A smart ship power grid parameter display and analysis device

Abstract

This application discloses an intelligent ship power grid parameter display and analysis device. It collects and monitors ship power grid parameters in real time, including three-phase voltage, current, active power, reactive power, apparent power, frequency, and power factor, and displays them visually on a display screen. Simultaneously, a data processing and analysis algorithm is introduced at the back end to perform a time-series collaborative analysis of these ship power grid parameters. This allows for visual display of the ship power grid status and anomaly detection, providing effective reference information for engine room personnel. This facilitates the timely discovery and resolution of potential problems in the ship power grid, ensuring its normal operation.

Description

Technical Field

[0001] The present application relates to the field of intelligent detection, and more specifically, to an intelligent ship power grid parameter display and analysis device. Background Art

[0002] The ship's power grid is a vital component of a vessel, providing power to various devices and ensuring their normal operation. However, due to its complexity and uncertainty, the ship's power grid may exhibit anomalies such as voltage fluctuations, frequency deviations, and harmonic distortion. These anomalies can affect the stability and reliability of the ship's power grid and even cause equipment damage or accidents. Therefore, real-time display and analysis of ship power grid parameters is a crucial means of improving the safety and efficiency of ship power grids.

[0003] However, traditional grid parameter analysis methods usually rely on dashboards and data loggers for display, and rely on threshold monitoring of various grid parameter data to detect and diagnose grid faults. This method cannot pay attention to the correlation and temporal changes between multiple grid parameters, and thus cannot detect grid quality anomalies in a timely manner, which is not conducive to the diagnosis and prevention of grid faults.

[0004] Therefore, an intelligent ship power grid parameter display and analysis device is desired. Summary of the Invention

[0005] In order to solve the above-mentioned technical problems, the present application is proposed. An embodiment of the present application provides an intelligent ship power grid parameter display and analysis device, which collects ship power grid parameters, including three-phase voltage, current, active power, reactive power, apparent power, frequency, and power factor, through real-time monitoring and visualization on a display screen. At the same time, a data processing and analysis algorithm is introduced at the back end to perform a time-series collaborative analysis of these ship power grid parameters, thereby achieving a visual display of the ship power grid status and anomaly detection, providing effective reference information for engine room personnel. This can facilitate the timely discovery and resolution of potential problems in the ship power grid, ensuring the normal operation of the ship power grid.

[0006] According to one aspect of the present application, a smart ship power grid parameter display and analysis device is provided, comprising:

[0007] A ship power grid parameter data acquisition module is used to obtain a time series of ship power grid parameters, wherein the ship power grid parameters include three-phase voltage, current, active power, reactive power, apparent power, frequency and power factor;

[0008] a ship power grid parameter embedding coding module, configured to extract features of each ship power grid parameter in the time series of the ship power grid parameter by using a power grid parameter embedding encoder based on a deep neural network to obtain a sequence of ship power grid parameter embedding coding feature vectors;

[0009] a nonlinear response optimization module, configured to perform nonlinear response optimization on each ship power grid parameter embedded coding feature vector in the sequence of ship power grid parameter embedded coding feature vectors to obtain a sequence of optimized ship power grid parameter embedded coding feature vectors;

[0010] a joint clustering analysis module, configured to perform a joint clustering analysis on the sequence of the optimized ship power grid parameter embedded coding feature vectors to obtain clustered ship power grid parameter semantic features;

[0011] The ship power grid state detection module is used to determine whether the ship power grid state is abnormal based on the clustered ship power grid parameter semantic features.

[0012] Compared to existing technologies, the intelligent ship grid parameter display and analysis device provided in this application monitors and collects ship grid parameters, including three-phase voltage, current, active power, reactive power, apparent power, frequency, and power factor, in real time, and displays them visually on a display screen. Simultaneously, a data processing and analysis algorithm is introduced at the back end to perform a time-series collaborative analysis of these ship grid parameters. This enables visual display of the ship grid status and anomaly detection, providing effective reference information for engine room personnel. This facilitates the timely identification and resolution of potential issues with the ship grid, ensuring its normal operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The above and other purposes, features, and advantages of the present application will become more apparent through a more detailed description of the embodiments of the present application in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the drawings, the same reference numerals generally represent the same components or steps.

[0014] Figure 1 1 is a block diagram of a device for displaying and analyzing parameters of a smart ship power grid according to an embodiment of the present application;

[0015] Figure 2 This is a system architecture diagram of a smart ship power grid parameter display and analysis device according to an embodiment of the present application;

[0016] Figure 3 This is a block diagram of the training phase of the smart ship power grid parameter display and analysis device according to an embodiment of the present application.

[0017] Figure 4 Schematic diagram of the power supply and communication modules in the hardware configuration of the smart ship power grid parameter display and analysis device according to an embodiment of the present application.

[0018] Figure 5 This is a schematic diagram of a circuit block diagram in the hardware configuration of a smart ship power grid parameter display and analysis device according to an embodiment of the present application.

[0019] Figure 6 This is a schematic diagram of a main control display panel in the hardware configuration of the smart ship power grid parameter display and analysis device according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0021] As used in this application and the claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not intended to refer to the singular but may include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0022] Although the present application makes various references to certain modules in the system according to embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are illustrative only, and different aspects of the system and method can use different modules.

[0023] Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the various steps may be processed in reverse order or simultaneously, as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0024] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described herein.

[0025] Traditional grid parameter analysis methods typically rely on dashboards and data loggers for display, and rely on threshold monitoring of individual grid parameter data to detect and diagnose grid faults. This approach fails to address the correlations and temporal variations between multiple grid parameters, making it difficult to detect grid quality anomalies in a timely manner, hindering the diagnosis and prevention of grid faults. Therefore, a device for displaying and analyzing smart ship grid parameters is desired.

[0026] In the technical solution of the present application, a smart ship power grid parameter display and analysis device is proposed. Figure 1 4 is a block diagram of a smart ship power grid parameter display and analysis device according to an embodiment of the present application. Figure 2 FIG. 1 is a system architecture diagram of a smart ship power grid parameter display and analysis device according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the intelligent ship power grid parameter display and analysis device according to an embodiment of the present application includes: a ship power grid parameter data acquisition module 310, which is used to obtain a time series of ship power grid parameters, wherein the ship power grid parameters include three-phase voltage, current, active power, reactive power, apparent power, frequency and power factor; a ship power grid parameter embedding coding module 320, which is used to extract features of each ship power grid parameter in the time series of the ship power grid parameters through a grid parameter embedding encoder based on a deep neural network to obtain a sequence of ship power grid parameter embedded coding feature vectors; a nonlinear response optimization module 330, which is used to perform nonlinear response optimization on each ship power grid parameter embedded coding feature vector in the sequence of ship power grid parameter embedded coding feature vectors to obtain a sequence of optimized ship power grid parameter embedded coding feature vectors; a joint clustering analysis module 340, which is used to perform joint clustering analysis on the sequence of optimized ship power grid parameter embedded coding feature vectors to obtain clustered ship power grid parameter semantic features; and a ship power grid status detection module 350, which is used to determine whether the ship power grid status is abnormal based on the clustered ship power grid parameter semantic features.

[0027] In particular, the ship power grid parameter data acquisition module 310 is used to obtain a time series of ship power grid parameters, wherein the ship power grid parameters include three-phase voltage, current, active power, reactive power, apparent power, frequency, and power factor. It should be understood that in a specific example of the present application, low voltage may cause difficulty in starting the motor, equipment overload, and power system instability. High voltage may damage equipment and increase energy consumption. Unbalanced three-phase voltage may cause unstable operation and damage to equipment; excessively high or low current may indicate equipment failure, overload, or short circuit; abnormal current may cause equipment damage, increased fire risk, and power system instability; abnormal active power may indicate that the power system is overloaded or the equipment is faulty. Abnormal active power may cause equipment overheating, overload, and reduced system efficiency; abnormal reactive power may cause the power factor of the power system to decrease, affecting the stability and efficiency of the system. Abnormal reactive power may cause voltage fluctuations and equipment damage; abnormal apparent power may indicate that the total power demand of the power system exceeds the system capacity, which may cause grid overload, voltage instability and equipment damage; abnormal frequency may indicate unbalanced load in the power system or generator operation problems; abnormal frequency may cause equipment damage, grid instability and equipment performance reduction. A reduced power factor may indicate that there is a large amount of reactive power in the power system, affecting system efficiency and equipment performance. Abnormal power factor may cause low grid efficiency and equipment damage. Therefore, in the technical solution of the present application, it collects ship power grid parameters, including three-phase voltage, current, active power, reactive power, apparent power, frequency and power factor, through real-time monitoring and visualization on the display screen, and introduces data processing and analysis algorithms at the back end to perform time-series collaborative analysis of these ship power grid parameters, so as to achieve visual display of the ship power grid status and abnormality detection, and provide effective reference information for engine room personnel. Specifically, first, the time series of ship power grid parameters is obtained.

[0028] In particular, the ship grid parameter embedding encoding module 320 is configured to extract features from each ship grid parameter in the time series of the ship grid parameters using a grid parameter embedding encoder based on a deep neural network to obtain a sequence of ship grid parameter embedding encoding feature vectors. Considering that the ship grid parameters include three-phase voltage, current, active power, reactive power, apparent power, frequency, and power factor, and that these data exhibit a dynamic temporal change pattern in the time dimension, and that there are mutual temporal synergistic correlations between these ship grid parameters, it is necessary to perform embedding association analysis on the ship grid parameters at each time point in order to capture the temporal synergistic correlation semantics between these parameters. Specifically, in the technical solution of the present application, each ship grid parameter in the time series of the ship grid parameters is processed through a grid parameter embedding encoder based on a fully connected layer to obtain a sequence of ship grid parameter embedding encoding feature vectors. Processing by the grid parameter embedding encoder based on a fully connected layer enables mapping each ship grid parameter at each time point into the same space and capturing the semantic association features between these ship grid parameters, which helps to better capture the complex interactions and influences between grid parameters. It's worth mentioning that a fully connected layer is a common layer type in deep learning neural networks, also known as a densely connected layer. Each neuron in a fully connected layer is connected to all neurons in the previous layer, and there is a weight between each input and each output.

[0029] Accordingly, in one possible implementation, each ship power grid parameter in the time series of the ship power grid parameter can be respectively embedded in an encoder based on a fully connected layer to obtain a sequence of ship power grid parameter embedded coding feature vectors through the following steps, for example: inputting the time series of the ship power grid parameter; performing necessary preprocessing on the time series of the ship power grid parameter, such as normalization, standardization or other data processing operations, so as to better apply it to the neural network model; designing a fully connected layer encoder, which takes the ship power grid parameter at each time step as input and outputs the corresponding embedded coding feature vector. The encoder may include one or more fully connected layers, each followed by an activation function to introduce nonlinearity; for each time step of the ship power grid parameter, it is sequentially input into the fully connected layer encoder for processing; the embedded coding feature vectors obtained at each time step are combined into a feature vector sequence in chronological order to obtain a sequence of ship power grid parameter embedded coding feature vectors.

[0030] In particular, the nonlinear response optimization module 330 is configured to perform nonlinear response optimization on each of the ship power grid parameter embedded encoding feature vectors in the sequence of ship power grid parameter embedded encoding feature vectors to obtain a sequence of optimized ship power grid parameter embedded encoding feature vectors. Considering that different sensors may have differences in sensitivity, response speed, and other aspects, resulting in deviations in the collected data, and that the data collected by the sensors also exhibits nonlinear responses, in order to eliminate or reduce measurement errors between different sensors during data processing and perform nonlinear response correction to ensure data accuracy and reliability, the technical solution of the present application further performs nonlinear response optimization on each of the ship power grid parameter embedded encoding feature vectors in the sequence of ship power grid parameter embedded encoding feature vectors to obtain a sequence of optimized ship power grid parameter embedded encoding feature vectors. It should be understood that the correction processing of the nonlinear response optimization can eliminate the differences and nonlinear effects between these sensors, avoid redundant or unnecessary information in the feature vectors, and reduce information loss, thereby making the data more accurate and reliable, further improving the expressive power of the feature vectors and enabling them to better capture the complex relationships and patterns in the data, which helps to improve the distinguishability and characterization capabilities of the associated features of the ship power grid parameter data. Specifically, in a specific example of the present application, nonlinear response optimization is performed on each ship power grid parameter embedded coding feature vector in the sequence of the ship power grid parameter embedded coding feature vector using the following nonlinear response optimization formula to obtain the sequence of optimized ship power grid parameter embedded coding feature vectors; wherein the nonlinear response optimization formula is:

[0031]

[0032] in, is the eigenvalue of the i-th position of the s-th ship power grid parameter embedded coding feature vector in the sequence of the ship power grid parameter embedded coding feature vector, A, B, C and D are adjustment hyperparameters, v i It is the eigenvalue of the i-th position of the s-th optimized ship power grid parameter embedded coding feature vector in the optimized ship power grid parameter embedded coding feature vector.

[0033] In particular, the joint clustering analysis module 340 is configured to perform a joint clustering analysis on the sequence of optimized ship power grid parameter embedded coded feature vectors to obtain clustered ship power grid parameter semantic features. It should be understood that since each optimized ship power grid parameter embedded coded feature vector in the sequence of optimized ship power grid parameter embedded coded feature vectors represents the semantic associations and features between each ship power grid parameter at different time points, in order to gain a deeper understanding of the semantic associations and temporal synergistic relationships between different ship power grid parameters at each time point, thereby revealing the grid state characteristics represented by the temporal synergistic changes of the ship power grid parameters, in the technical solution of the present application, the sequence of optimized ship power grid parameter embedded coded feature vectors is further input into a joint clustering analysis network to obtain clustered ship power grid parameter semantic feature vectors. It should be understood that the joint clustering analysis network can cluster similar ship power grid parameter feature vectors together to form group features with certain semantic meanings, which facilitates more detailed analysis of power grid parameters and dissemination of semantic understanding of power grid status. In other words, joint clustering analysis can help discover potential hidden patterns and structures in data, thereby revealing the correlations and regularities between different power grid parameters. The semantic feature vector of the ship power grid parameter obtained by clustering can help identify different power grid states and modes, thereby more accurately judging whether there is an abnormality in the power grid. Specifically, in a specific example of the present application, the sequence of the optimized ship power grid parameter embedded coding feature vector is input into the joint clustering analysis network to obtain the clustered ship power grid parameter semantic feature vector as the clustered ship power grid parameter semantic feature, including: inputting the sequence of the optimized ship power grid parameter embedded coding feature vector into the joint clustering analysis network to process it with the following clustering formula to obtain the clustered ship power grid parameter semantic feature vector; wherein, the clustering formula is:

[0034]

[0035]

[0036] Among them, v i (x) and v j (x) are respectively the i-th and j-th optimized ship power grid parameter embedded coding feature vectors in the sequence of the optimized ship power grid parameter embedded coding feature vectors, V k is the sequence of the optimized ship power grid parameter embedded coding feature vectors, log represents the logarithmic function value with base 2, L is the length of each optimized ship power grid parameter embedded coding feature vector, M is the number of vectors in the sequence of the optimized ship power grid parameter embedded coding feature vector - 1, D i is the eigenvalue of each position in the ship power grid parameter time series semantic fluctuation eigenvector, Ns is the length of the ship power grid parameter time series semantic fluctuation feature vector, exp(·) is the exponential operation, and V is the clustered ship power grid parameter semantic feature vector.

[0037] In particular, the ship power grid state detection module 350 is used to determine whether there is an abnormality in the ship power grid state based on the clustered ship power grid parameter semantic features. In particular, in a specific example of the present application, the clustered ship power grid parameter semantic feature vector is passed through a classifier-based power grid state analyzer to obtain an analysis result, and the analysis result is used to indicate whether there is an abnormality in the ship power grid state. That is, the clustered semantic features of the ship power grid parameters are classified and processed to analyze the power grid state, and the analysis results are visualized on the display screen so that the engine room personnel can promptly discover and solve potential problems of the ship power grid and ensure the normal operation of the ship power grid. Specifically, the clustered ship power grid parameter semantic feature vector is fully connected encoded using multiple fully connected layers of the classifier to obtain an encoded classification feature vector; and the encoded classification feature vector is passed through the Softmax classification function of the classifier to obtain the classification result.

[0038] That is, in the technical solution of the present application, the labels of the classifier include whether the ship power grid state is abnormal (first label), and whether the ship power grid state is not abnormal (second label), wherein the classifier determines to which classification label the clustered ship power grid parameter semantic feature vector belongs through a soft maximum function. It is worth noting that the first label p1 and the second label p2 here do not contain artificially set concepts. In fact, during the training process, the computer model does not have the concept of "whether the ship power grid state is abnormal". It only has two classification labels and the probability of the output feature under these two classification labels, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether the ship power grid state is abnormal is actually converted into a binary classification class probability distribution that conforms to natural laws through the classification label. In essence, what is used is the physical meaning of the natural probability distribution of the label, rather than the linguistic text meaning of "whether the ship power grid state is abnormal".

[0039] A classifier is a machine learning model or algorithm that is used to classify input data into different categories or labels. Classifiers are part of supervised learning and perform classification tasks by learning a mapping from input data to output categories.

[0040] The Softmax classification function is a commonly used activation function for multi-classification problems. It converts each element of the input vector into a probability value between 0 and 1, where the sum of these probabilities equals 1. The Softmax function is often used in the output layer of neural networks and is particularly well-suited for multi-classification problems because it maps the network output into a probability distribution for each category. During training, the output of the Softmax function is used to calculate the loss function and update the network parameters through the backpropagation algorithm. It is worth noting that the output of the Softmax function does not change the relative size of the elements; it simply normalizes them. Therefore, the Softmax function does not change the characteristics of the input vector; it simply converts it into a probability distribution.

[0041] It should be understood that before using the above-mentioned neural network model for inference, it is necessary to train the fully connected layer-based grid parameter embedding encoder, the joint clustering analysis network, and the classifier-based grid status analyzer. In other words, the smart ship grid parameter display and analysis device 300 according to the present application also includes a training stage 400 for training the fully connected layer-based grid parameter embedding encoder, the joint clustering analysis network, and the classifier-based grid status analyzer.

[0042] Figure 3 FIG. 1 is a block diagram of the training phase of the smart ship power grid parameter display and analysis device according to an embodiment of the present application. Figure 3As shown, the intelligent ship power grid parameter display and analysis device 300 according to the embodiment of the present application includes: a training stage 400, including: a training data acquisition module 410, for acquiring training data, wherein the training data includes a time series of training ship power grid parameters, wherein the training ship power grid parameters include three-phase voltage, current, active power, reactive power, apparent power, frequency and power factor; a training parameter embedding coding module 420, for extracting features of each training ship power grid parameter in the time series of the training ship power grid parameters by using a grid parameter embedding encoder based on a fully connected layer to obtain a sequence of training ship power grid parameter embedding coding feature vectors; a training nonlinear response optimization module 430, for extracting features of each training ship power grid parameter embedding coding feature vector in the sequence of the training ship power grid parameter embedding coding feature vectors. Linear response optimization is used to obtain a sequence of optimized training ship power grid parameter embedded coding feature vectors; a training joint clustering analysis module 440 is used to input the sequence of optimized training ship power grid parameter embedded coding feature vectors into the joint clustering analysis network to obtain a training clustered ship power grid parameter semantic feature vector; an optimization module 450 is used to perform clustering optimization on the training clustered ship power grid parameter semantic feature vector to obtain an optimized training clustered ship power grid parameter semantic feature vector; a classification loss module 460 is used to pass the optimized training clustered ship power grid parameter semantic feature vector through a classifier-based power grid status analyzer to obtain a classification loss function value; a training module 470 is used to train the fully connected layer-based power grid parameter embedding encoder, the joint clustering analysis network and the classifier-based power grid status analyzer based on the classification loss function value.

[0043] Among them, the training module is used to cluster the eigenvalues ​​of the training clustered ship power grid parameter semantic feature vector based on the distance between the eigenvalues ​​to obtain a cluster feature set; and optimize the training clustered ship power grid parameter semantic feature vector based on the feature in-class and out-of-class representations in the cluster feature set to obtain the optimized training clustered ship power grid parameter semantic feature vector.

[0044] In particular, in the technical solution of the present application, each optimized training ship power grid parameter embedded coding feature vector in the sequence of optimized training ship power grid parameter embedded coding feature vectors respectively represents the nonlinear response optimization presentation of the fully connected embedded coding features of the ship power grid parameters at each predetermined time point. However, considering that the ship power grid parameters at each predetermined time point have obvious source domain data distribution differences in the data source domain, this causes the sequence of optimized training ship power grid parameter embedded coding feature vectors to have inconsistency in embedded coding feature distribution. In this way, when the sequence of optimized training ship power grid parameter embedded coding feature vectors is input into the joint clustering analysis network, the joint clustering analysis network can capture the clustering features of the sequence of optimized training ship power grid parameter embedded coding feature vectors. However, due to the inconsistency in embedded coding feature distribution in the sequence of optimized training ship power grid parameter embedded coding feature vectors, the obtained training clustered ship power grid parameter semantic feature vector will have more significant local feature distribution discreteness. In this way, when the training clustered ship power grid parameter semantic feature vector is used as a whole for classification and regression through a classifier, the local feature distribution discreteness of the training clustered ship power grid parameter semantic feature vector will lead to difficulty in converging towards the predetermined class label when the classifier is used for classification and regression, thereby affecting the training speed of the classifier and the accuracy of the final classification result. Based on this, the applicant of this application performs clustering optimization on the training clustered ship power grid parameter semantic feature vector, that is, first clustering the various eigenvalues ​​of the training clustered ship power grid parameter semantic feature vector, for example, clustering based on the distance between eigenvalues, and then optimizing based on the in-class and out-of-class representation of the clustered features, expressed as:

[0045]

[0046] Wherein, f is each feature value of the training clustered ship power grid parameter semantic feature vector, n is the number of feature sets corresponding to the training clustered ship power grid parameter semantic feature vector, k is the number of cluster features, and C represents the cluster feature set. Specifically, by treating the in-class features and out-of-class features of the training clustered ship power grid parameter semantic feature vector as different instance roles to perform a class instance description based on clustering proportional distribution, and introducing a clustering response history based on in-class and out-of-class dynamic context, a coordinated global perspective is maintained on the in-class distribution and out-of-class distribution of the overall features of the training clustered ship power grid parameter semantic feature vector, so that the optimized feature clustering operation of the training clustered ship power grid parameter semantic feature vector can maintain a consistent response of the in-class and out-of-class features, thereby maintaining a consistent regression convergence path based on feature clustering in the classification and regression process, improving the convergence effect of the training clustered ship power grid parameter semantic feature vector towards the predetermined target classification label, and improving the training speed of the classifier and the accuracy of the classification results. In this way, the status of the ship's power grid can be visualized and abnormalities detected, providing effective reference information for engine room personnel. In other words, this is conducive to timely discovery and resolution of potential problems in the ship's power grid, ensuring the normal operation of the ship's power grid.

[0047] Hardware Description:

[0048] In particular, in the technical solution of this application, the smart ship power grid parameter display and analysis device is a digital and visual integrated device at the hardware level, which includes the main generator switchboard, emergency switchboard, and main control board and other devices related to power grid data, among which some hardware devices are shown in FIG. Figures 4 to 6 This helps shipboard engine personnel to provide an effective basis for visual display of power grid data and analysis of power grid quality.

[0049] Specifically, the hardware configuration of the smart ship power grid parameter display and analysis device includes:

[0050] 1. Use high-definition 8-inch TFT LCD display.

[0051] 2. Two-way independent isolated RS485 communication circuit structure.

[0052] 3. 8-way switch output with configurable alarm type.

[0053] 4. Independent real-time voltage and current waveform display clearly shows the quality of the grid voltage and current at that time.

[0054] 5. Display of all grid parameters, alarm status, and generator operating time.

[0055] 6. Use the Fast Fourier Transform algorithm to calculate and display the 1st to 31st harmonic components.

[0056] 7. Use fast unbalance algorithm to calculate and display the unbalance of voltage and current.

[0057] This patent shows that the ship power grid data parameters are complete, there are many alarm information, many harmonic orders, real-time voltage and current waveform visualization, the operation interface is clearly and intuitively displayed in Chinese and English, the communication rate is high, the transmission distance is long, and the anti-interference ability is strong.

[0058] While various embodiments of the present disclosure have been described above, the above descriptions are illustrative, non-exhaustive, and not intended to be limiting of the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A smart ship power grid parameter display and analysis device, characterized in that: include: A ship power grid parameter data acquisition module is used to obtain a time series of ship power grid parameters, wherein the ship power grid parameters include three-phase voltage, current, active power, reactive power, apparent power, frequency and power factor; a ship power grid parameter embedding coding module, configured to extract features of each ship power grid parameter in the time series of the ship power grid parameter by using a power grid parameter embedding encoder based on a deep neural network to obtain a sequence of ship power grid parameter embedding coding feature vectors; a nonlinear response optimization module, configured to perform nonlinear response optimization on each ship power grid parameter embedded coding feature vector in the sequence of ship power grid parameter embedded coding feature vectors to obtain a sequence of optimized ship power grid parameter embedded coding feature vectors; a joint clustering analysis module, configured to perform a joint clustering analysis on the sequence of the optimized ship power grid parameter embedded coding feature vectors to obtain clustered ship power grid parameter semantic features; A ship power grid state detection module, configured to determine whether the ship power grid state is abnormal based on the clustered ship power grid parameter semantic features; The nonlinear response optimization module is configured to perform nonlinear response optimization on each ship power grid parameter embedded coding feature vector in the sequence of ship power grid parameter embedded coding feature vectors using the following nonlinear response optimization formula to obtain the sequence of optimized ship power grid parameter embedded coding feature vectors; Wherein, the nonlinear response optimization formula is: in, is the eigenvalue of the i-th position of the s-th ship power grid parameter embedded coding feature vector in the sequence of the ship power grid parameter embedded coding feature vector, A, B, C and D are adjustment hyperparameters, v i It is the eigenvalue of the i-th position of the s-th optimized ship power grid parameter embedded coding feature vector in the optimized ship power grid parameter embedded coding feature vector.

2. The intelligent ship power grid parameter display and analysis device according to claim 1, characterized in that: The power grid parameter embedding encoder based on deep neural network is a power grid parameter embedding encoder based on fully connected layer.

3. The intelligent ship power grid parameter display and analysis device according to claim 2, characterized in that: The joint clustering analysis module is used to: input the sequence of the optimized ship power grid parameter embedded coding feature vectors into the joint clustering analysis network to obtain the clustered ship power grid parameter semantic feature vector as the clustered ship power grid parameter semantic feature.

4. The intelligent ship power grid parameter display and analysis device according to claim 3, characterized in that: The joint clustering analysis module is used to: input the sequence of the optimized ship power grid parameter embedded coding feature vectors into the joint clustering analysis network and process it according to the following clustering formula to obtain the clustered ship power grid parameter semantic feature vector; Wherein, the clustering formula is: Among them, v i (x) and v j (x) are respectively the i-th and j-th optimized ship power grid parameter embedded coding feature vectors in the sequence of the optimized ship power grid parameter embedded coding feature vectors, V k is the sequence of the optimized ship power grid parameter embedded coding feature vectors, log represents the logarithmic function value with base 2, L is the length of each optimized ship power grid parameter embedded coding feature vector, M is the number of vectors in the sequence of the optimized ship power grid parameter embedded coding feature vector - 1, D i is the eigenvalue of each position in the ship power grid parameter time series semantic fluctuation eigenvector, N s is the length of the ship power grid parameter time series semantic fluctuation feature vector, exp(·) is the exponential operation, and V is the clustered ship power grid parameter semantic feature vector.

5. The intelligent ship power grid parameter display and analysis device according to claim 4, characterized in that: The ship power grid state detection module is used to: pass the clustered ship power grid parameter semantic feature vector through a classifier-based power grid state analyzer to obtain an analysis result, and the analysis result is used to indicate whether the ship power grid state is abnormal.

6. The intelligent ship power grid parameter display and analysis device according to claim 5, characterized in that: It also includes a training module for training the fully connected layer-based power grid parameter embedding encoder, the joint clustering analysis network and the classifier-based power grid status analyzer.

7. The intelligent ship power grid parameter display and analysis device according to claim 6, characterized in that: The training module includes: A training data acquisition module is used to acquire training data, wherein the training data includes a time series of training ship power grid parameters, wherein the training ship power grid parameters include three-phase voltage, current, active power, reactive power, apparent power, frequency and power factor; a training parameter embedding coding module, configured to extract features of each training ship power grid parameter in the time series of the training ship power grid parameters by using a power grid parameter embedding encoder based on a fully connected layer to obtain a sequence of training ship power grid parameter embedding coding feature vectors; A training nonlinear response optimization module is used to perform nonlinear response optimization on each training ship power grid parameter embedded coding feature vector in the sequence of training ship power grid parameter embedded coding feature vectors to obtain a sequence of optimized training ship power grid parameter embedded coding feature vectors; A training joint clustering analysis module is used to input the sequence of the optimized training ship power grid parameter embedded coding feature vector into the joint clustering analysis network to obtain a training cluster ship power grid parameter semantic feature vector; An optimization module, configured to perform clustering optimization on the training clustered ship power grid parameter semantic feature vector to obtain an optimized training clustered ship power grid parameter semantic feature vector; A classification loss module, configured to pass the optimized trained clustered ship power grid parameter semantic feature vector through a classifier-based power grid state analyzer to obtain a classification loss function value; A training module is used to train the fully connected layer-based power grid parameter embedding encoder, the joint clustering analysis network and the classifier-based power grid status analyzer based on the classification loss function value.

8. The intelligent ship power grid parameter display and analysis device according to claim 7, characterized in that: The training module is used to: Clustering each eigenvalue of the training clustered ship power grid parameter semantic feature vector based on the distance between eigenvalues ​​to obtain a cluster feature set; The training clustered ship power grid parameter semantic feature vector is optimized based on the feature in-class and out-of-class representations in the clustered feature set to obtain the optimized training clustered ship power grid parameter semantic feature vector.

Images