An intelligent method for monitoring abnormal operation of underwater submersible pumps

Abstract

The present invention discloses an intelligent method for monitoring the abnormal operation of underwater submersible pumps, and the method includes data collection, data preprocessing, submersible pump operation monitoring, hyperparameter optimization and underwater submersible pump operation management. The present invention relates to the technical field of underwater submersible pump equipment data processing, and specifically refers to an intelligent method for monitoring the abnormal operation of underwater submersible pumps. The present invention innovatively proposes a mechanism for combining the design of the threshold value of abnormal operation data with the monitoring model of abnormal operation of submersible pumps, which can improve the efficiency and reliability of submersible pump monitoring; design an abnormal operation monitoring activation function and a submersible pump operation consistency loss function, improve the stability of the monitoring model, and thus improve the accuracy of the monitoring results of abnormal operation of submersible pumps; improve the algorithm for obtaining the optimal parameters of the monitoring model through the optimal proportional sine strategy, obtaining the dynamic convergence factor and the iterative dynamic adjustment strategy, so as to obtain the optimal parameter combination and improve the accuracy of the monitoring model results.

Description

Technical Field

[0001] The invention relates to the technical field of underwater submersible pump equipment data processing, and in particular to an intelligent method for monitoring abnormal operation of underwater submersible pumps. Background Art

[0002] The underwater submersible pump operation abnormality monitoring method uses an artificial intelligence algorithm to detect abnormalities in the operating status of the underwater submersible pump. This method can promptly detect abnormal conditions that may occur during the operation of the submersible pump, ensure the normal operation of the equipment, reduce downtime and maintenance costs, and improve the reliability and safety of the equipment.

[0003] However, traditional methods for monitoring abnormal operation of underwater submersible pumps have problems of slow response to abnormal situations and unstable response results, which affect the safe operation of underwater submersible pumps; existing models suitable for monitoring abnormal operation of underwater submersible pumps have technical problems of slow convergence speed and unbalanced output result categories, which lead to inaccurate monitoring results of submersible pump operation abnormality; existing monitoring models have technical problems of inappropriate built-in parameter settings, which lead to inaccurate model output results. Summary of the invention

[0004] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an intelligent underwater submersible pump operation abnormality monitoring method, which aims at the problems of slow response to abnormal situations and unstable response results in traditional underwater submersible pump operation abnormality monitoring methods, thereby affecting the safe operation of underwater submersible pumps; this scheme innovatively proposes a mechanism combining the design of operation abnormality data threshold with the submersible pump operation abnormality monitoring model, and optimizes the data processing process to reduce the risks of false alarms and missed alarms, thereby improving the efficiency and reliability of submersible pump monitoring, thereby effectively improving the safety of underwater submersible pump operation; in view of the problems of slow convergence speed and similar output results in the existing models suitable for underwater submersible pump operation abnormality monitoring In order to solve the technical problem of imbalance in the operation of submersible pumps, which leads to inaccurate monitoring results of abnormal operation of submersible pumps, this scheme innovatively designs the activation function for abnormal operation monitoring and the consistency loss function for submersible pump operation, which accelerates the convergence speed of the model and improves the stability of the monitoring model, thereby improving the accuracy of the monitoring results of abnormal operation of submersible pumps; in view of the technical problem that the existing monitoring models have inappropriate built-in parameter settings, which leads to inaccurate model output results, this scheme improves the algorithm for obtaining the optimal parameters of the monitoring model through the optimal proportional sine strategy, obtaining the dynamic convergence factor and iterative dynamic adjustment strategy, so as to obtain the optimal parameter combination and improve the accuracy of the monitoring model results.

[0005] The technical solution adopted by the present invention is as follows: The present invention provides an intelligent method for monitoring abnormal operation of an underwater submersible pump, the method comprising the following steps:

[0006] Step S1: data collection;

[0007] Step S2: data preprocessing;

[0008] Step S3: submersible pump operation monitoring;

[0009] Step S4: hyperparameter optimization;

[0010] Step S5: Underwater submersible pump operation management.

[0011] Furthermore, in step S1, the data collection is specifically to obtain the original operation data of the submersible pump from the submersible pump equipment operation system by collecting; the original operation data of the submersible pump includes historical operation data and real-time sensor data; the historical operation data and real-time sensor data both include equipment temperature data, equipment vibration data, equipment flow data, equipment current data, equipment pressure data and equipment surrounding environment monitoring data; the historical operation data also includes equipment operation abnormal fault type, equipment operation abnormal fault data and equipment abnormal fault handling results.

[0012] Further, in step S2, the data preprocessing is specifically to clean the original data of the submersible pump operation, set the threshold value of the abnormal operation data, judge the threshold value of the abnormal operation data, normalize the data and select the features to obtain the preliminary operation data of the submersible pump; specifically includes the following steps:

[0013] Step S21: data cleaning, specifically processing missing values, outliers and duplicate values ​​in the data;

[0014] Step S22: setting the abnormal operation data threshold, specifically, setting a reasonable data threshold range for the submersible pump in normal operation according to the abnormal equipment operation fault data in the historical operation data and combined with manual review;

[0015] Step S23: abnormal data threshold determination, specifically, comparing the real-time sensor data with the set threshold, marking only the data exceeding the set threshold as valid data, for filtering noise and invalid information;

[0016] Step S24: data normalization, which is used to normalize the data, specifically, normalize the data set by using the minimum-maximum method;

[0017] Step S25: feature selection, specifically, using an adaptive feature selection method to select important features related to abnormal operation of the underwater submersible pump from the original data of the submersible pump operation.

[0018] Further, in step S3, the submersible pump operation monitoring is specifically to establish a submersible pump operation abnormality monitoring model and train the monitoring model with historical operation data, and use the real-time sensor data in the preliminary operation data of the submersible pump as input data of the trained monitoring model to obtain the underwater submersible pump operation monitoring result; comprising the following steps:

[0019] Step S31: Establishing a submersible pump operation abnormality monitoring model, specifically establishing the submersible pump operation abnormality monitoring model by designing an operation abnormality monitoring activation function; comprising the following steps:

[0020] Step S311: Design an abnormal operation monitoring activation function, the formula used is as follows:

[0021] ;

[0022] In the formula, represents the activation function for running abnormal monitoring, x represents the input variable of the activation function, Represents parameters that can adjust the range and shape of the activation function output;

[0023] Step S312: LSTM unit output calculation, specifically, recursively calculating the hidden state and cell state of each time step, and linearly transforming the hidden state of each time step by running the abnormal monitoring activation function to obtain the monitoring result of a single time step; the formula used is as follows:

[0024] ;

[0025] ;

[0026] In the formula, Indicates the unit runs the function, represents the hidden state at the tth time step, represents the input data at the tth time step, Indicates that it is The hidden state of time steps, Indicates that it is The cell state at time steps, Represents the output result of the LSTM unit at the tth time step, represents the output weight matrix, represents the activation function, Represents the bias parameter of the output layer;

[0027] Step S313: Calculate the probability distribution, specifically by converting the LSTM unit output result into a probability distribution through a function to obtain the monitoring probability of each submersible pump operation abnormality category; the formula used is as follows:

[0028] ;

[0029] In the formula, Represents the probability distribution of the output result at the tth time step;

[0030] Step S314: Calculate the final output result of the entire time series, specifically, calculate the average output probability of the entire time series to obtain the final submersible pump operation abnormality monitoring result; the formula used is as follows:

[0031] ;

[0032] In the formula, Represents the final output result of the entire time series, and T represents the length of the time series;

[0033] Step S32: training a submersible pump operation abnormality monitoring model, specifically, using the historical operation data in the preliminary operation data of the submersible pump to train the submersible pump operation abnormality monitoring model to obtain a trained submersible pump operation abnormality monitoring model;

[0034] Step S321: Design the submersible pump operation consistency loss function, the formula used is as follows:

[0035] ;

[0036] In the formula, represents the submersible pump operation consistency loss function value, C represents the total number of submersible pump operation abnormality categories, represents the value of the i-th category of the real data, Represents the weight parameter between the control loss functions, represents the parameters that control the probability distribution of the output results, represents the output probability of the i-th category at the d-th time step, Output probability of the i-th category at the t-th time step;

[0037] Step S322: Update the weights, specifically by calculating the gradient of the weights through the back propagation algorithm and updating the weight matrix; the formula used is as follows:

[0038] ;

[0039] ;

[0040] In the formula, Denotes the loss function for the weight matrix The gradient of Represents the weight matrix of the LSTM unit output result at the tth time step The rate of change, represents the new output weight matrix, represents the old output weight matrix, represents the learning rate;

[0041] Step S323: iterative determination, specifically monitoring the change of the loss function. If the loss function is lower than the set threshold, the iterative training is stopped, otherwise the iterative training is continued, and finally the trained submersible pump operation abnormality monitoring model is obtained;

[0042] Step S33: Underwater submersible pump operation abnormality monitoring, specifically, using the real-time sensor data in the preliminary operation data of the submersible pump as the input data of the trained submersible pump operation abnormality monitoring model to obtain the underwater submersible pump operation abnormality monitoring results. The underwater submersible pump operation abnormality monitoring results can be divided into normal operation, equipment mechanical failure, equipment electrical failure, equipment fluid problem, equipment pressure abnormality, equipment temperature abnormality, equipment control system failure and other failures.

[0043] Furthermore, in step S4, the hyperparameter optimization specifically includes the following steps:

[0044] Step S41: Initialize parameters, specifically by constructing initial algorithm parameters; the initial algorithm parameters include the population size N and the maximum number of iterations ;

[0045] Step S42: Initialize the population, specifically, initialize the search individuals at random positions in the feasible search space, and the formula used is as follows:

[0046] ;

[0047] In the formula, represents the initial position of the i-th individual, and Respectively represent the search lower and upper limits, Indicates between A random number uniformly distributed within a range;

[0048] Step S43: Calculate the fitness value, specifically calculate the fitness value f of individuals in the population i ; The performance of the submersible pump operation abnormality monitoring model established based on the individual position is used as the individual fitness value, and the individuals are sorted from best to worst according to the fitness value to obtain the individual position with the highest current global fitness value ;

[0049] Step S44: Individual global search behavior, specifically, individual global search for the optimal solution through the optimal proportional sine strategy, the formula used is as follows:

[0050] ;

[0051] In the formula, t represents the current number of iterations, represents the position of the i-th individual in the t-th generation population, represents the position of the i-th individual in the t+1 generation population, represents the position of the individual with the best fitness value in the population in the tth iteration, Indicates between A random number uniformly distributed in the range, and Indicates between A random number uniformly distributed in the range, represents the golden ratio coefficient, Indicates between A random number uniformly distributed within a range;

[0052] Step S45: Obtain the dynamic convergence factor, the specific formula used is as follows:

[0053] ;

[0054] In the formula, represents the dynamic convergence factor, and π represents the circumference of a circle;

[0055] Step S46: optimal solution derived behavior, the specific formula used is as follows:

[0056] ;

[0057] ;

[0058] ;

[0059] In the formula, represents the position of the i-th derived individual in the t+1 generation population, represents the position of the i-th derived individual in the t-th generation population, represents the lower boundary of the derived individual position, represents the upper boundary of the derived individual position, and represent the maximum and minimum values ​​of the solution space respectively, and Indicates between A random number uniformly distributed within a range;

[0060] Step S47: Individual local exploration behavior, the specific formula used is as follows:

[0061] ;

[0062] ;

[0063] ;

[0064] In the formula, represents the lower boundary of individual local exploration, represents the upper boundary of individual local exploration, represents random numbers that follow a normal distribution; Indicates between A random number uniformly distributed within a range;

[0065] Step S48: Competitive global perturbation behavior, specifically using an iterative dynamic adjustment strategy to perform global individual perturbations; the formula used is as follows:

[0066] ;

[0067] In the formula, Indicates between A random number uniformly distributed within a range;

[0068] Step S49: search determination, specifically, by constructing a search termination condition, performing search determination on the optimal individual position, and obtaining the optimal individual position data setting;

[0069] The search termination conditions include threshold termination and iteration termination;

[0070] The threshold termination is specifically to set the fitness threshold, when the individual fitness value f i When it is above the fitness threshold, the hyperparameter search is completed;

[0071] The iteration termination specifically refers to terminating the iteration and obtaining the optimal individual position when the maximum number of iterations is reached;

[0072] The optimal individual position includes a combination of submersible pump operation abnormality monitoring model parameters.

[0073] Furthermore, in step S5, the underwater submersible pump operation management is specifically based on the output result of the submersible pump operation abnormality monitoring model. If the output result is that the operation is normal, the operation data of the underwater submersible pump will continue to be recorded and backed up. If the output result is not that the operation is normal, the alarm system will be triggered immediately to notify the underwater submersible pump maintenance personnel. The maintenance personnel will quickly formulate a maintenance plan based on the type of abnormality to ensure the smooth progress of underwater operations.

[0074] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0075] (1) Aiming at the problems of slow response to abnormal situations and unstable response results in traditional underwater submersible pump operation abnormality monitoring methods, which affect the safe operation of underwater submersible pumps, this scheme innovatively proposes a mechanism combining the design of operation abnormality data threshold with the submersible pump operation abnormality monitoring model. By optimizing the data processing process, the risk of false alarms and missed alarms is reduced, and the efficiency and reliability of submersible pump monitoring are improved, thereby effectively improving the safety of underwater submersible pump operation.

[0076] (2) In view of the technical problems of slow convergence speed and unbalanced output result categories in the existing models suitable for underwater submersible pump operation abnormality monitoring, which leads to inaccurate submersible pump operation abnormality monitoring results, this scheme innovatively designs the operation abnormality monitoring activation function and the submersible pump operation consistency loss function, which accelerates the convergence speed of the model and improves the stability of the monitoring model, thereby improving the accuracy of the submersible pump operation abnormality monitoring results.

[0077] (3) In order to solve the technical problem that the existing monitoring models have inappropriate built-in parameter settings, which leads to inaccurate model output results, this solution improves the algorithm for obtaining the optimal parameters of the monitoring model through the optimal proportional sine strategy, obtaining the dynamic convergence factor and iterative dynamic adjustment strategy, so as to obtain the optimal parameter combination and improve the accuracy of the monitoring model results. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 A schematic diagram of a flow chart of an intelligent method for monitoring abnormal operation of an underwater submersible pump provided by the present invention;

[0079] Figure 2 is a schematic flow chart of step S2;

[0080] Figure 3 is a schematic diagram of the process of step S3;

[0081] Figure 4 is a schematic flow chart of step S4;

[0082] Figure 5 is a flow chart of step S31;

[0083] Figure 6 is a schematic diagram of the process of step S32;

[0084] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0085] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0086] In the description of the present invention, it is necessary to understand that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred system or element must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0087] Example 1, see Figure 1 The technical solution adopted by the present invention is as follows: The present invention provides an intelligent method for monitoring abnormal operation of an underwater submersible pump, the method comprising the following steps:

[0088] Step S1: data collection, by collecting historical operation data and real-time sensor data from the submersible pump equipment operation system to obtain the submersible pump operation raw data;

[0089] Step S2: data preprocessing, by performing data cleaning, abnormal operation data threshold setting, abnormal operation data threshold judgment, data normalization and feature selection on the original operation data of the submersible pump, to obtain preliminary operation data of the submersible pump;

[0090] Step S3: submersible pump operation monitoring, by establishing a submersible pump operation abnormality monitoring model and training the monitoring model with historical operation data, using the real-time sensor data in the preliminary operation data of the submersible pump as input data of the trained monitoring model, and obtaining the underwater submersible pump operation monitoring result;

[0091] Step S4: Hyperparameter optimization, improving the algorithm for obtaining the optimal parameters of the monitoring model through the optimal proportional sine strategy, obtaining the dynamic convergence factor and iterative dynamic adjustment strategy, so as to obtain the optimal parameter combination of the submersible pump operation abnormality monitoring model;

[0092] Step S5: Underwater submersible pump operation management, output results through the submersible pump operation abnormality monitoring model, and notify maintenance personnel to take different treatment measures.

[0093] Example 2, see Figure 1This embodiment is based on the above embodiment. In step S1, the data collection is specifically to obtain the original data of the submersible pump operation from the submersible pump equipment operation system by collection; the original data of the submersible pump operation includes historical operation data and real-time sensor data; the historical operation data and real-time sensor data both include equipment temperature data, equipment vibration data, equipment flow data, equipment current data, equipment pressure data and equipment surrounding environment monitoring data; the historical operation data also includes equipment operation abnormal fault type, equipment operation abnormal fault data and equipment abnormal fault processing results.

[0094] Example 3, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. The data preprocessing is used to clean the original data of the submersible pump operation, set the threshold value of the abnormal operation data, judge the threshold value of the abnormal operation data, normalize the data and select features to obtain the preliminary operation data of the submersible pump; specifically, it includes the following steps:

[0095] Step S21: data cleaning, specifically processing missing values, outliers and duplicate values ​​in the data;

[0096] Step S22: setting the abnormal operation data threshold, specifically, setting a reasonable data threshold range for the submersible pump in normal operation according to the abnormal equipment operation fault data in the historical operation data and combined with manual review;

[0097] Step S23: abnormal data threshold determination, specifically, comparing the real-time sensor data with the set threshold, marking only the data exceeding the set threshold as valid data, for filtering noise and invalid information;

[0098] Step S24: data normalization, which is used to normalize the data, specifically, normalize the data set by using the minimum-maximum method;

[0099] Step S25: feature selection, specifically, using an adaptive feature selection method to select important features related to abnormal operation of the underwater submersible pump from the original data of the submersible pump operation.

[0100] By performing the above operations, the problems of slow response to abnormal situations and unstable response results in traditional underwater submersible pump operation abnormality monitoring methods, which affect the safe operation of underwater submersible pumps, are addressed. This solution innovatively proposes a mechanism that combines the design of operation abnormality data thresholds with the submersible pump operation abnormality monitoring model. By optimizing the data processing process, the risks of false alarms and missed alarms are reduced, the efficiency and reliability of submersible pump monitoring are improved, and the safety of underwater submersible pump operation is effectively improved.

[0101] Example 4, see Figure 1 , Figure 3, Figure 5 and Figure 6 This embodiment is based on the above embodiment. In step S3, the submersible pump operation monitoring is specifically to establish a submersible pump operation abnormality monitoring model and train the monitoring model with historical operation data, and use the real-time sensor data in the preliminary operation data of the submersible pump as the input data of the trained monitoring model to obtain the underwater submersible pump operation monitoring result; it includes the following steps:

[0102] Step S31: Establishing a submersible pump operation abnormality monitoring model, specifically establishing the submersible pump operation abnormality monitoring model by designing an operation abnormality monitoring activation function; comprising the following steps:

[0103] Step S311: Design an abnormal operation monitoring activation function, the formula used is as follows:

[0104] ;

[0105] In the formula, represents the activation function for running abnormal monitoring, x represents the input variable of the activation function, Represents parameters that can adjust the range and shape of the activation function output;

[0106] Step S312: LSTM unit output calculation, specifically, recursively calculating the hidden state and cell state of each time step, and linearly transforming the hidden state of each time step by running the abnormal monitoring activation function to obtain the monitoring result of a single time step; the formula used is as follows:

[0107] ;

[0108] ;

[0109] In the formula, Indicates the unit runs the function, represents the hidden state at the tth time step, represents the input data at the tth time step, Indicates that it is The hidden state of time steps, Indicates that it is The cell state at time steps, Represents the output result of the LSTM unit at the tth time step, represents the output weight matrix, represents the activation function, Represents the bias parameter of the output layer;

[0110] Step S313: Calculate the probability distribution, specifically by converting the LSTM unit output result into a probability distribution through a function to obtain the monitoring probability of each submersible pump operation abnormality category; the formula used is as follows:

[0111] ;

[0112] In the formula, Represents the probability distribution of the output result at the tth time step;

[0113] Step S314: Calculate the final output result of the entire time series, specifically, calculate the average output probability of the entire time series to obtain the final submersible pump operation abnormality monitoring result; the formula used is as follows:

[0114] ;

[0115] In the formula, Represents the final output result of the entire time series, and T represents the length of the time series;

[0116] Step S32: training a submersible pump operation abnormality monitoring model, specifically, using the historical operation data in the preliminary operation data of the submersible pump to train the submersible pump operation abnormality monitoring model to obtain a trained submersible pump operation abnormality monitoring model;

[0117] Step S321: Design the submersible pump operation consistency loss function, the formula used is as follows:

[0118] ;

[0119] In the formula, represents the submersible pump operation consistency loss function value, C represents the total number of submersible pump operation abnormality categories, represents the value of the i-th category of the real data, Represents the weight parameter between the control loss functions, represents the parameters that control the probability distribution of the output results, represents the output probability of the i-th category at the d-th time step, Output probability of the i-th category at the t-th time step;

[0120] Step S322: Update the weights, specifically by calculating the gradient of the weights through the back propagation algorithm and updating the weight matrix; the formula used is as follows:

[0121] ;

[0122] ;

[0123] In the formula, Denotes the loss function for the weight matrix The gradient of Represents the weight matrix of the LSTM unit output result at the tth time step The rate of change, represents the new output weight matrix, represents the old output weight matrix, represents the learning rate;

[0124] Step S323: iterative determination, specifically monitoring the change of the loss function. If the loss function is lower than the set threshold, the iterative training is stopped, otherwise the iterative training is continued, and finally the trained submersible pump operation abnormality monitoring model is obtained;

[0125] Step S33: Underwater submersible pump operation abnormality monitoring, specifically, using the real-time sensor data in the preliminary operation data of the submersible pump as the input data of the trained submersible pump operation abnormality monitoring model to obtain the underwater submersible pump operation abnormality monitoring results. The underwater submersible pump operation abnormality monitoring results can be divided into normal operation, equipment mechanical failure, equipment electrical failure, equipment fluid problem, equipment pressure abnormality, equipment temperature abnormality, equipment control system failure and other failures.

[0126] By performing the above operations, in order to solve the technical problems of slow convergence speed and unbalanced output result categories in the existing models suitable for abnormal operation monitoring of underwater submersible pumps, which leads to inaccurate abnormal operation monitoring results of submersible pumps, this scheme innovatively designs the abnormal operation monitoring activation function and the submersible pump operation consistency loss function, which accelerates the convergence speed of the model, improves the stability of the monitoring model, and thus improves the accuracy of the abnormal operation monitoring results of submersible pumps.

[0127] Example 5, see Figure 1 and Figure 4 This embodiment is based on the above embodiment. In step S4, the hyperparameter optimization specifically includes the following steps:

[0128] Step S41: Initialize parameters, specifically by constructing initial algorithm parameters; the initial algorithm parameters include the population size N and the maximum number of iterations ;

[0129] Step S42: Initialize the population, specifically, initialize the search individuals at random positions in the feasible search space, and the formula used is as follows:

[0130] ;

[0131] In the formula, represents the initial position of the i-th individual, and Respectively represent the search lower and upper limits, Indicates between A random number uniformly distributed within a range;

[0132] Step S43: Calculate the fitness value, specifically calculate the fitness value f of individuals in the population i ; The performance of the submersible pump operation abnormality monitoring model established based on the individual position is used as the individual fitness value, and the individuals are sorted from best to worst according to the fitness value to obtain the individual position with the highest current global fitness value ;

[0133] Step S44: Individual global search behavior, specifically, individual global search for the optimal solution through the optimal proportional sine strategy, the formula used is as follows:

[0134] ;

[0135] In the formula, t represents the current number of iterations, represents the position of the i-th individual in the t-th generation population, represents the position of the i-th individual in the t+1 generation population, represents the position of the individual with the best fitness value in the population in the tth iteration, Indicates between A random number uniformly distributed in the range, and Indicates between A random number uniformly distributed in the range, represents the golden ratio coefficient, Indicates between A random number uniformly distributed within a range;

[0136] Step S45: Obtain the dynamic convergence factor, the specific formula used is as follows:

[0137] ;

[0138] In the formula, represents the dynamic convergence factor, and π represents the circumference of a circle;

[0139] Step S46: optimal solution derived behavior, the specific formula used is as follows:

[0140] ;

[0141] ;

[0142] ;

[0143] In the formula, represents the position of the i-th derived individual in the t+1 generation population, represents the position of the i-th derived individual in the t-th generation population, represents the lower boundary of the derived individual position, represents the upper boundary of the derived individual position, and represent the maximum and minimum values ​​of the solution space respectively, and Indicates between A random number uniformly distributed within a range;

[0144] Step S47: Individual local exploration behavior, the specific formula used is as follows:

[0145] ;

[0146] ;

[0147] ;

[0148] In the formula, represents the lower boundary of individual local exploration, represents the upper boundary of individual local exploration, represents random numbers that follow a normal distribution; Indicates between A random number uniformly distributed within a range;

[0149] Step S48: Competitive global perturbation behavior, specifically using an iterative dynamic adjustment strategy to perform global individual perturbations; the formula used is as follows:

[0150] ;

[0151] In the formula, Indicates between A random number uniformly distributed within a range;

[0152] Step S49: search determination, specifically, by constructing a search termination condition, performing search determination on the optimal individual position, and obtaining the optimal individual position data setting;

[0153] The search termination conditions include threshold termination and iteration termination;

[0154] The threshold termination is specifically to set the fitness threshold, when the individual fitness value f i When it is above the fitness threshold, the hyperparameter search is completed;

[0155] The iteration termination specifically refers to terminating the iteration and obtaining the optimal individual position when the maximum number of iterations is reached;

[0156] The optimal individual position includes a combination of submersible pump operation abnormality monitoring model parameters.

[0157] In order to address the technical problem that the existing monitoring models have inappropriate built-in parameter settings, which leads to inaccurate model output results, this solution improves the algorithm for obtaining the optimal parameters of the monitoring model through the optimal proportional sine strategy, obtaining the dynamic convergence factor and the iterative dynamic adjustment strategy, so as to obtain the optimal parameter combination and improve the accuracy of the monitoring model results.

[0158] Example 6, see Figure 1 This embodiment is based on the above embodiment. In step S5, the underwater submersible pump operation management is specifically based on the output result of the submersible pump operation abnormality monitoring model. If the output result is normal operation, the underwater submersible pump operation data will continue to be recorded and backed up. If the output result is not normal operation, the alarm system will be immediately triggered to notify the underwater submersible pump maintenance personnel. The maintenance personnel will quickly formulate a maintenance plan based on the abnormality type to ensure the smooth progress of underwater operations.

[0159] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0160] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0161] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. An intelligent method for monitoring abnormal operation of underwater submersible pumps, characterized in that: The method comprises the following steps: Step S1: data collection, by collecting historical operation data and real-time sensor data from the submersible pump equipment operation system to obtain the submersible pump operation raw data; Step S2: data preprocessing, by performing data cleaning, abnormal operation data threshold setting, abnormal operation data threshold judgment, data normalization and feature selection on the original operation data of the submersible pump, to obtain preliminary operation data of the submersible pump; Step S3: Submersible pump operation monitoring, by designing an abnormal operation monitoring activation function and a submersible pump operation consistency loss function to establish a submersible pump operation abnormality monitoring model and training the monitoring model with historical operation data, using the real-time sensor data in the preliminary operation data of the submersible pump as the input data of the trained monitoring model, and obtaining the underwater submersible pump operation monitoring result; Step S4: Hyperparameter optimization, improving the algorithm for obtaining the optimal parameters of the monitoring model through the optimal proportional sine strategy, obtaining the dynamic convergence factor and iterative dynamic adjustment strategy, so as to obtain the optimal parameter combination of the submersible pump operation abnormality monitoring model; Step S5: Underwater submersible pump operation management, outputting results through the submersible pump operation abnormality monitoring model to notify maintenance personnel to take different treatment measures; The submersible pump operation monitoring specifically includes the following steps: Step S31: Establishing a submersible pump operation abnormality monitoring model, specifically establishing the submersible pump operation abnormality monitoring model by designing an operation abnormality monitoring activation function; comprising the following steps: Step S311: Design an abnormal operation monitoring activation function, the formula used is as follows: ； In the formula, represents the activation function for running abnormal monitoring, x represents the input variable of the activation function, Represents parameters that can adjust the range and shape of the activation function output; Step S312: LSTM unit output calculation, specifically, recursively calculating the hidden state and cell state of each time step, and linearly transforming the hidden state of each time step by running the abnormal monitoring activation function to obtain the monitoring result of a single time step; the formula used is as follows: ； ； In the formula, Indicates the unit runs the function, represents the hidden state at the tth time step, represents the input data at the tth time step, Indicates that it is The hidden state of time steps, Indicates that it is The cell state at time steps, Represents the output result of the LSTM unit at the tth time step, represents the output weight matrix, represents the activation function, Represents the bias parameter of the output layer; Step S313: Calculate the probability distribution, specifically by converting the LSTM unit output result into a probability distribution through a function to obtain the monitoring probability of each submersible pump operation abnormality category; the formula used is as follows: ； In the formula, Represents the probability distribution of the output result at the tth time step; Step S314: Calculate the final output result of the entire time series, specifically, calculate the average output probability of the entire time series to obtain the final submersible pump operation abnormality monitoring result; the formula used is as follows: ； In the formula, Represents the final output result of the entire time series, and T represents the length of the time series; Step S32: training a submersible pump operation abnormality monitoring model, specifically, using the historical operation data in the preliminary operation data of the submersible pump to train the submersible pump operation abnormality monitoring model to obtain a trained submersible pump operation abnormality monitoring model; Step S321: Design the submersible pump operation consistency loss function, the formula used is as follows: ； In the formula, represents the submersible pump operation consistency loss function value, C represents the total number of submersible pump operation abnormality categories, represents the value of the i-th category of the real data, Represents the weight parameter between the control loss functions, represents the parameters that control the probability distribution of the output results, represents the output probability of the i-th category at the d-th time step, Output probability of the i-th category at the t-th time step; Step S322: Update the weights, specifically by calculating the gradient of the weights through the back propagation algorithm and updating the weight matrix; the formula used is as follows: ； ； In the formula, Denotes the loss function for the weight matrix The gradient of Represents the weight matrix of the LSTM unit output result at the tth time step The rate of change, represents the new output weight matrix, represents the old output weight matrix, represents the learning rate; Step S323: iterative determination, specifically monitoring the change of the loss function. If the loss function is lower than the set threshold, the iterative training is stopped, otherwise the iterative training is continued, and finally the trained submersible pump operation abnormality monitoring model is obtained; Step S33: Underwater submersible pump operation abnormality monitoring, specifically, using the real-time sensor data in the preliminary operation data of the submersible pump as the input data of the trained submersible pump operation abnormality monitoring model to obtain the underwater submersible pump operation abnormality monitoring results. The underwater submersible pump operation abnormality monitoring results can be divided into normal operation, equipment mechanical failure, equipment electrical failure, equipment fluid problem, equipment pressure abnormality, equipment temperature abnormality, equipment control system failure and other failures.

2. The intelligent method for monitoring abnormal operation of underwater submersible pumps according to claim 1 is characterized by: In step S4, the hyperparameter optimization specifically includes the following steps: Step S41: Initialize parameters, specifically by constructing initial algorithm parameters; the initial algorithm parameters include the population size N and the maximum number of iterations ; Step S42: Initialize the population, specifically, initialize the search individuals at random positions in the feasible search space, and the formula used is as follows: ； In the formula, represents the initial position of the i-th individual, and Respectively represent the search lower and upper limits, Indicates between A random number uniformly distributed within a range; Step S43: Calculate the fitness value, specifically calculate the fitness value f of individuals in the population i ; The performance of the submersible pump operation abnormality monitoring model established based on the individual position is used as the individual fitness value, and the individuals are sorted from best to worst according to the fitness value to obtain the individual position with the highest current global fitness value ; Step S44: Individual global search behavior, specifically, individual global search for the optimal solution through the optimal proportional sine strategy, the formula used is as follows: ； In the formula, t represents the current number of iterations, represents the position of the i-th individual in the t-th generation population, represents the position of the i-th individual in the t+1 generation population, represents the position of the individual with the best fitness value in the population in the tth iteration, Indicates between A random number uniformly distributed in the range, and Indicates between A random number uniformly distributed in the range, represents the golden ratio coefficient, Indicates between A random number uniformly distributed within a range; Step S45: Obtain the dynamic convergence factor, the specific formula used is as follows: ； In the formula, represents the dynamic convergence factor, and π represents the circumference of a circle; Step S46: optimal solution derived behavior, the specific formula used is as follows: ； ； ； In the formula, represents the position of the i-th derived individual in the t+1 generation population, represents the position of the i-th derived individual in the t-th generation population, represents the lower boundary of the derived individual position, represents the upper boundary of the derived individual position, and represent the maximum and minimum values ​​of the solution space respectively, and Indicates between A random number uniformly distributed within a range; Step S47: Individual local exploration behavior, the specific formula used is as follows: ； ； ； In the formula, represents the lower boundary of individual local exploration, represents the upper boundary of individual local exploration, represents random numbers that follow a normal distribution; Indicates between A random number uniformly distributed within a range; Step S48: Competitive global perturbation behavior, specifically using an iterative dynamic adjustment strategy to perform global individual perturbations; the formula used is as follows: ； In the formula, Indicates between A random number uniformly distributed within a range; Step S49: search determination, specifically, by constructing a search termination condition, performing search determination on the optimal individual position, and obtaining the optimal individual position data setting; The search termination conditions include threshold termination and iteration termination; The threshold termination is specifically to set the fitness threshold, when the individual fitness value f i When it is above the fitness threshold, the hyperparameter search is completed; The iteration termination specifically refers to terminating the iteration and obtaining the optimal individual position when the maximum number of iterations is reached; The optimal individual position includes a combination of submersible pump operation abnormality monitoring model parameters.

3. The intelligent method for monitoring abnormal operation of underwater submersible pumps according to claim 1 is characterized in that: In step S5, the underwater submersible pump operation management is specifically based on the output result of the submersible pump operation abnormality monitoring model. If the output result is normal operation, the underwater submersible pump operation data will continue to be recorded and backed up. If the output result is not normal operation, the alarm system will be triggered immediately to notify the underwater submersible pump maintenance personnel. The maintenance personnel will quickly formulate a maintenance plan based on the type of abnormality to ensure the smooth progress of underwater operations.

4. The intelligent method for monitoring abnormal operation of underwater submersible pumps according to claim 1 is characterized in that: In step S1, the data collection specifically involves obtaining the original operation data of the submersible pump from the submersible pump equipment operation system by collecting data; the original operation data of the submersible pump includes historical operation data and real-time sensor data; the historical operation data and real-time sensor data both include equipment temperature data, equipment vibration data, equipment flow data, equipment current data, equipment pressure data and equipment surrounding environment monitoring data; the historical operation data also includes equipment operation abnormal fault type, equipment operation abnormal fault data and equipment abnormal fault handling results.

5. The intelligent method for monitoring abnormal operation of underwater submersible pumps according to claim 1 is characterized by: In step S2, the data preprocessing is specifically to clean the original data of the submersible pump operation, set the threshold value of the abnormal operation data, judge the threshold value of the abnormal operation data, normalize the data and select the features to obtain the preliminary operation data of the submersible pump; specifically includes the following steps: Step S21: data cleaning, specifically processing missing values, outliers and duplicate values ​​in the data; Step S22: setting the abnormal operation data threshold, specifically, setting a reasonable data threshold range for the submersible pump in normal operation according to the abnormal equipment operation fault data in the historical operation data and combined with manual review; Step S23: abnormal data threshold determination, specifically, comparing the real-time sensor data with the set threshold, marking only the data exceeding the set threshold as valid data, for filtering noise and invalid information; Step S24: data normalization, which is used to normalize the data, specifically, normalize the data set by using the minimum-maximum method; Step S25: feature selection, specifically, using an adaptive feature selection method to select important features related to abnormal operation of the underwater submersible pump from the original data of the submersible pump operation.

Images