A multi-source sensing information fusion and expansion method based on system state representation

Abstract

The application provides a kind of multi-source sensing information fusion and expansion method based on system state representation, only with the key failure factors in the sliding bearing system, gear box system, rolling bearing system that cannot be repaired as the ship power system state evaluation index system, can better reflect the running state of ship power system, is conducive to the effective evaluation of power system state;Through the fusion method of multi-state parameter characteristics, through data cleaning operation for data missing value filling and denoising processing, through feature selection and feature extraction operation to remove multi-source data redundant information, the problem of the existing ship power system relying on single parameter threshold alarm method can be overcome, and the deterioration process before the failure of the ship power system can be identified in advance, the state change of the ship power system during operation is accurately evaluated, and the accuracy of weight distribution is guaranteed by comprehensive empowerment method, thereby effectively improving the accuracy of the subsequently constructed state evaluation benchmark model.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ship state evaluation, and in particular to a multi-source sensing information fusion and expansion method based on system state representation. BACKGROUND

[0002] With the continuous development of science and technology, the structure of the ship becomes more and more complex, at the same time, the connection between each subsystem of the ship becomes more and more close, once a subsystem fails, it will cause the whole ship to fail, which will bring a major threat to people's life and property safety, and even bring disastrous consequences. The ship power system is the core subsystem of the ship, and its sensing data sources are extensive and complex, so it is particularly necessary to fuse and process multi-source sensing information and accurately control the state evaluation of the ship power system, which is not only an important basis for state maintenance, but also a prerequisite for fault prediction and fault diagnosis.

[0003] At present, the state evaluation method for ship power system at home and abroad often uses a small amount of monitoring information for state evaluation, usually relies on the subjective experience of people to set a fixed threshold, or uses a large amount of monitoring information, but due to its high complexity, mainly manifested as data instability, high coupling and strong correlation, it is difficult to obtain the state information of the system from the data, and due to the complex and variable working conditions of the ship power system, the change range of the conventional state evaluation characteristic parameter threshold is large, which affects the accuracy of state evaluation. SUMMARY

[0004] Therefore, the present application provides a multi-source sensing information fusion and expansion method based on system state representation, device, electronic equipment and storage medium, to solve the problem that the change range of the conventional state evaluation characteristic parameter threshold of the ship power system is large due to the complex and variable working conditions, which affects the accuracy of state evaluation.

[0005] To solve the above technical problems, in a first aspect, the present application embodiment provides a multi-source sensing information fusion and expansion method based on system state representation, comprising:

[0006] Obtaining the running state parameters of each subsystem of the ship power system, performing data cleaning, feature extraction and principal component analysis on the running state parameters to obtain state evaluation characteristic parameters; the subsystems include sliding bearing system, gear box system and rolling bearing system;

[0007] Dividing the running conditions of the ship power system into multiple categories, extracting the state evaluation characteristic parameters under each category of running condition, and constructing a state evaluation benchmark model corresponding to the running condition based on the state evaluation characteristic parameters;

[0008] Obtain the running condition and state evaluation benchmark model of the current time, construct a state evaluation index based on the running condition and the Mahalanobis distance of the state evaluation benchmark model, and perform state evaluation on the ship power system based on the state evaluation index.

[0009] Further, the running state parameters of the sliding bearing system include temperature parameters and pressure parameters; the temperature parameters include a first bearing temperature parameter of a driving end sliding bearing and a second bearing temperature parameter of a non-driving end sliding bearing, and the running state parameters of the sliding bearing system further include lubricating oil temperature and cooling water temperature; the pressure parameters include a first lubricating oil pressure parameter of a non-driving sliding bearing and a second lubricating oil pressure parameter of a non-driving end sliding bearing;

[0010] The running state parameters of the gear box system include a first acceleration parameter, and the first acceleration parameter includes acceleration collected at multiple measuring points on the gear box;

[0011] The running state parameters of the rolling bearing system include a second acceleration parameter, and the second acceleration parameter includes acceleration collected at each rolling bearing;

[0012] After obtaining the running state parameters of each subsystem in the ship power system, the method further includes:

[0013] A parameter group based on time series is established for each of the subsystems, the parameter group including time series and numerical values of each running state parameter corresponding to the time series; the parameter group includes a sliding bearing system parameter group, a gear box system parameter group, and a rolling bearing system parameter group.

[0014] Further, the running state parameters are subjected to data cleaning, feature extraction, and feature analysis to obtain state evaluation feature parameters, specifically including:

[0015] The missing values in the parameter group are subjected to interpolation processing based on the Lagrange interpolation method, noise data are screened out and removed based on the 3σ criterion, and the missing values after removing the noise data are subjected to interpolation filling based on the Lagrange interpolation method;

[0016] Each running state parameter in the parameter group is subjected to feature selection based on the Pearson correlation coefficient method to extract state evaluation feature parameters;

[0017] The state evaluation feature parameters are subjected to standardization processing and secondary feature extraction processing, and the weights of each state evaluation feature parameter in the parameter group are determined.

[0018] Further, each running state parameter in the parameter group is subjected to feature selection based on the Pearson correlation coefficient method to extract state evaluation feature parameters, specifically including:

[0019] In the sliding bearing system parameter group, the first Pearson correlation coefficient of the remaining temperature parameters and the reference temperature is determined based on the bearing temperature of the driving end sliding bearing as the reference temperature, the temperature parameters with the first Pearson correlation coefficient lower than the preset first correlation coefficient threshold are removed, and the state evaluation characteristic parameters corresponding to the sliding bearing system parameter group are obtained;

[0020] In the gear box system parameter group and the rolling bearing system parameter group, a random acceleration is selected as a reference acceleration, the second Pearson correlation coefficient of the remaining accelerations in the gear box system parameter group and the rolling bearing system parameter group and the reference acceleration is determined, and the accelerations with the second Pearson correlation coefficient higher than the second correlation coefficient threshold are removed; the time domain characteristic parameters and the frequency domain characteristic parameters of each acceleration in the gear box system and the rolling bearing system are extracted, and the state evaluation characteristic parameters corresponding to the gear box system parameter group and the rolling bearing system parameter group are obtained.

[0021] Further, the state evaluation characteristic parameters are standardized and subjected to secondary feature extraction, and the weights of each operating state parameter in the parameter group are determined, specifically including:

[0022] Based on the multi-index comprehensive evaluation method and the preset index interval, the state evaluation characteristic parameters in each parameter group are divided into positive indicators, moderate indicators and reverse indicators; and standardized processing is performed based on the corresponding strategy;

[0023] If the state evaluation characteristic parameter is determined to be a positive indicator, and the measured value of the operating state parameter is not greater than the average value; or if the state evaluation characteristic parameter is determined to be a reverse indicator, and the measured value of the operating state parameter is greater than the average value; the operating state parameter after standard deviation processing is determined to be 0;

[0024] If the state evaluation characteristic parameter is determined to be a positive indicator, and the measured value of the operating state parameter is greater than the average value; or if the state evaluation characteristic parameter is determined to be a moderate indicator; or if the state evaluation characteristic parameter is determined to be a reverse indicator, and the measured value of the operating state parameter is not greater than the average value; the operating state parameter after standard deviation processing is determined to be: the difference between the measured value and the average value divided by the standard deviation of the operating state parameter;

[0025] Secondary feature extraction is performed on the parameter group based on the principal component analysis method, the initial weights of each state evaluation characteristic parameter in the parameter group are calculated based on the entropy method, and the initial weights are corrected based on the weighted average method.

[0026] Further, a state evaluation reference model corresponding to the operating condition is constructed based on the state evaluation characteristic parameters, specifically including:

[0027] Obtain the set of state assessment feature parameters corresponding to the current category of operating conditions, and form a state assessment feature vector based on the state assessment feature parameters;

[0028] Determine the weight of each state evaluation feature parameter in the state evaluation feature vector, and model the distribution of the state evaluation feature parameters as a Gaussian probability density function;

[0029] A baseline model for the current category of operating conditions is constructed based on the weight of each of the aforementioned state assessment feature parameters and the corresponding Gaussian probability density function.

[0030] Furthermore, based on the operating conditions and the Mahalanobis distance of the state assessment benchmark model, a state assessment index is constructed. The state assessment of the ship's propulsion system is then performed based on this state assessment index, specifically including:

[0031] Based on the current state assessment feature vector, the corresponding weight, the mean vector of the Gaussian distribution of the state assessment benchmark model under each operating condition, and the covariance matrix, the first Mahalanobis distance between the current state assessment feature vector and each Gaussian distribution is determined.

[0032] Based on the weighting coefficients of the Gaussian probability density function and the first Mahalanobis distance, the second Mahalanobis distance between the current state assessment feature vector and the state assessment benchmark model under each operating condition is determined.

[0033] Based on the second Mahalanobis distance and the probability that the current operating condition of the ship's power system belongs to each category, the third Mahalanobis distance between the current state assessment feature vector and the state assessment benchmark model of T operating conditions is determined, where T is the number of operating condition categories.

[0034] A state evaluation index is constructed based on the product of the third Mahalanobis distance and the normal operating state constant value of the ship's propulsion system.

[0035] The ship's power system is assessed based on the average state evaluation index within a sliding window. If the average state evaluation index of multiple consecutive sampling points is greater than a first index threshold, or the state evaluation index of at least one sampling point is greater than a second index threshold, then it is determined to be an abnormal state. If the average state evaluation index of multiple consecutive sampling points is greater than the second index threshold, or the state evaluation index of at least one sampling point is greater than a third index threshold, then it is determined that immediate shutdown is required. The first index threshold is less than the second index threshold, and the second index threshold is less than the third index threshold.

[0036] Secondly, embodiments of the present invention provide a multi-source sensor information fusion and expansion device based on system state representation, comprising:

[0037] A multi-source data fusion module acquires operation state parameters of each subsystem in a ship power system, performs data cleaning, feature extraction and principal component analysis on the operation state parameters to obtain state evaluation feature parameters; the subsystems include a sliding bearing system, a gear box system and a rolling bearing system;

[0038] A working condition division module divides the operation working conditions of the ship power system into multiple categories, extracts state evaluation feature parameters under each category of operation working condition, and constructs a state evaluation benchmark model for the corresponding operation working condition based on the state evaluation feature parameters;

[0039] A state evaluation module acquires the operation working condition at the current time and the state evaluation benchmark model, constructs a state evaluation index based on the Mahalanobis distance of the operation working condition and the state evaluation benchmark model, and performs state evaluation on the ship power system based on the state evaluation index.

[0040] In a third aspect, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the method for multi-source sensor information fusion and expansion based on system state representation according to the embodiment of the first aspect of the present application when executing the program.

[0041] In a fourth aspect, a non-transitory computer readable storage medium is provided, which stores a computer program, and the computer program is executable on a processor to implement the steps of the method for multi-source sensor information fusion and expansion based on system state representation according to the embodiment of the first aspect of the present application.

[0042] The method, device, electronic device and storage medium provided by the embodiments of the present application can better reflect the operation state of the ship power system by taking the groups of parameters of the sliding bearing system, the gear box system and the rolling bearing system as the state evaluation index system of the ship power system, which are key failure factors that cannot be repaired, and are conducive to effectively evaluating the state of the power system; the method can overcome the shortcomings of the existing ship power system relying on a single parameter threshold alarm method, and can identify the deterioration process before the failure of the ship power system, accurately evaluate the state change of the ship power system during operation, by fusing multiple state parameters, filling and denoising data missing values through data cleaning operations, and removing redundant information of multi-source data through feature selection and feature extraction operations; the method can effectively improve the accuracy of the subsequently constructed state benchmark model, and thus improve the accuracy of state evaluation, and further effectively improve the rationality of the evaluation result. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0044] Figure 1 This is a flowchart of a multi-source sensor information fusion and augmentation method based on system state representation according to an embodiment of the present invention;

[0045] Figure 2 This is a diagram illustrating the composition of a multi-source data group according to an embodiment of the present invention;

[0046] Figure 3 This is a schematic diagram illustrating the specific process of the multi-source sensor information fusion and expansion method based on system state representation according to an embodiment of the present invention;

[0047] Figure 4 This is a schematic diagram of the physical structure according to an embodiment of the present invention. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0049] Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor systems and / or microcontroller systems.

[0050] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0051] In order to effectively analyze the state of the ship power system and predict the possibility of failure of the ship power system, multi-source sensing information of the ship power system needs to be subjected to data cleaning, feature selection, feature extraction, feature fusion and expansion and the like for state evaluation, so as to ensure continuous, safe and reliable operation of the ship power system. The method can reduce the experience requirement of testers, improve the efficiency and accuracy of state evaluation, and has certain engineering application value.

[0052] Therefore, the embodiment of the present application provides a multi-source sensing information fusion and expansion method based on system state representation, device, electronic equipment and storage medium, only with the sliding bearing system parameter group, the gear box system parameter group, the rolling bearing system parameter group these non-repairable key failure factors as the ship power system state evaluation index system, can better reflect the running state of the ship power system, is conducive to realizing the effective evaluation of the state of the power system. The following will be expanded and introduced through multiple embodiments.

[0053] Figure 1 The multi-source sensing information fusion and expansion method based on system state representation provided by the embodiment of the present application, that is, the multi-source sensing information fusion and expansion method based on system state representation, comprises:

[0054] Step S1, obtaining the running state parameters of each subsystem in the ship power system, performing data cleaning, feature extraction and principal component analysis on the running state parameters to obtain state evaluation feature parameters; the subsystems include a sliding bearing system, a gear box system and a rolling bearing system.

[0055] Step S11, obtaining the running state parameters of each subsystem in the ship power system;

[0056] In order to drive the ship to sail at a certain speed, it must be given a thrust. This thrust is generated by the propeller when it works, and the propeller can be a propeller, a planar wheel or a paddle wheel. The propeller is driven by a prime mover, and the prime mover includes a diesel engine, a steam engine, a gas turbine and a combined power device composed of them. The above-mentioned prime mover is called main engine, and the main engine plus auxiliary machines, pipelines, equipment required to ensure the work of these main engines, and shafting for transmitting power from the main engine to the propeller are collectively referred to as ship power system. The ship power system mainly includes main engine, transmission equipment and shafting, machinery equipment and power pipeline in engine room; due to the complex and variable working conditions of the ship power system, the change range of the conventional state characteristic parameter threshold is large, which affects the accuracy of state evaluation, and in the embodiment of the present application, only the sliding bearing system, the gear box system and the rolling bearing system are used as key failure factors, and a ship power system state evaluation index system is established, which can better reflect the running state of the ship power system, and is conducive to realizing the effective evaluation of the state of the power system.

[0057] The sliding bearing is an important component of the ship power system, and whether the bearing system can work normally will directly affect the economic benefit, power performance and service life of the whole ship. The working condition environment of the marine sliding bearing is more complex than that of the general radial sliding bearing, mainly because the ship body will sway and tilt due to the influence of wind and waves during the sea voyage, and the working plane of the sliding bearing will also be tilted to different degrees due to the influence of the ship body sway, which puts higher requirements on the safety and reliability of the whole sliding bearing system. The oil ring is often used in the sliding bearing to bring oil into the shaft and bearing bush, and if the oil ring is deformed or severely worn, the oil ring will not rotate or rotate slowly, which will reduce the oil amount between the shaft and the bearing bush and cause overheating. As long as the power system is running, the lubricating system must ensure a certain system oil pressure, and once the lubricating oil pressure is insufficient, it cannot guarantee its functions of lubrication, cooling, cleaning, sealing, corrosion prevention, shock absorption and the like, thereby causing excessive wear of the crank connecting rod mechanism, valve train and other friction pairs, and even causing cylinder pulling and bearing holding, so that the ship power system is scrapped in advance. If the lubricating oil pressure is too high, it will cause oil leakage of various oil seals, oil leakage of oil sump, oil leakage of main oil way, oil leakage of oil filter and the like. Therefore, the failure reasons of the sliding bearing system generally include bearing overheating, poor heat dissipation of lubricating oil, water temperature of cooling water, lubricating oil pressure and the like. In the embodiment, as shown in Figure 2 , the operating state parameters of the sliding bearing system include temperature parameters and pressure parameters; the temperature parameters include a first bearing temperature parameter of the driving end sliding bearing (i.e. bearing 1 temperature in Figure 2 ), a second bearing temperature parameter of the non-driving end sliding bearing (i.e. bearing 2 temperature in Figure 2 ), and further include lubricating oil temperature and cooling water temperature; the pressure parameters include a first lubricating oil pressure parameter of the non-driving sliding bearing (i.e. bearing 1 lubricating oil pressure in Figure 2 ), and a second lubricating oil pressure parameter of the non-driving end sliding bearing (i.e. bearing 2 lubricating oil pressure in Figure 2 ).

[0058] The gearbox is responsible for transmitting power to the propeller and providing a reduction ratio, because the rotation speed of the internal combustion engine is too high and the torque is too small, which is not suitable for ship propulsion, so the gearbox is needed to reduce the rotation speed and increase the torque; the gearbox is mainly composed of a transmission shaft, gears and a housing, and its main performance indicators are transmission efficiency, water resistance coefficient and reliability and durability. Before the gearbox fails, the vibration signal of the housing will gradually increase, and the fault can be judged by installing a vibration acceleration sensor on the gearbox housing to test the vibration signal. In the embodiment, the running state parameters of the gearbox system include a first acceleration parameter, and the first acceleration parameter includes the acceleration collected at multiple measurement points on the gearbox. In the embodiment, the first acceleration parameter is vibration acceleration, and the acceleration indicators (i.e. vibration acceleration) of 8 measurement points on the gearbox are as follows: Figure 3 The vibration signal of the housing is detected by using an acceleration sensor, and the real-time monitoring of the gearbox is realized by time domain analysis, and the acceleration is used as the running state parameter.

[0059] The rolling bearing system is one of the important components of the ship power system, and its working state directly determines the performance and operation of the ship power system. In engineering practice, a small fault of the rolling bearing system can cause the shutdown of the production line or damage to the equipment, causing serious economic losses. Therefore, it is of great significance to monitor the working state of the rolling bearing system in real time, find faults in time and develop reliable maintenance strategies; the running state parameters of the rolling bearing system include a second acceleration parameter, and the second acceleration parameter includes the acceleration collected at each rolling bearing, and the second acceleration parameter is vibration acceleration; in the embodiment, the acceleration of 4 rolling bearings is collected, and the vibration signal of the rolling bearing is a high-frequency signal, so an acceleration sensor is used to pick up the signal.

[0060] After obtaining the running state parameters of each subsystem in the ship power system, as shown in Figure 3 It also includes:

[0061] A parameter group based on time series is established for each of the subsystems, and the parameter group includes time series and the values of each running state parameter corresponding to the time series; the parameter group includes a sliding bearing system parameter group, a gearbox system parameter group and a rolling bearing system parameter group, and the sliding bearing system parameter group, the gearbox system parameter group and the rolling bearing system parameter group together constitute the multi-source data group of the embodiment.

[0062] In step S12, the missing values of each running state parameter in the parameter group are interpolated based on the Lagrange interpolation method; the approximate value of the missing value of each running state parameter is calculated according to the following formula:

[0063]

[0064]

[0065] In the above formula, x1, x2, …, x l represent time series, i∈(1,2,…,l), j∈(1,2,…,l), and j≠i; y1, y2, … y i represent the values of each operating state parameter in each parameter group, L l represents the number of operating state parameters in each parameter group; x i (x) represents the interpolation basis function of i sample points (x1, y1), (x2, y2), (x3, y3) … (x l , y l ); f l (x) is an l-order Lagrange interpolation function obtained by using the interpolation basis function, and the missing value of the corresponding point can be obtained by bringing the point corresponding to the missing function value into the interpolation function to perform interpolation filling.

[0066] On the basis of the above embodiment, as a preferred embodiment, noise data is screened out and removed based on the 3σ criterion; the 3σ criterion is as follows:

[0067]

[0068]

[0069] In the above formula, is the arithmetic mean of each operating state parameter, n is the number of initial sample points to be denoised; σ is the standard deviation, R n is the residual error, if the residual error is greater than 3σ, Z is noise data, which is removed, if the residual error is less than or equal to 3σ, it is retained, Z(k) represents the value of the operating state parameter at each time of the time series, k∈(1,2,…,n).

[0070] And based on the Lagrange interpolation method, the missing values after removing the noise data are interpolated and filled.

[0071] Step S13, based on the Pearson correlation coefficient method, each operating state parameter in the parameter group is selected for feature selection, and a state evaluation feature parameter is extracted.

[0072] The greater the absolute value of Pearson correlation coefficient is, the higher the correlation degree is. In the parameter group of the sliding bearing system, the types of the indexes are inconsistent. In order to accurately represent the system state, the bearing temperature of the driving end sliding bearing is taken as a reference temperature, the first Pearson correlation coefficient of the remaining temperature parameters and the reference temperature is determined, the temperature parameters whose first Pearson correlation coefficient is lower than a preset first correlation coefficient threshold are removed, and the state evaluation characteristic parameters corresponding to the parameter group of the sliding bearing system are obtained. In the parameter group of the gearbox system and the parameter group of the rolling bearing system, the types of the indexes are acceleration signals. In order to eliminate the redundancy of the data, one acceleration is randomly selected as a reference acceleration, the second Pearson correlation coefficient of the remaining accelerations in the parameter group of the gearbox system and the parameter group of the rolling bearing system and the reference acceleration is determined, and the accelerations whose second Pearson correlation coefficient is higher than a second correlation coefficient threshold are removed.

[0073] Based on the feature selection, the state evaluation characteristic parameters are preliminarily selected. Because the acceleration sampling frequency in the parameter group of the gearbox system and the parameter group of the rolling bearing system is high, changes quickly and data is complex, the time domain characteristic parameters and the frequency domain characteristic parameters are further extracted, so as to obtain the state evaluation characteristic parameters corresponding to the parameter group of the gearbox system and the parameter group of the rolling bearing system, and form the parameter group including the state evaluation parameters.

[0074] On the basis of the above embodiment, as a preferred embodiment, the extracted time domain and frequency domain characteristic parameters are as follows:

[0075] x max = max (|x(n)|)

[0076]

[0077]

[0078] x p-p = x max -x min

[0079]

[0080] In the formula, x(n) represents a signal after acceleration sampling and discretization, N represents the number of sampling points, K represents the number of spectral lines in the spectrum corresponding to x(n), Y(k) represents the amplitude corresponding to the kth spectral line, x max 、 x rms , x p-p , and Mf respectively represent the maximum value, the absolute mean value, the root mean square value, the peak-peak value, and the mean frequency.

[0081] On the basis of the above embodiments, as a preferred embodiment, the parameter matrix of each parameter group after feature selection and preliminary feature extraction is as follows:

[0082]

[0083] In the formula, n represents the number of samples, and p represents the number of initially selected state feature parameters.

[0084] The state evaluation feature parameters are subjected to standardization processing and secondary feature extraction processing, and the weights of the state evaluation feature parameters in the parameter group are determined.

[0085] In the embodiment, through the fusion method of multi-state parameter features, data missing value filling and denoising processing are performed through data cleaning operation, and multi-source data redundant information is removed through feature selection and feature extraction operation, which can overcome the shortcomings of the existing ship power system relying on single parameter threshold alarm method, and can identify the deterioration process before the failure of the ship power system, and accurately evaluate the state change of the ship power system during operation.

[0086] In step S14, based on a multi-index comprehensive evaluation method and a preset index interval, the state evaluation feature parameters in each parameter group are divided into positive indexes, moderate indexes and reverse indexes; and based on a corresponding strategy, the state evaluation feature parameters are subjected to standardization processing. The parameter matrix of each parameter group is subjected to standardization processing based on system evaluation indexes to improve the accuracy of subsequent state evaluation. The system evaluation indexes mainly include three types. One is a positive index, that is, the larger the index value, such as greater than a preset first threshold, the better the working state. One is a reverse index, that is, the smaller the index value, such as less than a preset second threshold, the better the working state. The other is a moderate index, and the parameter value is within a certain preset range, and too small or too large will adversely affect the health state of the system.

[0087] If it is judged that the state evaluation feature parameter is a positive index, and the measured value of the running state parameter is not greater than the average value; or if it is judged that the state evaluation feature parameter is a reverse index, and the measured value of the running state parameter is greater than the average value; then the running state parameter after standard deviation processing is determined to be 0.

[0088] If it is judged that the state evaluation feature parameter is a positive index, and the measured value of the running state parameter is greater than the average value; or if it is judged that the state evaluation feature parameter is a moderate index; or if it is judged that the state evaluation feature parameter is a reverse index, and the measured value of the running state parameter is not greater than the average value; then the running state parameter after standard deviation processing is determined to be: the difference between the measured value and the average value divided by the standard deviation of the running state parameter.

[0089] On the basis of the above embodiments, as a preferred embodiment, for the positive indicators, the following formula is used for standardization processing:

[0090]

[0091] For the moderate indicators, the following formula is used for standardization processing:

[0092]

[0093] For the reverse indicators, the following formula is used for standardization processing:

[0094]

[0095] In the above formula, x' ij is the standardized value of the jth operating state parameter, x ij is the measured value of the operating state parameter, is the average value of the jth operating state parameter, is the standard deviation of the jth operating state parameter.

[0096] The data matrix parameter matrix of the kth index group after standardization becomes:

[0097]

[0098] Step S15, performing secondary feature extraction on the parameter group based on the principal component analysis method. Since the state evaluation characteristic parameters in the gear box parameter group and the rolling bearing parameter group are too many, the PCA principal component analysis is used for secondary feature extraction operation to reduce the dimension of the features.

[0099]

[0100]

[0101]

[0102] In the formula, x1, x2,..., x p are original parameter variables, y1, y2,..., y m (m≤p) are principal variables after dimension reduction, a mp represents the coefficient value of the original parameter variable in the principal variable after dimension reduction; y i and y j (i≠j; j=1, 2,..., m) are mutually independent, λ iCPV represents the ratio of each principal component variable to the total variable, i.e. the contribution rate of the principal component to the total variance of the sample, m is the number of selected principal component variables, the number of principal component variables m is selected according to the contribution rate, and in the embodiment, CPV≥90% is taken as the standard, and α i represents the weight of each principal component.

[0103] On the basis of the above embodiments, as a preferred embodiment, the initial weight of each state evaluation characteristic parameter in the parameter group is calculated based on the entropy value method, as follows:

[0104]

[0105]

[0106]

[0107] K = 1 / ln(n)

[0108]

[0109] In the above formula, X k is the data matrix of the kth parameter group after standardization and PCA feature dimension reduction, n is the number of samples, and m is the number of finally selected state evaluation characteristic parameters, i.e. the principal component variable selected in the above step, x ij is the i-th row and j-th column element in X k , P ij is the contribution degree of the i-th sample under the j-th state evaluation characteristic parameter, and if P ij = 0, P ij ·lnP ij is taken as 0, is the information entropy of each state evaluation characteristic parameter of the kth parameter group, is the j-th column index in X k , and W

[0110] The initial weight is corrected based on the weighted average method, as shown in the following formula:

[0111]

[0112]

[0113] In the formula, represents the final objective weight of the state evaluation characteristic parameter of the kth parameter group, and respectively represent the objective weight calculated by principal component analysis and the entropy value method, w represents the comprehensive weight vector, and m represents the number of finally selected state evaluation characteristic parameters, i.e. the number of principal component variables.

[0114] In the embodiment, the comprehensive weighting method based on principal component analysis and entropy value weighted summation ensures the accuracy of weight distribution, thereby effectively improving the accuracy of the subsequently constructed state reference model, and further improving the accuracy of state evaluation.

[0115] In step S2, the operating conditions of the ship power system are divided into multiple categories, state evaluation characteristic parameters under each category of operating condition are extracted, and a state evaluation reference model corresponding to the operating condition is constructed based on the state evaluation characteristic parameters.

[0116] In the embodiment, the operating condition characteristic parameters of the ship power system are selected, and the historical operating condition data is subjected to operating condition division operation. The K-means algorithm is used to divide the operating conditions of the ship power system, and the operating condition characteristic parameters include motor speed and environmental temperature. According to the collected historical data, the operating conditions of the system are preliminarily divided into T operating conditions.

[0117] The Smote method is used to expand the data of each operating condition, and the following formula is used:

[0118]

[0119] Where x 合成 is a new sample after synthesis, is a near neighbor sample point of sample point x i , and δ∈[0,1] is a random number. The new sample x 合成 synthesized according to the above formula is a point randomly selected on the line segment connecting the sample point x i and the sample point . Through data expansion, the number of data set samples under each operating condition is balanced.

[0120] On the basis of the above embodiments, as a preferred embodiment, the BP neural network can be used to identify the real-time operating condition of the ship power system. Assuming that the given training sample data set is {(x (1) , y (1) ), (x (2) , y (2) ),..., (x (T) , y (T) )}, where y (i) ∈{1,2,...,T}, T is the number of operating condition categories, the conditional probability of y (i) =t, t={1,2,...,T} can be calculated as α1,α2,...,α T .

[0121] Based on the above embodiments, as a preferred implementation, a state assessment benchmark model is constructed based on the expanded multi-source sensor data:

[0122]

[0123]

[0124] Where K represents the number of Gaussian distributions in the Gaussian mixture model, k represents the iteration over the K Gaussian distributions from 1 to K; p(x|Θ) is the output of the state evaluation baseline model, ω k It is the weight of the k-th Gaussian distribution, and μ k and C k Let N represent the Gaussian probability density function respectively. k =(x|μ k C k The mean and covariance of ). x is an eigenvector composed of selected state assessment feature parameters for the operating condition. Θ={ω k ,μ k C k} represents the set of all parameters of the baseline model for state assessment. N k =(x|μ k C k ) is the k-th Gaussian probability density function.

[0125] Step S3: Obtain the current operating conditions and state assessment benchmark model, construct a state assessment index based on the Mahalanobis distance of the operating conditions and the state assessment benchmark model, and conduct a state assessment of the ship's power system based on the state assessment index.

[0126] Step S31: Based on the current state evaluation feature vector x, the corresponding comprehensive weight vector w, the mean vector μ of the k-th Gaussian distribution of the state evaluation benchmark model under the j′-th operating condition, and the covariance matrix C, determine the first Mahalanobis distance d between the current state evaluation feature vector and each Gaussian distribution. k (x).

[0127]

[0128] Weighting coefficients ω based on Gaussian probability density function k The first Mahalanobis distance is used to determine the second Mahalanobis distance D between the current state evaluation feature vector and the state evaluation baseline model under each operating condition. j’ (x).

[0129]

[0130] In the above formula, D1(x), D2(x), …, D j' (x), …D T (x) respectively represent the second Mahalanobis distance of the state evaluation feature vector at the current time from the reference model of the operating condition j'; ω k represents the probability of each Gaussian distribution, and

[0131] Based on the second Mahalanobis distance, and the probability α j’ , j'∈{T}; determine the third Mahalanobis distance between the state evaluation feature vector at the current time and the state evaluation reference model of T operating conditions, wherein T is the number of operating condition categories.

[0132]

[0133]

[0134] Based on the product of the third Mahalanobis distance and the normal operating state constant value c of the ship power system, a state evaluation index sa(t) is constructed:

[0135] sa(t)=cD(x)

[0136]

[0137] The value range of sa(t) is limited to [0, 1], and the closer the value of sa(t) is to 0, the worse the state of the ship power system at the current time is; the closer the value is to 1, the better the state of the system at the current time is. s represents the number of sliding windows, that is, the number of finally selected state evaluation feature parameters selected in the embodiment is set, and SA(t) represents the average state evaluation index in the sliding window.

[0138] Based on the average state evaluation index in the sliding window, the state of the ship power system is evaluated, wherein if the average state evaluation index of a plurality of consecutive sampling points (such as 8) is greater than a first index threshold, or the state evaluation index of at least one sampling point is greater than a second index threshold, it is judged as an abnormal state; if the average state evaluation index of a plurality of consecutive sampling points is greater than a second index threshold, or the state evaluation index of at least one sampling point is greater than a third index threshold, it is judged as needing immediate shutdown processing; wherein the first index threshold is less than the second index threshold, and the second index threshold is less than the third index threshold. In this embodiment, the first index threshold is 1, the second index threshold is 2, and the third index threshold is 3.

[0139] The embodiment of the present application also provides a multi-source sensing information fusion and expansion device based on system state representation, based on the multi-source sensing information fusion and expansion method based on system state representation in each of the above embodiments, comprising:

[0140] A multi-source data fusion module obtains operation state parameters of each subsystem in the ship power system, performs data cleaning, feature extraction and principal component analysis on the operation state parameters to obtain state evaluation feature parameters; the subsystems include a sliding bearing system, a gear box system and a rolling bearing system;

[0141] A working condition division module divides the operation working condition of the ship power system into multiple categories, extracts state evaluation feature parameters under each category of operation working condition, and constructs a state evaluation reference model of the corresponding operation working condition based on the state evaluation feature parameters;

[0142] A state evaluation module obtains the operation working condition at the current time and the state evaluation reference model, constructs a state evaluation index based on the Mahalanobis distance of the operation working condition and the state evaluation reference model, and performs state evaluation on the ship power system based on the state evaluation index.

[0143] Based on the same concept, the embodiment of the present application also provides an entity structure schematic diagram, as shown in Figure 4 The server can include a processor 810, a communications interface 820, a memory 830 and a communications bus 840, wherein the processor 810, the communications interface 820 and the memory 830 complete mutual communication through the communications bus 840. The processor 810 can call the logic instructions in the memory 830 to execute the steps of the multi-source sensing information fusion and expansion method based on system state representation as described in each of the above embodiments. For example, comprising:

[0144] Obtaining operation state parameters of each subsystem in the ship power system, performing data cleaning, feature extraction and principal component analysis on the operation state parameters to obtain state evaluation feature parameters; the subsystems include a sliding bearing system, a gear box system and a rolling bearing system;

[0145] Dividing the operation working condition of the ship power system into multiple categories, extracting state evaluation feature parameters under each category of operation working condition, and constructing a state evaluation reference model of the corresponding operation working condition based on the state evaluation feature parameters;

[0146] Obtaining the operation working condition at the current time and the state evaluation reference model, constructing a state evaluation index based on the Mahalanobis distance of the operation working condition and the state evaluation reference model, and performing state evaluation on the ship power system based on the state evaluation index.

[0147] Further, the logic instructions in the memory 830 described above can be implemented in the form of software functional units and sold or used as independent products, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or parts of the present application that essentially contribute to the prior art or parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0148] Based on the same concept, the embodiments of the present application also provide a non-transitory computer readable storage medium, which stores a computer program containing at least one code, which can be executed by a host device to control the host device to implement the steps of the multi-source sensing information fusion and expansion method based on system state representation as described in the above embodiments. For example, it includes:

[0149] Obtaining the running state parameters of each subsystem in the ship power system, performing data cleaning, feature extraction and principal component analysis on the running state parameters to obtain state evaluation feature parameters; the subsystems include a sliding bearing system, a gear box system and a rolling bearing system;

[0150] Divide the running conditions of the ship power system into multiple categories, extract the state evaluation feature parameters under each category of running condition, and construct a state evaluation reference model corresponding to the running condition based on the state evaluation feature parameters;

[0151] Obtain the running condition at the current time and the state evaluation reference model, construct a state evaluation index based on the Mahalanobis distance of the running condition and the state evaluation reference model, and perform state evaluation on the ship power system based on the state evaluation index.

[0152] Based on the same technical concept, the embodiments of the present application also provide a computer program, which when executed by a host device, is used to implement the above method embodiments.

[0153] The program can be stored in whole or in part on a storage medium packaged together with the processor, or in part or in whole on a storage medium not packaged together with the processor.

[0154] Based on the same technical concept, the embodiment of the present application further provides a processor for implementing the above method embodiment.

[0155] To sum up, the method and system for multi-source sensor information fusion and expansion based on system state representation provided by the embodiment of the present application take the key fault factors of the slide bearing system parameter group, the gear box system parameter group and the rolling bearing system parameter group, which are not repairable, as the ship power system state evaluation index system, can better reflect the running state of the ship power system, and is conducive to realizing effective evaluation of the power system state; through the fusion method of multi-state parameter characteristics, data missing value filling and denoising processing are performed through data cleaning operation, and multi-source data redundant information is removed through feature selection and feature extraction operation, which can overcome the shortcomings of the existing ship power system relying on single parameter threshold alarm method, and can identify the deterioration process before the failure of the ship power system, accurately evaluate the state change of the ship power system during operation; the standardization processing of each parameter of the parameter group based on the system state index is performed, and meanwhile, the comprehensive weighting method based on principal component analysis and entropy value weighted summation is adopted to ensure the accuracy of weight distribution, thereby effectively improving the accuracy of the subsequently constructed state benchmark model, and further improving the accuracy of the state evaluation, and further effectively improving the rationality of the evaluation result.

[0156] The embodiments of the present application can be combined arbitrarily to achieve different technical effects.

[0157] In the above embodiments, all or part of the embodiments can be realized by software, hardware, firmware or any combination thereof. When realized by software, all or part of the embodiments can be realized in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available medium can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), optical medium (for example, DVD), or semiconductor medium (for example, solid state disk) and the like.

[0158] Those skilled in the art can understand that all or part of the processes in the above-mentioned method embodiments can be implemented by a computer program instructing the relevant hardware to complete, and the program can be stored in a computer readable storage medium. When the program is executed, it can include the processes of the above-mentioned method embodiments. The aforementioned storage medium includes ROM or random storage memory RAM, magnetic disc or optical disc and various storage code medium.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for fusing and expanding multi-source sensor information based on system state representation, characterized in that, include: The operational status parameters of each subsystem in the ship's power system are obtained. These parameters are then cleaned, feature extracted, and subjected to principal component analysis to obtain state assessment feature parameters. Specifically, this includes interpolating missing values ​​in the parameter group using Lagrange interpolation, and using 3D modeling... σ The criteria are used to screen out and remove noisy data, and the missing values ​​after removing noisy data are filled by interpolation based on the Lagrange interpolation method; feature selection is performed on each operating state parameter in the parameter group based on the Pearson correlation coefficient method to extract state assessment feature parameters; the state assessment feature parameters are standardized and subjected to secondary feature extraction, and the weight of each state assessment feature parameter in the parameter group is determined; the subsystem includes a sliding bearing system, a gearbox system and a rolling bearing system, and the parameter group includes a time series and the values ​​of each operating state parameter corresponding to the time series; The operating conditions of the ship's power system are divided into multiple categories. State assessment feature parameters are extracted for each category of operating conditions. Based on the state assessment feature parameters, a state assessment benchmark model for the corresponding operating conditions is constructed. Obtain the current operating conditions and state assessment benchmark model, construct a state assessment index based on the Mahalanobis distance of the operating conditions and the state assessment benchmark model, and conduct a state assessment of the ship's power system based on the state assessment index. The state assessment feature parameters are standardized and subjected to secondary feature extraction, and the weights of each operating state parameter in the parameter group are determined, specifically including: Based on the multi-index comprehensive evaluation method and the preset index range, the state evaluation characteristic parameters in each parameter group are divided into positive indicators, moderate indicators and negative indicators; and standardized processing is performed based on the corresponding strategies. If the state assessment feature parameter is determined to be a positive indicator and the measured value of the operating state parameter is not greater than the average value; or if the state assessment feature parameter is determined to be a negative indicator and the measured value of the operating state parameter is greater than the average value; then the operating state parameter after standard deviation processing is determined to be 0. If the state assessment feature parameter is determined to be a positive indicator, and the measured value of the operating state parameter is greater than the average value; or if the state assessment feature parameter is determined to be an appropriate indicator; or if the state assessment feature parameter is determined to be a negative indicator, and the measured value of the operating state parameter is not greater than the average value; then the operating state parameter after standard deviation processing is determined to be: the difference between the measured value and the average value divided by the standard deviation of the operating state parameter. The parameter group is subjected to secondary feature extraction based on principal component analysis, and the initial weights of each state evaluation feature parameter in the parameter group are calculated based on the entropy method. The initial weights are then corrected based on the weighted average method.

2. The method for multi-source sensor information fusion and expansion based on system state representation according to claim 1, characterized in that, The operating parameters of the sliding bearing system include temperature parameters and pressure parameters; the temperature parameters include the first bearing temperature parameter of the sliding bearing at the drive end and the second bearing temperature parameter of the sliding bearing at the non-drive end; the operating parameters of the sliding bearing system also include lubricating oil temperature and cooling water temperature; the pressure parameters include the first lubricating oil pressure parameter of the non-drive sliding bearing and the second lubricating oil pressure parameter of the non-drive sliding bearing. The operating status parameters of the gearbox system include a first acceleration parameter, which includes acceleration collected at multiple measuring points on the gearbox. The operating status parameters of the rolling bearing system include a second acceleration parameter, which includes the acceleration collected at each rolling bearing. After obtaining the operating status parameters of each subsystem in the ship's power system, the following is also included: Establish time-series-based parameter groups for each of the subsystems; The parameter group includes the sliding bearing system parameter group, the gearbox system parameter group, and the rolling bearing system parameter group.

3. The method for multi-source sensor information fusion and expansion based on system state representation according to claim 1, characterized in that, Based on the Pearson correlation coefficient method, feature selection is performed on each operating state parameter in the parameter group to extract state assessment feature parameters, specifically including: In the sliding bearing system parameter group, the bearing temperature of the sliding bearing at the drive end is used as the reference temperature. The first Pearson correlation coefficient between the remaining temperature parameters and the reference temperature is determined. Temperature parameters with a first Pearson correlation coefficient lower than a preset first correlation coefficient threshold are removed to obtain the state evaluation characteristic parameters corresponding to the sliding bearing system parameter group. In the parameter groups of the gearbox system and the rolling bearing system, an acceleration is randomly selected as a reference acceleration. The second Pearson correlation coefficient between the remaining accelerations in the gearbox system parameter group and the reference acceleration is determined. Accelerations with a second Pearson correlation coefficient higher than the second correlation coefficient threshold are removed. The time-domain and frequency-domain characteristic parameters of each acceleration in the gearbox system and the rolling bearing system are extracted to obtain the state evaluation characteristic parameters corresponding to the gearbox system parameter group and the rolling bearing system parameter group.

4. The method for multi-source sensor information fusion and expansion based on system state representation according to claim 1, characterized in that, Based on the aforementioned condition assessment feature parameters, a condition assessment benchmark model for the corresponding operating condition is constructed, specifically including: Obtain the set of state assessment feature parameters corresponding to the current category of operating conditions, and form a state assessment feature vector based on the state assessment feature parameters; Determine the weight of each state evaluation feature parameter in the state evaluation feature vector, and model the distribution of the state evaluation feature parameters as a Gaussian probability density function; A baseline model for the current category of operating conditions is constructed based on the weight of each of the aforementioned state assessment feature parameters and the corresponding Gaussian probability density function.

5. The method for multi-source sensor information fusion and expansion based on system state representation according to claim 4, characterized in that, Based on the operating conditions and the Mahalanobis distance of the condition assessment benchmark model, a condition assessment index is constructed. Based on this index, a condition assessment of the ship's propulsion system is performed, specifically including: Based on the current state assessment feature vector, the corresponding weight, the mean vector of the Gaussian distribution of the state assessment benchmark model under each operating condition, and the covariance matrix, the first Mahalanobis distance between the current state assessment feature vector and each Gaussian distribution is determined. Based on the weighting coefficients of the Gaussian probability density function and the first Mahalanobis distance, the second Mahalanobis distance between the current state assessment feature vector and the state assessment benchmark model under each operating condition is determined. Based on the second Mahalanobis distance and the probability that the current operating condition of the ship's power system belongs to each category, the third Mahalanobis distance between the current state assessment feature vector and the state assessment benchmark model of T operating conditions is determined, where T is the number of operating condition categories. A state evaluation index is constructed based on the product of the third Mahalanobis distance and the normal operating state constant value of the ship's propulsion system. The ship's power system is assessed based on the average state evaluation index within a sliding window. If the average state evaluation index of multiple consecutive sampling points is greater than a first index threshold, or the state evaluation index of at least one sampling point is greater than a second index threshold, then it is determined to be an abnormal state. If the average state evaluation index of multiple consecutive sampling points is greater than the second index threshold, or the state evaluation index of at least one sampling point is greater than a third index threshold, then it is determined that immediate shutdown is required. The first index threshold is less than the second index threshold, and the second index threshold is less than the third index threshold.

6. A multi-source sensor information fusion and expansion device based on system state representation, characterized in that, include: The multi-source data fusion module acquires the operating status parameters of each subsystem in the ship's power system, performs data cleaning, feature extraction, and principal component analysis on these parameters to obtain state assessment feature parameters. Specifically, this includes: interpolating missing values ​​in the parameter group using Lagrange interpolation, and... σ The criteria are used to screen out and remove noisy data, and the missing values ​​after removing noisy data are filled by interpolation based on the Lagrange interpolation method; feature selection is performed on each operating state parameter in the parameter group based on the Pearson correlation coefficient method to extract state assessment feature parameters; the state assessment feature parameters are standardized and subjected to secondary feature extraction, and the weight of each state assessment feature parameter in the parameter group is determined; the subsystem includes a sliding bearing system, a gearbox system and a rolling bearing system, and the parameter group includes a time series and the values ​​of each operating state parameter corresponding to the time series; The operating condition classification module divides the operating conditions of the ship's power system into multiple categories, extracts the state assessment feature parameters for each category of operating conditions, and constructs a state assessment benchmark model for the corresponding operating conditions based on the state assessment feature parameters. The status assessment module obtains the current operating conditions and status assessment benchmark model, constructs status assessment indices based on the Mahalanobis distance of the operating conditions and the status assessment benchmark model, and performs status assessment on the ship's power system based on the status assessment indices. The state assessment feature parameters are standardized and subjected to secondary feature extraction, and the weights of each operating state parameter in the parameter group are determined, specifically including: Based on the multi-index comprehensive evaluation method and the preset index range, the state evaluation characteristic parameters in each parameter group are divided into positive indicators, moderate indicators and negative indicators; and standardized processing is performed based on the corresponding strategies. If the state assessment feature parameter is determined to be a positive indicator and the measured value of the operating state parameter is not greater than the average value; or if the state assessment feature parameter is determined to be a negative indicator and the measured value of the operating state parameter is greater than the average value; then the operating state parameter after standard deviation processing is determined to be 0. If the state assessment feature parameter is determined to be a positive indicator, and the measured value of the operating state parameter is greater than the average value; or if the state assessment feature parameter is determined to be an appropriate indicator; or if the state assessment feature parameter is determined to be a negative indicator, and the measured value of the operating state parameter is not greater than the average value; then the operating state parameter after standard deviation processing is determined to be: the difference between the measured value and the average value divided by the standard deviation of the operating state parameter. The parameter group is subjected to secondary feature extraction based on principal component analysis, and the initial weights of each state evaluation feature parameter in the parameter group are calculated based on the entropy method. The initial weights are then corrected based on the weighted average method.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the multi-source sensor information fusion and expansion method based on system state representation as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the multi-source sensor information fusion and expansion method based on system state representation as described in any one of claims 1 to 5.

Images