Energy-saving motor operation optimization analysis method, system and storage medium

Abstract

The present invention relates to the field of motor control technology, and specifically to an energy-saving motor operation optimization analysis method, the analysis method comprising the following steps: S1: real-time monitoring of the load and operating state of the motor, and obtaining the monitoring data of the motor; S2: establishing an energy consumption prediction model through a machine learning algorithm and the monitoring data, and determining the predicted energy consumption of the motor under different control strategies; S3: the predicted energy consumption of the motor is determined by an optimization algorithm to determine the motor operating parameters that can achieve the maximum energy efficiency ratio; S4: adjusting the operating parameters of the motor in real time according to the motor operating parameters, and repeating steps S1‑S3. The present invention ensures the high accuracy and reliability of the monitoring data by setting a multi-dimensional sensor and performing anti-interference processing, and at the same time adopts an AR model to determine the temporal dependency between the data, which can improve the accuracy of predicting the dynamic changes of the motor operating state, thereby facilitating a more accurate prediction of the motor energy consumption trend.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control, and in particular to an energy-saving motor operation optimization analysis method, system and storage medium. Background Art

[0002] In today's rapidly developing industrial automation and smart home fields, motors are an indispensable source of power, and their application range covers a wide range of fields from heavy industrial machinery to household appliances. With the advancement of technology and the enhancement of environmental protection awareness, energy efficiency has become one of the key indicators for measuring motor performance. The energy efficiency performance of motors under different working conditions, especially in the face of dynamically changing load conditions, is directly related to the energy utilization efficiency and operating costs, which in turn affects the economy and environmental sustainability of the system.

[0003] Traditional motor control strategies are mostly based on fixed modes or simple feedback adjustment mechanisms. These methods are incapable of handling complex and changing load requirements. For example, under light load or variable load conditions, motors often cannot efficiently convert electrical energy, which can easily cause a large amount of unnecessary energy consumption. This non-optimal operation mode not only increases energy costs, but may also cause problems such as motor overheating and shortened life, thereby limiting the overall performance of the system.

[0004] In order to overcome the above challenges, researchers and engineers have been committed to exploring more intelligent and flexible motor control solutions in recent years. These efforts focus on developing advanced technologies that can monitor the working status of the motor in real time, accurately evaluate the load characteristics, and automatically adjust the motor operating parameters (such as voltage, current, frequency, etc.) accordingly. The goal is to ensure that the motor always operates at the optimal operating point, maximize the energy efficiency ratio, and maintain the required output performance and stability. However, there are multiple technical obstacles to achieving this goal, including but not limited to the use of high-precision sensors, the design and real-time processing capabilities of complex algorithms, and the efficient integration of control systems and motor hardware. Therefore, there is an urgent need for a new method of motor operation optimization to solve the current problem. Summary of the invention

[0005] The object of the present invention is to provide an energy-saving motor operation optimization analysis method, system and storage medium to solve the problems raised in the above background technology.

[0006] The technical solution of the present invention is: an energy-saving motor operation optimization analysis method, the analysis method comprises the following steps:

[0007] S1: monitor the load and operating status of the motor in real time and obtain monitoring data of the motor;

[0008] S2: Establishing an energy consumption prediction model through a machine learning algorithm and the monitoring data to determine the predicted energy consumption of the motor under different control strategies;

[0009] S3: The motor predicted energy consumption is determined by an optimization algorithm to achieve the motor operating parameters that can achieve the maximum energy efficiency ratio;

[0010] S4: According to the motor operating parameters, adjust the operating parameters of the motor in real time, and repeat steps S1-S3.

[0011] Furthermore, the monitoring data of the motor is obtained as follows:

[0012] S1.1: According to the structure and operating environment of the motor, a multi-dimensional sensor is set;

[0013] S1.2: performing anti-interference processing on the data information acquired by the multi-dimensional sensor;

[0014] S1.3: The data information after anti-interference processing is analyzed through time series to determine the temporal dependency between the data.

[0015] Furthermore, the data information obtained by the multi-dimensional sensor is subjected to anti-interference processing, specifically as follows:

[0016] S1.2.1: Determine the filtered data value corresponding to each data according to the data information obtained by the multi-dimensional sensor, specifically: ;

[0017] in: is the filtered signal at the current time The value of is the attenuation factor, The data obtained by the multi-dimensional sensor, is the filtered signal at the previous moment The value of

[0018] S1.2.2: Repeat step S1.2.1 and continuously update the filtered data value.

[0019] Going further, determine the temporal dependencies between data, as follows:

[0020] S1.3.1: Determine the order of the AR model according to the data information after the anti-interference processing, and the AR model is specifically: ;

[0021] in: For a time series at a point in time The value of is the autoregression coefficient at time point The value of For a time series at a point in time The value of is the autoregression coefficient at time point The value of For a time series at a point in time The value of is the autoregression coefficient at time point The value of For a time series at a point in time The value of is white noise;

[0022] S1.3.2: Estimate the parameters of the AR model using statistical modeling tools and the order of the AR model;

[0023] S1.3.3: Determine the order of the autoregressive term of the AR model through the autocorrelation diagram, specifically:

[0024] ;

[0025] in: is the first-order autocorrelation coefficient of the autoregressive term of the AR model, is the sequence length, For a time series at a point in time The value of is the mean value of the series, is the value of the time series at time point t-1, for time point;

[0026] S1.3.4: According to the order of the autoregressive term of the AR model, the AR model is established, and the temporal dependency relationship between the data is determined through the AR model.

[0027] Furthermore, the energy consumption prediction model is established as follows:

[0028] S2.1: Standardize the monitoring data;

[0029] S2.2: Select and train a machine learning model and determine the final machine learning model;

[0030] S2.3: Use the standardized monitoring data as the input of the final machine learning model to establish the energy consumption prediction model.

[0031] Furthermore, the monitoring data is standardized as follows:

[0032] S2.1.1: Clean the monitoring data using the interquartile range rule, specifically:

[0033] ;

[0034] in: is the interquartile range, is the upper quartile, is the lower quartile;

[0035] S2.1.2: The cleaned monitoring data is standardized by the Z-score standardization algorithm, specifically:

[0036] ;

[0037] in: is the standardized monitoring data. For the monitoring data after cleaning, is the mean value of the data set, is the standard deviation.

[0038] Furthermore, the final machine learning model is determined as follows:

[0039] S2.2.1: Selecting a primary machine learning model based on the monitoring data after anti-interference processing;

[0040] S2.2.2: Determine the optimal hyperparameter combination corresponding to the primary machine learning model through the Bayesian optimization method;

[0041] S2.2.3: The machine learning model corresponding to the optimal hyperparameter combination is verified by the K-fold cross-validation method to determine the final machine learning model.

[0042] Going further, the motor operating parameters that can achieve the maximum energy efficiency ratio are determined as follows:

[0043] S3.1: Convert the parameters to be optimized into the maximum energy efficiency ratio;

[0044] S3.2: Obtain the gradient vector corresponding to the parameter to be optimized by using the gradient ascent formula, wherein the gradient ascent formula is specifically: ;

[0045] in: is the gradient vector corresponding to the parameter to be optimized, is a parameter to be optimized, is another parameter to be optimized.

[0046] An energy-saving motor operation optimization analysis system, the energy-saving motor operation optimization analysis system comprising:

[0047] A sensor unit, used to obtain monitoring data of the motor;

[0048] A first processing unit, used for processing the monitoring data of the motor;

[0049] A second processing unit, used for obtaining motor operating parameters capable of achieving a maximum energy efficiency ratio according to the processed motor monitoring data;

[0050] The execution unit is used to adjust the operating parameters of the motor in real time according to the motor operating parameters that can achieve the maximum energy efficiency ratio.

[0051] A storage medium stores a computer program, wherein the computer program is configured to execute the method when running.

[0052] The present invention provides an energy-saving motor operation optimization analysis method, system and storage medium through improvement, which have the following improvements and advantages compared with the prior art:

[0053] Firstly, the present invention ensures the high accuracy and reliability of monitoring data by setting up multi-dimensional sensors and performing anti-interference processing. At the same time, the AR model is used to determine the time series dependency between data, which can improve the prediction accuracy of the dynamic change of the motor running state, thereby facilitating a more accurate prediction of the motor energy consumption trend.

[0054] Second: The present invention ensures the accuracy of machine learning model training through standardized processing and data cleaning, improves the generalization ability and prediction effect of the model, and realizes intelligent selection and hyperparameter optimization of motor control strategy through machine learning and Bayesian optimization, thereby ensuring the efficiency and accuracy of the energy consumption prediction model and further improving the motor energy efficiency ratio;

[0055] Thirdly, the present invention automatically searches for and adjusts the motor operating parameters to the maximum energy efficiency ratio point through a gradient ascent optimization algorithm, thereby significantly reducing the energy waste under non-full load conditions and improving energy utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The present invention will be further explained below in conjunction with the accompanying drawings and embodiments:

[0057] Figure 1 It is a flow chart of the energy-saving motor operation optimization analysis method of the present invention;

[0058] Figure 2 It is a system block diagram of the energy-saving motor operation optimization analysis system of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] It should be noted that, in the description of the present invention, the terms "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc. 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, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation on the present invention.

[0061] Furthermore, it should be understood that for the sake of ease of description, the sizes of the various components shown in the drawings are not drawn according to actual proportions. For example, the thickness or width of certain layers may be exaggerated relative to other layers.

[0062] It should be noted that like reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined or described in one drawing, it will not require further detailed discussion and description in the description of the subsequent drawings. Example 1

[0063] refer to Figure 1 This embodiment provides an energy-saving motor operation optimization analysis method, which dynamically adjusts the working parameters of the motor through intelligent analysis and control strategies, thereby reducing the energy consumption of the motor under different load conditions and improving the overall energy efficiency. Specifically, the energy-saving motor operation optimization analysis method includes the following steps:

[0064] Step S1: monitor the load and operating status of the motor in real time and obtain the monitoring data of the motor. In this embodiment, the monitoring data of the motor includes but is not limited to the current, voltage, speed, temperature, vibration frequency and other related parameters of the motor. It mainly obtains the relevant operating parameters of the motor to provide a more comprehensive information basis for the establishment of the subsequent model, thereby improving the accuracy of the model and making the prediction ability of the model more accurate. The monitoring data of the motor is obtained as follows:

[0065] Step S1.1: Set up multi-dimensional sensors according to the structure and operating environment of the motor. In this embodiment, the sensors include but are not limited to current sensors, voltage sensors, speed sensors, temperature sensors, and vibration sensors, so as to fully cover and obtain key parameters of the motor operation. It is worth noting that in the process of setting up the sensors, it is necessary to ensure the comprehensiveness and accuracy of data collection, and at the same time reduce interference to avoid data uncertainty and large errors.

[0066] Step S1.2: Anti-interference processing is performed on the data information obtained by the multi-dimensional sensor to eliminate or reduce errors or distortions introduced by the characteristics of the sensor itself, environmental factors or system noise. Anti-interference processing can usually be processed by a digital filtering method. In this embodiment, low-pass filtering in the digital filtering method is used as an example for specific explanation, as follows:

[0067] Step S1.2.1: According to the data information obtained by the multi-dimensional sensor in step S1.1, determine the filtered data value corresponding to each data, specifically: ;

[0068] in: is the filtered signal at the current time The value of is the attenuation factor, The data obtained by the multi-dimensional sensor, is the filtered signal at the previous moment The value of .

[0069] Specifically, the attenuation factor The value range is , which determines the cutoff frequency and steepness of the filter, and usually needs to be selected according to the specific application scenario and noise characteristics. is the weight of the past value, is the weight of the current value. When it is close to 1, the filter responds faster to high-frequency changes, but allows more high-frequency noise to pass. When it is close to 0, the filtering effect of the filter will be stronger, but the real-time performance of the signal will be sacrificed.

[0070] Step S1.2.2: Repeat step S1.2.1 and continuously update the filtered data value.

[0071] Step S1.3: The data information after anti-interference processing in step S1.2.2 is analyzed through time series to determine the time series dependency between the data. The details are as follows:

[0072] Step S1.3.1: Determine the order of the AR model based on the data information after anti-interference processing. The AR model is specifically: ;

[0073] in: For a time series at a point in time The value of is the autoregression coefficient at time point The value of For a time series at a point in time The value of is the autoregression coefficient at time point The value of For a time series at a point in time The value of is the autoregression coefficient at time point The value of For a time series at a point in time The value of is white noise.

[0074] Specifically, here It represents the order of the AR model, which takes into account the previous The impact of the value at a time point on the current value. Regression coefficient It can reflect the influence of the value of the time series at the corresponding previous time point on the value at the current time point. White noise Represents unpredictable random disturbances. Usually, when performing specific calculations, it generally satisfies the conditions of mean 0 and variance Normal distribution, that is, white noise The following conditions must be met: ;

[0075] It is worth noting that can be obtained by dividing the residual sum of squares by the degrees of freedom.

[0076] Step S1.3.2: Estimate the parameters of the AR model through statistical modeling tools and the order of the AR model. Specifically, after the order of the AR model is determined, the AR model is determined according to the characteristics of the time series data and the analysis requirements. Then, the parameters of the AR model are preliminarily estimated by the maximum likelihood estimation method, and the residuals are checked to see whether they meet statistical assumptions, such as cubic distribution, zero mean, etc., so as to adjust the parameters of the AR model or reconstruct the AR model, so that the performance of the final AR model can meet the predetermined standard or optimal criterion.

[0077] Step S1.3.3: In time series analysis, the applicability of the AR model determined in step S1.3.2 can be tested by residual analysis, autocorrelation diagram, partial autocorrelation diagram, etc. to ensure that the determined AR model can fully extract the autocorrelation in the data. In this embodiment, the order of the autoregressive term of the AR model is determined by the autocorrelation diagram, specifically:

[0078] ;

[0079] in: is the first-order autocorrelation coefficient of the autoregressive term of the AR model, is the sequence length, For a time series at a point in time The value of is the mean value of the series, is the value of the time series at time point t-1, For time point.

[0080] Step S1.3.4: After verification through step S1.3.3, the order of the autoregressive term of the final AR model can be obtained, and the final AR model can be established. At the same time, the time series dependency between the data can be determined through the final AR model.

[0081] Step S2: Establish an energy consumption prediction model through machine learning algorithms and monitoring data, and predict the operation of the motor through the established energy consumption prediction model, that is, predict the motor energy consumption under different control strategies. The details are as follows:

[0082] Step S2.1: Standardize the monitoring data after anti-interference processing in step S1.2.2, as follows:

[0083] Step S2.1.1: Clean the monitoring data using the interquartile range rule, specifically:

[0084] ;

[0085] in: is the interquartile range, is the upper quartile, The lower quartile.

[0086] Specifically, the interquartile range It can measure the dispersion of the middle 50% of the data set, that is, the difference between the two quartiles in the middle of the data set. The 75th percentile is the value at the 75th position after the monitoring data is arranged from small to large. In other words, 75% of the data values ​​are less than or equal to the upper quartile. , and the remaining 25% of the data values ​​are greater than the upper quartile . Lower quartile The 25th percentile is the value at the 25th position after the monitoring data is arranged from small to large. In other words, 25% of the data values ​​are less than or equal to the lower quartile. , and the remaining 75% of the data values ​​are greater than the lower quartile .

[0087] Step S2.1.2: The cleaned monitoring data is standardized by the Z-score standardization algorithm, that is, the cleaned monitoring data in step S2.1.1 is scaled so that it falls into a small specific interval actually required. In this embodiment, the calculation formula of the Z-score standardization algorithm is specifically:

[0088] ;

[0089] in: is the standardized monitoring data. For the monitoring data after cleaning, is the mean value of the data set, is the standard deviation.

[0090] Step S2.2: Since there are many types of machine learning models, a machine learning model is selected and trained based on the standardized monitoring data in step S2.1.2 to determine the final machine learning model, as follows:

[0091] Step S2.2.1: Select a primary machine learning model based on the monitoring data after anti-interference processing in step S1.2.2.

[0092] Step S2.2.2: Through the Bayesian optimization method, combined with the primary machine learning model, the optimal hyperparameter combination corresponding to the primary machine learning model is determined, and the optimal hyperparameter combination is substituted into the primary machine learning model to obtain the machine learning model after primary training.

[0093] Step S2.2.3: The machine learning model after primary training obtained in step S2.2.2 is verified by the K-fold cross-validation method, wherein the machine learning model after primary training that passes the verification is the final machine learning model.

[0094] Step S2.3: Use the monitoring data after standardization in step S2.1.2 as the input of the final machine learning model in step S2.2.3 to establish the energy consumption prediction model, and predict the motor energy consumption under different control strategies through the established energy consumption prediction model.

[0095] Step S3: The predicted energy consumption of the motor obtained in step S2.3 is optimized by an optimization algorithm to determine the motor operating parameters that can achieve the maximum energy efficiency ratio, as follows:

[0096] Step S3.1: Convert the parameters to be optimized into the maximum energy efficiency ratio. and frequency Specifically, the output power With voltage and frequency The following relationship is satisfied:

[0097] ;

[0098] in: is the output power of the motor, is the operating voltage of the motor, is the operating frequency of the motor, are all constants.

[0099] Furthermore, the maximum energy efficiency ratio converted by the above formula is:

[0100] ;

[0101] in: is the output power of the motor, is the operating voltage of the motor, is the operating frequency of the motor, are all constants.

[0102] It is worth noting that in the above formula It can be expressed as the following formula, specifically:

[0103] ;

[0104] in: is the operating voltage of the motor, is the operating frequency of the motor, are all constant coefficients.

[0105] Step S3.2: The gradient vector corresponding to the parameter to be optimized obtained in step S3.1 is obtained by using the gradient ascent formula, wherein the gradient ascent formula is specifically:

[0106] ;

[0107] in: is the gradient vector corresponding to the parameter to be optimized, is the operating voltage of the motor, is the operating frequency of the motor.

[0108] Step S4: According to the motor operating parameters that can achieve the maximum energy efficiency ratio obtained in step S3.2, adjust the operating parameters of the motor in real time, and repeat steps S1-S3 to adjust the actual operating parameters of the motor in real time, so that the motor can continue to maintain efficient and stable operation. Example 2

[0109] The present embodiment provides an energy-saving motor operation optimization analysis system, which is used to implement the above-mentioned embodiment mode, and the description has been made no further. As used below, the term "unit" can implement a combination of software and / or hardware of a predetermined function. Although the system described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0110] Figure 2 is a system block diagram of the energy-saving motor operation optimization analysis system according to an embodiment of the present invention, Figure 2 The energy-saving motor operation optimization analysis system includes a sensor unit for obtaining monitoring data of the motor. A first processing unit for processing the monitoring data of the motor. A second processing unit for obtaining motor operating parameters that can achieve a maximum energy efficiency ratio based on the processed motor monitoring data. An execution unit for adjusting the operating parameters of the motor in real time based on the motor operating parameters that can achieve a maximum energy efficiency ratio.

[0111] In this embodiment, the process of adjusting the motor operation by the energy-saving motor operation optimization analysis system is as follows:

[0112] Step S1: Set the sensor unit at a key position of the motor to monitor the operating status of the motor in real time. It is worth noting that the operating parameters of the motor include but are not limited to key parameters such as current, voltage, speed, temperature, and load.

[0113] Step S2: After receiving the original motor operation data transmitted by the sensor unit, the first processing unit performs necessary data cleaning, noise removal and outlier removal operations, and then performs preliminary analysis to provide standard data for the subsequent operation of the second processing unit.

[0114] Step S3: The second processing unit uses a pre-trained machine learning model, such as a deep learning network, a support vector machine, etc., based on the monitoring data processed by the first processing unit to predict the energy consumption of the motor under different control strategies. It is worth noting that the model needs to combine the working characteristics and historical operating data of the motor to estimate the energy consumption more accurately. And by using the optimization algorithm and the prediction results, the working parameters that can enable the motor to achieve the maximum energy efficiency ratio under the current working conditions are determined.

[0115] Step S4: the execution unit adjusts the actual operating parameters of the motor in real time through the motor control system according to the optimal operating parameters calculated by the second processing unit, so as to maximize the motor operating energy efficiency.

[0116] It should be noted that the above-mentioned units can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to this: the above-mentioned units are all located in the same processor; or the above-mentioned units are located in different processors in any combination. Example 3

[0117] This embodiment provides a storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps in the above method embodiment when running.

[0118] Furthermore, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0119] Step S1: Acquire the monitoring data of the motor in real time.

[0120] Step S2: Based on the monitoring data of the motor, an energy consumption prediction model is established to determine the predicted energy consumption of the motor under different control strategies.

[0121] Step S3: Optimize the predicted energy consumption of the motor through an optimization algorithm to determine the motor operating parameters that can achieve the maximum energy efficiency ratio.

[0122] Step S4: adjusting the operating parameters of the motor in real time according to the motor operating parameters that can achieve the maximum energy efficiency ratio.

[0123] Specifically, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0124] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing and analyzing the operation of an energy-saving motor, characterized in that: The analytical method comprises the following steps: S1: monitor the load and operating status of the motor in real time and obtain monitoring data of the motor; S2: Establishing an energy consumption prediction model through a machine learning algorithm and the monitoring data to determine the predicted energy consumption of the motor under different control strategies; S3: The motor predicted energy consumption is determined by an optimization algorithm to achieve the motor operating parameters that can achieve the maximum energy efficiency ratio; S4: adjusting the operating parameters of the motor in real time according to the operating parameters of the motor, and repeating steps S1-S3; The monitoring data of the motor is obtained as follows: S1.1: According to the structure and operating environment of the motor, a multi-dimensional sensor is set; S1.2: performing anti-interference processing on the data information acquired by the multi-dimensional sensor; S1.3: The data information after anti-interference processing is analyzed through time series to determine the temporal dependency relationship between the data; The data information obtained by the multi-dimensional sensor is subjected to anti-interference processing, as follows S1.2.1: Determine the filtered data value corresponding to each data according to the data information obtained by the multi-dimensional sensor, specifically: ； in: is the value of the filtered signal at the current time n, is the attenuation factor, The data obtained by multi-dimensional sensors, is the value of the filtered signal at the previous moment n-1; S1.2.2: Repeat step S1.2.1 and continuously update the filtered data value; Determine the timing dependencies between data, as follows: S1.3.1: Determine the order of the AR model according to the data information after the anti-interference processing, and the AR model is specifically: ； in: is the value of the time series at time point t, is the autoregression coefficient at time point The value of For a time series at a point in time The value of is the autoregression coefficient at time point The value of For a time series at a point in time The value of is the autoregression coefficient at time point The value of For a time series at a point in time The value of is white noise; S1.3.2: Estimate the parameters of the AR model using statistical modeling tools and the order of the AR model; S1.3.3: Determine the order of the autoregressive term of the AR model through the autocorrelation diagram, specifically: ; in: is the first-order autocorrelation coefficient of the autoregressive term of the AR model, is the sequence length, is the value of the time series at time point t, is the mean value of the series, For a time series at a point in time The value of , t is the time point; S1.3.4: Establish the AR model according to the order of the autoregressive term of the AR model, and determine the temporal dependency relationship between the data through the AR model; The energy consumption prediction model is established as follows: S2.1: Standardize the monitoring data; S2.2: Select and train a machine learning model and determine the final machine learning model; S2.3: using the standardized monitoring data as input of the final machine learning model to establish the energy consumption prediction model; The monitoring data is standardized as follows: S2.1.1: Clean the monitoring data using the interquartile range rule, specifically: ; in: is the interquartile range, is the upper quartile, is the lower quartile; S2.1.2: The cleaned monitoring data is standardized by the Z-score standardization algorithm, specifically: ; in: is the standardized monitoring data. For the monitoring data after cleaning, is the mean value of the data set, is the standard deviation; The final machine learning model is determined as follows: S2.2.1: Selecting a primary machine learning model based on the monitoring data after anti-interference processing; S2.2.2: Determine the optimal hyperparameter combination corresponding to the primary machine learning model through the Bayesian optimization method; S2.2.3: The machine learning model corresponding to the optimal hyperparameter combination is verified by the K-fold cross-validation method to determine the final machine learning model.

2. The energy-saving motor operation optimization analysis method according to claim 1 is characterized in that: Determine the motor operating parameters that can achieve the maximum energy efficiency ratio, as follows: S3.1: Convert the parameters to be optimized into the maximum energy efficiency ratio; S3.2: Obtain the gradient vector corresponding to the parameter to be optimized by using the gradient ascent formula, wherein the gradient ascent formula is specifically: ; in: is the gradient vector corresponding to the parameter to be optimized, is a parameter to be optimized, is another parameter to be optimized.

3. A storage medium, characterized in that: The storage medium stores a computer program, wherein the computer program is configured to execute the method according to any one of claims 1 to 2 when executed.

Images