Optimization control method and system for intelligent speed regulator of ultra-large fan

Abstract

The invention discloses an optimization control method and system for an ultra-large fan intelligent speed regulator, and relates to the technical field of fan intelligent control, and the method comprises the steps: obtaining the current operation environment parameters of a fan, including but not limited to a wind speed signal, a motor input current, a voltage, a power factor, a rotating speed, an environment temperature and atmospheric pressure; monitoring whether the wind speed fluctuation amplitude exceeds a set threshold or not in a continuous sampling period; judging whether the current system enters a critical hysteresis zone state or not, wherein the critical hysteresis zone state is a working condition when the wind speed data of the sensor does not have stability and fan output has current fluctuation abnormity and a differential pressure rapid change trend; when the system enters a critical hysteresis zone state, the collected wind speed signal is paused as a speed regulation basis, an air volume estimation model based on current, voltage, rotating speed, temperature, humidity and air pressure is constructed, and equivalent air volume is output as a pseudo feedback signal; the problem that the fan speed regulation accuracy and the system stability are reduced in the plateau low-pressure environment is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of fans, and specifically to an optimized control method and system for an ultra-large fan intelligent speed governor. Background Art

[0002] With the continuous acceleration of the industrialization and urbanization processes in plateau areas, key equipment systems in many high-altitude environments have put forward higher requirements for fan speed control systems with high efficiency, stability, and low failure rates. When traditional industrial fans operate in medium and low altitude areas, by constructing a feedback loop to control the fan speed with the help of wind speed sensors, current sensors, or air pressure sensors, relatively precise closed-loop control can be achieved to a certain extent, thereby maintaining the ventilation, cooling, or air flow transportation functions of the target environment.

[0003] However, in a plateau low-pressure environment (such as areas above 2,500 meters above sea level), due to the significant decrease in air pressure and the drastic change in air density, conventional wind speed or air pressure sensors are extremely prone to signal drift, range loss, linear interval offset and other failure phenomena. In severe cases, it may even cause the feedback loop to break, thereby affecting the fan speed regulation accuracy and system stability.

[0004] To overcome this problem, in some engineering practices, the method of "multi-sensor redundant fusion" is tried, that is, multiple types of feedback sensors are set in the system, and when one signal fails, compensation is achieved by switching to the standby sensor signal. However, this solution also has significant limitations:

[0005] Contradiction of increased system complexity: Introducing multiple redundant signals and switching logics will inevitably lead to a significant increase in the complexity of the control system, with large controller resource consumption, long debugging cycles, and high maintenance costs, seriously violating the basic requirement of "equipment in remote areas needs to be simple and stable".

[0006] False triggering and oscillation contradiction: The determination mechanism between redundant systems often based on empirical thresholds is prone to false judgments or frequent switching due to noise interference or physical response lag between sensors, instead introducing secondary oscillations or false speed regulation phenomena and destroying system stability.

[0007] Response speed and control accuracy contradiction: There is an inevitable response delay in performing fault tolerance processing after sensor failure. Even if switching is performed through a fault detection algorithm, the fan has entered an unstable state, and the speed regulation lags behind the actual environmental requirements, resulting in air pressure imbalance or ventilation efficiency decline.

[0008] Therefore, it is very necessary to design an optimized control method and system for an ultra-large fan intelligent speed governor that can stably and predictably adjust the fan speed without relying on measured feedback signals in a plateau environment. Summary of the Invention

[0009] The object of the present invention is to provide an optimized control method and system for an ultra-large fan intelligent speed regulator to solve the problems raised in the above-mentioned background technology.

[0010] To solve the above technical problems, the present invention provides the following technical solution: An optimized control method for an ultra-large fan intelligent speed regulator, comprising the following steps:

[0011] Step S1: Obtain the current operating environment parameters of the fan, including but not limited to: wind speed signal, motor input current, voltage, power factor, rotation speed, ambient temperature and atmospheric pressure, and monitor whether the wind speed fluctuation amplitude exceeds the set threshold in a continuous sampling period;

[0012] Step S2: Determine whether the current system enters the critical hysteresis zone state, where the critical hysteresis zone state is the working condition when the sensor wind speed data is not stable and there are abnormal current fluctuations and a rapid change trend of the pressure difference in the fan output;

[0013] Step S3: When the system enters the critical hysteresis zone state, pause the collected wind speed signal as the speed regulation basis, construct an air volume estimation model based on current, voltage, rotation speed, temperature and humidity, and air pressure, and output the equivalent air volume as a pseudo-feedback signal, and this model is obtained by fitting with a training function;

[0014] Step S4: Calculate the rotation speed increment that the fan needs to adjust according to the difference between the target air volume and the pseudo-feedback air volume, and limit its change rate by a rate limiter not to exceed the maximum adjustment threshold to ensure the safety of the equipment and the smooth response of the fan;

[0015] Step S5: Control the fan to execute speed adjustment, and continuously and periodically detect the stability of the real sensor signal. When the detection result continuously meets the preset stability standard and lasts for a set time, the system will automatically switch from the pseudo-feedback control logic back to the real closed-loop feedback mode.

[0016] According to the above technical solution, in the step S1, it further includes a wind speed fluctuation trend model constructed based on the historical fluctuation trajectory, and this model is jointly constructed by the maximum amplitude difference within a continuous time window, the change frequency of periodic wave peaks and wave troughs, and the data stability index. When it is detected that the wind speed signal does not meet the stability threshold within three consecutive periods and the similarity of the fluctuation trend with the historical model exceeds the preset dynamic similarity factor, it is determined that the system enters the unstable hysteresis zone state.

[0017] According to the above technical solution, the air volume estimation model in the step S3 is a combined function Q f = f(I,U,ω,T,P), where Q fLet $\dot{V}$ denote the estimated air volume, $I$ denote the fan current, $U$ be the input voltage, $\omega$ denote the fan speed, $T$ be the ambient temperature, and $P$ be the ambient pressure. The function adopts a multi-variable quadratic regression form combined with a dynamic weighting factor $\lambda$. i , and this weighting factor is adaptively updated from historical data and satisfies the following functional form:

[0018]

[0019] where $\varphi$ i (x i ) is the non-linear transformation term of variable $x$ i , and $\epsilon$ is the model residual term.

[0020] According to the above technical solution, in step S4, during the process of calculating the rotational speed increment based on the difference between the target air volume and the pseudo-feedback air volume, an integral factor $\eta$ and an error threshold $\delta$ are introduced min . When the average error within consecutive sampling periods is lower than the threshold $\delta$ min , the controller maintains the current state unchanged; otherwise, the speed regulation logic is triggered. In addition, the controller has an offset correction factor for the air volume response curve to adapt to the non-linear effect of the air volume response under different plateau air pressure conditions and avoid unexpected speed regulation offsets caused by small deviations.

[0021] According to the above technical solution, in step S5, during the execution of the speed regulation method, a multi-source signal consistency determination module is called. This module compares the deviations among the true wind speed sensor values, pseudo-feedback estimation values, and backtest prediction values within three periods, and introduces a heterologous stability consistency index $\theta$. When $\theta$ is less than the set threshold and the deviation continuously converges for more than the specified sampling window time, the system automatically switches the feedback source. During the switching process, a transition buffer mechanism is used to perform linear interpolation transition on the fan speed regulation command to prevent the control oscillation phenomenon caused by "feedback source mutation" and improve the system stability and anti-interference ability.

[0022] According to the above technical solution, the method has an adaptive initialization learning mechanism at the initial stage of system operation. Before entering the critical hysteresis control process for the first time, the system operates in a safe low-speed mode to collect operation data under different combinations of environmental variables and electrical parameters, and generates a set of highly reliable initial fitting parameters in combination with the local air pressure conditions for establishing the first-round pseudo-feedback air volume estimation model, avoiding significant control disorders directly relying on the model with "large real-time errors" during the first operation stage of the system. At the same time, this initial training set is cached in a lightweight data structure and automatically updated regularly to ensure the gradual optimization and improvement of the estimation model during long-term operation and improve the long-term control reliability.

[0023] An optimized control system for an extra-large fan intelligent speed regulator, which includes: a fan main body, a sensor assembly, a pseudo-feedback modeling module, a control decision module, and an execution controller, where:

[0024] The sensor assembly includes a wind speed sensor, a current sensor, a voltage detection unit, a temperature and humidity sensor, and an atmospheric pressure sensing unit. The sensor assembly is used to collect data on the environment and operating state of the fan and transmit the collected signals to the control decision-making module;

[0025] The control decision-making module includes a critical hysteresis state determination unit, a pseudo-feedback control trigger logic unit, and a stability detection unit. After receiving all sensor values, the control decision-making module determines whether the system enters the air volume feedback critical hysteresis state and dynamically switches the feedback signal source according to the judgment result;

[0026] The pseudo-feedback modeling module is used to receive inputs of variables such as the collected current, voltage, rotational speed, ambient temperature, and air pressure, and calculate an equivalent air volume value through a set air volume prediction function;

[0027] The execution controller is used to calculate the target rotational speed required for adjustment and output it to the fan driver after receiving the error signal between the target air volume and the predicted air volume.

[0028] According to the above technical solution, the pseudo-feedback modeling module includes: a data sampling and caching unit, a signal preprocessing unit, a variable screening and weight configuration module, an air volume predictor, and an error feedback corrector; among them,

[0029] The variable screening and weight configuration module is used to evaluate the importance of the collected variables, screen the main variables through the entropy weight method or the information gain ratio algorithm, and set a dynamically adjustable weight factor for each variable. The weight factor changes dynamically with the fan operating environment according to the long-term trend adjustment mechanism.

[0030] According to the above technical solution, the control decision-making module further includes a pseudo-feedback trigger criterion fusion unit. The pseudo-feedback trigger criterion fusion unit receives the continuously fluctuating data provided by the wind speed sensor and the system electrical load state parameters, and determines whether to enable pseudo-feedback control through a multi-parameter collaborative trigger mechanism based on a time window. Among them, this trigger mechanism adopts a master control decision tree strategy, establishes a decision path according to three dimensions: the signal fluctuation amplitude, the system current spike abnormal threshold, and the air volume estimation stability degree, and automatically adjusts the trigger sensitivity according to the historical misjudgment rate; when only part of the trigger conditions are met, the system enters the "pre-pseudo-feedback buffer state", observes the state change in the next time period to decide whether to officially enable the pseudo-feedback mode.

[0031] Compared with the prior art, the beneficial effects achieved by the present invention are: the present invention,

[0032] (1) By setting up a recognition mechanism through introducing a critical hysteresis region, when abnormal fluctuations in the wind speed signal are detected and accompanied by characteristics of current and differential pressure changes, the system can promptly suspend its reliance on unstable signals, avoid the direct impact of sensor failures on control decisions, reduce the risk of system instability from the source, and enhance the stability and reliability of the fan speed regulation system in the plateau environment;

[0033] (2) By constructing an air volume estimation model based on multiple variables such as current, voltage, rotational speed, ambient temperature, and air pressure, and adopting a support vector machine optimization and dynamic weighting mechanism to generate a high-precision pseudo-feedback signal, even in the case where the actual wind speed is unmeasurable, the equivalent control of the air volume can be achieved, effectively maintaining the continuity and consistency of the fan output, and realizing precise speed regulation under the condition of no measured wind speed;

[0034] (3) By providing a multi-source signal consistency determination module and a transition buffer mechanism, when the wind speed signal returns to stability, the system can smoothly switch back to the true feedback closed-loop on the basis of fully verifying data consistency, effectively preventing secondary oscillations caused by feedback mutations, further enhancing the overall robustness of the control system, and realizing the smooth switching between pseudo-feedback control and true closed-loop control. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention and do not constitute a limitation to the present invention.

[0036] In the drawings:

[0037] Figure 1 is a schematic diagram of the overall flow of an optimized control method for an ultra-large fan intelligent speed regulator provided in Embodiment 1 of the present invention;

[0038] Figure 2 is a framework diagram of an optimized control system for an ultra-large fan intelligent speed regulator provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. 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.

[0040] Embodiment 1

[0041] Figure 1 is a schematic diagram of the overall flow of an optimized control method for an ultra-large fan intelligent speed regulator provided in Embodiment 1 of the present invention;

[0042] In this embodiment, the method includes the following steps:

[0043] Step S1: Obtain the current operating environment parameters of the fan, including but not limited to: wind speed signal, motor input current, voltage, power factor, rotational speed, ambient temperature, and atmospheric pressure, and monitor whether the amplitude fluctuation of the wind speed exceeds a set threshold during a continuous sampling period;

[0044] In the embodiment of the present invention, to further improve the accuracy of critical hysteresis state recognition, it further includes a wind speed fluctuation trend model constructed based on historical fluctuation trajectories. This model is jointly constructed by the maximum amplitude difference within a continuous time window, the change frequency of periodic wave peaks and valleys, and the data stability index. When it is detected that the wind speed signal does not meet the stability threshold within three consecutive periods and the similarity of the fluctuation trend with the historical model exceeds a preset dynamic similarity factor, it is determined that the system enters an unstable hysteresis zone state, thereby effectively avoiding misjudgment caused by short-term interference or individual anomalies and improving the accuracy and robustness of speed regulation triggering.

[0045] Step S2: Determine whether the current system enters the critical hysteresis zone state, where the critical hysteresis zone state is the operating condition when the sensor wind speed data lacks stability and there are abnormal current fluctuations and a rapid change trend in pressure difference in the fan output;

[0046] Step S3: When the system enters the critical hysteresis zone state, pause the collected wind speed signal as the basis for speed regulation, construct an air volume estimation model based on current, voltage, rotational speed, temperature, humidity, and air pressure, and output the equivalent air volume as a pseudo-feedback signal. This model is obtained by fitting through a training function;

[0047] The air volume estimation model is a combined function Q f = f(I, U, ω, T, P), where Q f represents the estimated air volume, I represents the fan current, U is the input voltage, ω represents the fan rotational speed, T is the ambient temperature, P is the ambient pressure, and the function adopts a multivariable quadratic regression form combined with a dynamic weighting factor λ i , and this weighting factor is adaptively updated by historical data and satisfies the following functional form:

[0048]

[0049] where φ i (x i ) is the non-linear transformation term of variable x i , ∈ is the model residual term, and this modeling process is optimized and solved through a support vector machine for training to ensure the valuation stability in the edge region and have a buffering ability for sudden inputs, improving the equivalence and stability when predicting the air volume as a pseudo-feedback;

[0050] Step S4: Calculate the rotational speed increment that the fan needs to adjust according to the difference between the target air volume and the pseudo-feedback air volume, and limit its change rate by a rate limiter not to exceed the maximum adjustment threshold to ensure the safety of the equipment and the smoothness of the fan response;

[0051] Exemplarily, in the embodiment of the present invention, to avoid frequent speed regulation of the system due to short-term errors, an integration factor η and an error threshold δ are introduced in the process of calculating the rotational speed increment according to the difference between the target air volume and the pseudo-feedback air volume. min When the average error within consecutive sampling periods is lower than the threshold δ min the controller maintains the current state unchanged; otherwise, the speed regulation logic is triggered. In addition, the controller has an offset correction factor for the air volume response curve to adapt to the non-linear effect of the air volume response under different plateau air pressure conditions and avoid unexpected speed regulation offsets caused by small deviations; an integration tracking error accumulation mechanism and an error threshold determination are introduced, combined with the offset correction of the response curve under the environmental air pressure conditions, to optimize the air volume and rotational speed adjustment logic, suppress frequent speed regulation caused by short-term fluctuations, improve the overall stability of the system, and at the same time adapt to the non-linear effect caused by the change of the plateau air pressure, and enhance the accuracy and reliability of the speed regulation.

[0052] Step S5: Control the fan to execute the speed adjustment, and continuously and periodically detect the stability of the real sensor signal. When the detection results continuously meet the preset stability standard and last for a set time, the system will automatically switch from the pseudo-feedback control logic back to the real closed-loop feedback mode; through continuous sampling and monitoring of the fan operating environment and electrical parameters, combined with the identification of the critical hysteresis zone state, the pseudo-feedback control logic is dynamically switched to ensure the safe, smooth and intelligent speed regulation of the fan even when the sensor fails or the data is unstable. Effectively avoid the speed regulation abnormal problems caused by sensor drift or failure in the plateau environment, and improve the adaptability, stability and reliability of the system in extreme environments;

[0053] In step S5, to ensure that the process of the system switching back to the real feedback signal has sufficient criteria and smoothness of action, during the execution of the speed regulation method, a multi-source signal consistency determination module is called. This module compares the deviations between the real wind speed sensor value, the pseudo-feedback estimation value and the backtest prediction value within three periods, and introduces a heterogeneous stability consistency index θ. When θ is less than the set threshold and the deviation continuously converges for more than the specified sampling window time, the system automatically switches the feedback source, and a transition buffer mechanism is used during the switching process to linearly interpolate and transition the fan speed regulation command to prevent the control oscillation phenomenon caused by the "mutation of the feedback source" and improve the stability and anti-interference ability of the system.

[0054] This method has an adaptive initialization learning mechanism in the initial stage of system operation. Before entering the critical hysteresis control process for the first time, the system operates at a safe low speed to collect operation data under different combinations of environmental variables and electrical parameters, and generates a set of highly reliable initial fitting parameters in combination with local air pressure conditions to establish the first-round pseudo-feedback air volume estimation model, avoiding significant control disorders caused by directly relying on the model with "large real-time errors" in the first operation stage of the system. At the same time, this initial training set is cached in a lightweight data structure and automatically updated regularly to ensure the gradual optimization and improvement of the estimation model during long-term operation and improve the long-term control reliability.

[0055] Embodiment 2

[0056] Figure 2 It is a frame diagram of an optimized control system for an extra-large fan intelligent speed regulator provided by Embodiment 2 of the present invention;

[0057] In this embodiment, the system includes: a fan main body, a sensor assembly, a pseudo-feedback modeling module, a control decision-making module, and an execution controller, where:

[0058] The sensor assembly includes a wind speed sensor, a current sensor, a voltage detection unit, a temperature and humidity sensor, and an atmospheric pressure sensing unit. The sensor assembly is used to collect data on the environment and operation state of the fan and transmit the collected signals to the control decision-making module;

[0059] The control decision-making module includes a critical hysteresis state determination unit, a pseudo-feedback control trigger logic unit, and a stability detection unit. After receiving all sensor values, the control decision-making module determines whether the system enters the air volume feedback critical hysteresis state and dynamically switches the feedback signal source according to the judgment result;

[0060] The pseudo-feedback modeling module is used to receive inputs of variables such as current, voltage, rotational speed, environmental temperature, and air pressure, and calculate the equivalent air volume value through a set air volume prediction function;

[0061] The execution controller is used to calculate the target rotational speed required for adjustment and output it to the fan driver after receiving the error signal between the target air volume and the predicted air volume; it not only effectively solves the key technical problems such as feedback failure, control lag, and oscillation faced by the fan speed regulation system in the plateau low-pressure environment, but also comprehensively improves the stability, accuracy, and adaptive ability of the extra-large fan intelligent speed regulation system in complex environments through innovative means such as introducing intelligent pseudo-feedback control, dynamic self-learning optimization, and multi-parameter collaborative determination.

[0062] The pseudo-feedback modeling module includes: a data sampling and caching unit, a signal preprocessing unit, a variable screening and weight configuration module, an air volume predictor, and an error feedback corrector; where,

[0063] The variable screening and weight configuration module is used to evaluate the importance of variables such as the collected voltage, current, rotational speed, ambient temperature, and air pressure. The main variables are screened through the entropy weight method or the information gain ratio algorithm, and a dynamically adjusted weight factor is set for each variable. The weight factor changes dynamically with the fan operating environment according to the long-term trend adjustment mechanism, so that the air volume predictor can continuously adapt to the characteristics of the plateau environment, improve the consistency and stability between the pseudo-feedback air volume and the actual air volume, and avoid misjudging the speed regulation under environmental disturbances.

[0064] The control decision module further includes a pseudo-feedback trigger criterion fusion unit. The pseudo-feedback trigger criterion fusion unit receives the continuously fluctuating data provided by the wind speed sensor and the system electrical load state parameters, and determines whether to enable pseudo-feedback control through a multi-parameter collaborative trigger mechanism based on a time window. Among them, this trigger mechanism adopts a master control decision tree strategy, establishes a decision path according to three dimensions: the signal fluctuation amplitude, the system current peak anomaly threshold, and the stability degree of the air volume estimation, and automatically adjusts the trigger sensitivity according to the historical misjudgment rate; when only part of the trigger conditions are met, the system enters the "pre-pseudo-feedback buffer state", observes the state change in the next time period to decide whether to officially enable the pseudo-feedback mode, so as to effectively avoid the instability caused by frequent state switching and improve the fault tolerance and self-recovery ability of the system in a complex airflow field environment.

[0065] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0066] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0067] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are executed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions for implementing the steps specified in one process or a plurality of processes and / or blocks Figure 1 in one block or a plurality of blocks Figure 1 in the functions specified.

[0068] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative rather than restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit of the present invention and the scope protected by the claims. These all fall within the protection scope of the present invention.

Claims

1. An optimization control method for an ultra-large fan intelligent speed regulator, characterized in that: The method comprises the following steps: Step S1, obtaining the current operating environment parameters of the fan, including but not limited to: wind speed signal, motor input current, voltage, power factor, speed, ambient temperature and atmospheric pressure, and monitoring whether the wind speed fluctuation amplitude exceeds the set threshold during the continuous sampling period; Step S2, determining whether the current system has entered a critical hysteresis zone state, wherein the critical hysteresis zone state is a condition where the wind speed data of the sensor is not stable and the fan output has abnormal current fluctuations and a rapid pressure difference change trend; Step S3: When the system enters the critical hysteresis zone, the collected wind speed signal is paused as the basis for speed regulation, and an air volume estimation model based on current, voltage, speed, temperature, humidity and air pressure is constructed to output the equivalent air volume as a pseudo feedback signal. The model is obtained by fitting the training function; Step S4, calculating the speed increment of the fan to be adjusted according to the difference between the target air volume and the pseudo feedback air volume, and limiting the change rate to not exceed the maximum adjustment threshold through the rate limiter to ensure the safety of the equipment and the stability of the fan response; Step S5, control the fan to perform speed adjustment, and continue to periodically detect the stability of the real sensor signal. When the detection result continuously meets the preset stability standard and lasts for a set period of time, the system will automatically switch from the pseudo-feedback control logic back to the real closed-loop feedback mode.

2. The optimization control method of an ultra-large fan intelligent speed regulator according to claim 1 is characterized in that: The step S1 further includes a wind speed fluctuation trend model constructed based on the historical fluctuation trajectory. The model is constructed by the maximum amplitude difference in a continuous time window, the frequency of periodic peak and trough changes, and the data stability index. When it is detected that the wind speed signal does not meet the stability threshold for three consecutive periods and the similarity with the fluctuation trend of the historical model exceeds the preset dynamic similarity factor, it is determined that the system enters an unstable hysteresis zone state.

3. The optimization control method of an ultra-large fan intelligent speed regulator according to claim 1 is characterized in that: The air volume estimation model in step S3 is a set of synthetic functions Q f =f(I,U,ω,T,P), where Q f represents the estimated air volume, I represents the fan current, U represents the input voltage, ω represents the fan speed, T represents the ambient temperature, and P represents the ambient pressure. The function adopts a multivariate quadratic regression form and combines a dynamic weighting factor λ i , the weighting factor is adaptively updated by historical data and satisfies the following function form: where φ i (x i ) is the variable x i is the nonlinear transformation term, and ∈ is the model residual term.

4. The optimization control method of a super-large fan intelligent speed regulator according to claim 1 is characterized in that: In step S4, in the process of calculating the speed increment according to the difference between the target air volume and the pseudo feedback air volume, the integral factor η and the error threshold δ are introduced. min , when the average error in consecutive sampling periods is lower than the threshold δ min When , the controller maintains the current state unchanged; Otherwise, the speed regulation logic is triggered. In addition, the controller sets an offset correction factor for the wind volume response curve to adapt to the nonlinear effect of wind volume response under different plateau pressure conditions to avoid unexpected speed regulation offset caused by small deviations.

5. The optimization control method of a super-large fan intelligent speed regulator according to claim 1 is characterized in that: In step S5, during the execution of the speed control method, a multi-source signal consistency judgment module is called, which compares the deviations between the real wind speed sensor value, the pseudo feedback estimation value and the backtest prediction value within three cycles, and introduces a heterogeneous stability consistency index θ. When θ is less than the set threshold and the deviation continues to converge for more than the specified sampling window time, the system automatically switches the feedback source, and uses a transition buffer mechanism during the switching process to perform linear interpolation transition on the fan speed control command to prevent control oscillation caused by "feedback source mutation" and improve system stability and anti-interference ability.

6. The optimization control method of a super-large fan intelligent speed regulator according to claim 1 is characterized in that: The method has an adaptive initialization learning mechanism in the initial stage of system operation. Before entering the critical hysteresis control process for the first time, the system collects operating data under different combinations of environmental variables and electrical parameters in a safe low-speed operation mode, and generates a set of high-confidence initial fitting parameters in combination with local air pressure conditions to establish the first round of pseudo-feedback air volume estimation model, avoiding significant control imbalance caused by direct reliance on the "large real-time error" model in the first operation stage of the system. At the same time, the initial training set is cached in a lightweight data structure and automatically updated regularly to ensure that the valuation model is gradually optimized and improved during long-term operation, thereby improving long-term control reliability.

7. An optimization control system for an ultra-large fan intelligent speed regulator, characterized in that: The system includes: a fan body, a sensor assembly, a pseudo feedback modeling module, a control decision module and an execution controller, wherein: The sensor assembly includes a wind speed sensor, a current sensor, a voltage detection unit, a temperature and humidity sensor, and an atmospheric pressure sensing unit. The sensor assembly is used to collect data about the environment and operating status of the fan, and transmit the collected signals to the control decision module; The control decision module includes a critical hysteresis state determination unit, a pseudo feedback control trigger logic unit and a stability detection unit. After receiving all sensor values, the control decision module determines whether the system enters a critical hysteresis state of air volume feedback, and dynamically switches the feedback signal source according to the determination result; The pseudo-feedback modeling module is used to receive the collected variable inputs such as current, voltage, speed, ambient temperature and air pressure, and calculate the equivalent air volume value through the set air volume prediction function; The execution controller is used to calculate the target rotation speed required for adjustment after receiving the error signal between the target air volume and the predicted air volume, and output it to the fan driver.

8. The optimization control system of the super-large fan intelligent speed regulator according to claim 7 is characterized in that: The pseudo feedback modeling module includes: a data sampling cache unit, a signal preprocessing unit, a variable screening and weight configuration module, an air volume predictor and an error feedback corrector; wherein, The variable screening and weight configuration module is used to evaluate the importance of the collected variables, screen the main variables through the entropy weight method or the information gain ratio algorithm, and set a dynamically adjusted weight factor for each variable. The weight factor changes dynamically with the fan operating environment based on the long-term trend adjustment mechanism.

9. The optimization control system of the super-large fan intelligent speed regulator according to claim 7 is characterized in that: The control decision module further includes a pseudo-feedback trigger criterion fusion unit, which receives the continuous fluctuation data provided by the wind speed sensor and the system electrical load state parameters, and determines whether to enable pseudo-feedback control through a multi-parameter collaborative trigger mechanism based on a time window. The trigger mechanism adopts a main control decision tree strategy, establishes a decision path according to three dimensions: signal fluctuation amplitude, system current spike abnormality threshold, and wind volume estimation stability, and automatically adjusts the trigger sensitivity according to the historical misjudgment rate; when the trigger condition is only partially met, the system enters a "pre-pseudo-feedback buffer state" to observe the state changes in the next time period to decide whether to formally enable the pseudo-feedback mode.

Images