Multi-source sensor fault detection method and computer equipment

Abstract

The invention relates to the technical field of intelligent monitoring, and aims to solve the problem that a traditional sensor fault detection method needs human intervention and cannot perform accurate fault detection, positioning and correction on a large number of multi-source sensors in a ship power system in real time. The invention discloses a multi-source sensor fault detection method, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring sensor data acquired by various types of sensors; performing fault identification on the sensor data through a convolutional neural network, and determining the type of a sensor fault; reconstructing sensor data collected by the faulted sensor through a convolution auto-encoder corresponding to the type of the sensor fault, and outputting correction data; and replacing the sensor data collected by the faulted sensor with the correction data, and outputting a reconstructed data matrix. By adopting the method, accurate fault detection, classification and fault data reconstruction can be carried out on the sensor.

Description

Technical Field

[0001] The present application relates to the field of intelligent monitoring technology, and more specifically, to a multi-source sensor fault detection method, a computer device, a computer-readable storage medium, and a computer program product. Background Art

[0002] The continuous advancement of sensing and detection technologies has led to the rapid development of intelligent monitoring systems for monitoring ship power systems. Accurate sensor data can provide information about the operation of the ship's power system to help predict the ship's operating status and help calculate the control parameters of the subsystem.

[0003] However, due to sensor aging, defects or environmental factors, sensors are prone to data loss, random noise, drift and other faults. Although traditional sensor fault detection methods can detect and isolate faults and reconstruct damaged or lost sensor data, they require a lot of human intervention and cannot accurately detect, locate and correct faults of a large number of multi-source sensors in a ship power system in real time. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a multi-source sensor fault detection method, computer equipment, computer-readable storage medium and computer program product, which can perform fault detection, fault classification and fault data reconstruction on sensors in real time, thereby improving the accuracy of sensor fault detection.

[0005] To achieve the above object, according to a first aspect of the present invention, a multi-source sensor fault detection method is provided, which is applied to a ship power system and comprises:

[0006] Obtain sensor data collected by various types of sensors;

[0007] Use convolutional neural networks to identify faults in sensor data and determine the type of sensor faults;

[0008] Reconstructing sensor data collected by the faulty sensor through convolutional autoencoders corresponding to the types of sensor faults, and outputting corrected data, wherein the number of convolutional autoencoders is the number of sensor fault types;

[0009] The sensor data collected by the faulty sensor is replaced with the corrected data, and the reconstructed data matrix is ​​output.

[0010] Furthermore, sensor data collected by various types of sensors are obtained, including obtaining sensor data collected by various types of sensors through a sliding window of a preset time length, and the sensor data include at least one of the flow, pressure, and temperature of steam and condensate generated by the ship's power system during the power conversion process.

[0011] Furthermore, sensor data collected by multiple types of sensors are obtained, including obtaining sensor data collected by multiple types of sensors, and the sensor data is input into a convolutional neural network in a matrix form.

[0012] Furthermore, the sensor data is fault identified through a convolutional neural network to determine the type of sensor fault, including detecting whether the sensor data is abnormal through a convolutional neural network; when abnormal sensor data is detected, the sensor fault is classified to determine the type of sensor fault.

[0013] Furthermore, the types of sensor faults include data missing faults, random noise and drift faults. When abnormal sensor data is detected, the sensor fault is classified to determine the type of the sensor fault, including when the sensor data is detected to be missing data, determining that the sensor fault is a data missing fault; when the sensor data is detected to be random data, determining that the sensor fault is random noise; when the sensor data is detected to be drift data, determining that the sensor fault is a drift fault.

[0014] Furthermore, the convolutional neural network includes a first convolutional layer, a first maximum pooling layer, a second convolutional layer, a second maximum pooling layer, a first fully connected layer, a second fully connected layer and a softmax layer.

[0015] Furthermore, the sensor data collected by the faulty sensor is replaced with the corrected data, and a reconstructed data matrix is ​​output, including calculating a differential metric between the sensor data collected by the faulty sensor and the corrected data; when the differential metric is greater than a preset value, the sensor data collected by the faulty sensor is replaced with the corrected data, and a reconstructed data matrix is ​​output.

[0016] According to a second aspect of the present invention, there is also provided a computer device, which includes a memory, a processor and a computer program stored in the memory, and the processor executes the computer program to implement the steps of any one of the above methods.

[0017] According to a third aspect of the present invention, there is also provided a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0018] According to a fourth aspect of the present invention, there is also provided a computer program product, comprising a computer program, which implements the steps of any of the above methods when executed by a processor.

[0019] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0020] A multi-source sensor fault detection method provided by the present invention utilizes the spatiotemporal correlation between sensor data collected by multi-source sensors, performs fault detection on the sensor data through a convolutional neural network, determines the fault type when a fault is detected, and then reconstructs the fault data through a convolutional autoencoder corresponding to the fault type, replaces the fault data with reconstructed corrected data, and finally outputs a reconstructed data matrix, thereby achieving the purpose of accurately detecting, classifying and reconstructing fault data of one or more sensors in a ship power system in real time and automatically, and avoiding manual intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 A schematic flow chart of a multi-source sensor fault detection method provided for one embodiment of the present application;

[0023] Figure 2 A schematic flow chart of a multi-source sensor fault detection method provided in another embodiment of the present application;

[0024] Figure 3 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0026] The terms "first", "second", "third", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0027] like Figure 1 As shown, a multi-source sensor fault detection method is provided, which is applied to a ship power system. As a complex power system, a ship power system requires a large number of sensors to detect the structural performance of the detector in order to realize the health monitoring of the ship power system. The method can be executed by a terminal or by a server that communicates with the terminal through a network. Among them, the terminal can be but is not limited to various personal computers, laptops, smart phones, tablets, Internet of Things devices, portable wearable devices, etc. The server can be an independent server, or a server cluster composed of multiple servers can be used. Taking the application of this method to a terminal as an example, it includes the following steps:

[0028] Step 101: Acquire sensor data collected by various types of sensors.

[0029] Among them, the sensor data collected by various types of sensors (also called multi-source sensors) have temporal and spatial correlation, that is, they have a correlation in time series and spatial position. The change of one sensor data over time can be analyzed to obtain the change of another sensor data over time, which helps to improve the accuracy of sensor data.

[0030] The sensor data includes the flow, pressure, temperature, etc. of steam and condensed water generated by the ship's power system during power conversion.

[0031] Exemplarily, the terminal obtains sensor data collected by various types of sensors through a sliding time window of a preset length (eg, 200 ms), and inputs the sensor data into a convolutional neural network (CNN) in a matrix form.

[0032] Step 102: Perform fault identification on the sensor data through a convolutional neural network to determine the type of sensor fault.

[0033] Among them, the convolutional neural network (CNN), also known as a classifier, is used to identify the type of sensor fault. Sensor data includes normal data, missing data (abbreviated as T1), random data (abbreviated as T2) and drift data (abbreviated as T3). The types of sensor faults include missing data fault (abbreviated as RC1), random noise (abbreviated as RC2) and drift fault (abbreviated as RC3).

[0034] Exemplarily, the terminal detects whether there is any sensor data abnormality through a convolutional neural network; when abnormal sensor data is detected, the sensor fault is classified to determine the type of sensor fault; when no abnormal sensor data is detected (i.e., the sensor data is all normal data), the sensor data is not processed.

[0035] When abnormal sensor data is detected, the sensor fault is classified and the type of the sensor fault is determined, including when the sensor data is detected as missing data (T1), determining the sensor fault as a missing data fault (RC1); when the sensor data is detected as random data (T2), determining the sensor fault as random noise (RC2); when the sensor data is detected as drift data (T3), determining the sensor fault as a drift fault (RC3).

[0036] Step 103, reconstructing the sensor data collected by the faulty sensor through a convolutional autoencoder corresponding to the type of sensor fault, and outputting corrected data, wherein the number of convolutional autoencoders is the number of sensor fault types.

[0037] Among them, Convolutional Auto-Encoders (CAE) are used for data reconstruction, and the number of Convolutional Auto-Encoders is consistent with the number of sensor fault types.

[0038] Exemplarily, when the type of sensor failure is determined, the terminal reconstructs the erroneous data through the convolutional autoencoder corresponding to the type and outputs the corrected data. Since the number of convolutional autoencoders is consistent with the number of sensor failure types, one type corresponds to one convolutional autoencoder, so that data reconstruction can be scalable. In the case of a large number of failed sensors, parallel processing of faults can also be achieved, that is, sensor data collected by multiple failed sensors can be reconstructed in parallel.

[0039] Step 104: Replace the sensor data collected by the faulty sensor with the corrected data, and output a reconstructed data matrix.

[0040] Among them, the reconstructed data matrix is ​​all the sensor data output by all convolutional autoencoders.

[0041] Since possible biases may be introduced in the sensor data of other sensors when the convolutional autoencoder reconstructs the sensor data collected by the faulty sensor, in order to avoid this, a differential metric is used to identify the faulty sensor and only the sensor data collected by the identified faulty sensor is replaced in the final output.

[0042] Exemplarily, the terminal calculates the differential metric between the sensor data collected by the faulty sensor and the corrected data. When the differential metric is greater than a preset value, it is determined that the sensor data is data collected by the faulty sensor. At this time, the sensor data collected by the faulty sensor is replaced with the corrected data, and the reconstructed data matrix is ​​output.

[0043] The calculation formula of the difference metric is as follows:

[0044]

[0045] Among them, the input is the sensor data collected by the sensor, and the output is the corrected data reconstructed by the convolutional autoencoder.

[0046] In the above-mentioned multi-source sensor fault detection method, compared with the traditional sensor fault detection method, which is difficult to apply to a large number of sensor fault detection, the temporal and spatial correlation between the sensor data collected by the multi-source sensors is utilized to perform fault detection on the sensor data through a convolutional neural network. When a fault is detected, the fault type is determined, and then the fault data is reconstructed through a convolutional autoencoder corresponding to the fault type. The fault data is replaced with the reconstructed corrected data, and finally the reconstructed data matrix is ​​output, thereby achieving the purpose of real-time automatic and accurate fault detection, fault classification and fault data reconstruction of one or more sensors in the ship power system and avoiding manual intervention.

[0047] In one embodiment, the convolutional neural network includes two alternating convolutional layers and maximum pooling layers (i.e., a first convolutional layer, a first maximum pooling layer, a second convolutional layer, a second maximum pooling layer), two fully connected layers (i.e., a first fully connected layer and a second fully connected layer) and a softmax layer, which are used to extract features across multiple time scales and classify the type of sensor fault based on the features.

[0048] Specifically, since sensor faults are mainly captured as temporal features (e.g., data missing faults have zero readings in the temporal direction), the kernel size is set to 1 to avoid feature averaging of the data along the spatial dimension. After the first convolutional layer, the first max pooling layer is introduced with the goal of retaining the most dominant features and reducing the number of parameters in the network. The second convolutional layer is then added to extract features at higher time scales, which is critical in some cases, e.g., drift or spike faults may not be easily perceived at smaller time scales but can be identified at longer time scales. Using the cross entropy loss as the objective function, the kernel size of the first convolutional layer is optimized to (11×1), the kernel size of the second convolutional layer is optimized to (5×1), and the filter size of the two max pooling layers is (2×1). Finally, a fully connected layer and a softmax layer are added to use the identified features for classification. The training process of the convolutional neural network aims to minimize the cross entropy loss function, which is:

[0049]

[0050] Where N is the number of samples, M is the total number of sensor fault categories, and y icIndicates the true label of the i-th sample belonging to category m (if it belongs, it is 1, otherwise it is 0), p ic Represents the probability that the convolutional neural network predicts that the i-th sample belongs to category m.

[0051] In one embodiment, multiple sensor data are input to the convolutional neural network in the form of a matrix. When multiple sensor data are stacked into an array, the sensor data is conceptually similar to a pixel, but has one dimension corresponding to the time evolution. It should be noted that although common machine learning algorithms have a certain accuracy in image classification tasks, they do not fully utilize the neighborhood features of specific points at multiple scales. In this application, the convolutional neural network is trained by combining multiple sensor data with spatiotemporal correlation, that is, the input data of the convolutional neural network has spatial and temporal correlation, thereby improving the accuracy of sensor data fault detection and classification.

[0052] In one embodiment, the architecture of all convolutional autoencoders is the same for different types of sensor faults. The size of the convolution kernel is selected by minimizing the loss function. Taking the types of sensor faults including data missing fault (RC1), random noise (RC2) and drift fault (RC3) as examples, the three CAE networks are trained independently, and the training process includes minimizing the mean square error, which is the difference between the true value and the CAE network predicted value. Each CAE network handles a specific type of sensor fault.

[0053] In this embodiment, at least one convolutional autoencoder with the same architecture is used to process the corresponding sensor failure. Compared with using only a single convolutional autoencoder to process all sensor failures, this can improve the accuracy of reconstructing sensor data collected by the faulty sensor and can also reduce the computational cost of the convolutional autoencoder during the training and use stages.

[0054] In one embodiment, Figure 2 As shown, a multi-source sensor fault detection method based on a convolutional neural network and a convolutional autoencoder includes:

[0055] Step 1: Use a convolutional neural network (classifier) ​​to identify sensor data collected by multi-source sensors and determine the type of sensor failure;

[0056] Step 2: Use a convolutional autoencoder (reconstructor) to reconstruct the sensor data collected by the faulty sensor.

[0057] Specifically, step one also includes a preprocessing step, which includes obtaining real-time sensor data through a sliding window of a preset time length (200ms) and passing it to the classifier as input. Step two uses three separate reconstructors (corresponding to three types of faults: data missing fault RC1, random noise RC2, and drift fault RC3). Step two also includes a post-processing step, which is used for fault isolation, including identifying the faulty sensor, determining whether it is a sensor fault or a system fault, and using the reconstructed data as the final output result.

[0058] like Figure 2 As shown at the top, five sensor data S1, S3, S5, S7 and S9 are obtained, of which S7 is missing data. The classifier detects the fault, classifies the fault type, identifies the data missing fault RC1, and calls the corresponding reconstructor according to the fault type (the reconstructor corresponds to Figure 2 The data missing fault RC1 indicated by the dashed arrow in the figure). Subsequently, the reconstructor reconstructs the data of S7 (fault tolerance), predicts the sensor data, and obtains the reconstructed data; and does not process the normal data of S1, S3, S5 and S9. The post-processing step identifies the only sensor with a fault, determines whether it is a sensor fault or a system fault, and outputs the reconstructed data as the final result, obtaining Figure 2 The new sensor data S7 is shown at the bottom.

[0059] In step 2, the number of convolutional autoencoders is consistent with the number of sensor fault types. The convolutional autoencoders can be expanded as the number of sensors increases. Therefore, the multi-source sensor fault detection method can be applied to a large number of sensor scenarios and can reconstruct data (correct erroneous data) in the presence of different types of sensor faults, so that erroneous data can be used directly to reduce human intervention. In addition, since sensor data is collected through a sliding window, this multi-source sensor fault detection method can handle sensor faults in real time.

[0060] The number of sensors that can be used in this multi-source sensor fault detection method is mainly limited by the computational cost of training CNN and CAE. Since step 1 realizes the classification of sensor faults, step 2 only requires r (r is the number of sensor fault types) CAEs to handle r types of faults. Assuming that n, m and r represent the total number of sensors, the maximum number of allowed faulty sensors and the number of sensor fault types, respectively, the traditional sensor fault detection method requires Compared with the multi-source sensor fault detection method provided in this embodiment, a CAE requires longer model training time and greater computing cost. Therefore, a multi-source sensor fault detection method provided in this embodiment can significantly reduce the computing cost, can be integrated into a data acquisition system or implemented using edge computing, and is particularly suitable for ship power systems with a large number of sensors.

[0061] The present application also provides a computer device, whose internal structure diagram can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface, the display unit and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a multi-source sensor fault detection method is implemented. The display unit of the computer device is used to form a visually visible image, and can be a display screen, a projection device or a virtual reality imaging device. The input device of the computer device can be a touch layer covered on the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse, etc.

[0062] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0063] like Figure 3 As shown, the present application also provides a computer device, which includes a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps in the above-mentioned method embodiments.

[0064] The present application also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented. The computer-readable storage medium may include, but is not limited to, any type of disk, including a floppy disk, an optical disk, a DVD, a CD-ROM, a micro drive, and a magneto-optical disk, a ROM, a RAM, an EPROM, an EEPROM, a DRAM, a VRAM, a flash memory device, a magnetic card or an optical card, a nanosystem (including a molecular memory IC), or any type of medium or device suitable for storing instructions and / or data.

[0065] The present application also provides a computer program product, including a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0066] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0067] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0068] The above is only an exemplary embodiment of the present disclosure, and the scope of the present disclosure cannot be limited thereto. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the specification and practicing the disclosure here, those skilled in the art will easily think of the implementation scheme of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

[0069] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0070] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A multi-source sensor fault detection method, applied to a ship power system, characterized in that: include: Obtain sensor data collected by various types of sensors; Performing fault identification on the sensor data through a convolutional neural network to determine the type of sensor fault; Reconstructing sensor data collected by the faulty sensor through convolutional autoencoders corresponding to the types of the sensor faults, and outputting corrected data, wherein the number of the convolutional autoencoders is the number of the types of the sensor faults; The sensor data collected by the faulty sensor is replaced with the correction data, and a reconstructed data matrix is ​​output.

2. The method according to claim 1, characterized in that The obtaining of sensor data collected by various types of sensors includes: Sensor data collected by various types of sensors are obtained through a sliding window of a preset time length, wherein the sensor data includes at least one of the flow, pressure, and temperature of steam and condensate generated by the ship power system during the power conversion process.

3. The method according to claim 1, characterized in that The obtaining of sensor data collected by various types of sensors includes: Sensor data collected by various types of sensors are acquired, and the sensor data are input into the convolutional neural network in a matrix form.

4. The method according to claim 1, characterized in that The method of performing fault identification on the sensor data through a convolutional neural network and determining the type of sensor fault includes detecting whether there is abnormal sensor data through a convolutional neural network, and classifying the sensor fault and determining the type of sensor fault when abnormal sensor data is detected.

5. The method according to claim 4, characterized in that The types of sensor faults include data missing faults, random noise and drift faults. When abnormal sensor data is detected, classifying the sensor faults and determining the types of sensor faults include: In the case where it is detected that the sensor data is missing data, determining that the sensor fault is a data missing fault; In the case where the sensor data is detected to be random data, determining that the sensor fault is random noise; When it is detected that the sensor data is drift data, it is determined that the sensor fault is a drift fault.

6. The method according to claim 1, characterized in that The convolutional neural network includes a first convolutional layer, a first maximum pooling layer, a second convolutional layer, a second maximum pooling layer, a first fully connected layer, a second fully connected layer and a softmax layer.

7. The method according to claim 1, characterized in that The step of replacing the sensor data collected by the failed sensor with the corrected data and outputting a reconstructed data matrix comprises: calculating a differential metric between sensor data collected by the failed sensor and the rectified data; When the differential metric is greater than a preset value, the sensor data collected by the faulty sensor is replaced with the corrected data, and a reconstructed data matrix is ​​output.

8. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images