates,

Represents the current memory state, it _represents the pair

forgetting ability.

The update of the output gate can be expressed as:

o _t ＝σ·(W _o h _t-1 +U _o x _t +b _o )

h _t ＝o _t ·tanh(C _t )

Among them, W _o , U _o and b _o are the coefficients and biases of the output gate, all of which are trainable parameters, h _t is the hidden output state of the tth cell, O _t represents the forgetting ability of the long-term state at the current moment.

So far, one cell of the LSTM neural network has achieved a complete forward propagation.

In a possible implementation manner, the SoftMax function of the SoftMax layer network is:

Among them, fc_out(m)[i] is the i-th data of the fully connected layer network output data of the m-th group of time-domain continuous voltage signals, and fc_out(m)[j] is the full connection of the m-th group of time-domain continuous voltage signals The jth data of the layer network output data, m is the number of time-domain continuous voltage signal acquisition groups, K is the number of fault types of the sensor circuit; s(fc_out(m)[i]) is the i-th fault diagnosis result of the sensor circuit When i>0, it means the probability that the sensor circuit element i fails, and when i=0, it means the probability that the sensor circuit does not fail.

Finally, take s(m)=[s(fc_out(m)[0]),s(fc_out(m)[1]),...,s(fc_out(m)[K])] as sensor circuit failure The diagnostic results are output to reflect the probability of failure of the sensor circuit and internal components.

The following are device embodiments of the present invention, which can be used to implement the method embodiments of the present invention. For details not disclosed in the device embodiment, please refer to the method embodiment of the present invention.

Referring to Fig. 4, in another embodiment of the present invention, a sensor circuit fault diagnosis system is provided, which can be used to implement the above sensor circuit fault diagnosis system method, specifically, the sensor circuit fault diagnosis system includes a data acquisition module, feature extraction modules and diagnostic modules.

Among them, the data acquisition module is used to obtain the time-domain continuous voltage signal output by the sensor circuit; the feature extraction module is used to perform time-domain feature extraction and frequency-domain feature extraction on the time-domain continuous voltage signal to obtain time-domain feature data and frequency-domain feature data ; The diagnostic module is used to invoke a preset sensor circuit fault diagnosis model according to the time-domain continuous voltage signal to obtain a sensor circuit fault diagnosis result.

In a possible implementation manner, the diagnostic module is specifically configured to: combine time-domain feature data and frequency-domain feature data to obtain time-frequency mixed data; normalize both the time-domain continuous voltage signal and the time-frequency mixed data processing to obtain normalized time-frequency voltage data and normalized time-frequency mixed data; input the normalized time-frequency voltage data and normalized time-frequency mixed data into the preset sensor circuit fault diagnosis model to obtain sensor circuit fault diagnosis result.

In a possible implementation, the sensor circuit fault diagnosis model includes a DNN neural network, an LSTM neural network, a fully connected layer network, and a SoftMax layer network; the DNN neural network is used to input normalized time-frequency mixed data, and output DNN The neural network outputs data to the fully connected layer network; the LSTM neural network is used to input normalized time-frequency voltage data, and the output data of the LSTM neural network is output to the fully connected layer network; the fully connected layer network is used to combine the output data of the DNN neural network with the LSTM The output data of the neural network is fully connected, and the output data of the fully connected layer network is obtained and output to the SoftMax layer network; the SoftMax layer network is used to output data according to the input fully connected layer network, and obtain and output the fault diagnosis result of the sensor circuit.

In a possible implementation, the feature extraction module is specifically configured to: obtain one or more of the maximum value, minimum value, average value, standard deviation, kurtosis and skewness of the continuous voltage signal in the time domain, and obtain Time-domain characteristic data: time-frequency conversion is performed on the time-domain continuous voltage signal to obtain a frequency-domain continuous voltage signal, and one or more of the bandwidth and center frequency of the frequency-domain continuous voltage signal is obtained to obtain frequency-domain characteristic data.

In a possible implementation manner, the SoftMax function of the SoftMax layer network is:

Among them, fc_out(m)[i] is the i-th data of the fully connected layer network output data of the m-th group of time-domain continuous voltage signals, and fc_out(m)[j] is the full connection of the m-th group of time-domain continuous voltage signals The jth data of the layer network output data, m is the number of time-domain continuous voltage signal acquisition groups, K is the number of fault types of the sensor circuit; s(fc_out(m)[i]) is the i-th fault diagnosis result of the sensor circuit When i>0, it means the probability that the sensor circuit element i fails, and when i=0, it means the probability that the sensor circuit does not fail.

All relevant content of each step involved in the foregoing embodiment of the sensor circuit fault diagnosis method can be referred to the functional description of the corresponding functional modules of the sensor circuit fault diagnosis system in the embodiment of the present invention, and will not be repeated here.

The division of modules in the embodiments of the present invention is schematic, and is only a logical function division. In actual implementation, there may be other division methods. In addition, each functional module in each embodiment of the present invention can be integrated into a processing In the controller, it can also be physically present separately, or two or more modules can be integrated into one module. The above-mentioned integrated modules can be implemented in the form of sensors or in the form of software function modules.

In yet another embodiment of the present invention, a sensor circuit fault diagnosis device is provided, the sensor circuit fault diagnosis device includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processing The device is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gates Array (Field-Programmable GateArray, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, are suitable for implementing one or more instructions, and are specifically suitable for To load and execute one or more instructions in the computer storage medium to realize the corresponding method flow or corresponding functions; the processor described in the embodiment of the present invention can be used for the operation of the sensor circuit fault diagnosis method.

In yet another embodiment of the present invention, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a computer device for storing programs and data. . It can be understood that the computer-readable storage medium here may include a built-in storage medium in the computer device, and of course may also include an extended storage medium supported by the computer device. The computer-readable storage medium provides storage space, and the storage space stores the operating system of the terminal. Moreover, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and these instructions may be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here may be a high-speed RAM memory, or a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. One or more instructions stored in the computer-readable storage medium can be loaded and executed by the processor, so as to realize the corresponding steps of the sensor circuit fault diagnosis method in the above-mentioned embodiments.

Those skilled in the art should understand that the embodiments of the present invention may be provided as methods, systems, or computer program products. Accordingly, the present invention can 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, etc.) 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 invention. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby The instructions provide steps for implementing the functions specified in the flow chart or blocks of the flowchart and/or the block or blocks of the block diagrams.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the present invention can still be Any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention shall fall within the protection scope of the claims of the present invention.

Claims (10)

1. A method of diagnosing a fault in a sensor circuit, comprising:

acquiring a time domain continuous voltage signal output by a sensor circuit;

performing time domain feature extraction and frequency domain feature extraction on the time domain continuous voltage signal to obtain time domain feature data and frequency domain feature data;

and calling a preset sensor circuit fault diagnosis model according to the time domain continuous voltage signal, the time domain characteristic data and the frequency domain characteristic data to obtain a sensor circuit fault diagnosis result.

2. The method for diagnosing the circuit fault of the sensor according to claim 1, wherein the performing time domain feature extraction and frequency domain feature extraction on the time domain continuous voltage signal to obtain time domain feature data and frequency domain feature data comprises:

acquiring one or more of the maximum value, the minimum value, the average value, the standard deviation, the kurtosis and the skewness of the time domain continuous voltage signal to obtain time domain characteristic data;

and performing time-frequency conversion on the time domain continuous voltage signal to obtain a frequency domain continuous voltage signal, and acquiring one or more of the bandwidth and the central frequency of the frequency domain continuous voltage signal to obtain frequency domain characteristic data.

3. The sensor circuit fault diagnosis method according to claim 1, wherein the obtaining of the sensor circuit fault diagnosis result by calling a preset sensor circuit fault diagnosis model according to the time-domain continuous voltage signal, the time-domain characteristic data and the frequency-domain characteristic data comprises:

combining the time domain characteristic data and the frequency domain characteristic data to obtain time-frequency mixed data;

normalizing the time domain continuous voltage signal and the time frequency mixed data to obtain normalized time frequency voltage data and normalized time frequency mixed data;

and inputting the normalized time-frequency voltage data and the normalized time-frequency mixed data into a preset sensor circuit fault diagnosis model to obtain a sensor circuit fault diagnosis result.

4. The sensor circuit failure diagnosis method according to claim 3, wherein the preset sensor circuit failure diagnosis model includes a DNN neural network, an LSTM neural network, a full connection layer network, and a SoftMax layer network;

the DNN neural network is used for inputting normalized time-frequency mixed data and outputting DNN neural network output data to the full-connection layer network;

the LSTM neural network is used for inputting normalized time-frequency voltage data and outputting LSTM neural network output data to the full-connection layer network;

the full connection layer network is used for performing full connection processing on DNN neural network output data and LSTM neural network output data to obtain full connection layer network output data and outputting the full connection layer network output data to a SoftMax layer network;

and the SoftMax layer network is used for outputting data according to the input full-connection layer network to obtain and output a fault diagnosis result of the sensor circuit.

5. The sensor circuit fault diagnostic method of claim 4, wherein the SoftMax function of the SoftMax tier network is:

the method comprises the steps that a first group of time domain continuous voltage signals are acquired, wherein fc _ out (m) i is the ith data of all-connection layer network output data of the mth group of time domain continuous voltage signals, fc _ out (m) j is the jth data of all-connection layer network output data of the mth group of time domain continuous voltage signals, m is the acquisition group number of the time domain continuous voltage signals, and K is the fault type number of a sensor circuit; s (fc _ out (m) [ i ]) is the i-th data of the sensor circuit failure diagnosis result, and indicates the probability of failure of the sensor circuit element i when i > 0, and indicates the probability of failure of the sensor circuit element i when i = 0.

6. A sensor circuit fault diagnostic system, comprising:

the data acquisition module is used for acquiring a time domain continuous voltage signal output by the sensor circuit;

the characteristic extraction module is used for carrying out time domain characteristic extraction and frequency domain characteristic extraction on the time domain continuous voltage signal to obtain time domain characteristic data and frequency domain characteristic data;

and the diagnosis module is used for calling a preset sensor circuit fault diagnosis model according to the time domain continuous voltage signal to obtain a sensor circuit fault diagnosis result.

7. The sensor circuit fault diagnostic system of claim 6, wherein the diagnostic module is specifically configured to:

combining the time domain characteristic data and the frequency domain characteristic data to obtain time-frequency mixed data;

carrying out normalization processing on the time domain continuous voltage signal and the time frequency mixed data to obtain normalized time frequency voltage data and normalized time frequency mixed data;

and inputting the normalized time-frequency voltage data and the normalized time-frequency mixed data into a preset sensor circuit fault diagnosis model to obtain a sensor circuit fault diagnosis result.

8. The sensor circuit fault diagnosis system according to claim 7, wherein the preset sensor circuit fault diagnosis model includes a DNN neural network, an LSTM neural network, a full connectivity layer network, and a SoftMax layer network;

the DNN neural network is used for inputting the normalized time-frequency mixed data and outputting the output data of the DNN neural network to a full-connection layer network;

the LSTM neural network is used for inputting normalized time-frequency voltage data and outputting LSTM neural network output data to the full-connection layer network;

the full connection layer network is used for performing full connection processing on DNN neural network output data and LSTM neural network output data to obtain full connection layer network output data and outputting the full connection layer network output data to a SoftMax layer network;

and the SoftMax layer network is used for outputting data according to the input full-connection layer network to obtain and output a fault diagnosis result of the sensor circuit.

9. A sensor circuit failure diagnosis apparatus comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the sensor circuit failure diagnosis method according to any one of claims 1 to 5 when executing the computer program.

10. A computer-readable storage medium, in which a computer program is stored, which, when being executed by a processor, carries out the steps of the sensor circuit failure diagnosis method according to any one of claims 1 to 5.

Images