Operating parameter optimization control system for engine coupler

Abstract

The invention belongs to the technical field of intelligent manufacturing, and discloses an engine coupling operation parameter optimization control system. Comprising a parameter acquisition module used for acquiring coupling operation parameters and coupling process parameters; the parameter classification module is used for dividing the coupling process parameters into a groups and matching each group with different parameters in the coupling operation parameters; the parameter identification module is used for carrying out anomaly analysis on the coupling operation parameters, identifying risk parameters in the coupling operation parameters, obtaining coupling process parameters corresponding to the risk parameters and marking the coupling process parameters as detection parameters; the parameter detection module is used for carrying out anomaly detection on the detection parameters and identifying to-be-adjusted parameters in the detection parameters; the parameter optimization module is used for formulating a process optimization strategy and optimizing the parameters to be adjusted; the integral performance and production efficiency of the coupler can be effectively improved, and higher fuel economy and durability are achieved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and more specifically, to an operating parameter optimization control system for an engine coupling. Background Art

[0002] In an automotive power system, as a key transmission component between the engine and the transmission, the performance of the engine coupling directly affects the power output, fuel economy, and durability of the whole vehicle; the operating state of the coupling is affected by various factors such as material properties, manufacturing processes, assembly accuracy, and vehicle operating conditions. Therefore, in the research and development and production processes of the coupling, optimizing its process parameters is crucial; currently, traditional coupling manufacturing processes mainly rely on experience and standardized parameter settings, with insufficient analysis of test data during the actual operation process, resulting in some products possibly showing problems such as low efficiency, high noise, and short fatigue life under actual working conditions; in addition, the production of the coupling involves multiple processes, including material selection, heat treatment, precision machining, and assembly, and minor deviations in process parameters will affect the quality of the final product; therefore, there is an urgent need for an intelligent parameter control system to improve the overall performance and production efficiency of the coupling.

[0003] In recent years, the rise of intelligent manufacturing has brought new opportunities to the production of automotive parts; by adopting advanced sensor technology, data acquisition and analysis technology, and automated control systems, real-time monitoring and optimization of parameters can be achieved; it can not only improve the accuracy and efficiency of the production process, but also significantly shorten the product development cycle and reduce production costs; for example, the patent with the publication number CN118795853A discloses an optimization system for the processing technology of automotive shock pad molds; and another example, the patent with the bulletin number CN116165988B discloses a production quality control method and system for an automotive center console; both have achieved the optimization control of process parameters during the production process of automotive parts.

[0004] However, the existing technologies mainly rely on preset process parameters, quality standards, or recommended models for process parameter optimization, and do not use the operating parameters of actual products to guide process adjustment, resulting in the optimization results not fully matching the actual usage, thus reducing the adaptability of process parameter optimization; at the same time, the existing technologies rely on fixed optimization strategies, such as quality score analysis, fixed processing parameter recommendations, etc., and fail to dynamically adjust the optimization plan in combination with actual operating parameters, resulting in a reduction in the accuracy and flexibility of the process parameter optimization process.

[0005] In view of this, the present invention proposes an operating parameter optimization control system for an engine coupling to solve the above problems. Summary of the Invention

[0006] To overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solution: An engine coupling operation parameter optimization control system, comprising:

[0007] A parameter acquisition module, configured to acquire coupling operation parameters and coupling process parameters;

[0008] A parameter classification module, configured to divide the coupling process parameters into a groups, and respectively match each group with different parameters in the coupling operation parameters;

[0009] A parameter identification module, configured to perform abnormal analysis on the coupling operation parameters, calculate a parameter abnormality coefficient, identify risk parameters in the coupling operation parameters based on the parameter abnormality coefficient, and obtain the coupling process parameters corresponding to the risk parameters according to the group corresponding to the risk parameters, and mark them as detection parameters;

[0010] A parameter detection module, configured to acquire historical parameters, construct an abnormal detection architecture based on the historical parameters, and perform abnormal detection on the detection parameters according to the abnormal detection architecture to identify the parameters to be adjusted in the detection parameters;

[0011] A parameter optimization module, configured to formulate a process optimization strategy, generate corresponding optimized values for each parameter to be adjusted by using the process optimization strategy, and optimize the parameter to be adjusted according to the optimized values.

[0012] Further, the coupling operation parameters are dynamic characteristics involved in the working process of the coupling; the coupling process parameters are operation parameters involved in the manufacturing process of the coupling;

[0013] The step of dividing the coupling process parameters into a groups includes:

[0014] Step S101: Adopt a pre-trained word embedding model to convert each parameter in the coupling process parameters into a corresponding text vector;

[0015] Step S102: Take each text vector as a node;

[0016] Step S103: Screen a nodes as center points, and take the nodes that are not used as center points as sample points, where a is the number of parameters in the coupling operation parameters;

[0017] Step S104: Construct corresponding a groups according to the a center points, and calculate the sample distance from each sample point to each center point in turn;

[0018] Step S105: Preset a distance threshold, compare each sample distance corresponding to each sample point with the distance threshold respectively, mark the center point corresponding to the sample distance with a value less than the distance threshold as a classification point, and do not mark the sample distance with a value greater than or equal to the distance threshold. Then divide each sample point into the group corresponding to the corresponding classification point in sequence;

[0019] Step S106: Calculate the new centroid corresponding to each group, and replace the center point of each group with the corresponding new centroid;

[0020] Step S107: Loop steps S104 to S106 until the new centroids of each group calculated in step S106 are the same as the corresponding new centroids calculated in the previous loop process, then the loop ends, and a groups are obtained.

[0021] Further, in the step S103, the steps of screening a nodes as center points include:

[0022] Step S201: Randomly screen a node as a center point;

[0023] Step S202: Calculate the sample distance from each sample point to each center point, compare the sample distances corresponding to the same sample point, and take the sample distance with the smallest value as the minimum distance of the corresponding sample point;

[0024] Step S203: Square each minimum distance of each sample point in sequence to obtain the squared distance;

[0025] Step S204: Add up the squared distances of each sample point in sequence to obtain the sum of squared distances; divide the squared distance of each sample point by the sum of squared distances respectively to obtain the screening probability corresponding to each sample point;

[0026] Step S205: Screen a node as a center point according to the screening probability;

[0027] Step S206: Loop steps S202 to S205 until a nodes are screened out, then the loop ends;

[0028] In the step S104, the method for calculating the sample distance is: calculate the cosine similarity between the sample point and the center point, and take the reciprocal of the cosine similarity as the sample distance;

[0029] In the step S106, the method for calculating the new centroid includes:

[0030] Count the number of nodes in each group and label it as the node count; add up the nodes in each group in sequence and then divide by the corresponding node count to obtain the average point corresponding to each group; calculate the sample distance from each node in each group to the corresponding average point and label it as the average distance; compare the same average distances of the corresponding groups, and take the node corresponding to the smallest average distance value as the new centroid of the corresponding group.

[0031] Further, in the step S205, the method for screening out nodes according to the screening probability includes:

[0032] Sort each node according to the corresponding screening probability from high to low to generate a node sorting table; according to the forward order of the node sorting table, add the screening probability of each node to the screening probability of the corresponding previous node in sequence to obtain the cumulative probability of each node; calculate the probability range of each node according to the cumulative probability of each node; generate a random number from the interval [0, 1], mark the probability range where the random number falls as the screening range, and screen out the nodes corresponding to the screening range; where, mark a node as the current node, and the previous nodes of the current node are all the nodes ranked in front of the current node in the node sorting table; the maximum value of the probability range corresponding to the current node is the cumulative probability corresponding to the current node, and the minimum value of the probability range corresponding to the current node is the cumulative probability corresponding to the node ranked one position in front of the current node in the node sorting table;

[0033] The method for matching each group with different parameters in the coupling operation parameters includes:

[0034] Adopt a pre-trained word embedding model to convert each parameter in the coupling operation parameters into a corresponding text vector and label it as the operation vector; calculate the sample distance from each operation vector to the center point of each group respectively and label it as the operation distance; sort the operation distances corresponding to each operation vector from small to large to generate a distance sorting table corresponding to each operation vector; mark the operation distance ranked first in each distance sorting table as the matching distance, match each operation vector with the group corresponding to the corresponding matching distance respectively, and match each group with the parameter corresponding to the corresponding operation vector respectively.

[0035] Further, the method for calculating the parameter anomaly coefficient includes:

[0036] Preset parameter thresholds, where the parameter thresholds include the thresholds corresponding to each parameter in the coupling operation parameters; each parameter in the coupling operation parameters and its corresponding threshold are used as a set of calculation sets, and the calculation sets correspond one by one to the parameters in the coupling operation parameters; each set of calculation sets is sequentially input into the trained coefficient calculation model to calculate the corresponding parameter anomaly coefficient; among them, the parameter anomaly coefficient is a 1×a matrix, and the elements in the matrix correspond one by one to the parameters in the coupling operation parameters, and the coefficient calculation model is a deep neural network model; the training process of the coefficient calculation model includes:

[0037] Pre-collect b sets of data sets, each set of data sets includes a sets of different calculation sets, and the b sets of data sets are all different. Set the corresponding parameter anomaly coefficients for the b sets of data sets, where b is an integer greater than 1. Convert the data sets and the corresponding parameter anomaly coefficients into a corresponding set of feature vectors; use each set of feature vectors as the input of the coefficient calculation model. The coefficient calculation model outputs a set of predicted parameter anomaly coefficients corresponding to each data set, and uses the actual parameter anomaly coefficient corresponding to each data set as the prediction target. The actual parameter anomaly coefficient is the parameter anomaly coefficient preset corresponding to the data set; use minimizing the sum of the prediction errors of all data sets as the training target; train the coefficient calculation model until the sum of the prediction errors reaches convergence and then stop training;

[0038] The method for identifying risk parameters in the coupling operation parameters includes:

[0039] Preset an anomaly threshold, and compare each element in the parameter anomaly coefficient with the anomaly threshold respectively; mark the elements with values greater than or equal to the anomaly threshold as abnormal elements, and do not mark the elements with values less than the anomaly threshold; use the parameters corresponding to the abnormal elements in the coupling operation parameters as risk parameters.

[0040] Further, the historical parameters are the normal parameters obtained at historical moments, and the normal parameters are the coupling process parameters in the normal state;

[0041] The anomaly detection architecture includes d detection sub-architectures, where d is the number of parameters in the coupling process parameters, and the detection sub-architectures correspond one by one to the parameters in the coupling process parameters; obtain the detection sub-architecture corresponding to the detection parameter in the anomaly detection architecture and mark it as the running sub-architecture; the steps for performing anomaly detection on the corresponding detection parameter based on one running sub-architecture include:

[0042] Step S301: Mark the detection parameter for anomaly detection as the current parameter, screen out the parameter corresponding to the current parameter from the historical parameters and mark it as the reference parameter, and collectively call the reference parameter and the current parameter the analysis parameters;

[0043] Step S302: Obtain the neighborhood parameters corresponding to each analysis parameter, subtract each neighborhood parameter corresponding to each analysis parameter from the corresponding analysis parameter respectively, and obtain the parameter difference corresponding to each analysis parameter; the neighborhood parameter corresponding to the analysis parameter is the remaining r analysis parameters, and r is the number of reference parameters;

[0044] Step S303: Obtain the standard span of each analysis parameter according to the parameter difference;

[0045] Step S304: Use the neighborhood parameters corresponding to the parameter differences ranked in the top y positions in each element sorting table as the standard parameters corresponding to the analysis parameters corresponding to the corresponding difference set, where 1 < y < r + 1;

[0046] Step S305: Calculate the adjacent span between each analysis parameter and each corresponding standard parameter in turn according to the parameter difference and the standard span, and calculate the regional distribution degree corresponding to each analysis parameter;

[0047] Step S306: Calculate the outlier coefficient of the current parameter based on the regional distribution degree, compare the outlier coefficient with the preset recognition coefficient. If the outlier coefficient is greater than the recognition coefficient, mark the current parameter as a parameter to be adjusted. If the outlier coefficient is less than or equal to the recognition coefficient, do not mark the current parameter.

[0048] Further, in the step S303, the method for obtaining the standard span includes:

[0049] Take the parameter differences of each analysis parameter as a set of difference sets, and the difference sets correspond to the analysis parameters one by one; sort the parameter differences in each difference set from small to large to generate a difference sorting table corresponding to each difference set; take the parameter difference ranked in the y-th position in each difference sorting table as the standard span of the analysis parameter corresponding to the corresponding difference set;

[0050] In the step S305, the method for calculating the adjacent span includes:

[0051] Mark the parameter differences between each analysis parameter and each corresponding standard parameter as adjacent differences; take one adjacent difference corresponding to each analysis parameter and the standard span as a set of span sets, that is, each set of span sets includes one adjacent difference and one standard span; collectively call the adjacent difference and the standard span in each span set as the analysis span, compare each analysis span in each span set respectively, and take the largest analysis span value as the adjacent span between the corresponding analysis parameter and the corresponding standard parameter;

[0052] The method for calculating the regional distribution degree includes:

[0053] Add the adjacent spans corresponding to each analysis parameter in sequence to obtain the comprehensive span corresponding to each analysis parameter; multiply the reciprocal of the comprehensive distance corresponding to each analysis parameter by y to obtain the regional distribution degree corresponding to each analysis parameter.

[0054] In the step S306, the method for calculating the outlier coefficient includes:

[0055] Obtain the standard parameter corresponding to the real-time parameter and mark it as the calculation parameter; divide the regional distribution degree corresponding to each calculation parameter by the regional distribution degree corresponding to the real-time element to obtain the relative distribution degree corresponding to each calculation parameter; add the relative densities of each evaluation element in sequence and then divide by y to obtain the outlier coefficient corresponding to the real-time parameter.

[0056] Further, the steps for generating the corresponding optimization value for each parameter to be adjusted include:

[0057] Step S401: Obtain the parameter range, and construct m groups of optimization sets according to the parameter range. Each group of optimization sets includes the adjustment values corresponding to each parameter to be adjusted.

[0058] Step S402: Define the initialization set. The initialization set includes candidate solutions, iteration thresholds, removal strategy sets, reconstruction strategy sets, strategy weight sets, and update factors, and set the number of iterations to 0; among them, the candidate solution is a group of optimization sets in the m groups of optimization sets.

[0059] Step S403: According to the strategy weights in the strategy weight set, calculate the removal selection probability of each removal strategy in the removal strategy set, and the reconstruction selection probability corresponding to each reconstruction strategy in the reconstruction strategy set.

[0060] Step S404: Select a removal strategy from the removal strategy set according to the removal selection probability, and obtain the strategy weight corresponding to the removal strategy from the strategy weight set and mark it as the removal weight.

[0061] Step S405: Perform a removal operation on the candidate solution using the removal strategy to obtain a partial solution.

[0062] Step S406: Select a reconstruction strategy from the reconstruction strategy set according to the reconstruction selection probability, and obtain the strategy weight corresponding to the reconstruction strategy from the strategy weight set and mark it as the reconstruction weight.

[0063] Step S407: Perform a reconstruction operation on the partial solution using the reconstruction strategy to obtain a reconstructed solution.

[0064] Step S408: Calculate the optimization effect of the reconstructed solution, calculate the corresponding update probability based on the optimization effect, and determine whether to update the candidate solution to the reconstructed solution according to the update probability.

[0065] Step S409: Adjust the removal weight and the reconstruction weight respectively according to the update result;

[0066] Step S410: Determine whether the number of iterations is less than the iteration threshold. If so, increment the number of iterations by one and return to Step S403. If not, proceed to Step S411;

[0067] Step S411: Obtain the optimization set corresponding to the candidate solution, and use each adjustment value in the optimization set as the optimized value of the corresponding parameter to be adjusted.

[0068] Furthermore, in Step S401, the parameter range includes the normal range corresponding to each parameter to be adjusted. The maximum value of the normal range corresponding to each parameter to be adjusted is the normal parameter with the largest value among the corresponding historical parameters, and the minimum value of the normal range corresponding to each parameter to be adjusted is the normal parameter with the smallest value among the corresponding historical parameters; Randomly select a value from each normal range and construct a set of optimization sets. A total of m sets of optimization sets are constructed, and the m sets of optimization sets are all different;

[0069] In Step S403, the method for calculating the removal selection probability is: Add up the strategy weights corresponding to each removal strategy in turn to obtain the total removal value; Divide the strategy weight corresponding to each removal strategy by the total removal value to obtain the removal selection probability corresponding to each removal strategy;

[0070] The method for calculating the reconstruction selection probability is: Add up the strategy weights corresponding to each reconstruction strategy in turn to obtain the total reconstruction value; Divide the strategy weight corresponding to each reconstruction strategy by the total reconstruction value to obtain the reconstruction selection probability corresponding to each reconstruction strategy;

[0071] In Step S404, the method for selecting a removal strategy from the removal strategy set is the same as the method for screening out nodes according to the screening probability in Step S205;

[0072] In Step S406, the method for selecting a reconstruction strategy from the reconstruction strategy set is the same as the method for screening out nodes according to the screening probability in Step S205.

[0073] Furthermore, in Step S408, the method for calculating the optimization effect of the reconstruction solution includes:

[0074] Obtain the optimized set corresponding to the reconstructed solution and mark it as the current set; replace the value of each parameter to be adjusted in the coupling process parameters with the corresponding adjustment value in the current set, and mark the coupling process parameters after replacement as the optimized process parameters; use the optimized process parameters, coupling process parameters, and coupling operation parameters as prediction parameters, and input the prediction parameters into the trained parameter prediction model to predict the corresponding optimized operation parameters; the training process of the parameter prediction model is the same as that of the coefficient calculation model, and both are deep neural network models; input the optimized operation parameters and the parameter threshold into the coefficient calculation model to calculate the corresponding parameter anomaly coefficient and mark it as the optimized anomaly coefficient; compare each element in the optimized anomaly coefficient with the anomaly threshold respectively to identify the abnormal elements in the optimized anomaly coefficient; count the number of abnormal elements in the optimized anomaly coefficient and mark it as the first abnormal number; count the number of abnormal elements in the parameter anomaly coefficient and mark it as the second abnormal number; subtract the first abnormal number from the second abnormal number to obtain the optimization effect.

[0075] The method for calculating the update probability is as follows: obtain the iteration number corresponding to the current iteration process and mark it as the current number; multiply the current number by the preset influence factor and then by the update factor to obtain the current factor; calculate the optimization effect of the candidate solution, mark the optimization effect of the candidate solution as the candidate effect, and mark the optimization effect of the reconstructed solution as the reconstruction effect; compare the candidate effect with the reconstruction effect; if the candidate effect is less than or equal to the reconstruction effect, the corresponding update probability is 1; if the candidate effect is greater than the reconstruction effect, subtract the candidate effect from the reconstruction effect and divide by the current factor to obtain the update coefficient; use the natural constant as the base and the update coefficient as the exponent to perform an exponential operation to obtain the update probability.

[0076] The method for determining whether to update the candidate solution to the reconstructed solution includes:

[0077] If the update probability is 1, update the candidate solution to the reconstructed solution; if the update probability is not 1, set the update range [0, k] and the retention range (k, 1], where k is the update probability; randomly generate a judgment value from the interval [0, 1], if the judgment value is within the update range, update the candidate solution to the reconstructed solution, if the judgment value is within the retention range, do not update the candidate solution to the reconstructed solution.

[0078] In the step S409, the method for adjusting the removal weight and the reconstruction weight respectively includes:

[0079] Subtract the candidate effect from the reconstruction effect to obtain the adjustment effect; multiply the adjustment effect by the current factor to obtain the adjustment degree; adjust the removal weight and the reconstruction weight respectively according to the adjustment degree.

[0080] Technical effects and advantages of an engine coupling operation parameter optimization control system of the present invention:

[0081] By classifying the coupling process parameters and matching them with the coupling operation parameters, an association between the process parameters and operation parameters of the coupling is established; using deep technology to perform abnormal analysis and risk identification on the coupling operation parameters to accurately discover potential problems existing in the coupling operation process; constructing an abnormal detection architecture to perform real-time abnormal detection and identification on the identified detection parameters, being able to timely discover parameters that need to be optimized and adjusted, improving the pertinence of the optimization effect; adopting a dynamic process optimization strategy to generate corresponding optimization values for different parameters to be adjusted, realizing flexible optimization of the process parameters, effectively improving the overall performance and production efficiency of the coupling; not only enhancing the dynamic performance and manufacturing accuracy of the coupling, but also shortening the development cycle, achieving higher fuel economy and durability, and enhancing the overall performance of the automotive power system. Brief Description of the Drawings

[0082] Figure 1 It is a flowchart of an engine coupling operation parameter optimization control system according to Embodiment 1 of the present invention;

[0083] Figure 2 It is a schematic diagram of an engine coupling operation parameter optimization control system according to Embodiment 1 of the present invention. Detailed Embodiments

[0084] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0085] Embodiment 1

[0086] Please refer to Figure 1 and Figure 2 As shown, an engine coupling operation parameter optimization control system described in this embodiment includes a parameter acquisition module, a parameter classification module, a parameter identification module, a parameter detection module, and a parameter optimization module; each module is connected by wired and / or wireless means to realize data transmission between modules.

[0087] The parameter acquisition module is used to acquire the coupling operation parameters and the coupling process parameters.

[0088] The operating parameters of the coupling refer to the dynamic characteristics involved in the working process of the coupling, which are used to describe the performance of the coupling under working conditions; the operating parameters of the coupling include, for example, the maximum rotational speed, the maximum torque transmission capacity, the maximum vibration, the maximum temperature, etc.; among them, if the maximum rotational speed does not meet the rotational speed requirement, process parameters such as the quenching temperature and tempering time in the heat treatment stage, and the grinding accuracy in the finishing stage should be optimized and controlled to improve the hardness and balance of the coupling; if the maximum torque transmission capacity does not meet the torque transmission capacity requirement, process parameters such as the carburizing temperature and carburizing depth in the carburizing heat treatment stage, and the shot peening pressure in the surface strengthening stage should be optimized and controlled to improve the bearing capacity and fatigue resistance of the coupling; if the maximum vibration exceeds the vibration limit, process parameters such as the tool feed rate in the turning process stage and the dynamic balance grade in the dynamic balance correction stage should be optimized and controlled to improve the coaxiality of the coupling and the uniformity of the rotor mass distribution; if the maximum temperature exceeds the temperature limit, process parameters such as the carburizing temperature and tempering temperature in the heat treatment stage, and the interference fit of the shaft hole in the surface treatment stage should be optimized and controlled to improve the high-temperature resistance of the coupling and the mating stability in a high-temperature environment.

[0089] The process parameters of the coupling refer to the operating parameters involved in the manufacturing process of the coupling, which are used to describe the conditions that need to be controlled and optimized during the production and processing of the coupling, and directly affect the machining accuracy, mechanical properties, and service life of the coupling; the process parameters of the coupling include, for example, the quenching temperature, the carburizing temperature, the shot peening pressure, the grinding accuracy, etc.

[0090] A parameter classification module is used to divide the process parameters of the coupling into a groups, and each group is respectively matched with different parameters in the operating parameters of the coupling.

[0091] The steps of dividing the process parameters of the coupling into a groups include:

[0092] Step S101: Using a pre-trained word embedding model (such as Word2Vec, GloVe, BERT, etc.), convert each parameter in the process parameters of the coupling into a corresponding text vector;

[0093] Step S102: Take each text vector as a node;

[0094] Step S103: Select a nodes as the center points, and take the nodes that are not used as the center points as the sample points, where a is the number of parameters in the operating parameters of the coupling;

[0095] Step S104: Construct a corresponding a groups according to the a center points, and calculate the sample distance from each sample point to each center point in turn;

[0096] Step S105: Preset a distance threshold, compare each sample distance corresponding to each sample point with the distance threshold respectively, mark the center point corresponding to the sample distance with a value less than the distance threshold as a classification point, and do not mark the sample distance with a value greater than or equal to the distance threshold. Then divide each sample point into the group corresponding to the corresponding classification point in sequence; the distance threshold is preset by those skilled in the art according to the actual situation.

[0097] Step S106: Calculate the new centroid corresponding to each group, and replace the center point of each group with the corresponding new centroid.

[0098] Step S107: Loop through steps S104 to S106 until the new centroids of each group calculated in step S106 are the same as the corresponding new centroids calculated in the previous loop, then the loop ends, and a groups are obtained.

[0099] In the above step S103, the steps of screening a nodes as center points include:

[0100] Step S201: Randomly screen a node as a center point.

[0101] Step S202: Calculate the sample distance from each sample point to each center point, compare the sample distances corresponding to the same sample point, and take the sample distance with the smallest value as the minimum distance of the corresponding sample point.

[0102] Step S203: Square each minimum distance of each sample point in sequence to obtain the squared distance.

[0103] Step S204: Add up the squared distances of each sample point in sequence to obtain the sum of squared distances; divide the squared distance of each sample point by the sum of squared distances respectively to obtain the screening probability corresponding to each sample point.

[0104] Step S205: Screen a node as a center point according to the screening probability.

[0105] Step S206: Loop through steps S202 to S205 until a nodes are screened, then the loop ends.

[0106] In the above step S104, the method for calculating the sample distance is: calculate the cosine similarity between the sample point and the center point, and take the reciprocal of the cosine similarity as the sample distance; the calculation method of the cosine similarity is prior art and will not be elaborated here.

[0107] In the above step S106, the method for calculating the new centroid includes:

[0108] Count the number of nodes in each group and label it as the number of nodes; add up the nodes in each group in sequence, and then divide by the corresponding number of nodes to obtain the average point corresponding to each group; calculate the sample distance from each node in each group to the corresponding average point and label it as the average distance; compare the average distances of the same group, and use the node corresponding to the smallest average distance value as the new centroid of the corresponding group.

[0109] In the above step S205, the method of screening nodes according to the screening probability includes:

[0110] Sort each node in descending order according to the corresponding screening probability to generate a node sorting table; according to the forward order of the node sorting table, add the screening probability of each node to the screening probability of the corresponding previous node in sequence to obtain the cumulative probability of each node; calculate the probability range of each node according to the cumulative probability of each node; generate a random number from the interval [0,1], mark the probability range where the random number falls as the screening range, and screen out the nodes corresponding to the screening range; where, mark a node as the current node, and the previous nodes of the current node are all the nodes ranked before the current node in the node sorting table; the maximum value of the probability range corresponding to the current node is the cumulative probability corresponding to the current node, and the minimum value of the probability range corresponding to the current node is the cumulative probability corresponding to the node ranked one position before the current node in the node sorting table.

[0111] The method of matching each group with different parameters in the coupling operation parameters includes:

[0112] Adopt a pre-trained word embedding model to convert each parameter in the coupling operation parameters into a corresponding text vector and label it as the operation vector; calculate the sample distance from each operation vector to the center point of each group respectively and label it as the operation distance; sort the operation distances corresponding to each operation vector from small to large to generate a distance sorting table corresponding to each operation vector; mark the operation distance ranked first in each distance sorting table as the matching distance, match each operation vector with the group corresponding to the corresponding matching distance respectively, and match each group with the parameter corresponding to the corresponding operation vector respectively.

[0113] The parameter identification module is used to perform abnormal analysis on the coupling operation parameters, calculate the parameter abnormal coefficient, identify the risk parameters in the coupling operation parameters based on the parameter abnormal coefficient, and obtain the coupling process parameters corresponding to the risk parameters according to the group corresponding to the risk parameters, and label them as the detection parameters.

[0114] The method of calculating the parameter abnormal coefficient includes:

[0115] Preset parameter thresholds, where the parameter thresholds include the thresholds corresponding to each parameter in the coupling operation parameters, and the parameter thresholds are preset by those skilled in the art according to the coupling performance requirements; each parameter in the coupling operation parameters is respectively used as a group to form a calculation set with the corresponding threshold, and the calculation sets correspond one by one to the parameters in the coupling operation parameters; each group of calculation sets is sequentially input into the trained coefficient calculation model to calculate the corresponding parameter anomaly coefficient; among them, the parameter anomaly coefficient is a 1×a matrix, and the elements in the matrix correspond one by one to the parameters in the coupling operation parameters, and the coefficient calculation model is a deep neural network model.

[0116] The training process of the coefficient calculation model includes:

[0117] Pre-collect b groups of data sets. Each group of data sets includes a groups of different calculation sets, and the b groups of data sets are all different. Corresponding parameter anomaly coefficients are set for the b groups of data sets. b is an integer greater than 1. The data sets and the corresponding parameter anomaly coefficients are converted into a corresponding group of feature vectors; the parameter anomaly coefficients corresponding to the data sets are collected by those skilled in the art during the process of calculating historical parameter anomaly coefficients. b groups of data sets are collected, and each group of a groups of calculation sets in each group of data sets is analyzed in combination with the actual situation, the anomaly degree corresponding to each group of calculation sets is evaluated, a matrix is constructed according to the anomaly degree of the a groups of calculation sets, and the matrix is used as the parameter anomaly coefficient corresponding to the data set. Corresponding parameter anomaly coefficients are sequentially set for the b groups of data sets;

[0118] Each group of feature vectors is used as the input of the coefficient calculation model. The coefficient calculation model outputs a group of predicted parameter anomaly coefficients corresponding to each group of data sets, and uses the actual parameter anomaly coefficient corresponding to each group of data sets as the prediction target. The actual parameter anomaly coefficient is the preset parameter anomaly coefficient corresponding to the data set; minimizing the sum of the prediction errors of all data sets is used as the training target; among them, the calculation formula for the prediction error is η w =(θ w -ε w ) 2 , where η w is the prediction error, w is the group number of the feature vector corresponding to the data set, θ w is the predicted parameter anomaly coefficient corresponding to the wth group of data sets, and ε w is the actual parameter anomaly coefficient corresponding to the wth group of data sets; the coefficient calculation model is trained until the sum of the prediction errors reaches convergence and then the training stops.

[0119] The method for identifying risk parameters in the coupling operation parameters includes:

[0120] A preset anomaly threshold, which is pre-set by those skilled in the art according to the actual situation; compare each element in the parameter anomaly coefficient with the anomaly threshold respectively; mark the elements with values greater than or equal to the anomaly threshold as anomaly elements, and do not mark the elements with values less than the anomaly threshold; use the parameters corresponding to the anomaly elements in the coupling operation parameters as risk parameters.

[0121] A parameter detection module, which is used to obtain historical parameters, construct an anomaly detection architecture based on the historical parameters, perform anomaly detection on the detected parameters according to the anomaly detection architecture, and identify the parameters to be adjusted in the detected parameters.

[0122] The historical parameters are normal parameters obtained at historical moments, and the normal parameters are the coupling process parameters in the normal state; among them, in the normal state, there are no risk parameters in the coupling operation parameters corresponding to the produced couplings.

[0123] The anomaly detection architecture includes d detection sub-architectures, where d is the number of parameters in the coupling process parameters, and the detection sub-architectures correspond one-to-one with the parameters in the coupling process parameters; obtain the detection sub-architecture corresponding to the detected parameters in the anomaly detection architecture and mark it as the running sub-architecture; the steps for performing anomaly detection on the corresponding detected parameters according to one running sub-architecture include:

[0124] Step S301: Mark the detected parameter for which anomaly detection is performed as the current parameter, screen out the parameter corresponding to the current parameter from the historical parameters and mark it as the reference parameter, and collectively call the reference parameter and the current parameter the analysis parameters;

[0125] Step S302: Obtain the neighborhood parameters corresponding to each analysis parameter, subtract each neighborhood parameter corresponding to each analysis parameter respectively, and obtain the parameter difference corresponding to each analysis parameter; the neighborhood parameter corresponding to the analysis parameter is the remaining r analysis parameters, where r is the number of reference parameters;

[0126] Step S303: According to the parameter difference, obtain the standard span of each analysis parameter;

[0127] Step S304: Use the neighborhood parameters corresponding to the parameter differences ranked in the top y positions in each element sorting table as the standard parameters corresponding to the analysis parameters corresponding to the corresponding difference set, 1 < y < r + 1, and the specific value of y is set by those skilled in the art according to the actual situation;

[0128] Step S305: According to the parameter difference and the standard span, calculate the adjacent spans between each analysis parameter and each corresponding standard parameter in turn, and calculate the regional distribution degree corresponding to each analysis parameter;

[0129] Step S306: Calculate the outlier coefficient of the current parameter based on the regional distribution degree, and compare the outlier coefficient with a preset recognition coefficient. If the outlier coefficient is greater than the recognition coefficient, mark the current parameter as a parameter to be adjusted; if the outlier coefficient is less than or equal to the recognition coefficient, do not mark the current parameter. The recognition coefficient is preset by those skilled in the art according to the actual situation, and the value range of the recognition coefficient is [1.5, 3].

[0130] In the above step S303, the method for obtaining the standard span includes:

[0131] Take the parameter difference of each analysis parameter as a set of difference sets, and the difference sets correspond to the analysis parameters one by one; sort the parameter differences in each difference set from small to large to generate a difference sorting table corresponding to each difference set; take the parameter difference ranked at the y-th position in each difference sorting table as the standard span of the analysis parameter corresponding to the corresponding difference set.

[0132] In the above step S305, the method for calculating the adjacent span includes:

[0133] Mark the parameter difference between each analysis parameter and its corresponding standard parameter as an adjacent difference; take one adjacent difference corresponding to each analysis parameter and the standard span as a set of span sets, that is, each set of span sets includes one adjacent difference and one standard span; collectively call the adjacent difference and the standard span in each span set as the analysis span, compare each analysis span in each span set respectively, and take the analysis span with the largest value as the adjacent span between the corresponding analysis parameter and the corresponding standard parameter.

[0134] The method for calculating the regional distribution degree includes:

[0135] Add up the adjacent spans corresponding to each analysis parameter in turn to obtain the comprehensive span corresponding to each analysis parameter; multiply the reciprocal of the comprehensive distance corresponding to each analysis parameter by y to obtain the regional distribution degree corresponding to each analysis parameter.

[0136] In the above step S306, the method for calculating the outlier coefficient includes:

[0137] Obtain the standard parameter corresponding to the real-time parameter and mark it as the calculation parameter; divide the regional distribution degree corresponding to each calculation parameter by the regional distribution degree corresponding to the real-time element to obtain the relative distribution degree corresponding to each calculation parameter; add up the relative densities of each evaluation element in turn and then divide by y to obtain the outlier coefficient corresponding to the real-time parameter.

[0138] The parameter optimization module is used to formulate a process optimization strategy, generate corresponding optimization values for each parameter to be adjusted by using the process optimization strategy, and optimize the parameters to be adjusted according to the optimization values.

[0139] The steps of generating corresponding optimized values for each parameter to be adjusted include:

[0140] Step S401: Obtain the parameter range, and construct m groups of optimization sets according to the parameter range. Each group of optimization sets includes the adjustment values corresponding to each parameter to be adjusted;

[0141] Step S402: Define the initialization set. The initialization set includes candidate solutions, iteration thresholds, removal strategy sets, reconstruction strategy sets, strategy weight sets, and update factors, and set the iteration count to 0; among them, the candidate solution is one of the m groups of optimization sets;

[0142] Step S403: According to the strategy weights in the strategy weight set, calculate the removal selection probability of each removal strategy in the removal strategy set, and the reconstruction selection probability corresponding to each reconstruction strategy in the reconstruction strategy set;

[0143] Step S404: Select a removal strategy from the removal strategy set according to the removal selection probability, and obtain the strategy weight corresponding to the removal strategy from the strategy weight set, and mark it as the removal weight;

[0144] Step S405: Perform a removal operation on the candidate solution using the removal strategy to obtain a partial solution;

[0145] Step S406: Select a reconstruction strategy from the reconstruction strategy set according to the reconstruction selection probability, and obtain the strategy weight corresponding to the reconstruction strategy from the strategy weight set, and mark it as the reconstruction weight;

[0146] Step S407: Perform a reconstruction operation on the partial solution using the reconstruction strategy to obtain a reconstructed solution;

[0147] Step S408: Calculate the optimization effect of the reconstructed solution, calculate the corresponding update probability based on the optimization effect, and determine whether to update the candidate solution to the reconstructed solution according to the update probability;

[0148] Step S409: Adjust the removal weight and the reconstruction weight respectively according to the update result;

[0149] Step S410: Determine whether the iteration count is less than the iteration threshold. If so, increment the iteration count by one and return to Step S403. If not, enter Step S411;

[0150] Step S411: Obtain the optimization set corresponding to the candidate solution, and use each adjustment value in the optimization set as the optimized value of the corresponding parameter to be adjusted.

[0151] In the above step S401, the parameter range includes the normal range corresponding to each parameter to be adjusted. The maximum value of the normal range corresponding to each parameter to be adjusted is the normal parameter with the largest value among the corresponding historical parameters, and the minimum value of the normal range corresponding to each parameter to be adjusted is the normal parameter with the smallest value among the corresponding historical parameters; a value is randomly selected from each normal range, and a set of optimization sets is constructed. A total of m sets of optimization sets are constructed, and the m sets of optimization sets are all different.

[0152] In the above step S402, the removal strategy set includes various removal strategies, such as random removal (i.e., randomly removing a part of the adjustment values corresponding to the parameters to be adjusted from the candidate solutions), cost-driven removal (i.e., removing a part of the adjustment values corresponding to the parameters to be adjusted that have a greater impact on the optimization effect) and so on; the reconstruction strategy set includes various reconstruction strategies, such as random value insertion (i.e., for the removed adjustment value, randomly selecting a new value from the corresponding normal range and inserting it into the candidate solution), neighborhood search insertion (i.e., making a small adjustment to the removed adjustment value and inserting the adjusted value into the candidate solution) and so on; the strategy weight set includes the strategy weights corresponding to each removal strategy and reconstruction strategy. The value range of the update factor is [0, 0.5]. The strategy weight set, the update factor and the iteration threshold are all preset by those skilled in the art according to the actual situation.

[0153] It should be understood that the purpose of adopting the removal strategy and the reconstruction strategy is to break the existing structure by removing a part of the candidate solution and reconstruct the removed part, avoid the search falling into local optimum, enhance the exploration of the process optimization strategy, and thus improve the global optimization ability.

[0154] In the above step S403, the method for calculating the removal selection probability is: adding the strategy weights corresponding to each removal strategy in turn to obtain the total removal value; dividing the strategy weights corresponding to each removal strategy by the total removal value respectively to obtain the removal selection probability corresponding to each removal strategy.

[0155] The method for calculating the reconstruction selection probability is: adding the strategy weights corresponding to each reconstruction strategy in turn to obtain the total reconstruction value; dividing the strategy weights corresponding to each reconstruction strategy by the total reconstruction value respectively to obtain the reconstruction selection probability corresponding to each reconstruction strategy.

[0156] In the above step S404, the method for selecting a removal strategy from the removal strategy set is the same as the method for screening out nodes according to the screening probability in step S205.

[0157] In the above step S406, the method for selecting a reconstruction strategy from the reconstruction strategy set is the same as the method for screening out nodes according to the screening probability in step S205.

[0158] In the above step S408, the method for calculating the optimization effect of the reconstructed solution includes:

[0159] Obtain the optimization set corresponding to the reconstructed solution and mark it as the current set; replace the value of each parameter to be adjusted in the coupling process parameters with the corresponding adjustment value in the current set, and mark the coupling process parameters after replacement as the optimized process parameters; use the optimized process parameters, coupling process parameters, and coupling operation parameters as prediction parameters, and input the prediction parameters into the trained parameter prediction model to predict the corresponding optimized operation parameters; among them, the optimized operation parameters are the coupling operation parameters corresponding to the coupling product obtained by producing the coupling according to the optimized process parameters; the training process of the parameter prediction model is the same as that of the coefficient calculation model, and both are deep neural network models; input the optimized operation parameters and the parameter threshold into the coefficient calculation model to calculate the corresponding parameter anomaly coefficient and mark it as the optimized anomaly coefficient; compare each element in the optimized anomaly coefficient with the anomaly threshold respectively to identify the abnormal elements in the optimized anomaly coefficient; count the number of abnormal elements in the optimized anomaly coefficient and mark it as the first anomaly number; count the number of abnormal elements in the parameter anomaly coefficient and mark it as the second anomaly number; subtract the first anomaly number from the second anomaly number to obtain the optimization effect.

[0160] The method for calculating the update probability includes:

[0161] Obtain the iteration number corresponding to the current iteration process and mark it as the current number; multiply the current number by the preset influence factor and then by the update factor to obtain the current factor; the value range of the influence factor is [0,1], and the influence factor is preset by those skilled in the art according to the actual situation; calculate the optimization effect of the candidate solution, mark the optimization effect of the candidate solution as the candidate effect, and mark the optimization effect of the reconstructed solution as the reconstruction effect; compare the candidate effect with the reconstruction effect; if the candidate effect is less than or equal to the reconstruction effect, the corresponding update probability is 1; if the candidate effect is greater than the reconstruction effect, subtract the candidate effect from the reconstruction effect and divide by the current factor to obtain the update coefficient; use the natural constant as the base and the update coefficient as the exponent to perform exponential operation to obtain the update probability.

[0162] The method for determining whether to update the candidate solution to the reconstructed solution includes:

[0163] If the update probability is 1, update the candidate solution to the reconstructed solution; if the update probability is not 1, set the update range [0,k] and the retention range (k,1], where k is the update probability; randomly generate a judgment value from the interval [0,1], if the judgment value is within the update range, update the candidate solution to the reconstructed solution, if the judgment value is within the retention range, do not update the candidate solution to the reconstructed solution.

[0164] It should be understood that the purpose of judging whether to update the candidate solution to the reconstructed solution according to the update probability is that if the candidate solution is only updated according to a better reconstructed solution, it will quickly fall into a local optimum, while updating the candidate solution based on the update probability can jump out of the local optimum, enhance the global search ability, and thus improve the search efficiency.

[0165] In the above step S409, the methods for adjusting the removal weight and the reconstruction weight respectively include:

[0166] Subtract the candidate effect from the reconstruction effect to obtain the adjustment effect; multiply the adjustment effect by the current factor to obtain the adjustment degree; adjust the removal weight and the reconstruction weight respectively according to the adjustment degree.

[0167] In this embodiment, by classifying the coupling process parameters and matching them with the coupling operation parameters, the association between the process parameters and the operation parameters of the coupling is established; the deep technology is used to perform abnormal analysis and risk identification on the coupling operation parameters, and the potential problems existing in the operation process of the coupling can be accurately found; an abnormal detection architecture is constructed to perform real-time abnormal detection and identification on the identified detection parameters, and the parameters that need to be optimized and adjusted can be found in time, improving the pertinence of the optimization effect; a dynamic process optimization strategy is adopted to generate corresponding optimization values for different parameters to be adjusted, realizing flexible optimization of the process parameters, effectively improving the overall performance and production efficiency of the coupling; not only improving the dynamic performance and manufacturing accuracy of the coupling, but also shortening the development cycle, achieving higher fuel economy and durability, and enhancing the overall performance of the automotive power system.

[0168] Embodiment 2

[0169] The present application also provides an electronic device. The electronic device may include one or more processors and one or more memories. Among them, computer-readable code is stored in the memory, and when the computer-readable code is run by one or more processors, it can execute an engine coupling operation parameter optimization control system as described above.

[0170] The method or system according to the embodiment of the present application can also be implemented by means of the architecture of the electronic device shown in the present application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to the network, input / output, a hard disk, etc. The storage device in the electronic device, such as ROM or a hard disk, can store an engine coupling operation parameter optimization control system provided by the present application. Further, the electronic device may also include a user interface. Of course, the architecture shown in the present application is only exemplary, and when implementing different devices, one or more components shown in the electronic device of the present application can be omitted according to actual needs.

[0171] Embodiment 3

[0172] One embodiment of the present application discloses a computer-readable storage medium. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are run by a processor, an engine coupling operation parameter optimization control system according to an embodiment of the present application described with reference to the above drawings can be executed. The storage medium includes but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory, etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0173] In addition, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be run by a processor to execute instructions corresponding to the method steps provided by the present application, such as: an engine coupling operation parameter optimization control system. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0174] The above is only the specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

[0175] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should all be included in the protection scope of the present invention.

Claims

1. An optimized control system for the operating parameters of an engine coupling, characterized in that, Including: A parameter acquisition module, configured to acquire the operating parameters and process parameters of the coupling; A parameter classification module, configured to divide the coupling process parameters into a groups, and match each group with different parameters in the coupling operating parameters respectively; A parameter identification module, configured to perform anomaly analysis on the coupling operating parameters, calculate the parameter anomaly coefficient, identify the risk parameters in the coupling operating parameters based on the parameter anomaly coefficient, and obtain the coupling process parameters corresponding to the risk parameters according to the groups corresponding to the risk parameters, and mark them as detected parameters; A parameter detection module, configured to acquire historical parameters, construct an anomaly detection framework based on the historical parameters, and perform anomaly detection on the detected parameters according to the anomaly detection framework to identify the parameters to be adjusted in the detected parameters; A parameter optimization module, configured to formulate a process optimization strategy, generate corresponding optimized values for each parameter to be adjusted by using the process optimization strategy, and optimize the parameters to be adjusted according to the optimized values.

2. The optimized control system for the operating parameters of an engine coupling according to claim 1, wherein The coupling operating parameters are the dynamic characteristics involved in the working process of the coupling; the coupling process parameters are the operating parameters involved in the manufacturing process of the coupling; The step of dividing the coupling process parameters into a groups includes: Step S101: Adopt a pre-trained word embedding model to convert each parameter in the coupling process parameters into a corresponding text vector; Step S102: Use each text vector as a node; Step S103: Select a nodes as the center points, and use the nodes that are not used as the center points as the sample points, where a is the number of parameters in the coupling operating parameters; Step S104: Construct corresponding a groups according to the a center points, and calculate the sample distance from each sample point to each center point in turn; Step S105: Preset a distance threshold, compare each sample distance corresponding to each sample point with the distance threshold respectively, mark the center point corresponding to the sample distance with a value less than the distance threshold as the classification point, and do not mark the sample distance with a value greater than or equal to the distance threshold, and divide each sample point into the group corresponding to the corresponding classification point in turn; Step S106: Calculate the new centroid corresponding to each group, and replace the center point of each group with the corresponding new centroid; Step S107: Loop steps S104 to S106 until the new centroids of each group calculated in step S106 are the same as the new centroids calculated in the previous loop process, then the loop ends, and a groups are obtained.

3. An optimized control system for the operating parameters of an engine coupling according to claim 2, characterized in that, In the step S103, the step of selecting a nodes as the center points includes: Step S201: Randomly select a node as the center point; Step S202: Calculate the sample distance from each sample point to each center point, compare the sample distances corresponding to the same sample point, and use the sample distance with the smallest value as the minimum distance of the corresponding sample point; Step S203: Square the minimum distance of each sample point in turn to obtain the squared distance; Step S204: Add up the squared distances of each sample point in turn to obtain the sum of squared distances; divide the squared distance of each sample point by the sum of squared distances respectively to obtain the screening probability corresponding to each sample point; Step S205: Select a node as the center point according to the screening probability; Step S206: Loop through steps S202 to S205 until a nodes are selected, and then the loop ends; In the said step S104, the method for calculating the sample distance is: calculate the cosine similarity between the sample point and the center point, and take the reciprocal of the cosine similarity as the sample distance; In the said step S106, the method for calculating the new centroid includes: Count the number of nodes in each group and mark it as the node number; add up the nodes in each group in turn, and then divide by the corresponding node number to obtain the average point corresponding to each group; calculate the sample distance from each node in each group to the corresponding average point and mark it as the average distance; compare the average distances corresponding to the same group, and take the node corresponding to the minimum average distance value as the new centroid of the corresponding group.

4. An optimization control system for operating parameters of an engine coupling according to claim 3, characterized in that, In the said step S205, the method for selecting a node according to the screening probability includes: Sort each node from high to low according to the corresponding screening probability to generate a node sorting table; according to the forward order of the node sorting table, add the screening probability of each node to the screening probability of the corresponding previous node in turn to obtain the cumulative probability of each node; calculate the probability range of each node according to the cumulative probability of each node; generate a random number from the interval [0, 1], mark the probability range where the random number falls as the screening range, and select the node corresponding to the screening range; where, mark a node as the current node, and the previous nodes of the current node are all the nodes ranked before the current node in the node sorting table; the maximum value of the probability range corresponding to the current node is the cumulative probability corresponding to the current node, and the minimum value of the probability range corresponding to the current node is the cumulative probability corresponding to the node ranked one position before the current node in the node sorting table; The method for matching each group with different parameters in the coupling operation parameters includes: Adopt a pre-trained word embedding model to convert each parameter in the coupling operation parameters into a corresponding text vector and mark it as an operation vector; calculate the sample distance from each operation vector to the center point of each group respectively and mark it as an operation distance; sort the operation distances corresponding to each operation vector from small to large to generate a distance sorting table corresponding to each operation vector; mark the operation distance ranked first in each distance sorting table as the matching distance, match each operation vector with the group corresponding to the corresponding matching distance respectively, and match each group with the parameter corresponding to the corresponding operation vector respectively.

5. The optimized control system for operating parameters of an engine coupling according to claim 4, wherein The said method for calculating the parameter anomaly coefficient includes: Preset parameter thresholds, where the parameter thresholds include the thresholds corresponding to each parameter in the coupling operation parameters; each parameter in the coupling operation parameters and its corresponding threshold are used as a set of calculation sets, and the calculation sets correspond one by one to the parameters in the coupling operation parameters; each set of calculation sets is sequentially input into the trained coefficient calculation model to calculate the corresponding parameter anomaly coefficient; among them, the parameter anomaly coefficient is a 1×a matrix, and the elements in the matrix correspond one by one to the parameters in the coupling operation parameters, and the coefficient calculation model is a deep neural network model; the training process of the coefficient calculation model includes: Pre-collect b sets of data sets, each set of data sets includes a sets of different calculation sets, and the b sets of data sets are all different. Set the corresponding parameter anomaly coefficients for the b sets of data sets, where b is an integer greater than 1. Convert the data sets and the corresponding parameter anomaly coefficients into a corresponding set of feature vectors; use each set of feature vectors as the input of the coefficient calculation model, and the coefficient calculation model outputs a set of predicted parameter anomaly coefficients corresponding to each data set, and uses the actual parameter anomaly coefficient corresponding to each data set as the prediction target. The actual parameter anomaly coefficient is the parameter anomaly coefficient preset corresponding to the data set; use minimizing the sum of the prediction errors of all data sets as the training target; train the coefficient calculation model until the sum of the prediction errors reaches convergence and then stop training; The method for identifying risk parameters in the coupling operation parameters includes: Preset an anomaly threshold, and compare each element in the parameter anomaly coefficient with the anomaly threshold respectively; mark the elements with values greater than or equal to the anomaly threshold as anomaly elements, and do not mark the elements with values less than the anomaly threshold; use the parameters corresponding to the anomaly elements in the coupling operation parameters as risk parameters.

6. The optimized control system for the operating parameters of an engine coupling according to claim 5, characterized in that, The historical parameters are normal parameters obtained at historical moments, and the normal parameters are the coupling process parameters in the normal state; The anomaly detection architecture includes d detection sub-architectures, where d is the number of parameters in the coupling process parameters, and the detection sub-architectures correspond one by one to the parameters in the coupling process parameters; obtain the detection sub-architecture corresponding to the detection parameter in the anomaly detection architecture and mark it as the running sub-architecture; The steps for performing anomaly detection on the detection parameter according to a running sub-architecture include: Step S301: Mark the detection parameter for anomaly detection as the current parameter, screen out the parameter corresponding to the current parameter from the historical parameters and mark it as the reference parameter, and collectively call the reference parameter and the current parameter the analysis parameter; Step S302: Obtain the neighborhood parameters corresponding to each analysis parameter, subtract each analysis parameter from its corresponding neighborhood parameter respectively, and obtain the parameter difference corresponding to each analysis parameter; the neighborhood parameter corresponding to the analysis parameter is the remaining r analysis parameters, where r is the number of reference parameters; Step S303: According to the parameter difference, obtain the standard span of each analysis parameter; Step S304: Use the neighborhood parameters corresponding to the parameter differences ranked in the top y positions in each element sorting table as the standard parameters of the analysis parameter corresponding to the corresponding difference set, where 1 < y < r + 1; Step S305: According to the parameter difference and the standard span, calculate the adjacent spans between each analysis parameter and its corresponding standard parameter in sequence, and calculate the regional distribution degree corresponding to each analysis parameter; Step S306: Based on the regional distribution degree, calculate the outlier coefficient of the current parameter. Compare the outlier coefficient with the preset recognition coefficient. If the outlier coefficient is greater than the recognition coefficient, mark the current parameter as a parameter to be adjusted. If the outlier coefficient is less than or equal to the recognition coefficient, do not mark the current parameter.

7. An optimization control system for the operating parameters of an engine coupling according to claim 6, characterized in that, In the said step S303, the method for obtaining the standard span includes: Take the parameter differences of each analysis parameter as a set of difference sets, and the difference sets correspond to the analysis parameters one by one; sort the parameter differences in each difference set from small to large to generate a difference sorting table corresponding to each difference set; take the parameter difference ranked at the y-th position in each difference sorting table as the standard span of the analysis parameter corresponding to the corresponding difference set; In the said step S305, the method for calculating the adjacent span includes: Mark the parameter differences between each analysis parameter and its corresponding standard parameter as adjacent differences; take one adjacent difference corresponding to each analysis parameter and the standard span as a set of span sets, that is, each set of span sets includes one adjacent difference and one standard span; collectively call the adjacent difference and the standard span in each span set as the analysis span, compare each analysis span in each span set respectively, and take the largest analysis span value as the adjacent span between the corresponding analysis parameter and the corresponding standard parameter; The method for calculating the regional distribution degree includes: Add up the adjacent spans corresponding to each analysis parameter in sequence to obtain the comprehensive span corresponding to each analysis parameter; multiply the reciprocal of the comprehensive distance corresponding to each analysis parameter by y to obtain the regional distribution degree corresponding to each analysis parameter; In the said step S306, the method for calculating the outlier coefficient includes: Obtain the standard parameter corresponding to the real-time parameter and mark it as the calculation parameter; divide the regional distribution degree corresponding to each calculation parameter by the regional distribution degree corresponding to the real-time element to obtain the relative distribution degree corresponding to each calculation parameter; add up the relative densities of each evaluation element in sequence and then divide by y to obtain the outlier coefficient corresponding to the real-time parameter.

8. An optimization control system for the operating parameters of an engine coupling according to claim 7, characterized in that, The steps for generating corresponding optimization values for each parameter to be adjusted include: Step S401: Obtain the parameter range, and construct m groups of optimization sets according to the parameter range. Each group of optimization sets includes the adjustment values corresponding to each parameter to be adjusted; Step S402: Define the initialization set. The initialization set includes candidate solutions, iteration thresholds, removal strategy sets, reconstruction strategy sets, strategy weight sets, and update factors, and set the number of iterations to 0; among them, the candidate solution is one group of optimization sets among the m groups of optimization sets; Step S403: According to the strategy weights in the strategy weight set, calculate the removal selection probability of each removal strategy in the removal strategy set and the reconstruction selection probability corresponding to each reconstruction strategy in the reconstruction strategy set; Step S404: Select a removal strategy from the set of removal strategies according to the removal selection probability, obtain the strategy weight corresponding to the removal strategy from the set of strategy weights, and mark it as the removal weight; Step S405: Perform a removal operation on the candidate solution using the removal strategy to obtain a partial solution; Step S406: Select a reconstruction strategy from the set of reconstruction strategies according to the reconstruction selection probability, obtain the strategy weight corresponding to the reconstruction strategy from the set of strategy weights, and mark it as the reconstruction weight; Step S407: Perform a reconstruction operation on the partial solution using the reconstruction strategy to obtain a reconstructed solution; Step S408: Calculate the optimization effect of the reconstructed solution, calculate the corresponding update probability based on the optimization effect, and determine whether to update the candidate solution to the reconstructed solution according to the update probability; Step S409: Adjust the removal weight and the reconstruction weight respectively according to the update result; Step S410: Determine whether the number of iterations is less than the iteration threshold. If so, increment the number of iterations by one and return to Step S403. If not, proceed to Step S411; Step S411: Obtain the optimization set corresponding to the candidate solution, and use each adjustment value in the optimization set as the optimization value of the corresponding parameter to be adjusted.

9. The optimized control system for the operating parameters of an engine coupling according to claim 8, wherein In the said Step S401, the parameter range includes the normal range corresponding to each parameter to be adjusted. The maximum value of the normal range corresponding to each parameter to be adjusted is the normal parameter with the largest value among the corresponding historical parameters, and the minimum value of the normal range corresponding to each parameter to be adjusted is the normal parameter with the smallest value among the corresponding historical parameters; Randomly select a value from each normal range and construct a set of optimization sets. A total of m sets of optimization sets are constructed, and the m sets of optimization sets are all different; In the said Step S403, the method for calculating the removal selection probability is as follows: successively add the strategy weights corresponding to each removal strategy to obtain the total removal value; divide the strategy weights corresponding to each removal strategy by the total removal value respectively to obtain the removal selection probability corresponding to each removal strategy; The method for calculating the reconstruction selection probability is as follows: successively add the strategy weights corresponding to each reconstruction strategy to obtain the total reconstruction value; divide the strategy weights corresponding to each reconstruction strategy by the total reconstruction value respectively to obtain the reconstruction selection probability corresponding to each reconstruction strategy; In the said Step S404, the method for selecting a removal strategy from the set of removal strategies is the same as the method for screening out nodes according to the screening probability in Step S205; In the said Step S406, the method for selecting a reconstruction strategy from the set of reconstruction strategies is the same as the method for screening out nodes according to the screening probability in Step S205.

10. The optimized control system for the operating parameters of an engine coupling according to claim 9, characterized in that, In the said Step S408, the method for calculating the optimization effect of the reconstructed solution includes: Obtain the optimized set corresponding to the reconstruction solution and mark it as the current set; replace the value of each parameter to be adjusted in the coupling process parameters with the corresponding adjustment value in the current set, and mark the coupling process parameters after replacement as the optimized process parameters; use the optimized process parameters, coupling process parameters, and coupling operation parameters as prediction parameters, and input the prediction parameters into the trained parameter prediction model to predict the corresponding optimized operation parameters; the training process of the parameter prediction model is the same as that of the coefficient calculation model, and both are deep neural network models; input the optimized operation parameters and parameter thresholds into the coefficient calculation model to calculate the corresponding parameter anomaly coefficient and mark it as the optimized anomaly coefficient; compare each element in the optimized anomaly coefficient with the anomaly threshold respectively to identify the abnormal elements in the optimized anomaly coefficient; count the number of abnormal elements in the optimized anomaly coefficient and mark it as the first anomaly number; count the number of abnormal elements in the parameter anomaly coefficient and mark it as the second anomaly number; subtract the first anomaly number from the second anomaly number to obtain the optimization effect. The method for calculating the update probability is as follows: obtain the iteration number corresponding to the current iteration process and mark it as the current number; multiply the current number by the preset influence factor and then by the update factor to obtain the current factor; calculate the optimization effect of the candidate solution, mark the optimization effect of the candidate solution as the candidate effect, and mark the optimization effect of the reconstruction solution as the reconstruction effect; compare the candidate effect with the reconstruction effect; if the candidate effect is less than or equal to the reconstruction effect, the corresponding update probability is 1; if the candidate effect is greater than the reconstruction effect, subtract the candidate effect from the reconstruction effect and divide by the current factor to obtain the update coefficient; use the natural constant as the base and the update coefficient as the exponent to perform an exponential operation to obtain the update probability. The method for determining whether to update the candidate solution to the reconstruction solution includes: If the update probability is 1, update the candidate solution to the reconstruction solution; if the update probability is not 1, set the update range [0, k] and the retention range (k, 1], where k is the update probability; randomly generate a judgment value from the interval [0, 1], if the judgment value is within the update range, update the candidate solution to the reconstruction solution, if the judgment value is within the retention range, do not update the candidate solution to the reconstruction solution. In step S409, the method for adjusting the removal weight and the reconstruction weight respectively includes: Subtract the candidate effect from the reconstruction effect to obtain the adjustment effect; multiply the adjustment effect by the current factor to obtain the adjustment degree; adjust the removal weight and the reconstruction weight respectively according to the adjustment degree.

Images