A Motor Energy Saving Optimization and Intelligent Regulation Platform

Abstract

The present invention relates to the technical field of energy conservation of motor systems, and discloses a motor energy conservation optimization and intelligent control platform. The platform includes: a motor data analysis module, a motor system simulation module, a characteristic parameter extraction module, and a control strategy generation module. Specifically: perform performance analysis on the acquired operation data to obtain performance data; construct a dynamic simulation model using the acquired motor system parameters; calculate the dynamic characteristics of the motor system based on the performance data and the dynamic simulation model, and extract characteristic parameters from the operation data based on the dynamic characteristics to obtain mode characteristic parameters; based on the mode characteristic parameters, perform control parameter optimization iteration on the dynamic simulation model, evaluate the iteration results, and screen out the optimal control parameters from the candidate parameter set according to the strategy evaluation results. The present invention can solve the problem that the current motor control system still has insufficient control accuracy in motor energy conservation optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of energy-saving technologies for motor systems, and particularly to a motor energy-saving optimization and intelligent control platform. Background Art

[0002] As one of the most important power devices in industrial production, the energy efficiency level of motors directly affects the energy consumption and carbon emissions of the entire industrial system. Therefore, improving the energy efficiency of motor systems is of great significance for achieving the goals of energy conservation and emission reduction.

[0003] Currently, most motor control systems still adopt traditional control strategies, such as PID controllers with fixed parameters. Although such control methods are simple and easy to implement, they often cannot achieve the optimal control effect when facing complex and changeable working environments. Especially in working conditions with high nonlinearity and uncertainty, problems such as overshoot, undershoot, or oscillation are likely to occur, resulting in reduced energy efficiency and energy waste. In addition, traditional control strategies lack self-adaptability and learning ability, and it is difficult to make real-time adjustments according to the actual operating state of the motor, and they cannot make full use of the data resources brought by modern information technology. In summary, the current motor control systems still have problems with insufficient control accuracy in motor energy-saving optimization. Summary of the Invention

[0004] The present invention provides a motor energy-saving optimization and intelligent control platform, and its main purpose is to solve the problem that the current motor control systems still have insufficient control accuracy in motor energy-saving optimization.

[0005] To achieve the above purpose, a motor energy-saving optimization and intelligent control platform provided by the present invention includes:

[0006] A motor data analysis module, a motor system simulation module, a characteristic parameter extraction module, and a control strategy generation module. Specifically:

[0007] The motor data analysis module is used to obtain the operation data of historical motors, perform pattern recognition on the operation data to obtain pattern probabilities, and calculate the performance data of the operation data according to the pattern probabilities. Among them, the following formula is used for pattern recognition:

[0008] ;

[0009] Among them, represents the operation data corresponding pattern probability, is the weight of the th pattern in the preset weight matrix, is the total number of weights in the weight matrix, is the th pattern corresponding probability mean value, is the covariance matrix of the probabilities corresponding to the th mode, is the probability that the given data belongs to the th mode;

[0010] The motor system simulation module is used to construct a dynamic simulation model by using the collected motor system parameters;

[0011] The characteristic parameter extraction module is used to calculate dynamic characteristics according to the performance data and the dynamic simulation model, and extract the mode characteristic parameters of the operation data based on the dynamic characteristics;

[0012] The control strategy generation module is used to optimize the control parameters of the simulation model based on the mode characteristic parameters, generate a candidate parameter set, and perform parameter screening on the candidate parameter set to obtain the optimal control parameters.

[0013] Optionally, when the motor system simulation module executes the function of constructing a dynamic simulation model by using the collected motor system parameters, it is specifically used for:

[0014] Fitting a preset electromagnetic model according to the motor system parameters to obtain a motor electromagnetic model;

[0015] Fitting a preset mechanical model according to the motor system parameters to obtain a motor mechanical model;

[0016] Fitting a preset control model according to the motor system parameters to obtain a motor control model;

[0017] Constructing a simulation environment according to the motor control model, the motor mechanical model and the motor electromagnetic model to obtain a dynamic simulation model.

[0018] Optionally, when the characteristic parameter extraction module executes the function of calculating dynamic characteristics according to the performance data and the dynamic simulation model, it is specifically used for:

[0019] Constructing a state space equation according to the dynamic simulation model and the performance data;

[0020] Discretizing the data of the state space equation to obtain a discrete time matrix;

[0021] Performing matrix parameter estimation by using the performance data and the discrete time matrix to obtain a characteristic matrix;

[0022] Performing eigenvalue decomposition on the characteristic matrix to obtain dynamic characteristics.

[0023] Optionally, when the feature parameter extraction module executes the function of constructing a state - space equation according to the dynamic simulation model and the performance data, it specifically is used for:

[0024] Extract the input variables and output variables of the motor system from the dynamic simulation model;

[0025] Obtain the relationship between the input variables and output variables of the motor system based on the performance data;

[0026] Define the state variables of the motor system by using the dynamic simulation model and the performance data;

[0027] Construct a state - space equation according to the relationship between the state variables and the input variables and output variables of the motor system.

[0028] Optionally, when the feature parameter extraction module executes the function of discretizing the state - space equation to obtain a discrete - time matrix, it specifically is used for:

[0029] Determine the time step of the state - space equation according to the motor system parameters;

[0030] Discretize the state - space equation by using the time step to generate a discretized continuous - time equation;

[0031] Calculate the discrete - time matrix according to the discretized continuous - time equation.

[0032] Optionally, when the feature parameter extraction module executes the function of extracting the mode feature parameters of the operation data based on the dynamic characteristics, it specifically is used for:

[0033] Segment the operation data by using the dynamic characteristics to obtain time - series data of different stages;

[0034] Calculate the stability index of the motor system by using the time - series data and the dynamic characteristics;

[0035] Calculate the critical points of the efficiency interval of the time - series data;

[0036] Calculate the dynamic response time of the time - series data;

[0037] Combine the dynamic response time, the critical points of the efficiency interval, the stability index, and the performance data corresponding to the operation data to obtain mode feature parameters.

[0038] Optionally, when the feature parameter extraction module executes the function of segmenting the operation data by using the dynamic characteristics to obtain time - series data of different stages, it specifically is used for:

[0039] Construct a state evaluation function according to the dynamic characteristics;

[0040] Use the state evaluation function to perform change point detection on the operation data to obtain a state change probability value. The change point detection can be performed using the following formula:

[0041] ;

[0042] where, is the state change probability value corresponding to the time point , is the base of the natural logarithm, is a preset constant, is the time point when the change amount of the operation data, is the preset significance threshold operation data;

[0043] Construct a dynamic window adjustment function based on the state change probability value;

[0044] Use the dynamic window adjustment function and the preset initial window to segment the operation data to obtain time series data.

[0045] Optionally, when the control strategy generation module executes the function of optimizing the control parameters of the simulation model based on the pattern feature parameters to generate a candidate parameter set, it specifically is used for:

[0046] Use the pattern feature parameters to screen out the control key parameters from the operation data;

[0047] Use the control key parameters and the dynamic simulation model to generate a simulation result;

[0048] Calculate the performance index of the simulation result;

[0049] Use the performance index to adjust the control key parameters to obtain candidate control parameters;

[0050] Return to the step of generating a simulation result using the control key parameters and the dynamic simulation model until the number of parameter adjustments reaches the preset threshold to obtain a candidate parameter set.

[0051] Optionally, when the control strategy generation module executes the function of screening out the control key parameters from the operation data using the pattern feature parameters, it specifically is used for:

[0052] Calculate the correlation score between the pattern feature parameters and the operation data;

[0053] Screen out the preliminary relevant data from the operation data by using the correlation score and a preset threshold value;

[0054] Perform multiple correlation verifications on the preliminary relevant data to obtain a verification result;

[0055] Screen out the control key parameters from the preliminary relevant data based on the verification result.

[0056] Optionally, when the control strategy generation module executes the function of performing parameter screening on the candidate parameter set to obtain the optimal control parameter, it specifically is used for:

[0057] Perform a simulation operation of the motor based on the candidate parameter set and calculate the corresponding energy-saving effect;

[0058] Evaluate the performance indicators of each group of candidate control parameters on the motor system stability, response speed, and energy consumption based on the energy-saving effect;

[0059] Score each performance indicator according to a preset weight formula to obtain a scoring result;

[0060] Screen out the candidate control parameter with the highest comprehensive score as the optimal control parameter according to the scoring result.

[0061] Through detailed motor system parameters and advanced simulation tools, the present invention constructs a highly realistic motor system model to ensure the accuracy and reliability of the model; by using input performance data and a dynamic simulation model, the dynamic characteristics are calculated by using the matrix expression method, and the key characteristic parameters that have a significant impact on the motor performance are extracted, providing accurate data support for the optimization of the control strategy. Therefore, a motor energy-saving optimization and intelligent regulation platform proposed by the present invention can solve the problem that the current motor control system still has insufficient control accuracy in motor energy-saving optimization. Brief Description of the Drawings

[0062] Figure 1 It is a functional module diagram of a motor energy-saving optimization and intelligent regulation platform provided by an embodiment of the present invention;

[0063] Figure 2 It is a flow schematic diagram of pattern recognition provided by an embodiment of the present invention;

[0064] Figure 3 It is a flow schematic diagram of generating dynamic characteristics provided by an embodiment of the present invention;

[0065] Figure 4 It is a flow schematic diagram of a motor energy-saving optimization and intelligent regulation method provided by an embodiment of the present invention.

[0066] The realization, functional characteristics, and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed implementation manners

[0067] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0068] Referring to Figure 1 As shown, it is a functional module diagram of a motor energy-saving optimization and intelligent control platform provided by an embodiment of the present invention. In this embodiment, the motor energy-saving optimization and intelligent control platform 100 can be installed in an electronic device. According to the functions achieved, the motor energy-saving optimization and intelligent control platform 100 can include a motor data analysis module 101, a state calculation module 102, a fault detection module 103, and an automatic adjustment module 104. The modules in the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0069] In the embodiment of the present invention, the motor data analysis module 101 includes obtaining the operation data of historical motors, performing pattern recognition on the operation data to obtain pattern probabilities, and calculating the performance data of the operation data according to the pattern probabilities;

[0070] In the embodiment of the present invention, the motor system simulation module 102 includes constructing a dynamic simulation model by using the collected motor system parameters;

[0071] In the embodiment of the present invention, the characteristic parameter extraction module 103 includes calculating dynamic characteristics according to the performance data and the dynamic simulation model, and extracting pattern characteristic parameters of the operation data based on the dynamic characteristics;

[0072] In the embodiment of the present invention, the control strategy generation module 104 includes optimizing the control parameters of the simulation model based on the pattern characteristic parameters, generating a candidate parameter set, and performing parameter screening on the candidate parameter set to obtain the optimal control parameters.

[0073] Specifically, in the embodiment of the present invention, each module in the motor energy-saving optimization and intelligent control platform 100 uses the same technical means as the motor energy-saving optimization and intelligent control platform described in the drawings and can produce the same technical effects, which will not be elaborated here.

[0074] Next, in combination with specific embodiments, the respective components and specific working processes of the motor energy-saving optimization and intelligent control platform will be described separately:

[0075] The motor data analysis module 101 is used to obtain the operation data of historical motors, perform pattern recognition on the operation data to obtain pattern probabilities, and calculate the performance data of the operation data according to the pattern probabilities.

[0076] In the embodiments of the present invention, by obtaining operation data, it helps to analyze the long-term performance trend and state change of the motor, identify the fluctuations and potential problems of the motor performance, and provide support for subsequent optimization.

[0077] In the embodiments of the present invention, the operation data refers to the data recording the actual operation of the motor in the past period of time, including operation time, start-up times, motor load data, motor temperature data, voltage and current data, as well as fault and alarm data, etc.

[0078] Referring to Figure 2 As shown, in the embodiments of the present invention, when the motor data analysis module executes the function of performing performance analysis on the operation data according to the pattern probability to obtain performance data, it specifically is used for:

[0079] S21. Perform pattern division on the operation data according to the pattern probability to obtain data patterns;

[0080] S22. Perform regression analysis on the data patterns to obtain a performance change trend;

[0081] S23. Perform structured processing on the operation data based on the performance change trend to obtain performance data.

[0082] In the embodiments of the present invention, the following formula is used for pattern recognition:

[0083] ;

[0084] Wherein, represents the operation data corresponding pattern probability, is the weight of the th pattern in the preset weight matrix, is the total number of weights in the weight matrix, is the probability mean value corresponding to the th pattern, is the covariance matrix of the probability corresponding to the th pattern, is the probability that the given data belongs to the th pattern.

[0085] Specifically, use a multivariate linear regression equation to fit the data patterns and the operation data, perform regression coefficient analysis on the fitted multivariate linear regression equation to obtain a performance change trend. Since the positive and negative values of the regression coefficient indicate the influence direction of the independent variable on the dependent variable, and the absolute value of the coefficient represents the strength of the influence, so analyze the regression coefficient to obtain a performance change trend.

[0086] In the embodiments of the present invention, based on the performance change trend, the data is reclassified. Specifically, the originally scattered and disordered motor operation data is grouped according to different performance trends (such as "normal operation", "overload operation", "abnormal state", etc.), and each classified data is organized according to certain rules (such as time, load, speed, etc.) to form a structured table or database, including summarizing the motor operation data for different time periods to form records with fields such as "date - load - temperature - speed - power", etc. After reclassification and structuring, a unified and representative performance dataset, that is, performance data, is formed.

[0087] In the embodiments of the present invention, the pattern recognition can calculate the probability that each operation data point belongs to different patterns, identify the complex distribution and multiple patterns in the motor operation data, provide the probability distribution information of the data points, and provide a basis for subsequent pattern division.

[0088] In the embodiments of the present invention, based on the pattern probability, dividing the cleaned data into different patterns is achieved by assigning each data point to the pattern with the highest probability, so that each pattern corresponds to an operation state of the motor (such as normal, overload, inefficient, etc.), making the data divided into meaningful operation states, facilitating subsequent analysis, providing the classification result of the motor operation state, and supporting state monitoring and fault diagnosis.

[0089] In the embodiments of the present invention, by discovering the motor performance decline or potential faults in advance, the risk of faults can be reduced, and the stability and reliability of the motor system can be improved.

[0090] The motor system simulation module 102 is used to construct a dynamic simulation model by using the collected motor system parameters.

[0091] In the embodiments of the present invention, an accurate dynamic simulation model is constructed through the motor system parameters to simulate the performance of the motor under different operating conditions, avoiding high - cost and high - risk actual tests, and conducting multiple simulation verifications.

[0092] In the embodiments of the present invention, the motor system parameters refer to various key parameters and structural information used to describe the motor and its related systems. These information are the basis for constructing the dynamic simulation model, determining the accuracy and functional performance of the model, including motor body information, control system motor system parameters, control parameters, motor system environment and load information, etc.

[0093] In the embodiments of the present invention, when the motor system simulation module executes the function of constructing a dynamic simulation model by using the collected motor system parameters, it is specifically used for:

[0094] Fitting a preset electromagnetic model according to the motor system parameters to obtain a motor electromagnetic model;

[0095] Fit a preset mechanical model according to the motor system parameters to obtain a motor mechanical model;

[0096] Fit a preset control model according to the motor system parameters to obtain a motor control model;

[0097] Build a simulation environment according to the motor control model, the motor mechanical model and the motor electromagnetic model to obtain a dynamic simulation model.

[0098] In the embodiment of the present invention, the mathematical expression of the electromagnetic model is:

[0099] ;

[0100] Wherein, is the stator voltage in the motor system parameters, is the stator resistance in the motor system parameters, is the stator current in the motor system parameters, is the stator inductance in the motor system parameters, is the preset back electromotive force, the time change rate of the stator current, is the differential symbol.

[0101] In the embodiment of the present invention, the mathematical expression of the mechanical model is:

[0102] ;

[0103] Wherein, is the moment of inertia in the motor system parameters, is the angular velocity obtained by fitting, is the angular acceleration, which describes the time change rate of the angular velocity, represents a small change in the angular velocity, is the electromagnetic torque in the motor system parameters, is the load torque in the motor system parameters, is the preset friction coefficient.

[0104] In the embodiment of the present invention, the mathematical expression of the mathematical model is:

[0105] ;

[0106] Wherein, is the preset error result, is the control signal at the th moment in the motor system parameters, is the proportional gain in the motor system parameters, is the integral gain in the motor system parameters, is the derivative gain in the motor system parameters.

[0107] In the embodiments of the present invention, a physical model and a mathematical model are integrated to simulate the real-time behavior of the motor system and output the changes of key variables (such as current, torque, speed).

[0108] In the embodiments of the present invention, by fitting these models, a more accurate dynamic simulation model can be established based on the actual motor system parameters (such as motor parameters, electrical characteristics, mechanical loads, etc.). For example, the electromagnetic model of the motor can accurately describe the relationship between current and voltage, the mechanical model can describe the influence of factors such as moment of inertia and friction on the movement of the motor, and the control model can predict the reaction of the control system, enabling the motor to still operate stably when the load changes.

[0109] The feature parameter extraction module 103 is configured to calculate dynamic characteristics according to the performance data and the dynamic simulation model, and extract the pattern feature parameters of the operation data based on the dynamic characteristics.

[0110] In the embodiments of the present invention, by calculating based on the performance data and the dynamic simulation model, the key feature parameters of the motor are extracted, which can accurately reflect the dynamic characteristics of the motor.

[0111] As shown in the figure Figure 3 when the feature parameter extraction module executes the function of calculating the dynamic characteristics according to the performance data and the dynamic simulation model, it is specifically configured to:

[0112] S31. Construct a state space equation according to the dynamic simulation model and the performance data;

[0113] S32. Discretize the data of the state space equation to obtain a discrete time matrix;

[0114] S33. Use the performance data and the discrete time matrix for matrix parameter estimation to obtain a feature matrix;

[0115] S34. Perform eigenvalue decomposition on the feature matrix to obtain the dynamic characteristics.

[0116] In the embodiments of the present invention, when the feature parameter extraction module executes the function of constructing a state space equation according to the dynamic simulation model and the performance data, it is specifically configured to:

[0117] Extract the input variables and output variables of the motor system from the dynamic simulation model;

[0118] Obtain the relationship between the input variables and output variables of the motor system based on the performance data;

[0119] Define the state variables of the motor system by using the dynamic simulation model and the performance data;

[0120] Construct a state - space equation based on the relationship between the state variables, the input variables, and the output variables of the motor system.

[0121] Specifically, according to the definitions in the dynamic simulation model, automatically identify the input variables (such as voltage, current), output variables (such as rotational speed, torque), and state variables (such as rotor position, speed), and store the identified variables in a list or dictionary.

[0122] Specifically, match the performance data with the input variables and output variables, and use statistical methods (such as correlation analysis, regression analysis, etc.) to find the relationship between the input variables and output variables, and the variation of these relationships over time.

[0123] Specifically, according to the data expression of the dynamic simulation model, preliminarily select state variables such as rotor position, speed, etc., and use the performance data to verify the effectiveness of the state variables to obtain the final state variables.

[0124] Specifically, the state - space equation is:

[0125] ;

[0126] ;

[0127] Wherein, is the derivative of the state variable, is the state variable, is the output variable, is the input variable, is the system matrix, representing the relationship between states.

[0128] Specifically, by extracting the input variables and output variables of the motor system from the dynamic simulation model and combining the performance data to determine the relationship between the two, the dynamic behavior of the motor system can be described more precisely. By constructing a state - space equation and using the extracted state variables and their relationships with the input and output variables, a more comprehensive state - space equation can be constructed, enhancing the control ability of the motor system in energy - efficiency optimization.

[0129] In the embodiment of the present invention, when the feature parameter extraction module executes the function of discretizing the data of the state - space equation to obtain a discrete - time matrix, it is specifically used for:

[0130] Determine the time step of the state - space equation according to the motor system parameters;

[0131] Discretize the state - space equation using a time step to generate a discretized continuous - time equation;

[0132] Calculate the discrete - time matrix according to the discretized continuous - time equation.

[0133] Specifically, according to the specific motor - system parameters of the motor system, take one - tenth of the fastest dynamic - change time constant in the motor - system parameters as the time step.

[0134] Specifically, use a zero - order hold (ZOH) or other discretization methods (such as bilinear transformation) to discretize the state - space equation. By performing equation transformation on the system matrix in the state - space equation using the bilinear - transformation formula, a discretized continuous - time equation is obtained.

[0135] Specifically, according to the discretized continuous - time equation and the state - space equation, calculate the matrix exponential through the Taylor - series expansion method, and then calculate the matrix integral through a numerical - integration method, such as the trapezoidal rule or Simpson's rule. Re - combine the calculated matrix exponential and matrix integral with the system matrix to obtain the discrete - time matrix.

[0136] Specifically, by discretizing the continuous - time state - space equation, the dynamic behavior of the motor system can be more accurately represented in a digital control system, and the discretized equation is applicable to computer - control algorithms. This helps to improve the control accuracy of the system, ensure an accurate match between the control signal and the motor output at discrete time steps, and avoid system instability or inaccurate control caused by discretization errors.

[0137] In an embodiment of the present invention, construct an error function according to the discrete - time matrix, and optimize the error function using the least - squares method or the recursive least - squares method based on the configuration data and the performance data to ensure that the discrete - time model can reflect the dynamic law in the performance data, and finally obtain the characteristic matrix.

[0138] In an embodiment of the present invention, perform eigenvalue decomposition using the following formula:

[0139] ;

[0140] Wherein, is the characteristic matrix, is the diagonal matrix corresponding to the eigenvalues, is the eigen - vector matrix of the eigen - vector matrix decomposition.

[0141] In an embodiment of the present invention, when the characteristic - parameter extraction module executes the function of extracting the mode - characteristic parameters of the operation data based on the dynamic characteristics, it is specifically used for:

[0142] Segment the operation data according to the dynamic characteristics to obtain time series data of different stages;

[0143] Calculate the stability index of the motor system by using the time series data and the dynamic characteristics;

[0144] Calculate the critical points of the efficiency interval of the time series data;

[0145] Calculate the dynamic response time of the time series data;

[0146] Combine the dynamic response time, the critical points of the efficiency interval, the stability index, and the performance data corresponding to the operation data to obtain mode characteristic parameters.

[0147] In the embodiments of the present invention, according to the dynamic characteristics of the motor system, historical data is divided into different time periods. Since the dynamic characteristics can help identify key events or state change points in the data, the operation data is segmented through a variable-length time window (the window length is adaptively adjusted according to the dynamic characteristics) to obtain time series data.

[0148] In the embodiments of the present invention, by segmenting the operation data and calculating the stability index, the critical points of the efficiency interval, and the dynamic response time in combination with the dynamic characteristics, the performance of the motor system in different operating states can be accurately identified. By calculating the stability index, the stability of the motor system in each operating stage can be analyzed, avoiding problems such as motor oscillation, overheating, and stall, and comprehensively understanding the working state of the motor in each stage, so as to provide key data support for system optimization, fault diagnosis, performance prediction, etc.

[0149] In the embodiments of the present invention, when the feature parameter extraction module executes the function of segmenting the operation data according to the dynamic characteristics to obtain time series data of different stages, it is specifically used for:

[0150] Construct a state evaluation function according to the dynamic characteristics;

[0151] Detect change points of the operation data by using the state evaluation function to obtain state change probability values;

[0152] Construct a dynamic window adjustment function based on the state change probability values;

[0153] Segment the operation data by using the dynamic window adjustment function and a preset initial window to obtain time series data.

[0154] Specifically, the state evaluation function is used to describe the operating state of the motor system at different time points. These states are usually evaluated based on some characteristic values (such as rotational speed, load, current, etc.). By performing weighted calculations on the data of the state characteristics, the state evaluation function is finally obtained. Specifically, change point detection refers to identifying significant changes (such as sudden load changes, speed regulation processes, etc.) that occur during the operation of the motor. Through the state evaluation function, the "state change probability value" of each time point is calculated, which reflects the degree of state change between the current time point and the previous time point.

[0155] Specifically, by constructing the state evaluation function and combining change point detection, the state changes (such as load changes, temperature fluctuations, etc.) that occur during the operation of the motor system can be accurately identified. This dynamic segmentation processing can divide the operating data of the system into multiple meaningful time windows, thereby providing detailed data support for further analysis (such as performance optimization, fault prediction, etc.).

[0156] Specifically, the following formula can be used for change point detection:

[0157] ;

[0158] where, is the state change probability value corresponding to the time point , is the base of the natural logarithm, is a preset constant, is the change amount of the operating data at the time point , is the preset significance threshold for the operating data.

[0159] Specifically, based on the calculated state change probability value and the preset threshold, a piecewise function is constructed to obtain the dynamic window adjustment function.

[0160] Specifically, the data segmentation of the operating data using the dynamic window adjustment function and the preset initialization window is carried out through the initialization window. At each moment, S(t) and the state change probability value and the current dynamic characteristics are calculated. The dynamic window adjustment function is used to accumulate the data of the initialization window. When the accumulated data reaches the last data, a time series segment is generated, and overlapping control is executed, that is, if the overlap of adjacent segments > 15%, they are merged into a single segment, and finally the time series data is obtained.

[0161] In the embodiments of the present invention, time series analysis is performed on the data of each time period, and the autocorrelation coefficient is calculated.

[0162] In the embodiments of the present invention, the system efficiency in each time period is calculated, which is usually defined as the ratio of the output power to the input power. Then, a sliding window is used to calculate the average efficiency in each time period, and the mutation points of the efficiency change are detected.

[0163] In the embodiments of the present invention, a step input (such as suddenly increasing the load) is applied to the motor system, the time required for the system to reach the stable state from the initial state is recorded, and the rise time, peak time, and adjustment time are calculated. The calculated times are stored in the form of a vector to obtain the dynamic response time.

[0164] In the embodiments of the present invention, the dynamic response time, the critical points of the efficiency interval, the stability index, and the performance data corresponding to the operation data (such as energy consumption, temperature, etc.) are integrated together. By combining all the characteristic parameters into a characteristic vector, each characteristic parameter serves as a dimension.

[0165] In the embodiments of the present invention, through these characteristic parameters, accurate monitoring of the motor state can be achieved, providing data support for early fault diagnosis and maintenance.

[0166] The control strategy generation module 104 is used to optimize the control parameters of the simulation model based on the pattern characteristic parameters, generate a candidate parameter set, and perform parameter screening on the candidate parameter set to obtain the optimal control parameters.

[0167] In the embodiments of the present invention, by optimizing the extracted characteristic parameters, the dynamic simulation model is adjusted and a candidate parameter set is generated to improve the effect of the control strategy.

[0168] In the embodiments of the present invention, when the control strategy generation module executes the function of optimizing the control parameters of the simulation model based on the pattern characteristic parameters and generating a candidate parameter set, it is specifically used for:

[0169] Screening out the control key parameters from the operation data by using the pattern characteristic parameters;

[0170] Generating a simulation result by using the control key parameters and the dynamic simulation model;

[0171] Calculating the performance index of the simulation result;

[0172] Adjusting the control key parameters by using the performance index to obtain candidate control parameters;

[0173] Returning to the step of generating a simulation result by using the control key parameters and the dynamic simulation model until the number of times of parameter adjustment reaches a preset threshold to obtain a candidate parameter set.

[0174] In the embodiments of the present invention, a feature selection method is used to select control parameters highly correlated with the pattern feature parameters from historical data. By calculating the correlation coefficient between each control parameter and the pattern feature parameters, parameters with a correlation coefficient greater than a preset correlation threshold are selected to obtain the key control parameters.

[0175] In the embodiments of the present invention, when the control strategy generation module executes the function of screening out the key control parameters from the operation data by using the pattern feature parameters, it is specifically used for:

[0176] Calculating the correlation score between the pattern feature parameters and the operation data;

[0177] Using the correlation score and a preset threshold to screen out the initially relevant data from the operation data;

[0178] Performing multiple correlation verifications on the initially relevant data to obtain a verification result;

[0179] Based on the verification result, screening out the key control parameters from the initially relevant data.

[0180] Specifically, align the pattern feature parameters and the operation data along the time axis, calculate the covariance between the pattern feature parameters and the operation data, the standard deviation of the pattern feature parameters, and the standard deviation of the operation data respectively, calculate the product of the standard deviation of the pattern feature parameters and the standard deviation of the operation data, and divide the covariance by the product of the standard deviations to obtain the correlation score.

[0181] Specifically, calculate the marginal probability distribution and the joint probability distribution of the pattern feature parameters and the initially relevant data, calculate the univariate entropy and the joint entropy of the pattern feature parameters and the initially relevant data respectively according to the marginal probability distribution and the joint probability distribution, calculate the mutual information entropy of the pattern feature parameters and the initially relevant data according to the univariate entropy and the joint entropy, establish a trend mapping rule between the pattern feature parameters and the initially relevant data according to the mutual information entropy, and match the initially relevant data according to the trend mapping rule and the mutual information entropy to obtain the verification result.

[0182] In the embodiments of the present invention, input the key control parameters into the dynamic simulation model to obtain a simulation result. In the embodiments of the present invention, for each simulated operation result, a series of performance indicators are calculated to evaluate the system performance under this setting. These performance indicators may include but are not limited to energy efficiency ratio, response time, stability, and energy consumption, etc.

[0183] In the embodiments of the present invention, each control parameter is applied to a dynamic simulation model, the performance indexes corresponding to each control parameter are recorded during operation, the fitness value of each control parameter is calculated according to the performance indexes, a part of the control parameters are selected as the parameters to be processed according to the fitness value, the parameters to be processed are randomly paired, new control parameter combinations are generated through crossover operations, and the control parameter combinations are used as candidate control parameters.

[0184] In the embodiments of the present invention, through the optimization and iteration of control parameters, the system can not only improve energy efficiency, but also maintain the stability of the motor system in a changing working environment. The optimized control strategy can effectively cope with the influences of different load fluctuations, temperature changes, etc., avoid the motor from malfunctioning due to overload or maladjustment, and thus improve the overall reliability of the system.

[0185] For example, in the historical operation data of the motor system, the mode characteristic parameters can help screen out the most critical control parameters, such as the motor speed, torque, current, etc. These parameters directly affect the power consumption and efficiency of the motor. By adjusting the control methods of the speed and torque, the simulation system can predict the response, energy efficiency, etc. of the motor under load changes. The simulation results are evaluated through a series of performance indexes (such as efficiency, power factor, energy consumption, etc.), that is, the performance of the motor under each group of candidate control parameters is calculated to find the best operating range. According to the calculated results of the motor energy efficiency, the current and speed control strategies in the control system are adjusted for energy-saving optimization. Then, through continuous iteration and adjustment, a candidate parameter set is generated. After several optimization iterations, a set of adjusted and verified candidate parameter sets is finally obtained. The candidate parameter sets can maximize the energy efficiency of the motor and ensure the stability and high efficiency of the motor system during actual operation.

[0186] In the embodiments of the present invention, when the control strategy generation module executes the function of screening parameters from the candidate parameter set to obtain the optimal control parameters, it is specifically used for:

[0187] Performing motor simulation operation based on the candidate parameter set and calculating the corresponding energy-saving effect;

[0188] Evaluating the performance indexes of each group of candidate control parameters on the stability, response speed and energy consumption of the motor system based on the energy-saving effect;

[0189] Scoring each performance index according to a preset weight formula to obtain a scoring result;

[0190] Selecting the candidate control parameter with the highest comprehensive score as the optimal control parameter according to the scoring result.

[0191] In the embodiments of the present invention, a group of candidate control parameters is first selected. Then, a motor simulation model is used to simulate the performance of these parameters during actual operation to obtain operation data, and the energy-saving ratio corresponding to each operation data is calculated. The energy-saving ratio is used as the energy-saving effect.

[0192] In the embodiments of the present invention, the influence of each group of candidate control parameters on the stability, response speed, and energy consumption of the motor system is evaluated. The evaluation is carried out by calculating the response speed, which refers to the reaction time of the motor simulation model to simulate the system after receiving an input signal, the electric energy consumed when performing tasks, and the results of whether different loads can work properly, so as to obtain the comprehensive performance.

[0193] In the embodiments of the present invention, the comprehensive performance is weighted and calculated to obtain a scoring result.

[0194] In the embodiments of the present invention, after obtaining the comprehensive scores of each group of candidate control parameters, they can be sorted according to the scores. The control parameter group with the highest comprehensive score will be selected as the optimal control parameter, which means that it achieves the best balance in multiple dimensions and can provide the best performance (such as energy saving, response speed, stability, etc.).

[0195] In the embodiments of the present invention, by screening the optimal control parameters, it is ensured that the motor system operates efficiently and stably under different working conditions, prolonging the service life of the motor and reducing the maintenance cost.

[0196] As Figure 4 shown, it is a schematic flowchart of a method for motor energy-saving optimization and intelligent regulation provided by an embodiment of the present invention. In the embodiments of the present invention, the method for motor energy-saving optimization and intelligent regulation includes:

[0197] S401. Obtain the operation data of the historical motor, perform pattern recognition on the operation data to obtain the pattern probability, and calculate the performance data of the operation data according to the pattern probability. Among them, the following formula is used for pattern recognition:

[0198] ;

[0199] Among them, represents the pattern probability corresponding to the operation data , is the weight of the th pattern in the preset weight matrix, is the total number of weights in the weight matrix, is the mean probability corresponding to the th pattern, is the covariance matrix of the probability corresponding to the th pattern, is the given data belongs to the The probability of a pattern;

[0200] S402. Construct a dynamic simulation model by using the collected motor system parameters;

[0201] S403. Calculate dynamic characteristics according to the performance data and the dynamic simulation model, and extract pattern feature parameters of the operation data based on the dynamic characteristics;

[0202] S404. Optimize the control parameters of the simulation model based on the pattern feature parameters to generate a candidate parameter set, and perform parameter screening on the candidate parameter set to obtain the optimal control parameters.

[0203] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology.

[0204] Among them, Artificial Intelligence (AI) is a theory, method, technology and application system that uses a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0205] In addition, obviously, the word "including" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or systems stated in the system can also be implemented by one unit or system through software or hardware. Words such as first and second are used to represent names and do not represent any specific order.

[0206] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A motor energy-saving optimization and intelligent control platform, characterized in that: The platform includes: a motor data analysis module, a motor system simulation module, a characteristic parameter extraction module, and a control strategy generation module. Specifically: The motor data analysis module is used to obtain historical motor operation data, perform pattern recognition on the operation data to obtain pattern probability, and calculate performance data of the operation data according to the pattern probability, wherein the pattern recognition is performed using the following formula: Among them, P(x) represents the mode probability corresponding to the running data x, π i is the weight of the i-th mode in the preset weight matrix, k is the total number of weights in the weight matrix, and u i is the probability mean corresponding to the ith mode, ∑i is the covariance matrix of the probability corresponding to the ith mode, is the probability that the running data x belongs to the i-th mode; A motor system simulation module is used to fit a preset electromagnetic model according to the motor system parameters to obtain a motor electromagnetic model; fit a preset mechanical model according to the motor system parameters to obtain a motor mechanical model; fit a preset control model according to the motor system parameters to obtain a motor control model; and build a simulation environment according to the motor control model, the motor mechanical model and the motor electromagnetic model to obtain a dynamic simulation model; A characteristic parameter extraction module is used to construct a state space equation according to the dynamic simulation model and the performance data; discretize the state space equation to obtain a discrete time matrix; use the performance data and the discrete time matrix to estimate matrix parameters to obtain a characteristic matrix; perform eigenvalue decomposition on the characteristic matrix to obtain dynamic characteristics, use the dynamic characteristics to segment the operating data to obtain time series data of different stages; use the time series data and the dynamic characteristics to calculate the stability index of the motor system; calculate the efficiency interval critical point of the time series data; calculate the dynamic response time of the time series data; combine the dynamic response time, the efficiency interval critical point, the stability index and the performance data corresponding to the operating data to obtain mode characteristic parameters; A control strategy generation module is used to use the mode characteristic parameters to filter out key control parameters from the operation data; use the key control parameters and the dynamic simulation model to generate simulation results; calculate the performance indicators of the simulation results; use the performance indicators to adjust the key control parameters to obtain candidate control parameters; return to the step of generating simulation results using the key control parameters and the dynamic simulation model until the number of parameter adjustments reaches a preset threshold, obtain a candidate parameter set, perform motor simulation operation based on the candidate parameter set, and calculate the corresponding energy-saving effect; evaluate the performance indicators of each group of candidate control parameters on motor system stability, response speed and energy consumption based on the energy-saving effect; score each performance indicator according to a preset weight formula to obtain a scoring result; and screen out the candidate control parameter with the highest comprehensive score as the optimal control parameter based on the scoring result.

2. A motor energy-saving optimization and intelligent control platform as claimed in claim 1, characterized in that: When executing the function of constructing the state space equation according to the dynamic simulation model and the performance data, the characteristic parameter extraction module is specifically used to: Extracting input variables and output variables of the motor system from the dynamic simulation model; Acquire the relationship between the input variable and the output variable of the motor system based on the performance data; Defining state variables of a motor system using the dynamic simulation model and the performance data; A state space equation is constructed according to the relationship between the state variable and the input variable and output variable of the motor system.

3. A motor energy-saving optimization and intelligent control platform as claimed in claim 1, characterized in that: When the characteristic parameter extraction module performs the function of discretizing the state space equation to obtain a discrete time matrix, it is specifically used to: Determine the time step of the state space equation based on the motor system parameters; Discretizing the state space equation using a time step to generate a discretized continuous time equation; A discrete time matrix is ​​calculated from the discretized continuous time equations.

4. A motor energy-saving optimization and intelligent control platform as claimed in claim 1, characterized in that: When the characteristic parameter extraction module performs the function of segmenting the operation data by using the dynamic characteristics to obtain time series data of different stages, it is specifically used to: Constructing a state evaluation function according to the dynamic characteristics; The state evaluation function is used to detect the change point of the operation data to obtain the state change probability value. The change point detection can be performed using the following formula: Where P(c) is the probability value of state change corresponding to time point c, e is the base of natural logarithm, o is the preset constant, ΔS(c) is the change of running data at time point c, S th is the preset significance threshold; Constructing a dynamic window adjustment function based on the state change probability value; The operating data is segmented using the dynamic window adjustment function and a preset initialization window to obtain time series data.

5. The motor energy-saving optimization and intelligent control platform according to claim 1, characterized in that: When the control strategy generation module performs the function of screening out key control parameters from the operation data using the mode characteristic parameters, it is specifically used to: calculating a correlation score between the pattern feature parameter and the operation data; Filtering out preliminary relevant data from the operation data using the relevance score and a preset threshold; Performing multiple correlation verifications on the preliminary correlation data to obtain verification results; Based on the verification results, key control parameters are screened out from the preliminary relevant data.

Images