A crew post deployment method and device based on pattern classification recognition

Abstract

This invention discloses a method and apparatus for crew deployment based on pattern classification and recognition. The method includes: collecting a set of operational status information and an operational environment information set for ship equipment; the operational status information set includes a set of operational technical status parameters for each piece of ship equipment; the set of operational technical status parameters includes a data sequence of all technical status parameters during the operation of the ship equipment; the operational environment information set includes a data sequence of all environmental parameters of the environment in which each piece of ship equipment is located; performing pattern recognition processing on the operational status information set and the operational environment information set to obtain a ship equipment operational mode value; and selecting a crew deployment scheme corresponding to the ship equipment operational mode value from a preset set of crew deployment schemes based on the ship equipment operational mode value.

Description

Technical Field

[0001] This invention relates to the fields of industrial data processing, strategy optimization, and data processing technology, specifically to a method and apparatus for deploying crew positions based on pattern classification and recognition. Background Technology

[0002] During ship navigation and operation, the rationality of crew deployment directly affects the safe operation of ship equipment, operational efficiency, and navigational safety. Traditional crew deployment methods mainly rely on manual experience, such as fixed assignments based on ship routes, equipment types, or crew qualifications, which presents the following technical problems: The disconnect between equipment status and deployment plan: The operating status of ship equipment (such as power systems, navigation equipment, and communication equipment) changes in real time, and its health level (such as wear, parameter deviation, and potential failures) directly determines the required maintenance and operation intensity. Traditional deployment methods make it difficult to dynamically adjust crew assignments based on the real-time health status of the equipment. This may result in equipment in poor health not being monitored by dedicated personnel, or equipment in good health being assigned too much manpower, leading to resource waste.

[0003] The impact of environmental factors on deployment is often overlooked: the operating environment of ship equipment is complex and variable, and fluctuations in environmental parameters such as temperature, humidity, and sea state can exacerbate equipment wear and tear or trigger sudden failures (e.g., power equipment requires more frequent inspections in high-temperature environments). Traditional deployment methods are mostly based on fixed environmental assumptions and cannot adjust the crew's focus on critical equipment in accordance with real-time environmental patterns (e.g., severe sea conditions, high humidity environments), thus increasing the risk of equipment failure.

[0004] Subjectivity in pattern recognition and decision-making: Even in some scenarios, when humans refer to equipment status and environmental data, their decision-making process relies on personal experience and lacks systematic analysis of multi-dimensional data (such as equipment multi-parameter sequences and environmental statistical characteristics). It is difficult to quantify the matching relationship between equipment operation mode and crew deployment, resulting in low accuracy and consistency of deployment plans. Summary of the Invention

[0005] This invention primarily addresses the problems in existing crew deployment that include the disconnect between equipment status and deployment plan, the neglect of the impact of environmental factors on crew deployment, and the lack of objectivity in ship operation mode identification and decision-making. This invention discloses a crew deployment method and apparatus based on pattern classification and identification.

[0006] In a first aspect, this invention discloses a method for deploying crew members based on pattern classification and recognition, comprising: S1, Collect the set of operational status information and the set of operational environment information of the ship's equipment; the set of operational status information includes the set of operational technical status parameters for each piece of ship equipment; the set of operational technical status parameters includes the data sequence of all technical status parameters during the operation of the ship equipment; the set of operational environment information includes the data sequence of all environmental parameters of the environment in which each piece of ship equipment is located; The environmental parameters include temperature, humidity, air pressure, oxygen content, sea state, etc., and each environmental parameter can be collected by the corresponding sensor.

[0007] S2, perform pattern recognition processing on the set of operating status information and the set of operating environment information to obtain the operating mode value of the ship equipment; S3. Based on the ship equipment operation mode value, select the crew deployment scheme corresponding to the ship equipment operation mode value from the preset crew deployment scheme set.

[0008] The process of performing pattern recognition processing on the set of operational status information and the set of operational environment information to obtain the ship equipment operational mode value includes: S21, Perform equipment health assessment processing on the set of operating status information to obtain the health assessment value of each ship equipment; S22, Perform environmental pattern classification processing on the set of operating environment information to obtain the environmental pattern value of each ship equipment; S23, perform pattern classification processing on the health assessment values ​​and environmental pattern values ​​of all ship equipment to obtain the ship equipment operation pattern values.

[0009] The process of performing equipment health assessment on the set of operational status information to obtain a health assessment value for each piece of ship equipment includes: S211, Obtain the standard technical status parameter set for each piece of ship equipment; the standard technical status parameter set includes the standard values ​​of all technical status parameters during the operation of the ship equipment; S212, for each set of operational technical status parameters of ship equipment, subtract the corresponding set of standard technical status parameters of the ship equipment to obtain a set of technical status difference sequences of the ship equipment; the set of technical status difference sequences includes a difference sequence for each technical status parameter; the difference sequence of the technical status parameters is obtained by subtracting the data sequence of each technical status parameter from the standard value of the corresponding technical status parameter; S213, evaluate the set of technical condition difference sequences for each piece of ship equipment to obtain the health assessment value of the ship equipment.

[0010] The calculation expression for the evaluation process is: in, q is the preset weighting value for the j-th technical state parameter, and q is the preset scaling factor. F represents the preset standard state value, J represents the intermediate state value, and F represents the health assessment value. Let xj0 be the i-th element of the difference sequence of the j-th technical state parameter, xj0 be the mean of the difference sequence of the j-th technical state parameter, n be the length of the data sequence of the technical state parameter, and m be the total number of technical state parameters.

[0011] The environmental pattern classification process performed on the set of operating environment information to obtain the environmental pattern value for each ship equipment includes: S221, For the data sequence of all environmental parameters of the environment in which each ship equipment is located in the operating environment information set, calculate the statistical feature vector of each data sequence; S222, each statistical feature vector of a ship equipment is used as a row vector to construct the statistical matrix of the ship equipment; S223, Perform eigenvalue decomposition on the statistical matrix to obtain a set of eigenvectors and a set of eigenvalues; S224, perform pattern value calculation on the feature vector set and feature value set to obtain the environmental pattern value of the ship equipment.

[0012] The expression for calculating the pattern value is: , in, Let be the mean vector of all eigenvectors in the eigenvector set. Let be the vector formed by all the eigenvalues ​​of the eigenvalue set. The covariance matrix is ​​calculated for all row vectors of the statistical matrix, and s is the environmental mode value of the ship's equipment.

[0013] The process of classifying the health assessment values ​​and environmental model values ​​of all ship equipment into model values ​​yields the ship equipment operating model values, including: S231, using the health assessment values ​​and environmental model values ​​of all ship equipment, a model vector is constructed; S232, Perform feature encoding processing on the pattern vector to obtain the encoded vector; S233, Perform Frobenius norm calculation on the encoded vector to obtain the norm value; S234. Using a preset ship equipment operation mode model, the range value is processed to obtain the ship equipment operation mode value.

[0014] A second aspect of the present invention discloses a crew deployment device based on pattern classification and recognition, the device comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the crew deployment method based on pattern classification and recognition.

[0015] In a third aspect, the present invention discloses a computer-readable storage medium storing computer instructions, which, when invoked by a computer, are used to execute the crew deployment method based on pattern classification and recognition.

[0016] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the aforementioned method for deploying crew positions based on pattern classification and recognition.

[0017] The beneficial effects of this invention are as follows: This invention accurately quantifies the real-time health status of equipment by calculating equipment health assessment values ​​(based on a quantitative model constructed using weighted, trigonometric, and exponential functions, using a sequence of differences between technical status parameters and standard values). Adjusting crew deployment based on these assessment values ​​ensures that equipment in poor health receives more maintenance and monitoring resources, while equipment in good health requires less manpower, thus avoiding resource waste or monitoring gaps.

[0018] This invention quantifies the impact of environmental factors on equipment by analyzing the statistical characteristics (mean, variance, skewness, kurtosis) of environmental parameter data sequences, combined with eigenvalue decomposition and Mahalanobis distance calculations to obtain environmental model values. Based on these environmental model values, deployment plans can be adjusted to prioritize crew allocation for critical equipment in harsh environments (such as high sea states and extreme temperatures), reducing the risk of equipment failure due to environmental fluctuations.

[0019] This invention integrates health assessment values ​​and environmental pattern values ​​from all equipment to construct a pattern vector. This vector is then processed using feature encoding methods such as KernelPCA, LLE, or UMAP. Combined with norm calculations and pre-set models, it achieves quantitative identification of the overall operating mode of the equipment. This process eliminates the subjectivity of human experience and, through mathematical modeling of multi-dimensional data, enables an objective judgment of the overall operating status of the ship, providing a scientific basis for crew deployment. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation

[0021] To better understand the content of this invention, an embodiment is provided here.

[0022] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention.

[0023] In a first aspect, this invention discloses a method for deploying crew members based on pattern classification and recognition, comprising: S1, Collect the set of operational status information and the set of operational environment information of the ship's equipment; the set of operational status information includes the set of operational technical status parameters for each piece of ship equipment; the set of operational technical status parameters includes the data sequence of all technical status parameters during the operation of the ship equipment; the set of operational environment information includes the data sequence of all environmental parameters of the environment in which each piece of ship equipment is located; The environmental parameters include temperature, humidity, air pressure, oxygen content, sea state, etc., and each environmental parameter can be collected by the corresponding sensor.

[0024] S2, perform pattern recognition processing on the set of operating status information and the set of operating environment information to obtain the operating mode value of the ship equipment; S3. Based on the ship equipment operation mode value, select the crew deployment scheme corresponding to the ship equipment operation mode value from the preset crew deployment scheme set.

[0025] The preset set of crew deployment schemes includes the crew deployment scheme corresponding to each ship equipment operation mode value; the crew deployment scheme includes information on the ship equipment to be operated by all crew members.

[0026] The process of performing pattern recognition processing on the set of operational status information and the set of operational environment information to obtain the ship equipment operational mode value includes: S21, Perform equipment health assessment processing on the set of operating status information to obtain the health assessment value of each ship equipment; S22, Perform environmental pattern classification processing on the set of operating environment information to obtain the environmental pattern value of each ship equipment; S23, perform pattern classification processing on the health assessment values ​​and environmental pattern values ​​of all ship equipment to obtain the ship equipment operation pattern values.

[0027] The process of performing equipment health assessment on the set of operational status information to obtain a health assessment value for each piece of ship equipment includes: S211, Obtain the standard technical status parameter set for each piece of ship equipment; the standard technical status parameter set includes the standard values ​​of all technical status parameters during the operation of the ship equipment; S212, for each set of operational technical status parameters of ship equipment, subtract the corresponding set of standard technical status parameters of the ship equipment to obtain a set of technical status difference sequences of the ship equipment; the set of technical status difference sequences includes a difference sequence for each technical status parameter; the difference sequence of the technical status parameters is obtained by subtracting the data sequence of each technical status parameter from the standard value of the corresponding technical status parameter; S213, evaluate the set of technical condition difference sequences for each piece of ship equipment to obtain the health assessment value of the ship equipment.

[0028] The calculation expression for the evaluation process is: in, q is the preset weighting value for the j-th technical state parameter, and q is the preset scaling factor. F represents the preset standard state value, J represents the intermediate state value, and F represents the health assessment value. Let xj0 be the i-th element of the difference sequence of the j-th technical state parameter, xj0 be the mean of the difference sequence of the j-th technical state parameter, n be the length of the data sequence of the technical state parameter, and m be the total number of technical state parameters.

[0029] The calculation expression for the assessment process introduces a weighting mechanism, which highlights the role of core parameters and improves the targeting of the assessment by considering the impact of different technical status parameters on equipment health (e.g., key parameters have higher weights). Combined with trigonometric functions to capture the fluctuation characteristics of the difference sequence, it can sensitively reflect the degree to which parameters deviate from standard values, avoiding the limitations of single threshold judgments. The exponential function term can constrain excessively deviating difference values, avoiding excessive interference from extreme outliers on the overall assessment results and enhancing assessment stability. The overall formula transforms multi-dimensional parameter differences into a single health assessment value, achieving an effective mapping from complex sequences to quantitative indicators, facilitating subsequent global pattern analysis.

[0030] The environmental pattern classification process performed on the set of operating environment information to obtain the environmental pattern value for each ship equipment includes: S221, For the data sequence of all environmental parameters of the environment in which each ship equipment is located in the operating environment information set, calculate the statistical feature vector of each data sequence; The statistical feature vector includes the mean of each data sequence. ,variance skewness and kurtosis The calculation formula is as follows: , in, Let N be the i-th element of the data sequence, and N be the length of the data sequence.

[0031] S222, using all the statistical feature vectors of each ship equipment as row vectors, to construct the statistical matrix of the ship equipment; S223, Perform eigenvalue decomposition on the statistical matrix to obtain a set of eigenvectors and a set of eigenvalues; S224, Perform pattern value calculation on the feature vector set and feature value set to obtain the environmental pattern value of the ship equipment; The statistical feature vector, constructed using four statistical measures—mean, variance, skewness, and kurtosis—comprehensively characterizes the distributional properties of environmental data sequences. The mean reflects the overall level of environmental parameters, while variance reflects the amplitude of data fluctuations; their combination captures environmental stability (e.g., whether temperature remains consistently high or fluctuates frequently). Skewness and kurtosis supplement the morphological characteristics of the data distribution (e.g., whether it is left-skewed / right-skewed, or whether there is a concentration of extreme values), identifying abnormal environmental patterns that are not normally distributed (e.g., a spike distribution of sudden high temperatures). These four statistical measures compress the original data sequence from different dimensions, significantly reducing data dimensionality while retaining key information, providing efficient input for subsequent environmental pattern classification. The calculation method is simple and the physical meaning is clear, directly reflecting the potential impact of environmental parameters on equipment operation (e.g., a high-variance environment may accelerate equipment aging).

[0032] The eigenvalue decomposition can be implemented using a matrix eigenvalue decomposition algorithm.

[0033] The expression for calculating the pattern value is: in, Let be the mean vector of all eigenvectors in the eigenvector set. Let be the vector formed by all the eigenvalues ​​of the eigenvalue set. The covariance matrix is ​​calculated for all row vectors of the statistical matrix, and s is the environmental mode value of the ship's equipment.

[0034] The expression for calculating the model value is based on the Mahalanobis distance of the eigenvector mean, eigenvalue vector, and covariance matrix, achieving accurate quantification of environmental models. The introduction of covariance matrix correction eliminates the interference of correlations between different environmental parameter features (such as the covariance between temperature and humidity), making the model values ​​more reflective of the independent characteristics of the environment. Combining the difference analysis of the eigenvector mean and eigenvalue vector integrates the statistical characteristics of environmental parameters with the core features after matrix decomposition, improving the comprehensiveness of pattern recognition. Mahalanobis distance itself is scale-independent, unifying the feature contributions of environmental parameters with different dimensions (e.g., the numerical differences between temperature and air pressure do not affect the comparability of model values). The output single environmental model value can be directly used for subsequent global model analysis, realizing the transformation from multi-parameter environmental data to standardized model indicators.

[0035] The covariance matrix calculated from all row vectors of the statistical matrix is ​​obtained by using all row vectors of the statistical matrix as random variables and calculating their cross-covariance. The elements in the i-th row and j-th column of the covariance matrix are the cross-covariance values ​​of the i-th row vector and the j-th row vector of the statistical matrix.

[0036] The process of classifying the health assessment values ​​and environmental model values ​​of all ship equipment into model values ​​yields the ship equipment operating model values, including: A pattern vector is constructed using the health assessment values ​​and environmental pattern values ​​of all ship equipment. The pattern vector is subjected to feature encoding processing to obtain an encoded vector; The Frobenius norm is calculated on the encoded vector to obtain the norm value; By using a pre-defined ship equipment operation mode model, the range values ​​are processed to obtain the ship equipment operation mode values.

[0037] Calculating the Frobenius norm of the encoding vector provides a reliable quantitative basis for determining the running mode value. The preset ship equipment operation mode model includes several ship equipment operation mode values ​​and corresponding norm value ranges. By determining the norm value range to which the norm value belongs, the ship equipment operation mode value corresponding to the norm value range is determined, which is the obtained ship equipment operation mode value.

[0038] The feature encoding process can be implemented using Kernel PCA, LLE, or UMAP.

[0039] The pattern vector is obtained by splicing together the health assessment values ​​and environmental pattern values ​​of all ship equipment.

[0040] In the proposed pattern classification scheme, the Frobenius norm comprehensively measures the overall magnitude of the encoded vector, compressing high-dimensional features into a single value, facilitating matching with a preset pattern range. Norm calculation performs square root processing on the features of each dimension in the encoded vector, preserving the contribution of each dimension (health assessment value, environmental pattern value) while amplifying key differences through squaring, thus improving pattern discriminability. Combined with encoding methods such as Kernel PCA, LLE, or UMAP, the norm can indirectly reflect the similarity or difference of global patterns in the original data, making the correlation between pattern values ​​and the actual operating status of the equipment closer. The norm calculation results are deterministic and comparable, providing an objective standard for judging the values ​​of the preset model and avoiding the subjectivity of pattern classification.

[0041] This invention is based on a set of pre-set crew deployment schemes that match global operating mode values, ensuring that crew members' tasks cover the collaborative needs of all equipment (such as synchronous monitoring of equipment in a linkage system), avoiding the limitations of single equipment deployment, and improving the coordination and efficiency of the overall ship operation.

[0042] In all embodiments of the present invention, the variables involved in all computational expressions or mathematical functions have been dimensionlessized before computation.

[0043] In all embodiments of the present invention, the values ​​of the independent variables in the input of all computational expressions or mathematical functions meet the reasonable requirements of the input range of the computational expressions or mathematical functions, and can ensure that the computational expressions or mathematical functions can be calculated smoothly without violating physical laws or mathematical rules.

[0044] A second aspect of the present invention discloses a crew deployment device based on pattern classification and recognition, the device comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the crew deployment method based on pattern classification and recognition.

[0045] In a third aspect, the present invention discloses a computer-readable storage medium storing computer instructions, which, when invoked by a computer, are used to execute the crew deployment method based on pattern classification and recognition.

[0046] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the aforementioned method for deploying crew positions based on pattern classification and recognition.

[0047] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for deploying crew members based on pattern classification and recognition, characterized in that, include: S1, collects the set of operational status information and operational environment information of ship equipment; The set of operational status information includes a set of operational technical status parameters for each piece of ship equipment; the set of operational technical status parameters includes a data sequence of all technical status parameters during the operation of the ship equipment; the set of operational environment information includes a data sequence of all environmental parameters of the environment in which each piece of ship equipment is located. S2, perform pattern recognition processing on the set of operating status information and the set of operating environment information to obtain the operating mode value of the ship equipment; S3. Based on the ship equipment operation mode value, select the crew deployment scheme corresponding to the ship equipment operation mode value from the preset crew deployment scheme set.

2. The crew deployment method based on pattern classification and recognition as described in claim 1, characterized in that, The process of performing pattern recognition processing on the set of operational status information and the set of operational environment information to obtain the ship equipment operational mode value includes: S21, Perform equipment health assessment processing on the set of operating status information to obtain the health assessment value of each ship equipment; S22, Perform environmental pattern classification processing on the set of operating environment information to obtain the environmental pattern value of each ship equipment; S23, perform pattern classification processing on the health assessment values ​​and environmental pattern values ​​of all ship equipment to obtain the ship equipment operation pattern values.

3. The method for deploying crew members based on pattern classification and recognition as described in claim 2, characterized in that, The process of performing equipment health assessment on the set of operational status information to obtain a health assessment value for each piece of ship equipment includes: S211, Obtain the standard technical status parameter set for each piece of ship equipment; the standard technical status parameter set includes the standard values ​​of all technical status parameters during the operation of the ship equipment; S212, for each set of operational technical status parameters of ship equipment, subtract the corresponding set of standard technical status parameters of the ship equipment to obtain a set of technical status difference sequences of the ship equipment; the set of technical status difference sequences includes a difference sequence for each technical status parameter; the difference sequence of the technical status parameters is obtained by subtracting the data sequence of each technical status parameter from the standard value of the corresponding technical status parameter; S213, evaluate the set of technical condition difference sequences for each piece of ship equipment to obtain the health assessment value of the ship equipment.

4. The crew deployment method based on pattern classification and recognition as described in claim 3, characterized in that, The calculation expression for the evaluation process is: in, q is the preset weighting value for the j-th technical state parameter, and q is the preset scaling factor. F represents the preset standard state value, J represents the intermediate state value, and F represents the health assessment value. Let xj0 be the i-th element of the difference sequence of the j-th technical state parameter, xj0 be the mean of the difference sequence of the j-th technical state parameter, n be the length of the data sequence of the technical state parameter, and m be the total number of technical state parameters.

5. The crew deployment method based on pattern classification and recognition as described in claim 2, characterized in that, The environmental pattern classification process performed on the set of operating environment information to obtain the environmental pattern value for each ship equipment includes: S221, For the data sequence of all environmental parameters of the environment in which each ship equipment is located in the operating environment information set, calculate the statistical feature vector of each data sequence; S222, each statistical feature vector of a ship equipment is used as a row vector to construct the statistical matrix of the ship equipment; S223, Perform eigenvalue decomposition on the statistical matrix to obtain a set of eigenvectors and a set of eigenvalues; S224, perform pattern value calculation on the feature vector set and feature value set to obtain the environmental pattern value of the ship equipment.

6. The crew deployment method based on pattern classification and recognition as described in claim 5, characterized in that, The expression for calculating the pattern value is: in, Let be the mean vector of all eigenvectors in the eigenvector set. Let be the vector formed by all the eigenvalues ​​of the eigenvalue set. The covariance matrix is ​​calculated for all row vectors of the statistical matrix, and s is the environmental mode value of the ship's equipment.

7. The crew deployment method based on pattern classification and recognition as described in claim 2, characterized in that, The process of classifying the health assessment values ​​and environmental model values ​​of all ship equipment into model values ​​yields the ship equipment operating model values, including: S231, using the health assessment values ​​and environmental model values ​​of all ship equipment, a model vector is constructed; S232, Perform feature encoding processing on the pattern vector to obtain the encoded vector; S233, Perform Frobenius norm calculation on the encoded vector to obtain the norm value; S234. Using a preset ship equipment operation mode model, the range value is processed to obtain the ship equipment operation mode value.

8. A crew member deployment device based on pattern classification and recognition, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the crew deployment method based on pattern classification and recognition as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked by a computer, are used to execute the crew deployment method based on pattern classification and recognition as described in any one of claims 1 to 7.

10. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the crew deployment method based on pattern classification and recognition as described in any one of claims 1 to 7.

Images