A method and device for detecting the synchronous working state of double-drive motors of an electric ship

Abstract

This invention discloses a method and apparatus for detecting the synchronous operation of dual drive motors in an electric boat, relating to the field of electric boat technology. The method includes: acquiring the motor time-frequency diagrams corresponding to the rotor electrical signals of the dual drive motors; segmenting the motor time-frequency diagrams along the time axis, and then further segmenting them sequentially along both the time and frequency axes to output multiple first motor sub-region time-frequency diagrams and multiple second motor sub-region time-frequency diagrams; filtering each sub-region using its respective filtering parameters and then performing vector transformation to determine the corresponding first motor vector and second motor vector; constructing a linear discriminant hyperplane based on the amplitude time gradient of each motor sub-region time-frequency diagram; and determining that the dual drive motors of the electric boat are synchronously operating if the centroids of the first motor vectors and the second motor vectors are located on the same side of the linear discriminant hyperplane. This scheme improves the reliability of synchronous detection of dual drive motors in electric boats.

Description

Technical Field

[0001] This invention relates to the field of electric boat technology, and in particular to a method and apparatus for detecting the synchronous operation of dual drive motors in an electric boat. Background Technology

[0002] As core actuators of a ship's propulsion system, the synchronous operation of dual drive motors in electric boats directly impacts the ship's propulsion efficiency, operational stability, and navigation safety. Therefore, real-time online monitoring of the synchronous operation status of these motors is of paramount importance. Existing detection methods rely on rotational speed sensors for synchronization, but improper sensor installation or malfunction can lead to significant errors in the data. Furthermore, multi-data fusion methods based on vibration acceleration calculations and current sensor signals, when using BP neural network algorithms, fail to consider noise interference, suffer from algorithmic complexity, and incur high computational costs. Therefore, there is an urgent need to improve the reliability of synchronous detection for dual drive motors in electric boats. Summary of the Invention

[0003] This invention provides a method and apparatus for detecting the synchronous operation of dual drive motors in an electric boat, which improves the reliability of synchronous detection of dual drive motors in an electric boat.

[0004] The first aspect of this invention provides a method for detecting the synchronous operation state of dual drive motors in an electric boat, comprising: The rotor electrical signals of the first electric boat drive motor and the second electric boat drive motor are obtained according to the sampling duration, and the corresponding time-frequency diagrams of the first motor and the second motor are determined. The time-frequency diagrams of the first motor and the second motor are divided along the time axis according to the first time segmentation length to construct multiple sub-time-frequency diagrams of the first motor and multiple sub-time-frequency diagrams of the second motor. For any first motor sub-time-frequency diagram and any second motor sub-time-frequency diagram, divide them sequentially along the time axis and frequency axis according to the second time segmentation length and frequency segmentation length, and output multiple first motor sub-region time-frequency diagrams of each first motor sub-time-frequency diagram and multiple second motor sub-region time-frequency diagrams of each second motor sub-time-frequency diagram; For any first motor sub-region time-frequency diagram and any second motor sub-region time-frequency diagram, respectively, the corresponding filtering parameters are used for filtering and vector transformation is performed to determine the corresponding first motor vector and second motor vector; A linear discriminant hyperplane is constructed based on the amplitude time gradient of the time-frequency maps of each of the first motor sub-regions and each of the second motor sub-regions; The centroid of the first motor is determined by calculating the centroid of each of the first motor vectors, and the centroid of the second motor is determined by calculating the centroid of each of the second motor vectors. If they belong to the same side of the centroid located on the linear discrimination hyperplane, then it is determined that the dual drive motors of the electric boat are working synchronously.

[0005] Further, the construction of a linear discriminant hyperplane based on the amplitude-time gradients of the time-frequency maps of each of the first motor sub-regions and each of the second motor sub-regions includes: Determine the time gradient of the first motor amplitude at the position of the maximum amplitude frequency in the time-frequency diagram of each first motor sub-region; The vector summation of the first motor amplitude time gradient associated with each of the first motor sub-time-frequency maps is performed to construct the corresponding first motor fusion gradient; The fusion gradient of each of the first motors is projected onto different preset mutually orthogonal directions to determine the discrimination weight; Calculate the second motor amplitude time gradient at the position of the largest amplitude frequency in the time-frequency map of each second motor sub-region and perform vector summation to output the total fused gradient of the second motor. Project the total fusion gradient of the second motor onto the fusion gradient of the first motor with the maximum gradient magnitude to determine the discrimination bias; A linear discriminant hyperplane is constructed using the discriminant weights and the discriminant biases.

[0006] Furthermore, the process of determining the length of the first time segment includes: Determine whether the difference between the maximum and minimum frequency values ​​in the first motor time-frequency diagram or the second motor time-frequency diagram is greater than a preset frequency threshold; If it is greater than 1, then the range of values ​​for the number of timeline segments is determined as follows: ; If none of them are greater than, then the range of values ​​for the number of time axis segments is determined as follows: ; The first time segment length is determined by calculating the ratio between the sampling duration and the number of time axis segments in one operation. in, Indicates the number of timeline segments in one operation. Indicates the sampling duration.

[0007] Furthermore, the process of determining the frequency segmentation length includes: when When the frequency axis segmentation number is within the range of values, the value range is: ; when When the frequency axis segmentation number is within the range of values, the value range is: ; The ratio of the maximum frequency value of the first motor time-frequency diagram to the number of frequency axis segments is taken as the corresponding frequency segment length; The ratio of the maximum frequency value of the second motor time-frequency diagram to the number of frequency axis segments is taken as the corresponding frequency segment length; in, Indicates the number of frequency axis segments. This indicates the maximum frequency value.

[0008] Furthermore, the process of determining the filtering parameters includes: The first motor sub-region time-frequency diagram or the second motor sub-region time-frequency diagram is used as the input to the filter parameter prediction model, and the corresponding filter parameters are output.

[0009] Furthermore, it also includes: For any first motor sub-region time-frequency diagram and any second motor sub-region time-frequency diagram, after filtering with the corresponding filtering parameters, the corresponding first motor filtered sub-region time-frequency diagram and second motor filtered sub-region time-frequency diagram are determined. Based on the location correspondence, the time-frequency maps of the first and second motor filter sub-regions are sequentially measured to determine the distance between the corresponding sub-regions. The total distance is determined by weighted summation of the distances to each of the sub-regions. When the total distance is less than or equal to a preset distance anomaly threshold, the dual drive motors of the electric boat are determined to operate synchronously.

[0010] Furthermore, the process of determining the distance weight of the sub-region distance includes: The time-frequency map of the first motor filter sub-region or the time-frequency map of the second motor filter sub-region is used as the input to the weight determination model, and the corresponding distance weight is output.

[0011] The second aspect of the present invention provides a detection device for the synchronous operation state of dual drive motors of an electric boat, comprising: The sampling transformation module is used to acquire the rotor electrical signals of the first electric boat drive motor and the second electric boat drive motor according to the sampling duration, and to determine the corresponding time-frequency diagrams of the first motor and the second motor. The first segmentation module is used to segment the first motor time-frequency diagram and the second motor time-frequency diagram along the time axis according to the first time segmentation length, and construct multiple first motor sub-time-frequency diagrams and multiple second motor sub-time-frequency diagrams; The second segmentation module is used to segment any first motor sub-time-frequency diagram and any second motor sub-time-frequency diagram sequentially along the time axis and frequency axis according to the second time segmentation length and frequency segmentation length, and output multiple first motor sub-region time-frequency diagrams of each first motor sub-time-frequency diagram and multiple second motor sub-region time-frequency diagrams of each second motor sub-time-frequency diagram; The filtering and classification module is used to filter the time-frequency map of any first motor sub-region and any second motor sub-region using their respective filtering parameters, and then perform vector transformation to determine the corresponding first motor vector and second motor vector. The plane construction module is used to construct a linear discriminant hyperplane based on the amplitude time gradient of the time-frequency maps of each of the first motor sub-regions and each of the second motor sub-regions; The synchronization discrimination module is used to determine the centroid of the first motor by calculating the centroid of each of the first motor vectors, and to determine the centroid of the second motor by calculating the centroid of each of the second motor vectors. If they belong to the same side centroid located on the linear discrimination hyperplane, then it is determined that the dual drive motors of the electric boat are working synchronously.

[0012] A computer device provided in a third aspect of the present invention includes a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method for detecting the synchronous working state of dual drive motors of an electric boat as described in any of the preceding claims.

[0013] The fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed, it implements the method for detecting the synchronous working state of dual drive motors of an electric boat as described in any of the preceding claims.

[0014] As can be seen from the above technical solutions, the present invention has the following advantages: The above-described solution of the present invention provides a method for detecting the synchronous operation state of dual drive motors of an electric boat, comprising: acquiring the rotor electrical signals of a first electric boat drive motor and a second electric boat drive motor according to the sampling duration, and determining the corresponding time-frequency diagrams of the first motor and the second motor; dividing the first motor time-frequency diagram and the second motor time-frequency diagram along the time axis according to a first time segmentation length, constructing multiple first motor sub-time-frequency diagrams and multiple second motor sub-time-frequency diagrams; dividing any first motor sub-time-frequency diagram and any second motor sub-time-frequency diagram sequentially along the time axis and the frequency axis according to the second time segmentation length and the frequency segmentation length, and outputting multiple first motor sub-time-frequency diagrams. Multiple time-frequency maps of the first motor sub-region and each second motor sub-region are generated. For any first motor sub-region time-frequency map and any second motor sub-region time-frequency map, the corresponding filtering parameters are used for filtering and vector transformation to determine the corresponding first motor vector and second motor vector. A linear discriminant hyperplane is constructed based on the amplitude time gradient of each first motor sub-region time-frequency map and each second motor sub-region time-frequency map. The centroid of the first motor is determined by calculating the centroid of each first motor vector, and the centroid of the second motor is determined by calculating the centroid of each second motor vector. If they belong to the same side centroid located on the linear discriminant hyperplane, it is determined that the dual drive motors of the electric boat are working synchronously. Based on the above scheme, by constructing a time-frequency map of the rotor electrical signals of the dual-drive motors and performing local region segmentation analysis, the differences in the operating status of the dual-drive motors in different time periods and frequency ranges can be extracted more meticulously. This effectively captures the minute differences between the two motors in local time periods and frequency ranges, and has higher detection sensitivity for early anomalies, weak differences, and instantaneous changes. The entire process takes the rotor electrical signals of the dual-drive motors as input, and completes the detection of the synchronous working status of the dual motors through time-frequency transformation, region segmentation and filtering, and linear discrimination hyperplane same-side discrimination. It does not require additional complex sensors or the construction of overly complicated multi-source information fusion models, and has the advantages of low implementation cost, easy engineering deployment and promotion, while improving the reliability of the synchronous discrimination of dual-drive motors. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 The flowchart illustrates the steps of a method for detecting the synchronous operation of dual drive motors in an electric boat, as provided in Embodiment 1 of the present invention. Figure 2This is a structural block diagram of a device for detecting the synchronous working state of dual drive motors of an electric boat, provided in Embodiment 2 of the present invention. Detailed Implementation

[0017] This invention provides a method and apparatus for detecting the synchronous operation of dual drive motors in an electric boat, which improves the reliability of synchronous detection of dual drive motors in an electric boat.

[0018] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0019] Please see Figure 1 The present invention provides a method for detecting the synchronous operation of dual drive motors in an electric boat, comprising: Step 101: Obtain the rotor electrical signals of the first electric boat drive motor and the second electric boat drive motor according to the sampling duration, and determine the corresponding time-frequency diagrams of the first motor and the second motor.

[0020] Sampling duration refers to the duration for which a segment of electrical signal is acquired.

[0021] It should be noted that for the two identical drive motors on the same electric vessel, referred to as the first electric vessel drive motor I and the second electric vessel drive motor II, voltage sensors or current sensors are respectively arranged on the rotor side of the two drive motors to measure the sampling duration during the same working period. The corresponding rotor voltage or rotor current signals are collected as the rotor electrical signals in this embodiment, and the signals are processed using short-time Fourier transform to obtain the time-frequency diagrams of the first motor corresponding to drive motor I. Time-frequency diagram of the second motor corresponding to drive motor II The motor time-frequency diagram represents the time-frequency characteristic distribution of the motor within the current time period.

[0022] Step 102: Divide the time-frequency diagram of the first motor and the time-frequency diagram of the second motor along the time axis according to the first time segmentation length, and construct multiple sub-time-frequency diagrams of the first motor and multiple sub-time-frequency diagrams of the second motor.

[0023] Time segment length refers to the length of time for data to be segmented along the time dimension. Terms such as "first," "second," etc., are used to distinguish different data.

[0024] It should be noted that this embodiment does not perform an overall comparison of the entire time-frequency graph, but rather considers region segmentation detection. To this end, firstly, the time axes of the first motor time-frequency graph and the second motor time-frequency graph are respectively divided according to the set first time segmentation length. By segmenting, a series of first motor sub-time-frequency diagrams with the same time width are obtained. Second motor sub-time frequency diagram , Indicates the number of sub-time-frequency maps. Indicates the first The first motor sub-time frequency diagram. Indicates the first The second motor's time-frequency diagram.

[0025] In one specific embodiment of this example, the process of determining the first time segment length includes: Determine whether the difference between the maximum and minimum frequency values ​​in the time-frequency diagram of the first motor or the second motor is greater than a preset frequency threshold. If it is greater than 1, then the range of values ​​for the number of timeline segments is determined as follows: ; If none of them are greater than, then the range of values ​​for the number of time axis segments is determined as follows: ; The length of the first time segment is determined by calculating the ratio between the sampling duration and the number of time axis segments in one operation.

[0026] The number of timeline segments refers to the amount of data obtained by dividing the data along the time dimension. Terms such as "first-order" and "second-order" are used to distinguish different numbers of timeline segments.

[0027] It should be noted that the first time segment length is ,in, Indicates the number of timeline segments in one operation. This indicates the sampling duration; in a preferred implementation, the number of time axis segments is determined by the maximum frequency value of either the first or second motor time-frequency graph. Minimum frequency value With preset frequency threshold Between satisfy hour, The value range is less than or equal to one-fifth of the sampling duration, and conversely, it is greater than one-fifth of the sampling duration.

[0028] Step 103: For any first motor sub-time-frequency diagram and any second motor sub-time-frequency diagram, divide them sequentially along the time axis and frequency axis according to the second time segmentation length and frequency segmentation length, and output multiple first motor sub-region time-frequency diagrams of each first motor sub-time-frequency diagram and multiple second motor sub-region time-frequency diagrams of each second motor sub-time-frequency diagram.

[0029] Frequency segment length refers to the length of the frequency segmentation of data in the frequency dimension.

[0030] It should be noted that, in order to more effectively capture subtle differences between the two motors within local time periods and local frequency ranges, and to improve the detection sensitivity for local anomalies and early asynchronous states, this embodiment performs a second segmentation of the time-frequency diagrams of each motor. During the segmentation process, the diagrams are sequentially segmented along both the time axis and frequency axis of each motor sub-time-frequency diagram. The second time segmentation length is... for Frequency segmentation length for , Indicates the number of times the time axis is divided twice. Indicates the number of frequency axis segments; For example, the multiple first motor sub-region time-frequency maps segmented by each first motor sub-time-frequency map and the multiple second motor sub-region time-frequency maps segmented by each second motor sub-time-frequency map can be represented as follows: ; ; In the formula, Indicates the first The first motor sub-frequency diagram The frequency segment length and the first The time-frequency diagram of the first motor sub-region corresponding to the second time segment length. Indicates the first The second motor sub-time frequency diagram of the first The frequency segment length and the first The time-frequency diagrams of the second motor sub-regions corresponding to each second time segment length, with a time axis width of [missing information]. The width of the frequency axis is .

[0031] In one specific embodiment of this example, the process of determining the frequency segmentation length includes: when When the frequency axis segmentation number is within the range of values, the value range is: ; when When the frequency axis segmentation number is within the range of values, the value range is: ; The ratio of the maximum frequency value of the first motor's time-frequency diagram to the number of frequency axis segments is taken as the corresponding frequency segment length; The ratio of the maximum frequency value of the second motor's time-frequency diagram to the number of frequency axis segments is taken as the corresponding frequency segment length.

[0032] Frequency axis segmentation number refers to the number of data points obtained by segmenting along the frequency dimension.

[0033] It should be noted that, The value of is any positive integer greater than or equal to 1; in a preferred implementation, The range of values The range of values ​​for is determined, that is, when ... The value of the frequency axis segmentation number is greater than or equal to one-third of the maximum frequency value. The value of the frequency axis segmentation number is less than one-third of the maximum frequency value; it can be understood that the first motor's time-frequency diagram and the second motor's sub-time-frequency diagram use the same frequency axis segmentation number, i.e. To simultaneously satisfy the maximum frequency value of either the first motor time-frequency diagram or the second motor time-frequency diagram. The determined range of values.

[0034] Step 104: For any first motor sub-region time-frequency diagram and any second motor sub-region time-frequency diagram, filter them with their respective filtering parameters and then perform vector transformation to determine the corresponding first motor vector and second motor vector.

[0035] Filtering parameters refer to the parameters that play a core role in the filtering effect of a filtering algorithm.

[0036] It should be noted that in this embodiment, the time-frequency maps of each first motor sub-region and each second motor sub-region are filtered to obtain the corresponding first motor filtered sub-region time-frequency maps and second motor filtered sub-region time-frequency maps. Vector transformation is then performed on each first motor filtered sub-region time-frequency map and each second motor filtered sub-region time-frequency map to obtain the corresponding multi-dimensional feature vectors, referred to as the first motor vector and the second motor vector. The obtained first motor vector and second motor vector can be represented as follows: ; In the formula, Indicates the first The first motor sub-frequency diagram The first motor vector of the time-frequency map of the first motor filter sub-region. Indicates the first The second motor sub-time frequency diagram of the first The second motor vector of the second motor filter sub-region time-frequency map.

[0037] It is understood that there are multiple ways to implement the above vector transformation, and this embodiment does not limit it. For example, it can be directly mapped into a multi-dimensional vector based on the data distribution characteristics of the time-frequency diagram of the motor filter sub-region, or the multi-dimensional operating condition vector can be determined by identifying the operating condition characteristics based on the time-frequency diagram of the motor filter sub-region. Here, the operating condition vector refers to the vector that characterizes the operating state of the motor, such as stator voltage, stator current, stator temperature, rotor voltage, rotor current, rotor temperature, and motor output torque. In a preferred implementation, the vector prediction model is input to the respective filtering parameters for vector transformation. The vector prediction model takes the time-frequency map as the model input and the motor vector as the model output. That is, the time-frequency maps of each first motor filtering sub-region and each second motor filtering sub-region are input into the vector prediction model for feature processing, thereby obtaining the first motor vector and the second motor vector representing the time-frequency map of each motor filtering sub-region. For example, a perceptron neural network can be used and the corresponding number of output layer neurons can be set to build the vector prediction model. A public training set is constructed using classic time-frequency maps and corresponding category labels, and the model is trained to determine the trained vector prediction model.

[0038] In one specific embodiment of this example, the process of determining the filtering parameters includes: The time-frequency diagram of the first motor sub-region or the time-frequency diagram of the second motor sub-region is used as the input to the filter parameter prediction model, and the corresponding filter parameters are output.

[0039] A filter parameter prediction model refers to a prediction model that takes a time-frequency graph as input and filter parameters as output.

[0040] It should be noted that in this embodiment, the same filtering algorithm is used to filter each motor sub-region's time-frequency graph individually. During the filtering process, the filtering parameters used for each motor sub-region's time-frequency graph are specified separately. The resulting filtered motor sub-region time-frequency graph can be represented as follows: ; ; In the formula, Indicates the first The first motor sub-frequency diagram The time-frequency diagram of the first motor sub-region corresponding to the time-frequency diagram of the first motor sub-region. Indicates the first The second motor sub-time frequency diagram of the first The time-frequency diagram of the second motor sub-region corresponding to the time-frequency diagram of the second motor sub-region; In a preferred implementation of this embodiment, the filtering parameters are predicted and determined by a filtering parameter prediction model built based on a neural network model. A filtering training set is constructed from historical running samples using classic time-frequency maps with different noise levels and texture features, along with corresponding filtering parameter labels. Model training is then conducted to establish a mapping relationship between time-frequency map features and filtering parameters, resulting in a trained filtering parameter prediction model. The model is then input with a time-frequency map of the motor sub-region, processed, and the corresponding filtering parameters are output for filtering. For example, a perceptron neural network can be used, with a set number of neurons in the output layer. It is understood that this embodiment does not limit the filtering algorithm; the filtering parameters can be adaptively set according to different filtering parameters. For example, when the filtering algorithm is a Gaussian filtering algorithm, the filtering standard deviation can be used as the filtering parameter.

[0041] Step 105: Construct a linear discriminant hyperplane based on the amplitude time gradient of the time-frequency maps of each first motor sub-region and each second motor sub-region.

[0042] It should be noted that this embodiment considers using a linear discriminant hyperplane to determine the motor synchronization state. Based on this, the expression for the linear discriminant hyperplane is used... , Indicates the first The discrimination weights of the time-frequency diagrams of individual motors Indicates the first The input motor sub-time-frequency diagram, The discriminant bias is indicated by the fact that, in this embodiment, the discriminant weights and discriminant biases in the linear discriminant hyperplane are determined based on the amplitude time gradient of the first motor sub-region time-frequency map and the second motor sub-region time-frequency map at a specific location, so as to determine the linear discriminant hyperplane.

[0043] In one specific embodiment of this example, step 105 includes the following sub-steps: Determine the time gradient of the first motor amplitude at the position of the maximum amplitude frequency in the time-frequency diagram of each first motor sub-region; The vector summation of the time gradient of the first motor amplitude associated with each first motor sub-time-frequency map is performed to construct the corresponding first motor fusion gradient; The fusion gradient of each first motor is projected onto different preset mutually orthogonal directions to determine the discrimination weight; Calculate the second motor amplitude time gradient at the position of the largest amplitude frequency in the time-frequency map of each second motor sub-region and perform vector summation to output the total fused gradient of the second motor. Project the total fusion gradient of the second motor onto the fusion gradient of the first motor with the maximum gradient magnitude to determine the discrimination bias; A linear discriminant hyperplane is constructed using the discriminant weights and discriminant biases.

[0044] The position of maximum amplitude frequency refers to the specific location corresponding to the point with the largest amplitude.

[0045] It should be noted that in the linear discriminant hyperplane Based on the time-frequency diagram of the first motor respectively Calculations show that Based on the time-frequency diagram of the second motor Calculations show that, in a preferred implementation, the specific position is preferably the position with the maximum amplitude frequency: Calculate the time-frequency diagram of any first motor sub-module respectively. Gradients corresponding to the maximum amplitude frequency positions in each region , This indicates the sub-time-frequency diagram index of the motor. Indicates the first The first motor sub-frequency diagram The frequency segment length and the first The time-frequency diagram of the first motor sub-region corresponding to the second time segment length. Represents the time-frequency diagram of the first motor sub-region The first motor amplitude time gradient; Using the time-frequency diagram of the first motor as the unit, the time gradients of the amplitudes of each first motor are vector-summed: ; In the formula, Indicates the first The first motor fusion gradient of the first motor sub-time frequency map. Indicates the frequency segment length index. This indicates the index of the second time segment length. Indicates the first The first motor sub-frequency diagram The frequency segment length and the first The first motor amplitude time gradient of the time frequency map of the first motor sub-region corresponding to the second time segment length; Preset A mutually orthogonal direction , will each Towards a pre-set direction The projection is performed, and the projected value is... ; calculate Gradients corresponding to the maximum amplitude frequency positions in each region , Represents the time-frequency diagram of the second motor sub-region The second motor amplitude time gradient; The time gradients of the amplitudes of each of the second motors are vector-summed: ; In the formula, This represents the total fusion gradient of the second motor. Indicates the first The second motor sub-time frequency diagram of the first The frequency segment length and the first The second motor amplitude time gradient of the time frequency map of the second motor sub-region corresponding to the second time segment length; Will The direction with the largest magnitude in the amplitude-time gradient direction of the first motor Projection, the value of the projection is .

[0046] Step 106: Determine the centroid of the first motor by using the centroid of each first motor vector, and determine the centroid of the second motor by using the centroid of each second motor vector. If they belong to the same side centroid located on the linear discriminant hyperplane, then determine that the dual drive motors of the electric boat are working synchronously.

[0047] It should be noted that, for Finding the center of mass of the first motor ,right Finding the centroid of the second motor ; if and On the same side of the linear discriminant hyperplane, it is considered that drive motor I and the second electric boat drive motor II are in a state of synchronous operation; otherwise, it is considered that drive motor I and the second electric boat drive motor II are not in a state of synchronous operation. Further, it is determined whether monitoring needs to continue. If so, step 101 is executed to continue to collect the rotor electrical signal in the next sampling time and perform short-time Fourier transform, time-frequency graph segmentation and other processes, so as to realize continuous online detection of the synchronous operation state of the two drive motors.

[0048] It is understood that the parameters such as sampling duration, number of time axis segments, length of the first time segment, length of frequency segment, number of frequency axis segments, and threshold can be adaptively set in this embodiment, and this embodiment does not impose any restrictions.

[0049] In another specific embodiment of this example, after step 103, the method further includes: For any first motor sub-region time-frequency diagram and any second motor sub-region time-frequency diagram, after filtering with the corresponding filtering parameters, the corresponding first motor filtered sub-region time-frequency diagram and second motor filtered sub-region time-frequency diagram are determined. Based on the location correspondence, the time-frequency maps of the first and second motor filter sub-regions are sequentially measured to determine the distance between the corresponding sub-regions. The total distance is determined by weighted summation of the distances to each sub-region. When the total distance is less than or equal to the preset distance anomaly threshold, the dual drive motors of the electric boat are determined to work synchronously.

[0050] It should be noted that, in addition to the linear discriminant hyperplane judgment mechanism, this embodiment also considers a distance metric judgment mechanism based on the time-frequency map of the motor filter sub-region; specifically: Based on the filtered time-frequency maps of the first and second motor filter sub-regions, distance measurements are performed on the time-frequency maps of the motor filter sub-regions corresponding to the positions of drive motor I and the second electric boat drive motor II, respectively. ; In the formula, Represents the time-frequency diagram of the first motor filter sub-region Time-frequency diagram of the second motor filter sub-region Subregion distance, Indicates row index, Indicates the total number of rows. Indicates column index, Indicates the total number of columns. Represents the time-frequency diagram of the first motor filter sub-region The first in Line number List the values ​​at each data point. Represents the time-frequency diagram of the second motor filter sub-region The first in Line number List the values ​​at each data point; Multiply the distances of all sub-regions by their respective distance weights and then sum them using a weighted method: ; In the formula, Indicates the total distance. Indicates the distance of the sub-region Distance weights, Represents the time-frequency diagram of the first motor filter sub-region Time-frequency diagram of the second motor filter sub-region Subregion distance, Indicates the first The first motor sub-frequency diagram The time-frequency diagram of the first motor sub-region corresponding to the time-frequency diagram of the first motor sub-region. Indicates the first The second motor sub-time frequency diagram of the first The time-frequency diagram of the second motor sub-region corresponding to the time-frequency diagram of the second motor sub-region; Will Distance anomaly threshold If a comparison is made, Greater than It is then assumed that the operating states of drive motor I and the second electric boat drive motor II are out of sync, triggering an alarm. Less than or equal to It is assumed that drive motor I and the second electric boat drive motor II are operating synchronously. In one specific implementation of this embodiment, the process of determining the distance weight includes: using the time-frequency map of the first motor filter sub-region or the time-frequency map of the second motor filter sub-region as input to the weight determination model, and outputting the corresponding distance weight.

[0051] It should be noted that the time-frequency map of the motor filtering sub-region can be extracted from typical synchronous and asynchronous state samples. The distance weight category is labeled according to the importance of each sub-region to the final synchronous state determination, forming a weight training set. The weight determination model is trained using this weight training set to establish the mapping relationship between the features of the time-frequency map of the filtered sub-region and the distance metric weight, thus obtaining the trained weight determination model. For example, a perceptron neural network or other neural network can be used to build the weight determination model.

[0052] Understandably, since different sub-regions have different degrees of importance in determining synchronization status, it can highlight the role of key time-frequency regions in the detection results. By adaptively determining the distance weight of each sub-region, the subjectivity brought about by manual weight allocation can be reduced, and the final judgment result can be more in line with the actual operating state, which helps to improve the objectivity and robustness of the detection method.

[0053] In this embodiment of the invention, by constructing a time-frequency map of the rotor electrical signals of the dual-drive motor and performing local region segmentation analysis, the differences in the operating state of the dual-drive motor in different time periods and frequency ranges can be extracted more meticulously. This effectively captures the minute differences between the dual motors in local time periods and frequency ranges, avoiding the problem of the overall comparison method being insensitive to local anomalies. Therefore, it has higher detection sensitivity for early anomalies, weak differences, and instantaneous changes. By adaptively determining the filtering parameters to adaptively adjust the filtering intensity according to the time-frequency characteristics of different sub-regions, the filtering process is made more consistent with the actual feature distribution of each local region, avoiding the filtering inconsistencies caused by unreasonable parameter settings when using fixed filtering parameters. This method addresses the issue of insufficient or excessive filtering, thereby improving adaptability to different operating conditions and signal characteristics. It can more accurately retain key features reflecting the operating status of dual motors while suppressing the interference of noise and irrelevant information on the synchronization status discrimination results. The entire process uses the rotor electrical signals of the dual drive motors as input, and completes the detection of the synchronous working status of dual motors through time-frequency transformation, region segmentation and filtering, as well as linear discrimination hyperplane same-side discrimination or distance metric threshold discrimination. It does not require additional complex sensors or the construction of overly cumbersome multi-source information fusion models, and has the advantages of low implementation cost, easy engineering deployment and promotion, while improving the accuracy and effectiveness of dual drive motor synchronization discrimination.

[0054] Please see Figure 2 The present invention provides a detection device for the synchronous operation of dual drive motors of an electric boat, comprising: The sampling and transformation module 201 is used to acquire the rotor electrical signals of the first electric boat drive motor and the second electric boat drive motor according to the sampling duration, and determine the corresponding time-frequency diagram of the first motor and the time-frequency diagram of the second motor. The first segmentation module 202 is used to segment the time-frequency diagram of the first motor and the time-frequency diagram of the second motor along the time axis according to the first time segmentation length, and construct multiple first motor sub-time-frequency diagrams and multiple second motor sub-time-frequency diagrams; The second segmentation module 203 is used to segment any first motor sub-time-frequency diagram and any second motor sub-time-frequency diagram along the time axis and frequency axis in sequence according to the second time segmentation length and frequency segmentation length, and output multiple first motor sub-region time-frequency diagrams of each first motor sub-time-frequency diagram and multiple second motor sub-region time-frequency diagrams of each second motor sub-time-frequency diagram; The filtering and classification module 204 is used to filter the time-frequency diagram of any first motor sub-region and any second motor sub-region using their respective filtering parameters, and then perform vector transformation to determine the corresponding first motor vector and second motor vector. The plane construction module 205 is used to construct a linear discriminant hyperplane based on the amplitude time gradient of the time-frequency maps of each first motor sub-region and each second motor sub-region. The synchronization discrimination module 206 is used to determine the centroid of the first motor by calculating the centroid of each first motor vector and to determine the centroid of the second motor by calculating the centroid of each second motor vector. If they belong to the same side centroid located on the linear discrimination hyperplane, then it is determined that the dual drive motors of the electric boat are working synchronously.

[0055] In one specific embodiment of this example, a distance synchronization discrimination module is further included, used for: For any first motor sub-region time-frequency diagram and any second motor sub-region time-frequency diagram, after filtering with the corresponding filtering parameters, the corresponding first motor filtered sub-region time-frequency diagram and second motor filtered sub-region time-frequency diagram are determined. Based on the location correspondence, the time-frequency maps of the first and second motor filter sub-regions are sequentially measured to determine the distance between the corresponding sub-regions. The total distance is determined by weighted summation of the distances to each sub-region. When the total distance is less than or equal to the preset distance anomaly threshold, the dual drive motors of the electric boat are determined to work synchronously.

[0056] Embodiment 3 of the present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program; when the computer program is executed by the processor, the processor performs the steps of the method for detecting the synchronous working state of the dual drive motors of an electric boat as described in Embodiment 1 of the present invention.

[0057] Embodiment 4 of the present invention also provides a computer-readable storage medium storing a computer program / instructions thereon, which, when executed by a processor, implements the steps of the method described in Embodiment 1 of the present invention.

[0058] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0059] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0060] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0061] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0062] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0063] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting a synchronous operation state of double drive motors of an electric ship, characterized by, include: The rotor electrical signals of the first electric boat drive motor and the second electric boat drive motor are obtained according to the sampling duration, and the corresponding time-frequency diagrams of the first motor and the second motor are determined. The time-frequency diagrams of the first motor and the second motor are divided along the time axis according to the first time segmentation length to construct multiple sub-time-frequency diagrams of the first motor and multiple sub-time-frequency diagrams of the second motor. For any first motor sub-time-frequency diagram and any second motor sub-time-frequency diagram, divide them sequentially along the time axis and frequency axis according to the second time segmentation length and frequency segmentation length, and output multiple first motor sub-region time-frequency diagrams of each first motor sub-time-frequency diagram and multiple second motor sub-region time-frequency diagrams of each second motor sub-time-frequency diagram; For any first motor sub-region time-frequency diagram and any second motor sub-region time-frequency diagram, respectively, the corresponding filtering parameters are used for filtering and vector transformation is performed to determine the corresponding first motor vector and second motor vector; A linear discriminant hyperplane is constructed based on the amplitude time gradient of the time-frequency maps of each of the first motor sub-regions and each of the second motor sub-regions; The centroid of the first motor is determined by calculating the centroid of each of the first motor vectors, and the centroid of the second motor is determined by calculating the centroid of each of the second motor vectors. If they belong to the same side of the centroid located on the linear discrimination hyperplane, then it is determined that the dual drive motors of the electric boat are working synchronously.

2. The method of claim 1, wherein the method further comprises: The construction of a linear discriminant hyperplane based on the amplitude time gradients of the time-frequency maps of each of the first motor sub-regions and each of the second motor sub-regions includes: Determine the time gradient of the first motor amplitude at the position of the maximum amplitude frequency in the time-frequency diagram of each first motor sub-region; The vector summation of the time gradient of the first motor amplitude associated with each of the first motor sub-time-frequency maps is performed to construct the corresponding first motor fusion gradient; The fusion gradient of each of the first motors is projected onto different preset mutually orthogonal directions to determine the discrimination weight; Calculate the second motor amplitude time gradient at the position of the largest amplitude frequency in the time-frequency map of each second motor sub-region and perform vector summation to output the total fused gradient of the second motor. Project the total fusion gradient of the second motor onto the fusion gradient of the first motor with the maximum gradient magnitude to determine the discrimination bias; A linear discriminant hyperplane is constructed using the discriminant weights and the discriminant biases.

3. The method of claim 1, wherein the method further comprises: The process of determining the length of the first time segment includes: Determine whether the difference between the maximum and minimum frequency values ​​in the time-frequency diagram of the first motor or the second motor is greater than a preset frequency threshold. If greater, it is determined that the value range of the number of time axis divisions is ; If none of them are greater than, then the range of values ​​for the number of time axis segments is determined as follows: ; The first time segment length is determined by calculating the ratio between the sampling duration and the number of time axis segments in one operation. in, Indicates the number of timeline segments in one operation. Indicates the sampling duration.

4. The method for detecting the synchronous operation of dual drive motors in an electric boat according to claim 3, characterized in that, The process of determining the frequency segment length includes: when When the frequency axis segmentation number is within the range of values, the value range is: ; when When the frequency axis segmentation number is within the range of values, the value range is: ; The ratio of the maximum frequency value of the first motor time-frequency diagram to the number of frequency axis segments is taken as the corresponding frequency segment length; The ratio of the maximum frequency value of the second motor time-frequency diagram to the number of frequency axis segments is taken as the corresponding frequency segment length; in, Indicates the number of frequency axis segments. This indicates the maximum frequency value.

5. The method for detecting the synchronous operation of dual drive motors in an electric boat according to claim 1, characterized in that, The process of determining the filtering parameters includes: The first motor sub-region time-frequency diagram or the second motor sub-region time-frequency diagram is used as the input to the filter parameter prediction model, and the corresponding filter parameters are output.

6. The method for detecting the synchronous operation of dual drive motors in an electric boat according to claim 1, characterized in that, Also includes: For any first motor sub-region time-frequency diagram and any second motor sub-region time-frequency diagram, after filtering with the corresponding filtering parameters, the corresponding first motor filtered sub-region time-frequency diagram and second motor filtered sub-region time-frequency diagram are determined. Based on the location correspondence, the time-frequency maps of the first and second motor filter sub-regions are sequentially measured to determine the distance between the corresponding sub-regions. The total distance is determined by weighted summation of the distances to each of the sub-regions. When the total distance is less than or equal to a preset distance anomaly threshold, the dual drive motors of the electric boat are determined to operate synchronously.

7. The method for detecting the synchronous operation of dual drive motors in an electric boat according to claim 6, characterized in that, The process of determining the distance weight of the sub-region distance includes: The time-frequency map of the first motor filter sub-region or the time-frequency map of the second motor filter sub-region is used as the input to the weight determination model, and the corresponding distance weight is output.

8. A detection device for the synchronous operation of dual drive motors in an electric boat, characterized in that, The apparatus for detecting the synchronous operation of dual drive motors in an electric boat as described in any one of claims 1-7 includes: The sampling transformation module is used to acquire the rotor electrical signals of the first electric boat drive motor and the second electric boat drive motor according to the sampling duration, and to determine the corresponding time-frequency diagrams of the first motor and the second motor. The first segmentation module is used to segment the first motor time-frequency diagram and the second motor time-frequency diagram along the time axis according to the first time segmentation length, and construct multiple first motor sub-time-frequency diagrams and multiple second motor sub-time-frequency diagrams; The second segmentation module is used to segment any first motor sub-time-frequency diagram and any second motor sub-time-frequency diagram sequentially along the time axis and frequency axis according to the second time segmentation length and frequency segmentation length, and output multiple first motor sub-region time-frequency diagrams of each first motor sub-time-frequency diagram and multiple second motor sub-region time-frequency diagrams of each second motor sub-time-frequency diagram; The filtering and classification module is used to filter the time-frequency map of any first motor sub-region and any second motor sub-region using their respective filtering parameters, and then perform vector transformation to determine the corresponding first motor vector and second motor vector. The plane construction module is used to construct a linear discriminant hyperplane based on the amplitude time gradient of the time-frequency maps of each of the first motor sub-regions and each of the second motor sub-regions; The synchronization discrimination module is used to determine the centroid of the first motor by calculating the centroid of each of the first motor vectors, and to determine the centroid of the second motor by calculating the centroid of each of the second motor vectors. If they belong to the same side centroid located on the linear discrimination hyperplane, then it is determined that the dual drive motors of the electric boat are working synchronously.

9. A computer device, characterized in that, The device includes a memory and a processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the method for detecting the synchronous operation state of dual drive motors of an electric boat as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method for detecting the synchronous working state of the dual drive motors of an electric boat as described in any one of claims 1-7.

Images