Marine engine room data monitoring method and system based on wavelet transform analysis

Abstract

The invention discloses a marine engine room data monitoring method and system based on wavelet transform analysis, and the method comprises the steps: obtaining the operation data of marine engine room equipment, and carrying out the data preprocessing, and obtaining the preprocessed operation data of the marine engine room equipment; performing time sequence wavelet analysis on the preprocessed operation data of the marine engine room equipment to obtain a time sequence of the operation data of the marine engine room equipment; performing time sequence abrupt change point detection on the time sequence of the operation data of the marine engine room equipment, and determining a time abrupt change point of the operation data of the marine engine room equipment; and based on the time variation point of the operation data of the marine engine room equipment, combining a preset early warning mechanism to realize marine engine room data monitoring. According to the embodiment of the invention, the method can accurately find the potential fault of the equipment, evaluates the state of the equipment in a period of time in the future, and improves the safety of ship navigation. The method can be widely applied to the technical field of ship automation.

Description

Technical Field

[0001] This application relates to the technical field of ship automation, and particularly to a method and system for monitoring ship engine room data based on wavelet transform analysis. Background Art

[0002] With the development of ship automation and intelligent technologies, the navigation safety management of ocean-going ships increasingly depends on the real-time monitoring of the operating states of internal systems and equipment of ships. The existing ship remote monitoring systems mainly transmit the data collected by the ship engine room monitoring and alarm systems to shore-based monitoring terminals via maritime satellites, but they are limited to the over-limit alarm of thermal parameters, lack an intuitive display of the actual operating conditions of the equipment, and cannot meet the remote monitoring requirements of unmanned ships. Moreover, the monitoring and alarm systems in related technologies cannot effectively predict the faults or potential faults of the equipment and systems, making it difficult to detect abnormal situations in a timely manner. Once a fault occurs, it will seriously affect the safe navigation of the ship. In addition, its monitoring and alarm systems are limited to the over-limit alarm of thermal parameters, lack an intuitive display of the actual operating conditions of the equipment, and cannot meet the remote monitoring requirements of unmanned ships and evaluate the states of each equipment in a future period of time.

[0003] In summary, the technical problems existing in the related technologies need to be improved. Summary of the Invention

[0004] The main purpose of the embodiments of this application is to propose a method and system for monitoring ship engine room data based on wavelet transform analysis, which can accurately detect potential faults of equipment, evaluate the states of equipment in a future period of time, and thus improve the safety of ship navigation.

[0005] To achieve the above purpose, on the one hand, an embodiment of this application proposes a method for monitoring ship engine room data based on wavelet transform analysis, and the method includes:

[0006] Obtain the operation data of ship engine room equipment and perform data preprocessing to obtain the preprocessed operation data of ship engine room equipment;

[0007] Perform time series wavelet analysis on the preprocessed operation data of ship engine room equipment to obtain the time series of the operation data of ship engine room equipment;

[0008] Perform time series mutation point detection on the time series of the operation data of ship engine room equipment to determine the time mutation points of the operation data of ship engine room equipment;

[0009] Based on the time mutation points of the operation data of ship engine room equipment, combined with a preset early warning mechanism, realize the monitoring of ship engine room data.

[0010] In some embodiments, obtaining the operation data of the ship's engine room equipment and performing data preprocessing to obtain the preprocessed operation data of the ship's engine room equipment includes:

[0011] Obtaining the operation data of the ship's engine room equipment through sensors;

[0012] Performing data normalization processing on the operation data of the ship's engine room equipment through the mapminmax function to obtain the normalized operation data of the ship's engine room equipment;

[0013] Performing wavelet denoising processing on the normalized operation data of the ship's engine room equipment to obtain the preprocessed operation data of the ship's engine room equipment.

[0014] In some embodiments, performing time series wavelet analysis on the preprocessed operation data of the ship's engine room equipment to obtain the time series of the operation data of the ship's engine room equipment includes:

[0015] Performing discrete wavelet transform on the preprocessed operation data of the ship's engine room equipment according to a preset wavelet function to obtain the wavelet coefficients of the operation data of the ship's engine room equipment;

[0016] Drawing a coefficient diagram with the same level according to the wavelet coefficients of the operation data of the ship's engine room equipment;

[0017] Calculating the energy of the wavelet coefficients according to the coefficient diagram and extracting the periodic information to obtain the time series of the operation data of the ship's engine room equipment.

[0018] In some embodiments, the expression of the preset wavelet function is specifically as follows:

[0019]

[0020] In the above formula, ω f (·) represents the wavelet transform coefficient, a represents the scale parameter, b represents the translation parameter, f(t) represents the data to be analyzed, ψ(·) represents the mother wavelet, R represents the real number field, and t represents the time.

[0021] In some embodiments, performing time series mutation point detection on the time series of the operation data of the ship's engine room equipment to determine the time variation points of the operation data of the ship's engine room equipment includes:

[0022] Judging the periodic characteristics of the time series of the operation data of the ship's engine room equipment to obtain the time series of the operation data of the ship's engine room equipment with mutations;

[0023] Perform time series trend detection on the time series of the operation data of the ship's engine room equipment with mutations by the Mann-Kendall method in combination with the intersection of the forward and reverse sequences and the preset critical curve, and obtain the time series of the operation data of the ship's engine room equipment with significant change trends;

[0024] Decompose the time series of the operation data of the ship's engine room equipment with significant change trends through the Haar wavelet transform function and extract local features to obtain the time variation feature sequence of the operation data of the ship's engine room equipment;

[0025] Perform mutation point detection on the time variation feature sequence of the operation data of the ship's engine room equipment through the KS test algorithm to obtain the time variation points of the operation data of the ship's engine room equipment.

[0026] In some embodiments, the performing time series trend detection on the time series of the operation data of the ship's engine room equipment with mutations by the Mann-Kendall method in combination with the intersection of the forward and reverse sequences and the preset critical curve, and obtaining the time series of the operation data of the ship's engine room equipment with significant change trends includes:

[0027] Determine the time series region of significant change trends of the time series of the operation data of the ship's engine room equipment with mutations by the Mann-Kendall method;

[0028] Divide the time series of the operation data of the ship's engine room equipment with significant change trends into a positive sequence and a reverse sequence, and obtain the intersection of the positive sequence and the reverse sequence;

[0029] Define a critical value and draw the preset critical curve;

[0030] Based on the intersection and the preset critical curve, judge the time series region of significant change trends to obtain the time series of the operation data of the ship's engine room equipment with significant change trends.

[0031] In some embodiments, the realizing ship engine room data monitoring by combining with a preset warning mechanism based on the time variation points of the operation data of the ship's engine room equipment includes:

[0032] Determine the data end point based on the time series of the operation data of the ship's engine room equipment;

[0033] Connect the time variation point and the data end point, determine a straight line, and take the absolute value of the slope of the straight line to obtain the mutation rate;

[0034] Determine the alarm critical line according to the mutation rate;

[0035] According to a preset warning mechanism, combining the alarm critical line and the straight line, the monitoring of the data in the ship's engine room is realized.

[0036] In some embodiments, the realization of the monitoring of the data in the ship's engine room by combining the alarm critical line and the straight line according to a preset warning mechanism includes:

[0037] Set a first-level warning point and a second-level warning point to construct the preset warning mechanism;

[0038] Extend the straight line to intersect with the alarm critical line, and define the intersection point as the alarm point;

[0039] When the straight line reaches the first-level warning point, it is prompted that there is an abnormality in the data of the ship's engine room;

[0040] When the straight line reaches the second-level warning point, it is prompted that there is an abnormality in the data of the ship's engine room and equipment operation and maintenance are required;

[0041] When the straight line reaches the alarm point, it is prompted that there is an abnormality in the data of the ship's engine room and an alarm is issued to realize the monitoring of the data in the ship's engine room.

[0042] In some embodiments, the first-level warning point is located at the intermediate moment between the data end point and the alarm point, and the second-level warning point is located at the intermediate moment between the first-level warning point and the alarm point.

[0043] To achieve the above object, on the other hand, an embodiment of the present application proposes a ship engine room data monitoring system based on wavelet transform analysis. The system includes:

[0044] A first module for acquiring the operation data of the ship's engine room equipment and performing data preprocessing to obtain the preprocessed operation data of the ship's engine room equipment;

[0045] A second module for performing time series wavelet analysis on the preprocessed operation data of the ship's engine room equipment to obtain the time series of the operation data of the ship's engine room equipment;

[0046] A third module for detecting the time series mutation points of the time series of the operation data of the ship's engine room equipment to determine the time variation points of the operation data of the ship's engine room equipment;

[0047] A fourth module for realizing the monitoring of the data in the ship's engine room based on the time variation points of the operation data of the ship's engine room equipment and combining a preset warning mechanism.

[0048] The embodiments of the present application at least include the following beneficial effects: The present application provides a method and system for monitoring ship engine room data based on wavelet transform analysis. This solution improves the robustness of subsequent modeling by acquiring the operation data of ship engine room equipment and performing data preprocessing, and then performs time series wavelet analysis on the preprocessed operation data of ship engine room equipment. The time series wavelet analysis method is used to test the periodicity of the data, verify the existence of mutations in the sequence, and analyze the periodic intensity, providing preparation for subsequent data analysis. Further, it reduces the workload of data analysis and saves the time for data analysis. Then, it detects the time series mutation points of the operation data of ship engine room equipment, and combines with a preset warning mechanism to be able to timely and accurately discover the faults or potential faults of the equipment and system, accurately evaluate the state of each equipment in a future period of time, and issue early warning or alarm information at all levels, effectively improving the safety of ship navigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 is a flowchart of a method for monitoring ship engine room data based on wavelet transform analysis provided by an embodiment of the present application;

[0050] Figure 2 is a schematic structural diagram of a system for monitoring ship engine room data based on wavelet transform analysis provided by an embodiment of the present application;

[0051] Figure 3 is a schematic diagram of data early warning based on a preset warning mechanism provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of systems and methods consistent with some aspects of the embodiments of the present application detailed in the appended claims.

[0053] It will be understood that the terms "first", "second", etc. used in the present application may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if", "when" as used herein may be interpreted as "when...", "while...", or "in response to determining".

[0054] The terms "at least one", "a plurality of", "each", "any one", etc. used in the present application, at least one includes one, two or more than two, a plurality of includes two or more than two, each refers to each of the corresponding plurality, and any one refers to any one of the plurality.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0056] Refer to Figure 1 , Figure 1 is a flowchart of a method for monitoring ship engine room data based on wavelet transform analysis provided by an embodiment of the present invention. Refer to Figure 1 ,the method includes the following steps:

[0057] S100. Obtain the operation data of the ship engine room equipment and perform data preprocessing to obtain the preprocessed operation data of the ship engine room equipment;

[0058] It should be noted that in some embodiments, step S100 may include steps S110 to S130;

[0059] S110. Obtain the operation data of the ship engine room equipment through sensors;

[0060] In this embodiment, the operation data of the ship engine room equipment is collected. The operation parameter data of relevant equipment, such as temperature, pressure, flow rate, etc., can be collected through various sensors installed in the ship engine room. The collected data is sampled at a preset time interval to form time series data.

[0061] S120. Perform data normalization processing on the operation data of the ship engine room equipment through the mapminmax function to obtain the normalized operation data of the ship engine room equipment;

[0062] S130. Perform wavelet denoising processing on the normalized operation data of the ship engine room equipment to obtain the preprocessed operation data of the ship engine room equipment.

[0063] In some specific embodiments, the collected data is preprocessed, including data normalization and wavelet denoising. Among them, the data normalization uses the mapminmax function to convert the sample data to the range of [-1, 1], which is beneficial to subsequent data processing. The wavelet denoising uses the wavelet denoising method of digital signal processing technology to improve the robustness of subsequent modeling. Specifically, the Mallat algorithm and the db6 wavelet function are used to perform 8-layer wavelet analysis on the time spectrum, and the trend changes and abrupt components are extracted.

[0064] S200. Perform time series wavelet analysis on the preprocessed operation data of the ship's engine room equipment to obtain the time series of the operation data of the ship's engine room equipment;

[0065] Specifically, perform time series wavelet analysis on the preprocessed data to check the periodicity of the data, verify the existence of mutations in the sequence, and analyze the periodic intensity. Through wavelet analysis, the periodic characteristics of the data can be found, and whether there are mutation points can be judged.

[0066] It should be noted that in some embodiments, step S200 may include steps S210 to S230;

[0067] S210. Perform discrete wavelet transform on the preprocessed operation data of the ship's engine room equipment according to a preset wavelet function to obtain the wavelet coefficients of the operation data of the ship's engine room equipment;

[0068] S220. Draw a coefficient diagram with the same level according to the wavelet coefficients of the operation data of the ship's engine room equipment;

[0069] S230. Calculate the energy of the wavelet coefficients according to the coefficient diagram, extract the period information, and obtain the time series of the operation data of the ship's engine room equipment.

[0070] In some specific embodiments, wavelet transform is a representation method for decomposing a signal into different frequency components. It is similar to the Fourier transform but has the characteristic of time-frequency localization. This makes wavelet transform more effective in processing non-stationary signals (such as time series data, images, etc.).

[0071] The expression of the continuous wavelet transform function is specifically as follows:

[0072]

[0073] In the above formula, ω f (·) represents the wavelet transform coefficient, a represents the scale parameter, b represents the translation parameter, f(t) represents the data to be analyzed, ψ(·) represents the mother wavelet, R represents the real number field, and t represents the time.

[0074] Further, it should be noted that f(t) is the signal or function to be analyzed, ψ(t) is the mother wavelet, which is a function satisfying specific conditions. It is a double-window function, one is the time window and the other is the frequency spectrum. a is the scale parameter, which controls the stretching (i.e., frequency) of the wavelet function in the real number domain; b is the translation parameter, which controls the translation (i.e., time) of the wavelet function. ω f (a, b) are the wavelet transform coefficients, representing the wavelet translation components of the signal f(y) at frequency a and time b. It can be seen from this that the wavelet transform function is obtained by stretching and translating the mother wavelet.

[0075] Through wavelet theory, it can be found that wavelet transform can make the following analyses:

[0076] 1) The resolution is adjustable. When the local structure needs to be studied carefully, a very small part can be locally magnified. In this way, the local structure and its surrounding oscillation characteristics can be studied.

[0077] 2) Singularity point characterization. Singularity points may exist where the wavelet coefficients show oscillations. By combining the time positions of different scale changes of the sequence, a mutation signal can be provided for the system. That is, wavelet analysis can prepare for future data analysis.

[0078] 3) Using wavelet variance, it is possible to more accurately diagnose the vibration with the strongest period of a certain length. This periodicity discrimination, combined with wavelet variance, can infer the period length and the vibration degree within the period during a certain time period.

[0079] The main purpose of using wavelet analysis for engine room data is to test the periodicity of the data, verify the existence of mutations in the sequence, and analyze the period intensity. When analyzing a set or a series of data, the first thing to do is to test through its periodicity. If the data has a certain periodic distribution, the corresponding analysis only needs to analyze the data within one period, which reduces the workload and saves time to a certain extent. Especially when the data volume reaches several thousand or even tens of thousands of pieces, this periodic analysis is particularly important.

[0080] S300. Detect the time series mutation points of the operation data of the ship's engine room equipment to determine the time mutation points of the operation data of the ship's engine room equipment;

[0081] Specifically, for the data with mutations, adopt the time series mutation point detection method, such as the Mann-Kendall method, to determine the mutation points. The Mann-Kendall method combines the intersection points of the forward and reverse sequences and the critical curve to accurately determine the mutation points. At the same time, use the mutation point detection algorithm that combines the improved KS test and the Haar wavelet to achieve rapid and accurate detection of the mutation points in the data. Adopt a top-down search strategy and use the idea of halving to search, and quickly analyze the data information.

[0082] It should be noted that in some embodiments, step S300 may include steps S310 to S340;

[0083] S310. Periodically judge the characteristics of the time series of the operation data of the ship's engine room equipment, and obtain the time series of the operation data of the ship's engine room equipment with mutations;

[0084] S320. Detect the trend of the time series of the operation data of the ship's engine room equipment with mutations through the Mann-Kendall method in combination with the intersection point of the forward and reverse sequences and the preset critical curve, and obtain the time series of the operation data of the ship's engine room equipment with a significant change trend;

[0085] Furthermore, in some embodiments, step S320 may include steps S321 to S324;

[0086] S321. Determine the time series region of the significant change trend of the time series of the operation data of the ship's engine room equipment with mutations through the Mann-Kendall method;

[0087] S322. Divide the time series of the operation data of the ship's engine room equipment with a significant change trend into a positive sequence and a reverse sequence, and obtain the intersection point of the positive sequence and the reverse sequence;

[0088] S323. Define a critical value and draw a preset critical curve;

[0089] S324. Based on the intersection point and the preset critical curve, judge the time series region of the significant change trend to obtain the time series of the operation data of the ship's engine room equipment with a significant change trend.

[0090] In this embodiment, the Mann-Kendall test is defined, and the Mann-Kendall method is used to detect the trend of the time series. This method calculates a sign statistic (S) by comparing the magnitude relationship of any two observed values in the series to judge whether there is a significant upward or downward trend in the time series. Calculate the variance, and calculate the Z value according to the value of the statistic S and the variance. The larger (or smaller) the Z value, the more significant the trend.

[0091] Further judge the significance, and judge the significance of the trend according to the Z value and the corresponding critical value. If the Z value exceeds the critical value range, it indicates that there is a significant trend, which may point to a potential mutation point.

[0092] Construct forward and reverse sequences, and divide the time series into a positive sequence (arranged in chronological order) and a reverse sequence (arranged in reverse chronological order). The positive sequence reflects the natural development of the time series, while the reverse sequence can provide another perspective for detecting changes.

[0093] Calculate the intersection points. By comparing the values of the positive sequence and the reverse sequence, identify the intersection points. The intersection points represent the switching between the positive sequence and the reverse sequence at a certain moment and can be regarded as potential points of abnormal changes.

[0094] Set the critical curve and define the critical value. Set one or more critical values for evaluating whether the data points are abnormal. The critical value can usually be set by methods such as the mean, median, or standard deviation of historical data.

[0095] Plot the critical curve to visualize the critical value on the time series graph for visually identifying the upper and lower limits of the data points.

[0096] Combine the intersection points with the critical curve for judgment. By combining the intersection points, the results of the Mann-Kendall method, and the critical curve, identify the mutation points. For example, if a certain intersection point appears above the critical value or the trend is significant, then this point can be considered a mutation point.

[0097] S330. Decompose the time series of the operation data of the ship engine room equipment with a significant change trend through the Haar wavelet transform function and extract local features to obtain the time mutation feature sequence of the operation data of the ship engine room equipment;

[0098] S340. Detect the mutation points of the time mutation feature sequence of the operation data of the ship engine room equipment through the KS test algorithm to obtain the time mutation points of the operation data of the ship engine room equipment.

[0099] In some specific embodiments, the present invention detects mutation points by combining the Mann-Kendall method, the intersection points of the positive and reverse sequences, the critical curve, the improved KS test, and the Haar wavelet transform.

[0100] The first stage of mutation point detection: Mann-Kendall method;

[0101] First, perform trend detection. Apply the Mann-Kendall method to calculate the trend statistic, judge the overall trend of the data, and use it to initially identify possible mutation points. Calculate the Z value and the p value to determine the significance of the trend.

[0102] Further construct the positive and reverse sequences. Divide the data into a positive sequence and a reverse sequence. Calculate the intersection points between the positive and reverse sequences and mark the points where mutations may occur.

[0103] Even further, set the critical curve. Based on the mean and standard deviation of historical data, set the critical value and the critical curve. Visualize the critical curve in the time series and analyze the relationship between the intersection points and the critical value in combination with the Mann-Kendall statistical results to confirm possible mutation points.

[0104] The second stage of anomaly point detection: improved KS test and Haar wavelet;

[0105] First, perform Haar wavelet transform. Decompose the time series through Haar wavelet transform to extract the multi-scale information of the data. Use wavelet transform to capture the mutation characteristics in the data.

[0106] Furthermore, through the improved KS test, for the data after wavelet transform, introduce the improved Kolmogorov-Smirnov (KS) test to judge the location and significance of the mutation points. Identify the mutation points by testing the similarity of the data distribution.

[0107] S400. Based on the time anomaly points of the operation data of the ship's engine room equipment, combined with the preset warning mechanism, realize the monitoring of the ship's engine room data;

[0108] It should be noted that in some embodiments, step S400 may include steps S410 to S440;

[0109] S410. Determine the data end point based on the time series of the operation data of the ship's engine room equipment;

[0110] S420. Connect the time anomaly point and the data end point, determine a straight line, and take the absolute value of the slope of the straight line to obtain the mutation rate;

[0111] S430. Determine the alarm critical line according to the mutation rate;

[0112] S440. Based on the preset warning mechanism, combined with the alarm critical line and the straight line, realize the monitoring of the ship's engine room data.

[0113] In some specific embodiments, determine the mutation rate according to the anomaly point and predict the possible alarm time. Specifically, connect the anomaly point to the data end point to determine a straight line, and the absolute value of the slope of the straight line is the mutation rate. Extend the straight line to intersect with the limit line of over-limit alarm, and the intersection point is the predicted alarm time.

[0114] As Figure 3 shown, set the first-level warning point and the second-level warning point. The first-level warning point is located at the middle time between the data end point and the predicted alarm point, and the second-level warning point is located between the first-level warning point and the predicted alarm point. Among them, the first-level warning point prompts the management personnel that the data has a significant change trend; the second-level warning point reminds the manager that there is no sign of change in the trend and the equipment needs to be inspected.

[0115] In summary, the embodiments of the present invention have the following advantages compared with the prior art:

[0116] 1) It can timely and accurately detect faults or potential faults of equipment and systems, and issue early warning or alarm messages at all levels, providing timely and accurate technical guidance for ocean-going ships and effectively improving the safety of ship navigation.

[0117] 2) By analyzing the working characteristics and operating rules of relevant equipment and systems, applying mathematical methods to establish an alarm or early warning model, analyzing and processing the obtained ship engine room data, and discovering alarm or early warning information, it can meet the remote monitoring requirements of unmanned ships.

[0118] 3) The time series wavelet analysis method is used to test the periodicity of data, verify the existence of mutations in the sequence, analyze the cycle intensity, and prepare for subsequent data analysis; when mutations are found in the data, the time series mutation point detection method is used to determine the mutation points, the mutation rate is determined according to the mutation points, the possible alarm time is predicted, and the first-level early warning point and the second-level early warning point are set, improving the accuracy and timeliness of fault early warning.

[0119] 4) It can intuitively display the operating state of equipment by using technologies such as three-dimensional virtual reality, meet the requirements of remote monitoring, and improve the visualization level of the monitoring system.

[0120] Please refer to Figure 2 , the embodiment of the present application also provides a ship engine room data monitoring system based on wavelet transform analysis, which can implement the above-mentioned ship engine room data monitoring method based on wavelet transform analysis. The system includes:

[0121] The first module 201 is used to obtain the operation data of the ship engine room equipment and perform data preprocessing to obtain the preprocessed operation data of the ship engine room equipment;

[0122] The second module 202 is used to perform time series wavelet analysis on the preprocessed operation data of the ship engine room equipment to obtain the time series of the operation data of the ship engine room equipment;

[0123] The third module 203 is used to perform time series mutation point detection on the time series of the operation data of the ship engine room equipment to determine the time mutation points of the operation data of the ship engine room equipment;

[0124] The fourth module 204 is used to implement ship engine room data monitoring based on the time mutation points of the operation data of the ship engine room equipment in combination with a preset early warning mechanism.

[0125] It can be understood that the content in the above method embodiments is applicable to the system embodiments of the present application. The functions specifically implemented by the system embodiments of the present application are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0126] The preferred embodiments of the embodiments of the present application have been described above with reference to the accompanying drawings, which do not limit the scope of rights of the embodiments of the present application. Any modifications, equivalent replacements, and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall fall within the scope of rights of the embodiments of the present application.

Claims

1. A ship engine room data monitoring method based on wavelet transform analysis, characterized in that: The method comprises the following steps: Acquire the operation data of the ship engine room equipment and perform data preprocessing to obtain the preprocessed operation data of the ship engine room equipment; Performing time series wavelet analysis on the preprocessed ship engine room equipment operation data to obtain a time series of the ship engine room equipment operation data; Performing time series mutation point detection on the time series of the ship engine room equipment operation data to determine the time abnormal mutation point of the ship engine room equipment operation data; Based on the time variation points of the ship engine room equipment operation data and combined with a preset early warning mechanism, ship engine room data monitoring is achieved.

2. The method according to claim 1, characterized in that The obtaining of the operation data of the ship engine room equipment and performing data preprocessing to obtain the preprocessed operation data of the ship engine room equipment includes: Obtaining the operating data of ship engine room equipment through sensors; Performing data normalization processing on the operation data of the ship engine room equipment by using the mapminmax function, to obtain normalized operation data of the ship engine room equipment; The normalized ship engine room equipment operation data is subjected to wavelet denoising processing to obtain the pre-processed ship engine room equipment operation data.

3. The method according to claim 1, characterized in that The performing of time series wavelet analysis on the pre-processed ship engine room equipment operation data to obtain the time series of the ship engine room equipment operation data comprises: Performing discrete wavelet transform on the pre-processed ship engine room equipment operation data according to a preset wavelet function to obtain wavelet coefficients of the ship engine room equipment operation data; Drawing a coefficient graph with the same level according to the wavelet coefficients of the ship engine room equipment operation data; The energy of the wavelet coefficients is calculated according to the coefficient graph, the period information is extracted, and the time series of the ship engine room equipment operation data is obtained.

4. The method according to claim 3, characterized in that The expression of the preset wavelet function is specifically as follows: In the above formula, ω f (·) represents the wavelet transform coefficient, a represents the scale parameter, b represents the translation parameter, f(t) represents the data to be analyzed, ψ(·) represents the mother wavelet, R represents the real number domain, and t represents the time.

5. The method according to claim 1, characterized in that The performing time series mutation point detection on the time series of the ship engine room equipment operation data to determine the time abnormality point of the ship engine room equipment operation data includes: Performing periodic feature judgment on the time series of the ship engine room equipment operation data to obtain the time series of the ship engine room equipment operation data with mutations; Performing time series trend detection on the time series of the ship engine room equipment operation data with mutations by using the Mann-Kendall method and combining the intersection of the positive and negative sequences with a preset critical curve to obtain the time series of the ship engine room equipment operation data with significant change trends; Decomposing the time series of the ship engine room equipment operation data with a significant change trend and extracting local features through the Haar wavelet transform function, to obtain a time variation feature series of the ship engine room equipment operation data; The KS test algorithm is used to perform mutation point detection on the time variation characteristic sequence of the ship engine room equipment operation data to obtain the time variation point of the ship engine room equipment operation data.

6. The method according to claim 5, characterized in that The time series trend detection is performed on the time series of the ship engine room equipment operation data with mutations by using the Mann-Kendall method and combining the intersection of the positive and negative sequences with the preset critical curve to obtain the time series of the ship engine room equipment operation data with significant change trends, including: Determine the significant change trend time series area of ​​the time series of the ship engine room equipment operation data with mutation by the Mann-Kendall method; Dividing the time series of the ship engine room equipment operation data with a significant change trend into a positive series and a reverse series, and obtaining the intersection of the positive series and the reverse series; Defining a critical value and drawing the preset critical curve; Based on the intersection point and the preset critical curve, the time series area of ​​the significant change trend is judged to obtain the time series of the ship engine room equipment operation data with the significant change trend.

7. The method according to claim 1, characterized in that The time variation point based on the ship engine room equipment operation data is combined with a preset early warning mechanism to realize ship engine room data monitoring, including: Determining a data endpoint based on a time series of the ship engine room equipment operation data; Connect the time mutation point and the data end point to determine a straight line, and take the absolute value of the slope of the straight line to obtain the mutation rate; Determine an alarm critical line according to the mutation rate; According to the preset early warning mechanism, the alarm critical line and the straight line are combined to realize the ship engine room data monitoring.

8. The method according to claim 7, characterized in that The method of implementing ship engine room data monitoring based on the preset early warning mechanism and combining the alarm critical line with the straight line includes: Setting up first-level warning points and second-level warning points to build the preset warning mechanism; Extending the straight line to intersect the alarm critical line, and defining the intersection point as the alarm point; When the straight line reaches the first-level warning point, it indicates that the ship's engine room data is abnormal; When the straight line reaches the second-level warning point, it indicates that the ship's engine room data is abnormal and equipment operation and maintenance is required; When the straight line reaches the alarm point, it is prompted that the ship engine room data is abnormal and an alarm is issued, thereby realizing ship engine room data monitoring.

9. The method according to claim 8, characterized in that The first-level warning point is located at the middle time between the data end point and the alarm point, and the second-level warning point is located at the middle time between the first-level warning point and the alarm point.

10. A ship engine room data monitoring system based on wavelet transform analysis, characterized in that: The system comprises: The first module is used to obtain the operation data of the ship's engine room equipment and perform data preprocessing to obtain the preprocessed ship's engine room equipment operation data; The second module is used to perform time series wavelet analysis on the pre-processed ship engine room equipment operation data to obtain the time series of the ship engine room equipment operation data; The third module is used to perform time series mutation point detection on the time series of the ship engine room equipment operation data to determine the time abnormal change point of the ship engine room equipment operation data; The fourth module is used to realize ship engine room data monitoring based on the time change points of the ship engine room equipment operation data combined with a preset early warning mechanism.

Images