An online remote real-time status monitoring method for an exhaust fan unit

Abstract

The present invention relates to the technical field of unit status monitoring, and specifically discloses an online remote real-time status monitoring method for an exhaust fan unit, including: obtaining real-time wind pressure data of the exhaust fan unit, calculating the single-unit deviation rate and the deviation change rate as the wind pressure imbalance characteristics, identifying the wind-grabbing and wind-losing units based on a preset determination rule, and analyzing the linkage relationship between the two; if the linkage is confirmed, inputting the real-time characteristics and historical fault characteristics into a Gaussian mixture model, calculating the wind pressure imbalance probability and dividing the risk level; further constructing a steering amplitude matrix by the sliding window method, calculating the steering distribution entropy to determine whether intervention and regulation are required; for the cases that need intervention, screening the priority regulation units, and realizing the regulation of the motor speed in combination with the average steering amplitude. The present invention can effectively monitor the wind pressure balance state of the exhaust fan unit, timely identify the risks of wind-grabbing and wind-losing and intervene, which is beneficial to ensuring the coordinated and efficient operation of the parallel units.

Description

Technical Field

[0001] The present invention relates to the technical field of unit status monitoring, and particularly relates to an online remote real-time status monitoring method for an exhaust fan unit. Background Art

[0002] In large-scale scenarios such as industrial ventilation, tunnel ventilation, and mine exhaust, exhaust fan units often operate in parallel to meet the ventilation requirements of large flow rates and high reliability. However, affected by factors such as uneven distribution of pipe network resistance, differences in fan equipment characteristics, and dynamic changes in working conditions, parallel units are prone to the "wind robbing" phenomenon - some fans experience overloading due to abnormal increases in wind pressure, while other fans operate inefficiently due to passively reduced wind pressure, forming a vicious cycle of "one strong and one weak". This imbalance not only causes energy waste but also may pose risks such as aging of the motor winding insulation, increased equipment vibration, and decreased system stability.

[0003] Existing methods for monitoring the status of exhaust fan units mostly rely on single indicators and are difficult to capture the dynamic characteristics of wind pressure imbalance and the linkage effects between units; the existing technology lacks systematic quantitative analysis means, resulting in lagging risk assessment and crude control strategies, and it is difficult to meet the control requirements under complex working conditions.

[0004] Therefore, the present invention provides an online remote real-time status monitoring method for an exhaust fan unit. Summary of the Invention

[0005] The purpose of the present invention is to provide an online remote real-time status monitoring method for an exhaust fan unit to solve the above background problems.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] An online remote real-time status monitoring method for an exhaust fan unit includes the following steps:

[0008] Obtain the real-time output wind pressure data of the exhaust fan unit and extract the wind pressure imbalance characteristics of the exhaust fan unit;

[0009] Based on the wind pressure imbalance characteristics, judge the wind pressure balance state of the exhaust fan unit, identify the wind-robbing and underperforming units, conduct a linkage analysis on the wind pressure imbalance characteristics of the wind-robbing and underperforming units, and determine the linkage relationship between the wind-robbing and underperforming units;

[0010] If the linkage relationship shows linkage, construct a fan detection model through the wind pressure imbalance characteristics corresponding to the historical faults of the exhaust fan unit, use the real-time wind pressure imbalance characteristics as the input of the fan detection model, output the wind pressure imbalance probability, and divide the unit imbalance level according to the wind pressure imbalance probability;

[0011] Based on the determined imbalance level, according to the constructed wind pressure deviation sequence, extract the wind pressure imbalance characteristics of the robbing fan units and the losing fan units within the sliding window, construct a steering amplitude matrix and calculate the steering distribution entropy to determine whether intervention control is required for the robbing fan units and the losing fan units;

[0012] If necessary, screen the average steering amplitudes of different robbing or losing fan units in the steering amplitude matrix to obtain the priority regulation units, and construct a self-regulation model to perform intervention control on the priority regulation units.

[0013] As a further technical solution of the present invention: the method for extracting the wind pressure imbalance characteristics of the exhaust fan units is as follows:

[0014] Obtain the wind pressure of the fan units when all single exhaust fan units are operating in parallel, calculate the deviation ratio of the wind pressure of a single exhaust fan unit to the average wind pressure of all exhaust fan units to obtain the single-unit deviation rate;

[0015] Set multiple time sampling points, extract the single-unit deviation rates of multiple sampling points, and perform linear fitting analysis on the single-unit deviation rates of multiple sampling points to obtain the deviation change rate;

[0016] Extract the single-unit deviation rates and deviation change rates of all exhaust fan units as the wind pressure imbalance characteristics.

[0017] As a further technical solution of the present invention: the method for determining the linkage relationship between the robbing fan units and the losing fan units is as follows:

[0018] Based on the single-unit deviation rate and the deviation change rate, combine the determination rules to judge the wind pressure balance state of the exhaust fan units, and identify the losing fan units and the robbing fan units;

[0019] Perform linkage analysis on the losing fan units and the robbing fan units. If the correlation coefficients of the deviation change rates of the robbing fan units and the losing fan units, and the correlation coefficients of the single-unit deviation rates of the robbing fan units and the losing fan units are all negatively correlated, that is, the robbing fan units and the losing fan units show a linkage relationship.

[0020] As a further technical solution of the present invention: the method for obtaining the correlation coefficients of the deviation change rates of the robbing fan units and the losing fan units, and the correlation coefficients of the single-unit deviation rates of the robbing fan units and the losing fan units is as follows:

[0021] Calculate the Pearson correlation coefficients of the deviation change rates of the robbing fan units and the losing fan units, and the Pearson correlation coefficients of the single-unit deviation rates respectively, to obtain the correlation coefficients of the deviation change rates of the robbing fan units and the losing fan units and the correlation coefficients of the single-unit deviation rates.

[0022] As a further technical solution of the present invention: the method for obtaining the wind pressure imbalance probability is as follows:

[0023] Obtain the corresponding air pressure imbalance characteristics when the exhaust fan unit has historical failures, construct a feature dataset and perform normalization processing to obtain a standard feature dataset;

[0024] Through the Gaussian mixture model algorithm, use the standard feature dataset to construct and train a fan monitoring model;

[0025] Input the real-time air pressure imbalance characteristics of the robbing fan units and the losing fan units into the fan monitoring model to obtain the air pressure imbalance probability.

[0026] As a further technical solution of the present invention: the method for determining whether it is necessary to perform intervention control on the robbing fan units and the losing fan units is:

[0027] Extract the single-unit deviation rate corresponding to the air pressure imbalance characteristics of multiple sampling points within the sliding window, and construct an air pressure deviation sequence;

[0028] Perform a turning calculation on the air pressure deviation sequence to obtain the average turning amplitude of the single-unit deviation rate;

[0029] Obtain the average turning amplitude of all robbing fan units and losing fan units within all sliding windows, and establish a turning amplitude matrix;

[0030] By calculating the turning distribution entropy of the turning amplitude matrix, compare the turning distribution entropy with a preset turning distribution threshold to determine whether it is necessary to perform intervention control on the robbing fan units and the losing fan units.

[0031] As a further technical solution of the present invention: the method for performing a turning calculation on the air pressure deviation sequence is:

[0032] Perform a turning frequency analysis on the air pressure deviation sequence to obtain the turning frequency of the single-unit deviation rate;

[0033] Perform an amplitude calculation on the turning frequency of the single-unit deviation rate to obtain the average turning amplitude of the single-unit deviation rate.

[0034] As a further technical solution of the present invention: the method for calculating the turning distribution entropy of the turning amplitude matrix is:

[0035] Perform normalization processing on the turning amplitude matrix to obtain a standard turning amplitude matrix;

[0036] Perform an entropy value calculation on the standard turning amplitude matrix to obtain the turning distribution entropy.

[0037] As a further technical solution of the present invention: the method for the self-regulation model to perform intervention control on the priority regulation units is:

[0038] Arrange the maximum values of the average turning amplitudes of the robbing fan units or the losing fan units in descending order, and based on the sorting result, determine the priority regulation units;

[0039] Construct a self-regulation model through the proportional-integral-derivative algorithm, and correct the proportional coefficient of the proportional-integral-derivative algorithm in combination with the average turning amplitude;

[0040] Calculate the regulation amount of the motor speed through the self-regulation model, and regulate the motor speed of the priority-regulated unit.

[0041] As a further technical solution of the present invention: the method for correcting the proportional coefficient of the proportional-integral-derivative algorithm is:

[0042] Obtain the maximum value of the average turning amplitude of the priority-regulated unit as the input parameter of the self-tuning model;

[0043] Take the input parameter as the feed-forward compensation factor of the proportional-integral-derivative algorithm to correct the parameters of the proportional-integral-derivative algorithm.

[0044] The beneficial effects of the present invention:

[0045] (1) By collecting the wind pressure of a single unit in real time and calculating the single-unit deviation rate and deviation change rate, a wind pressure imbalance characteristic system including spatio-temporal dimensions is constructed from two dimensions of static and dynamic; it can capture the wind pressure anomaly at a certain moment and identify the change trend of the deviation degree, providing multi-dimensional data support for subsequent identification of wind stealing and wind loss, avoiding the one-sidedness of a single index; based on the preset positive and negative deviation ranges and dynamic change trends, the negative correlation between the single-unit deviation rate and change rate of the wind stealing unit and the wind loss unit is analyzed through the Pearson correlation coefficient, determining the "one strong and one weak" linkage relationship, clearly distinguishing potential wind stealing units and wind loss units, and at the same time revealing the coupling effect between parallel units through correlation analysis, providing a basis for system-level risk assessment.

[0046] (2) Use the Gaussian mixture model to model the historical fault characteristics, input the real-time data into the model to calculate the probability of wind pressure imbalance, and divide the low, medium, and high risk levels; quantify the risk based on the statistical laws of historical data to achieve a gradient assessment of the unit imbalance state; construct a turning amplitude matrix through the sliding window method, calculate the turning distribution entropy to quantify the regulation difference between units, the higher the entropy value, the more dispersed the fluctuation and the more complex the regulation, convert the multi-unit dynamic fluctuation characteristics into a single entropy value index, and automatically judge whether intervention is needed in combination with a preset threshold, which is beneficial to improving the pertinence of the regulation strategy.

[0047] (3)Screen high-fluctuation units as the priority control objects, adjust the proportional coefficient by combining the PID algorithm with the average steering amplitude to dynamically adjust the motor speed to balance the wind pressure; use the fluctuation intensity reflected by the steering amplitude to optimize the control parameters in real time, which is beneficial to suppressing the risk of motor overload caused by wind grabbing and extending the equipment life; form a closed-loop of the whole process from feature extraction to intervention control, and establish an intelligent monitoring and operation and maintenance system of "monitoring - analysis - decision - execution". Through multi-parameter coupling analysis and hierarchical control strategies, it is beneficial to ensure the balance of the unit wind pressure, improve the operation efficiency of the parallel system, and reduce the risk of chain failures caused by wind pressure imbalance, especially applicable to scenarios with high requirements for fan reliability such as industrial ventilation and tunnel ventilation. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention will be further described below with reference to the accompanying drawings.

[0049] Figure 1 is a flowchart of an online remote real-time status monitoring method for an exhaust fan unit of the present invention;

[0050] Figure 2 is a flowchart of an extraction method for the characteristics of wind pressure imbalance in the present invention;

[0051] Figure 3 is a block diagram of an online remote real-time status monitoring system for an exhaust fan unit. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment 1 Please refer to Figure 1 As shown, the present invention is an online remote real-time status monitoring method for an exhaust fan unit, including the following steps:

[0054] In scenarios such as industrial ventilation and tunnel ventilation, exhaust fan units often operate in a parallel network to meet the large-flow ventilation requirements. However, when exhaust fan units operate in parallel, the phenomenon of "wind grabbing" of the exhaust fan units is likely to occur due to problems such as uneven distribution of pipe network resistance and differences in equipment characteristics, that is, some units have too high a load while other units operate inefficiently, which not only causes energy waste but also may lead to risks such as motor overload and shortened equipment life. Therefore, monitoring the real-time output wind pressure of parallel units and evaluating its balance is the key entry point for ensuring the efficient operation of multi-machine collaboration and avoiding system-level failures.

[0055] S1. When the exhaust fan units are operating in parallel, obtain the real-time output air pressure data of the exhaust fan units, calculate the single-unit deviation rate of the average air pressure of a single unit and all units based on the air pressure data, and extract the air pressure imbalance characteristics of the exhaust fan units;

[0056] Among them, the method for obtaining the single-unit deviation rate is:

[0057] In some preferred embodiments, install air pressure sensors on the exhaust fan units to collect the air pressure of a single exhaust fan unit in real time;

[0058] Obtain the air pressure of the fan units when all single exhaust fan units are operating in parallel, and calculate the average air pressure value of all exhaust fan units;

[0059] Calculate the deviation ratio of the air pressure of a single exhaust fan unit to the average air pressure value of all exhaust fan units to obtain the single-unit deviation rate;

[0060] As Figure 2 shown, the method for extracting the air pressure imbalance characteristics is:

[0061] S101. Set multiple time sampling points, extract the single-unit deviation rates of multiple sampling points, perform linear fitting analysis on the single-unit deviation rates of multiple sampling points, and obtain the deviation change rate;

[0062] Those skilled in the art can understand that perform linear fitting analysis on the single-unit deviation rates of multiple sampling points, obtain the fitting equation and calculate the slope of the fitting equation as the deviation change rate;

[0063] The single-unit deviation rate reflects the deviation degree of the air pressure of a single fan at a certain moment from the average air pressure of all units, which belongs to a static index;

[0064] The deviation change rate reflects the change of this deviation degree over time, and identifies whether the air pressure imbalance is intensifying, alleviating or remaining stable;

[0065] S102. Extract the single-unit deviation rate and deviation change rate of all exhaust fan units as the air pressure imbalance characteristics;

[0066] Transmit the collected air pressure imbalance characteristics remotely to the fan remote control unit.

[0067] It can be understood that by extracting the air pressure imbalance characteristics, analyzing the single-unit deviation rate and deviation change rate, and based on the preset positive and negative deviation ranges and judgment rules, the air pressure balance state of the exhaust fan units can be judged, and the robbing fan units and losing fan units can be identified.

[0068] S2. Based on the air pressure imbalance characteristics, judge the air pressure balance state of the exhaust fan units, identify the robbing and losing fan units, perform linkage analysis on the air pressure imbalance characteristics of the robbing and losing fan units, and determine the linkage relationship between the robbing and losing fan units;

[0069] Extract the air pressure imbalance characteristics of all exhaust fan groups from the remote control unit of the fan, identify the air grabbing behavior of a single fan. If a single fan exhibits air grabbing behavior, then the single fan is regarded as the air-grabbing fan group.

[0070] It can be understood that when air grabbing occurs in the exhaust fan group, due to the sudden increase in the air pressure of a single fan, it may cause the motor current to exceed the rated value. Long-term overload will accelerate the aging of the motor winding insulation and even burn out the motor.

[0071] If a single fan in a parallel-connected fan group grabs air, it will cause the air pressure of other fans to decrease passively, that is, the phenomenon of air loss occurs, forming a vicious cycle of "one strong and one weak", and destroying the air pressure balance of the exhaust fan group.

[0072] Among them, the air pressure imbalance characteristics are used to judge the air pressure balance state of the exhaust fan group. The methods for identifying the air-grabbing and air-loss fan groups are as follows:

[0073] Extract the single-machine deviation rate and deviation change rate of the air pressure imbalance characteristics, and judge the air pressure balance state of the exhaust fan group based on the single-machine deviation rate and deviation change rate in combination with the judgment rules, and identify the air-loss fan group and the air-grabbing fan group.

[0074] Among them, the judgment rules are as follows:

[0075] Judgment rule one: Obtain the preset positive and negative deviation ranges. If the single-machine deviation rate of a single fan is positive and the single-machine deviation rate deviates from the positive deviation range, that is, the air pressure of the single fan deviates from the balanced state, then the single unit is a potential air-grabbing fan group.

[0076] Among them, the positive and negative deviation ranges include the positive deviation range and the negative deviation range.

[0077] If the single-machine deviation rate of a single fan is negative and the single-machine deviation rate deviates from the negative deviation range, that is, the air pressure of the single fan deviates from the balanced state, then the single unit is a potential air-loss fan group.

[0078] Judgment rule two: If it is a potential air-grabbing fan group and the deviation change rate is positive at the same time, that is, the potential air-grabbing fan group is an air-grabbing fan group.

[0079] If it is a potential air-loss fan group and the deviation change rate is negative at the same time, that is, the potential air-grabbing fan group is an air-loss fan group.

[0080] It should be explained that: the preset positive and negative deviation ranges are set by those skilled in the art according to the historical operating state of the exhaust fan group. Deviating from the positive and negative deviation ranges means that potential air loss or air grabbing phenomena occur.

[0081] Among them, the method for performing a linkage analysis on the air pressure imbalance characteristics of the air-grabbing and air-loss fan groups is as follows:

[0082] Obtain the deviation change rate and single-unit deviation rate of the wind-grabbing fan units and wind-losing fan units, perform a correlation calculation on the deviation change rate and single-unit deviation rate of the wind-grabbing fan units and wind-losing fan units, and obtain the correlation coefficient of the deviation change rate of the wind-grabbing fan units and wind-losing fan units, as well as the correlation coefficient of the single-unit deviation rate of the wind-grabbing fan units and wind-losing fan units;

[0083] The method of performing the correlation calculation is as follows:

[0084] Preferably, by calculating the Pearson correlation coefficient of the single-unit deviation rate of the wind-grabbing fan units and wind-losing fan units, as the correlation coefficient of the deviation change rate of the wind-grabbing fan units and wind-losing fan units;

[0085] Calculate the Pearson correlation coefficient of the deviation change rate of the wind-grabbing fan units and wind-losing fan units, as the correlation coefficient of the single-unit deviation rate of the wind-grabbing fan units and wind-losing fan units;

[0086] If the correlation coefficient of the deviation change rate of the wind-grabbing fan units and wind-losing fan units, as well as the correlation coefficient of the single-unit deviation rate of the wind-grabbing fan units and wind-losing fan units, both show a negative correlation, that is, the wind-grabbing fan units and wind-losing fan units show a linkage relationship;

[0087] It should be explained that if the correlation coefficient of the deviation change rate of the wind-grabbing fan units and wind-losing fan units and the correlation coefficient of the single-unit deviation rate both show a negative correlation, it indicates that the two show a linkage relationship;

[0088] Among them, when calculating the linkage relationship between the wind-grabbing fan units and wind-losing fan units, it is necessary to correspond the wind-grabbing fan units and wind-losing fan units one by one, that is, calculate the correlation of specific two fan units, rather than the overall correlation of all exhaust fan units;

[0089] At the operation level of the exhaust fan unit system: when the exhaust fan unit is operating normally, the air pressure between each fan is in a relatively balanced state; when the wind-grabbing fan units and wind-losing fan units show a linkage relationship, it means that this balance has been broken, the air pressure of the wind-grabbing fan units increases, occupying more system air volume and air pressure resources, causing the air pressure of the wind-losing fan units to decrease accordingly, reflecting the imbalance of the air pressure of the exhaust fan unit system;

[0090] A high negative correlation reflects the connection between wind-grabbing and wind-losing, and the wind-grabbing fan units and wind-losing fan units form a "one strong and one weak" vicious cycle; it further shows that in a parallel fan system, the operating states of each fan affect each other, and the abnormal operation of one fan will have a chain reaction on other fans;

[0091] The technical solution of this embodiment is: obtain the real-time output air pressure data of the exhaust fan unit, extract the air pressure imbalance characteristics of the exhaust fan unit; based on the air pressure imbalance characteristics, judge the air pressure balance state of the exhaust fan unit, identify the wind-grabbing and wind-losing fan units, perform a linkage analysis on the air pressure imbalance characteristics of the wind-grabbing and wind-losing fan units, determine the linkage relationship between the wind-grabbing and wind-losing fan units, reveal the coupling influence between parallel units, and provide a basis for system-level risk assessment.

[0092] Example 2

[0093] As Figure 1 shown, an on-line remote real-time status monitoring method for an exhaust fan unit further includes the following steps:

[0094] S3. If the linkage relationship shows linkage, construct a fan detection model based on the air pressure imbalance characteristics corresponding to the historical faults of the exhaust fan unit, use the real-time air pressure imbalance characteristics as the input of the fan detection model, output the air pressure imbalance probability, and divide the unit imbalance level according to the air pressure imbalance probability;

[0095] If the linkage relationship shows linkage, obtain the air pressure imbalance characteristics corresponding to the historical faults of the exhaust fan unit and construct a feature data set ;

[0096] Exemplarily, represents the first feature vector composed of the air pressure imbalance characteristics corresponding to the historical faults of the exhaust fan unit, and n is the total number of feature vectors;

[0097] Normalize the feature data set to obtain a standard feature data set ;

[0098] Construct and train a fan monitoring model using the standard feature data set through the Gaussian mixture model algorithm;

[0099] Among them, the Gaussian model probability density function is: ;

[0100] Among them, are the model parameters of the Gaussian mixture model. K represents the total number of Gaussian distributions in the mixture components, that is, the Gaussian mixture model algorithm is composed of K Gaussian distributions mixed together, and k is the number of the Gaussian distribution;

[0101] represents the weight ratio of the k-th Gaussian distribution in the Gaussian mixture model algorithm;

[0102] is the mean vector of the k-th Gaussian distribution, which is used to describe the central position of the Gaussian distribution;

[0103] is the covariance matrix of the k-th Gaussian distribution, which is used to describe the dispersion degree and correlation of the data in each dimension;

[0104] is the Gaussian distribution probability density function, is the feature vector of the standard feature data set, and N is the general representation symbol of the Gaussian distribution probability density function;

[0105] Input the real-time wind pressure imbalance characteristics of the boosting fan unit and the losing fan unit into the fan monitoring model to obtain the wind pressure imbalance probability;

[0106] Through the formula: Obtain the wind pressure imbalance probability p, where are the real-time wind pressure imbalance characteristics of the boosting fan unit and the losing fan unit;

[0107] It should be noted that before inputting into the fan monitoring model, the real-time wind pressure imbalance characteristics of the boosting fan unit and the losing fan unit need to be normalized;

[0108] Divide the unit risk level according to the numerical interval of the wind pressure imbalance probability p;

[0109] Exemplarily, divide the unit risk level into low risk, medium risk, and high risk;

[0110] When is low risk, p1 is the first-level risk probability threshold, reflecting that the difference between the real-time wind pressure imbalance characteristics and the historical failure mode is significant, and the probability of unit imbalance is low;

[0111] When is medium risk, p2 is the second-level risk probability threshold, reflecting that the characteristics partially match the failure mode, and it is necessary to closely monitor the real-time wind pressure imbalance characteristics;

[0112] When is high risk, reflecting that the real-time wind pressure imbalance characteristics highly match the historical failure mode characteristics, and the probability of unit imbalance is high;

[0113] It should be noted that the first-level risk probability threshold and the second-level risk probability threshold are obtained by those skilled in the art based on the historical working data of the exhaust fan unit;

[0114] S4. Based on the determined imbalance level, according to the constructed wind pressure deviation sequence, extract the wind pressure imbalance characteristics of the boosting fan unit and the losing fan unit within the sliding window, construct a turning amplitude matrix and calculate the turning distribution entropy to determine whether it is necessary to perform intervention control on the boosting fan unit and the losing fan unit;

[0115] If there is an exhaust fan unit with a high risk level in the exhaust fan unit, through the sliding window method, obtain the wind pressure imbalance characteristics of the boosting fan unit and the losing fan unit at multiple sampling points within the sliding window;

[0116] It can be understood that the operating state of the exhaust fan unit changes dynamically with time, and the wind pressure imbalance characteristics will also change accordingly. The sliding window can capture the changes of these characteristics in different time intervals and reflect the dynamic evolution process of the wind pressure imbalance;

[0117] For example, the window size can be set to 10 sampling points; the sliding step is the number of sampling points that the window moves each time. Assuming the sliding step is 2 sampling points, starting from the starting position of the wind pressure data sequence, select consecutive sampling points with a quantity equal to the window size to form the first sliding window. For example, for the wind pressure data in order: according to the set sliding step, move the window backward in turn. When the sliding step is 2, the second window contains the 3rd to 12th sampling points (that is, the first window moves backward by 2 sampling points), and the third window contains the 5th to 14th sampling points, and so on until the entire wind pressure data sequence is traversed. Arranged in the order of sequence, the first window contains the 1st to 10th sampling points;

[0118] Extract the single - unit deviation rate corresponding to the wind pressure imbalance characteristics of multiple sampling points within the sliding window, and construct a wind pressure deviation sequence , the value range of i is [1, m];

[0119] where i represents the number of the sampling point, m is the total number of sampling points, and y represents the single - unit deviation rate;

[0120] Through the formula: Obtain the turning frequency of the single - unit deviation rate ;

[0121] Those skilled in the art can understand that "#" represents the counting under the condition that For the i - th point x in the sequence i , if the change direction before and after x i is opposite, that is, from "rising to falling" or "falling to rising", then x i is called a turning point;

[0122] Exemplarily, if there are 5 positions in the wind pressure deviation sequence that meet , then the turning frequency = 5;

[0123] Through the formula: Obtain the average turning amplitude A of the single - unit deviation rate;

[0124] It should be explained that the number of turns from "rising to falling" or "falling to rising" within the wind pressure deviation sequence is the turning frequency of the single - unit deviation rate, and the average value of the absolute amplitude of each turn; by calculating the turning frequency and average turning amplitude of the single - unit deviation rate, the fluctuation degree of the wind pressure deviation sequence can be quantified;

[0125] Obtain the average turning amplitude of all scavenging fan groups and losing fan groups within all sliding windows, and establish a turning amplitude matrix M a,b ;

[0126] It can be understood that M a,bDenote the average turning amplitude of the \(a\)th robbing fan unit or losing fan unit within the \(b\)th sliding window, where \(a\) is the number of the robbing fan unit or losing fan unit, and \(b\) is the number of the sliding window;

[0127] Quantify the distribution complexity of the turning amplitudes of the robbing fan units and losing fan units by calculating the turning distribution entropy of the turning amplitude matrix;

[0128] Among them, the calculation method of the turning distribution entropy is as follows:

[0129] Perform normalization processing on the turning amplitude matrix to obtain the standard turning amplitude matrix ;

[0130] Through the formula: Obtain the turning distribution entropy ;

[0131] Among them, \(Z_a\) is the total number of robbing fan units and losing fan units in the turning amplitude matrix, and \(Z_b\) is the total number of sliding windows;

[0132] It can be understood that the numerical value of the turning distribution entropy reflects the distribution complexity of the turning amplitudes of the robbing fan units and losing fan units;

[0133] If = 0, the turning amplitudes of all robbing fan units and losing fan units within all windows are 0, indicating that there is no fluctuation in the turning amplitudes of all robbing fan units and losing fan units within all windows, and the regulation is the simplest;

[0134] If the larger the numerical value, the more dispersed the turning amplitude distribution, the greater the regulation difference between different units, and the higher the regulation complexity of the robbing fan units and losing fan units;

[0135] Compare the turning distribution entropy with a preset turning distribution threshold. If the turning distribution entropy is higher than or equal to the preset turning distribution threshold, it is necessary to intervene and regulate the air pressure of the exhaust fan units;

[0136] If the turning distribution entropy is lower than the preset turning distribution threshold, it indicates that the current regulation difference between different units is within the expected range, and the regulation can be corrected through the preset program of the exhaust fan units.

[0137] S5. If necessary, screen the average turning amplitudes of different robbing or losing fan units in the turning amplitude matrix to obtain the priority regulation units, and construct a self-regulation model to intervene and control the priority regulation units.

[0138] If necessary, extract the maximum value of the average turning amplitudes of the robbing fan units or losing fan units in the turning amplitude matrix;

[0139] Sort the maximum values of the average turning amplitudes of the scavenging fan units or the lost fan units in descending order, and determine the priority control units based on the sorting results;

[0140] Construct a self-regulation model through the proportional-integral-derivative (PID) algorithm, and correct the proportional coefficient of the PID algorithm in combination with the average turning amplitude. Calculate the regulation amount of the motor speed through the self-regulation model;

[0141] Regulate the air pressure of the priority control units based on the regulation amount of the motor speed;

[0142] Those skilled in the art can understand that the air pressure of the priority control units is mainly controlled by the speed of the motor;

[0143] Among them, the output of the standard PID controller is: ;

[0144] is the regulation amount of the motor speed of the priority control unit, and t represents time;

[0145] are the proportional gain coefficient, integral gain coefficient, and derivative gain coefficient respectively, which are used for the intensity of the proportional control action, the intensity of the integral control action, and the intensity of the derivative control action, set by those skilled in the art;

[0146] e(t) is the system error signal, which represents the difference between the motor speed set value and the actual output value at the current time;

[0147] represents the derivative of the error signal with respect to time, reflecting the change rate of the error;

[0148] represents the integral of the error signal in the time range [0, t], reflecting the cumulative effect of the integral link on the error, is a function of the variable is the integration variable;

[0149] Among them, the method of correcting the proportional coefficient of the PID algorithm in combination with the average turning amplitude is as follows:

[0150] Obtain the maximum value of the average turning amplitude of the priority control unit as the input parameter M of the self-tuning model max-avg ;

[0151] Take the input parameter M max-avg as the feedforward compensation factor of the proportional-integral-derivative (PID) algorithm to correct the PID parameters;

[0152] Through the formula: Obtain the proportional correction factor of the proportional-integral-derivative algorithm , where M ref is the reference value of the preset steering amplitude, and α is the preset proportional coefficient;

[0153] It should be noted that during the operation of the exhaust fan group, the maximum average steering amplitude M in the steering amplitude matrix max-avg can reflect the fluctuation degree of the air pressure imbalance characteristics;

[0154] When M max-avg数值 is relatively large, it means that the air pressure fluctuation is relatively intense. At this time, the PID controller needs to adjust the system more quickly and effectively to restore the air pressure balance. By incorporating M max-avg into the correction formula of the proportional coefficient, the proportional coefficient can be adjusted according to the actual situation of the air pressure fluctuation;

[0155] When the air pressure fluctuation intensifies (i.e., M max-avg is relatively large), the proportional correction factor will increase accordingly, thereby enhancing the proportional control effect of the PID controller and enabling the system to respond more quickly to the air pressure change to cope with complex operating conditions;

[0156] The preset steering amplitude reference value M ref provides a relative measurement standard for the adjustment of the proportional coefficient. It represents a reference level of the air pressure fluctuation under normal or expected operating conditions and is determined by those skilled in the art based on the historical operating condition data of the exhaust fan group;

[0157] Substitute the corrected proportional coefficient into the output formula of the standard PID controller to obtain the control quantity for preferentially regulating the motor speed of the unit;

[0158] Based on the control quantity for preferentially regulating the motor speed of the unit, the motor speed is regulated through the motor control unit of the exhaust fan group.

[0159] The technical solution of this embodiment is as follows: If the linkage relationship is shown as linkage, a fan detection model is constructed based on the wind pressure imbalance characteristics corresponding to the historical faults of the exhaust fan group. The real-time wind pressure imbalance characteristics are used as the input of the fan detection model to output the wind pressure imbalance probability, and the unit imbalance level is divided according to the wind pressure imbalance probability. Based on the determined imbalance level, according to the constructed wind pressure deviation sequence, the wind pressure imbalance characteristics of the robbing fans and losing fans within the sliding window are extracted, a steering amplitude matrix is constructed and the steering distribution entropy is calculated to determine whether intervention control is required for the robbing fans and losing fans. If necessary, the average steering amplitudes of different robbing or losing fans in the steering amplitude matrix are screened to obtain the priority regulation units, and a self-regulation model is constructed to perform intervention control on the priority regulation units. Through multi-parameter coupling analysis and hierarchical control strategies, it is beneficial to ensure the wind pressure balance of the unit, improve the operation efficiency of the parallel system, and reduce the risk of chain faults caused by wind pressure imbalance.

[0160] Embodiment III

[0161] As Figure 3 shown, an on-line remote real-time status monitoring system for an exhaust fan group includes the following modules:

[0162] Feature extraction module: When the exhaust fan groups are operating in parallel, it is used to obtain the real-time output wind pressure data of the exhaust fan groups, calculate the single-unit deviation rate of a single unit from the average wind pressure of all units according to the wind pressure data, and extract the wind pressure imbalance characteristics of the exhaust fan groups;

[0163] Among them, the method for obtaining the single-unit deviation rate is as follows:

[0164] In some preferred embodiments, wind pressure sensors are installed on the exhaust fan groups to collect the wind pressure of a single exhaust fan group in real time;

[0165] Obtain the wind pressure of the fan groups when all single exhaust fan groups are operating in parallel, and calculate the average wind pressure value of all exhaust fan groups;

[0166] Calculate the deviation ratio of the wind pressure of a single exhaust fan group to the average wind pressure of all exhaust fan groups to obtain the single-unit deviation rate;

[0167] Among them, the method for extracting the wind pressure imbalance characteristics is as follows:

[0168] Set multiple time sampling points, extract the single-unit deviation rates of multiple sampling points, and perform linear fitting analysis on the single-unit deviation rates of multiple sampling points to obtain the deviation change rate;

[0169] The single-unit deviation rate reflects the degree of deviation of the wind pressure of a single fan at a certain moment from the average wind pressure of all units, and belongs to a static index;

[0170] The deviation change rate reflects the change of this deviation degree over time, identifying whether the air pressure imbalance is intensifying, alleviating or remaining stable;

[0171] Extract the single - unit deviation rate and deviation change rate of all exhaust fan units as the air pressure imbalance characteristics;

[0172] Transmit the collected air pressure imbalance characteristics remotely to the fan remote control unit.

[0173] Linkage discrimination module: Based on the air pressure imbalance characteristics, judge the air pressure balance state of the exhaust fan unit, used to identify the air - robbing and air - losing fan units, conduct a linkage analysis on the air pressure imbalance characteristics of the air - robbing and air - losing fan units, and determine the linkage relationship between the air - robbing and air - losing fan units;

[0174] Extract the air pressure imbalance characteristics of all exhaust fan units from the fan remote control unit, identify the air - robbing behavior of a single fan. If a single fan has air - robbing behavior, then the single fan is regarded as an air - robbing fan unit;

[0175] Among them, the judgment method of the air pressure balance state of the exhaust fan unit is as follows:

[0176] Extract the single - unit deviation rate and deviation change rate of the air pressure imbalance characteristics, and judge the air pressure balance state of the exhaust fan unit based on the single - unit deviation rate and deviation change rate combined with the judgment rules, identifying the air - losing fan units and air - robbing fan units;

[0177] Among them, the judgment rules are as follows:

[0178] Judgment rule one: Obtain the preset positive and negative deviation ranges. If the single - unit deviation rate of a single fan is positive and the single - unit deviation rate deviates from the positive deviation range, that is, the air pressure of the single fan deviates from the balanced state, then the single unit is a potential air - robbing fan unit;

[0179] Among them, the positive and negative deviation ranges include the positive deviation range and the negative deviation range;

[0180] If the single - unit deviation rate of a single fan is negative and the single - unit deviation rate deviates from the negative deviation range, that is, the air pressure of the single fan deviates from the balanced state, then the single unit is a potential air - losing fan unit;

[0181] Judgment rule two: If it is a potential air - robbing fan unit and the deviation change rate is positive at the same time, that is, the potential air - robbing fan unit is an air - robbing fan unit;

[0182] If it is a potential air - losing fan unit and the deviation change rate is negative at the same time, that is, the potential air - robbing fan unit is an air - losing fan unit;

[0183] Among them, the method of conducting a linkage analysis on the air pressure imbalance characteristics of the air - robbing and air - losing fan units is as follows:

[0184] Obtain the deviation change rate and single - unit deviation rate of the competing fan units and the failed fan units, perform a correlation calculation on the deviation change rate and single - unit deviation rate of the competing fan units and the failed fan units, and obtain the correlation coefficient of the deviation change rate of the competing fan units and the failed fan units, as well as the correlation coefficient of the single - unit deviation rate of the competing fan units and the failed fan units;

[0185] The method of performing the correlation calculation is as follows:

[0186] Preferably, calculate the Pearson correlation coefficient of the single - unit deviation rate of the competing fan units and the failed fan units, and calculate the Pearson correlation coefficient of the deviation change rate of the competing fan units and the failed fan units;

[0187] If the correlation coefficient of the deviation change rate of the competing fan units and the failed fan units, and the correlation coefficient of the single - unit deviation rate of the competing fan units and the failed fan units both show negative correlation, that is, the competing fan units and the failed fan units show a linkage relationship.

[0188] Level - division module: If the linkage relationship shows linkage, it is used to construct a fan detection model through the wind - pressure imbalance characteristics corresponding to the historical faults of the exhaust fan units, take the real - time wind - pressure imbalance characteristics as the input of the fan detection model, output the wind - pressure imbalance probability, and divide the unit imbalance level according to the wind - pressure imbalance probability;

[0189] If the linkage relationship shows linkage, obtain the wind - pressure imbalance characteristics corresponding to the historical faults of the exhaust fan units and construct a feature data set ;

[0190] Take the feature data set Perform normalization processing to obtain a standard feature data set ;

[0191] Construct a fan monitoring model through the Gaussian mixture model algorithm and calculate the wind - pressure imbalance probability;

[0192] Among them, the Gaussian model probability density function is: ;

[0193] Among them, are the model parameters of the Gaussian mixture model, K represents the total number of Gaussian distributions in the mixture components, that is, the Gaussian mixture model algorithm is composed of K Gaussian distributions mixed together, and k is the number of the Gaussian distribution; represents the weight ratio of the k - th Gaussian distribution in the Gaussian mixture model algorithm;

[0194] is the mean vector of the k - th Gaussian distribution, which is used to describe the central position of the Gaussian distribution;

[0195] is the covariance matrix of the k - th Gaussian distribution, which is used to describe the degree of dispersion and correlation of the data in each dimension;

[0196] is the probability density function of the Gaussian distribution;

[0197] Input the real-time wind pressure imbalance characteristics of the boosting fan units and the losing fan units into the fan monitoring model to obtain the wind pressure imbalance probability;

[0198] Through the formula: Obtain the wind pressure imbalance probability , where are the real-time wind pressure imbalance characteristics of the boosting fan units and the losing fan units;

[0199] Divide the unit risk level according to the numerical interval of the wind pressure imbalance probability p.

[0200] Intervention discrimination module: Based on the determined risk level, it is used to extract the wind pressure imbalance characteristics of the boosting fan units and the losing fan units within the sliding window, construct the steering amplitude matrix and calculate the steering distribution entropy, and judge whether it is necessary to perform intervention control on the boosting fan units and the losing fan units based on the steering distribution entropy;

[0201] Based on the determined risk level, use the sliding window method to obtain the wind pressure imbalance characteristics of the boosting fan units and the losing fan units at multiple sampling points within the sliding window;

[0202] Extract the single-unit deviation rate corresponding to the wind pressure imbalance characteristics of multiple sampling points within the sliding window, and construct the wind pressure deviation sequence , where the value range of i is [1, m];

[0203] where i represents the number of the sampling point, m is the total number of sampling points, and y represents the single-unit deviation rate;

[0204] Through the formula: Obtain the steering frequency of the single-unit deviation rate ;

[0205] Through the formula: Obtain the average steering amplitude A of the single-unit deviation rate;

[0206] Obtain the average steering amplitude of all boosting fan units and losing fan units within all sliding windows, and establish the steering amplitude matrix M a,b ;

[0207] Quantify the distribution complexity of the steering amplitudes of the boosting fan units and the losing fan units by calculating the steering distribution entropy of the steering amplitude matrix;

[0208] Among them, the calculation method of the steering distribution entropy is:

[0209] Normalize the steering amplitude matrix to obtain the standard steering amplitude matrix ;

[0210] Through the formula: Obtain the steering distribution entropy ;

[0211] where Za is the total number of wind turbine units that grab wind and lose wind in the steering amplitude matrix, and Zb is the total number of sliding windows;

[0212] If = 0, the steering amplitudes of all wind turbine units that grab wind and lose wind in all windows are 0, indicating that there is no fluctuation in the steering amplitudes of all wind turbine units that grab wind and lose wind in all windows, and the regulation is the simplest;

[0213] If The larger the value, the more dispersed the steering amplitude distribution, the greater the regulation difference between different units, and the higher the regulation complexity for wind turbine units that grab wind and lose wind;

[0214] Compare the steering distribution entropy with a preset steering distribution threshold. If the steering distribution entropy is higher than or equal to the preset steering distribution threshold, it is necessary to intervene and regulate the air pressure of the exhaust fan units;

[0215] If the steering distribution entropy is lower than the preset steering distribution threshold, it indicates that the current regulation difference between different units is within the expected range, and the regulation can be corrected through the preset program of the exhaust fan units.

[0216] Intervention control module: If necessary, it is used to screen the average steering amplitudes of different wind turbine units that grab wind or lose wind in the steering amplitude matrix to obtain the priority regulation units, and construct a self-regulation model to intervene and control the priority regulation units;

[0217] If necessary, extract the maximum value of the average steering amplitudes of the wind turbine units that grab wind or lose wind in the steering amplitude matrix;

[0218] Sort the maximum values of the average steering amplitudes of the wind turbine units that grab wind or lose wind in descending order, and based on the sorting results, determine the priority regulation units;

[0219] Construct a self-regulation model through the proportional integral derivative (PID) algorithm, calculate the regulation amount of the motor speed, and correct the proportional coefficient of the PID algorithm in combination with the average steering amplitude;

[0220] Regulate the motor speed of the priority regulation units based on the regulation amount of the motor speed;

[0221] where the output of the standard PID controller is: ; is the regulation amount of the motor speed of the priority regulation units, and t represents time;

[0222] They are the proportional gain coefficient, integral gain coefficient, and derivative gain coefficient respectively, which are used for the intensity of proportional control action, the intensity of integral control action, and the intensity of derivative control action respectively;

[0223] e(t) is the system error signal, representing the difference between the motor speed set value and the actual output value at the current time;

[0224] It represents the reciprocal of the error signal with respect to the event, reflecting the rate of change of the error;

[0225] It represents the integral of the error signal in the time range [0, t], reflecting the cumulative effect of the integral link on the error. It is a function of the variable ; is the integration variable;

[0226] Among them, the method of correcting the proportional coefficient of the proportional-integral-derivative algorithm in combination with the average steering amplitude is as follows:

[0227] Obtain the maximum value of the average steering amplitude of the priority regulation unit as the input parameter M of the self-tuning model max-avg ;

[0228] Take the input parameter M max-avg as the feedforward compensation factor of the proportional-integral-derivative (PID) algorithm to correct the PID parameters;

[0229] Through the formula: Obtain the proportional correction factor of the proportional-integral-derivative algorithm , where M ref is the reference value of the preset steering amplitude, and α is the preset proportional coefficient;

[0230] Substitute the corrected proportional coefficient into the output formula of the standard PID controller to obtain the regulation amount of the motor speed of the priority regulation unit ;

[0231] Based on the regulation amount of the motor speed of the priority regulation unit , regulate the motor speed through the motor control unit of the exhaust fan unit.

[0232] The above has described an embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the present invention.

Claims

1. An online remote real-time status monitoring method for an exhaust fan unit, characterized in that: It includes the following steps: Obtain the real-time output air pressure data of the exhaust fan group, and extract the air pressure imbalance characteristics of the exhaust fan group; The method for extracting the air pressure imbalance characteristics of the exhaust fan group is as follows: Obtain the air pressure of the fan group when all single exhaust fan groups are operating in parallel, calculate the deviation ratio of the air pressure of a single exhaust fan group to the average air pressure of all exhaust fan groups to obtain the single-unit deviation rate; Set multiple time sampling points, extract the single-unit deviation rates of multiple sampling points, and perform linear fitting analysis on the single-unit deviation rates of multiple sampling points to obtain the deviation change rate; Extract the single-unit deviation rate and deviation change rate of all exhaust fan groups as the air pressure imbalance characteristics; Based on the air pressure imbalance characteristics, judge the air pressure balance state of the exhaust fan group, identify the air-grabbing and air-loss fan groups, and perform linkage analysis on the air pressure imbalance characteristics of the air-grabbing and air-loss fan groups to determine the linkage relationship between the air-grabbing and air-loss fan groups; The method for determining the linkage relationship between the air-grabbing and air-loss fan groups is as follows: Based on the single-unit deviation rate and deviation change rate, combine the judgment rules to judge the air pressure balance state of the exhaust fan group, and identify the air-loss fan groups and air-grabbing fan groups; Perform linkage analysis on the air-loss fan groups and air-grabbing fan groups. If the correlation coefficients of the deviation change rates of the air-grabbing fan groups and air-loss fan groups, and the correlation coefficients of the single-unit deviation rates of the air-grabbing fan groups and air-loss fan groups are all negatively correlated, that is, the air-grabbing fan groups and air-loss fan groups show a linkage relationship; If the linkage relationship shows linkage, construct a fan detection model through the air pressure imbalance characteristics corresponding to the historical faults of the exhaust fan group, use the real-time air pressure imbalance characteristics as the input of the fan detection model, output the air pressure imbalance probability, and divide the unit imbalance level according to the air pressure imbalance probability; Based on the determined imbalance level, according to the constructed air pressure deviation sequence, extract the air pressure imbalance characteristics of the air-grabbing fan groups and air-loss fan groups within the sliding window, construct a steering amplitude matrix and calculate the steering distribution entropy to judge whether it is necessary to perform intervention control on the air-grabbing fan groups and air-loss fan groups; The method for judging whether it is necessary to perform intervention control on the air-grabbing fan groups and air-loss fan groups is as follows: Extract the single-unit deviation rates corresponding to the air pressure imbalance characteristics of multiple sampling points within the sliding window, and construct an air pressure deviation sequence; Perform steering calculation on the air pressure deviation sequence to obtain the average steering amplitude of the single-unit deviation rate; Obtain the average steering amplitude of all air-grabbing fan groups and air-loss fan groups within all sliding windows, and establish a steering amplitude matrix; By calculating the steering distribution entropy of the steering amplitude matrix, compare the steering distribution entropy with the preset steering distribution threshold to determine whether it is necessary to perform intervention control on the air-grabbing fan groups and air-loss fan groups; If necessary, screen different air-grabbing or air-loss fan groups in the steering amplitude matrix to obtain the priority regulation units, and construct a self-regulation model to perform intervention control on the priority regulation units.

2. The online remote real-time status monitoring method for an exhaust fan unit according to claim 1, wherein: The method for obtaining the correlation coefficients of the deviation change rates of the air-grabbing fan groups and air-loss fan groups, and the correlation coefficients of the single-unit deviation rates of the air-grabbing fan groups and air-loss fan groups is as follows: Calculate the Pearson correlation coefficients of the deviation change rates of the air-grabbing fan groups and air-loss fan groups, and the Pearson correlation coefficients of the single-unit deviation rates respectively to obtain the correlation coefficients of the deviation change rates of the air-grabbing fan groups and air-loss fan groups and the correlation coefficients of the single-unit deviation rates.

3. An on-line remote real-time status monitoring method for an exhaust fan unit according to claim 1, characterized in that: The method for obtaining the air pressure imbalance probability is as follows: Obtain the corresponding wind pressure imbalance characteristics during the historical faults of the exhaust fan group, construct a feature dataset and perform normalization processing to obtain a standard feature dataset; Through the Gaussian mixture model algorithm, use the standard feature dataset to construct and train a fan monitoring model; Input the real-time wind pressure imbalance characteristics of the robbing fan group and the losing fan group into the fan monitoring model to obtain the wind pressure imbalance probability.

4. The on-line remote real-time status monitoring method for an exhaust fan unit according to claim 1, characterized in that: The method for performing turning calculation on the wind pressure deviation sequence is as follows: Perform turning frequency analysis on the wind pressure deviation sequence to obtain the turning frequency of the single-machine deviation rate; Perform amplitude calculation on the turning frequency of the single-machine deviation rate to obtain the average turning amplitude of the single-machine deviation rate.

5. An online remote real-time status monitoring method for an exhaust fan unit according to claim 1, characterized in that: The method for calculating the turning distribution entropy of the turning amplitude matrix is as follows: Perform normalization processing on the turning amplitude matrix to obtain a standard turning amplitude matrix; Perform entropy value calculation on the standard turning amplitude matrix to obtain the turning distribution entropy.

6. The on-line remote real-time status monitoring method for an exhaust fan unit according to claim 1, characterized in that: The method for constructing a self-regulation model to perform intervention control on the priority regulation unit is as follows: Sort the maximum values of the average turning amplitudes of the robbing fan group or the losing fan group, and based on the sorting results, determine the priority regulation unit; Construct a self-regulation model through the proportional integral derivative algorithm, and correct the proportional coefficient of the proportional integral derivative algorithm in combination with the average turning amplitude; Calculate the regulation amount of the motor speed through the self-regulation model, and regulate the motor speed of the priority regulation unit.

7. An online remote real-time status monitoring method for an exhaust fan unit according to claim 6, characterized in that: The method for correcting the proportional coefficient of the proportional integral derivative algorithm is as follows: Obtain the maximum value of the average turning amplitude of the priority regulation unit as the input parameter of the self-tuning model; Use the input parameter as the feedforward compensation factor of the proportional integral derivative algorithm to correct the parameters of the proportional integral derivative algorithm.

Images