Multi-fan airflow control method and system based on multi-point modeling and synchronous triggering

Abstract

The invention relates to the technical field of fan airflow control, in particular to a multi-fan airflow control method and system based on multi-point modeling and synchronous triggering, and the method comprises the steps: building a control signal prediction model; generating a prediction control signal time sequence according to the target wind speed function; s3, issuing a predictive control signal time sequence and caching the predictive control signal time sequence as a local control signal time sequence; the synchronous global clock outputs an initial multi-point control signal at a preset time point; the actual wind speed is measured in real time, a compensation control signal is fed back and generated, and the initial multi-point control signal is superposed to output a dynamic multi-point control signal; the control system is applied to the control method, accurate cooperative control can be carried out on the multiple fans, high-consistency output of the multiple fans and synchronous generation of various complex time-varying airflows are achieved, and meanwhile dynamic accurate compensation of airflow control of the multiple fans can be achieved in the face of performance degradation of the fans.

Description

Technical Field

[0001] The present invention relates to the technical field of fan airflow control, and in particular to a multi-fan airflow control method and system based on multi-point modeling and synchronous triggering. Background Art

[0002] In fields such as aviation simulation, fluid experiments, wind tunnel testing, and virtual natural wind simulation, multiple fan arrays are required to collaboratively output time-varying airflows with wind speed waveforms such as sinusoidal and disturbance. However, due to poor fan manufacturing consistency, nonlinear control signal output, and delay uncertainty in network communication, traditional unified fan control methods find it difficult to achieve highly consistent output and synchronous generation of time-varying airflows for thousands of fans. At the same time, traditional unified fan control methods are unable to perform closed-loop adaptive compensation for fan performance degradation caused by fan aging and environmental disturbances such as temperature, humidity, and load, to ensure precise dynamic control and long-term consistent output of multi-fan airflows.

[0003] Therefore, how to accurately coordinate and dynamically control multiple fans, achieve highly consistent output from multiple fans and synchronous generation of time-varying airflow, and combat fan performance degradation is an urgent problem that needs to be solved. Summary of the Invention

[0004] In response to the above problems, the present invention provides a multi-fan airflow control method and system based on multi-point modeling and synchronous triggering to solve the technical problems raised by the above background technology.

[0005] The technical solution adopted by the present invention to solve its technical problems is as follows:

[0006] A multi-fan airflow control method based on multi-point modeling and synchronous triggering includes the following steps:

[0007] S1: Sample the multi-point control signals and wind speed responses of each fan, establish a control signal wind speed fitting model, and reversely infer the model to obtain a control signal prediction model;

[0008] S2: Establish a target wind speed function based on the time-varying airflow characteristics to be generated, discretely generate a target wind speed time series, and generate a predicted control signal time series based on the control signal prediction model;

[0009] S3: Use the TCP protocol to send the predicted control signal time series to each fan controller, and each fan controller caches the predicted control signal time series as a local control signal time series;

[0010] S4: Synchronize the global clock through the NTP protocol, and based on the local control signal time sequence, output the initial multi-point control signal synchronously at the preset time point through the local timer;

[0011] S5: Measure the actual wind speed in real time, compare it with the target wind speed time series, generate a compensation control signal based on the PI / PID feedback adjustment algorithm, and superimpose the initial multi-point control signal to output a dynamic multi-point control signal.

[0012] Furthermore, the specific steps of establishing the control signal wind speed fitting model in S1 are: performing interpolation fitting on the sampled data to obtain the control signal wind speed fitting model, and the control signal wind speed fitting model is specifically V=f i (PWM), where V is the fan speed, PWM is the multi-point control signal, i is the fan number, and f i The mapping relationship between the multi-point control signal of each fan and the fan speed, the control signal prediction model is PWM=f i -1 (V).

[0013] Furthermore, the specific steps for establishing the target wind speed function in S2 are: according to the time-varying airflow characteristics to be generated, confirm the time-varying airflow waveform change curve, the time-varying airflow waveform change curve includes sine wave, square wave and triangle wave, and establish the target wind speed function according to the time-varying airflow waveform change curve.

[0014] Furthermore, when the time-varying airflow waveform curve is a sine wave, the target wind speed function of the continuous time variable t is V(t)=V 0,1 +A1sin(2πω1t+φ1), where V 0,1 is the sinusoidal reference wind speed, A1 is the sinusoidal amplitude, ω1 is the sinusoidal period frequency, and φ1 is the sinusoidal phase angle. When the time-varying airflow waveform curve is a square wave, the target wind speed function is V(t) = V 0,2 +A2sgn[sin(2πω2t+φ2)], where sgn(x) is the sign function, when x>0, sgn(x)=1, when x=0, sgn(x)=0, when x<0, sgn(x)=-1, V 0,2 is the square wave reference wind speed, A2 is the square wave amplitude, ω2 is the square wave period frequency, and φ2 is the square wave phase angle. When the time-varying airflow waveform curve is a triangular wave, the target wind speed function is: Where V 0,3 is the triangular wave reference wind speed, A3 is the triangular wave amplitude, T is the triangular wave period, and modT represents the position of the time variable within one period.

[0015] Furthermore, the specific steps of generating the prediction control signal time series in S2 are: sampling the target wind speed function at a uniform time interval Δt to obtain the target wind speed time series Where k is the sampling point number, n is the total length of the sequence, k = 0, 1, 2, ..., n, V kis the target wind speed at the sampling point, V k =V(kΔt); bring the target wind speed time series into the control signal prediction model to obtain the predicted control signal time series Among them, PWM 0，k =f i -1 (V k ).

[0016] Furthermore, the specific steps of S3 are: splitting the predicted control signal time series into multiple predicted control signal time segments, and using the TCP transmission protocol to send the predicted control signal time segments to each fan controller in batches; after each fan controller caches the predicted control signal time segments locally, it is reorganized in sequence into a local control signal time series.

[0017] Furthermore, the specific steps of S4 are as follows: each fan controller generates pre-execution data based on the local control signal time series and waits for the synchronization trigger instruction; aligns the global clock using the NTP protocol through the network to establish a unified time base for each fan controller, and each fan controller triggers the synchronization trigger instruction at a preset time point through a local timer based on the unified time base; each fan controller executes the pre-execution data and synchronously outputs the initial multi-point control signal, and the airflow speed of the multiple fans changes synchronously with the output of the initial multi-point control signal, and the initial multi-point control signal is f i -1 (V k ).

[0018] Furthermore, the specific steps of S5 are: measuring the actual wind speed V of each fan and each area in real time meas , calculate the actual wind speed error ΔV=V k -V meas Based on the PI / PID feedback regulation algorithm, a compensation multi-point control signal ΔPWM is generated, and the initial multi-point control signal is superimposed to generate a dynamic multi-point control signal PWM. a , PWM a =f i -1 (V k )+ΔPWM.

[0019] A multi-fan airflow control system based on multi-point modeling and synchronous triggering, applied to the multi-fan airflow control method based on multi-point modeling and synchronous triggering described above, comprises:

[0020] Multi-point modeling module: used to sample the multi-point control signals and wind speed responses of each fan, and generate a control signal wind speed fitting model and a control signal prediction model;

[0021] Control signal prediction module: used to generate target wind speed time series and predict control signal time series;

[0022] Fan multi-point control module: includes multiple fan controllers, each of which caches the predicted control signal time series issued in batches and outputs the initial multi-point control signal;

[0023] Local synchronous execution module: includes a local timer, which is used to trigger the synchronous trigger instruction at a preset time point under a unified time base;

[0024] Wind speed feedback adjustment module: includes a wind speed sensor, which is used to measure the actual wind speed in real time and feed it back to the fan multi-point control module. The fan multi-point control module outputs a dynamic multi-point control signal.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] 1. The present invention provides a multi-fan airflow control method based on multi-point modeling and synchronous triggering. By establishing a control signal prediction model, it can predict multi-point control signals for any target wind speed. By aligning the global clock through the NTP protocol, multi-point control signals are synchronously triggered at preset time points, and multiple fans are precisely coordinated to achieve highly consistent output from multiple fans and the synchronous generation of various complex time-varying airflows.

[0027] 2. The present invention provides a multi-fan airflow control method based on multi-point modeling and synchronous triggering. It outputs dynamically adjusted multi-point control signals based on real-time feedback adjustment of actual wind speed, thereby realizing dynamic and precise compensation of multi-fan airflow control, thereby combating fan performance degradation and ensuring long-term consistent output of multiple fans.

[0028] 3. The present invention provides a multi-fan airflow control system based on multi-point modeling and synchronous triggering. By uniformly planning control signals, aligning global clocks over the network, and caching and synchronously executing control signals locally, it can reduce dependence on real-time communication while achieving precise coordinated control of multiple fans. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0030] Figure 1 This is a flow chart of the multi-fan airflow control method based on multi-point modeling and synchronous triggering according to the present invention;

[0031] Figure 2This is a structural diagram of the multi-fan airflow control system based on multi-point modeling and synchronous triggering according to the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] The present invention provides a multi-fan airflow control method based on multi-point modeling and synchronous triggering, such as Figure 1 As shown, the method includes the following steps:

[0034] S1: Sample the multi-point control signals and wind speed responses of each fan, establish a control signal wind speed fitting model, and reversely infer the model to obtain a control signal prediction model;

[0035] Specifically, after sampling the multi-point control signal and wind speed response of each fan, a correlation database of the multi-point control signal and the corresponding wind speed response of each fan is established based on the sampled data. After data cleaning, deletion and correction of abnormal data, the correlation database is carefully interpolated and fitted to obtain the control signal wind speed fitting model V = f i (PWM), where V is the fan speed, PWM is the multi-point control signal, i is the fan number, and f i is the mapping relationship between the multi-point control signal of each fan and the fan speed; the control signal prediction model is specifically PWM=f i -1 (V); Through multi-point modeling and inverse control signal prediction model, multi-point control signal prediction for any target wind speed of each fan can be achieved.

[0036] S2: Establish a target wind speed function based on the time-varying airflow characteristics to be generated, discretely generate a target wind speed time series, and generate a predicted control signal time series based on the control signal prediction model;

[0037] Specifically, according to the time-varying airflow characteristics to be generated, the time-varying airflow waveform change curve is determined. The time-varying airflow waveform change curve includes a sine wave, a square wave, and a triangular wave. The target wind speed function is established according to the time-varying airflow waveform change curve. When the time-varying airflow waveform change curve is a sine wave, the target wind speed function is V(t)=V 0,1 +A1sin(2πω1t+φ1), where t is a continuous time variable, V(t) is the target wind speed at different times, and V 0,1is the sinusoidal reference wind speed, A1 is the sinusoidal amplitude, ω1 is the sinusoidal period frequency, and φ1 is the sinusoidal phase angle. When the time-varying airflow waveform curve is a square wave, the target wind speed function is V(t) = V 0,2 +A2sgn[sin(2πω2t+φ2)], where sgn(x) is the sign function, when x>0, sgn(x)=1, when x=0, sgn(x)=0, when x<0, sgn(x)=-1, V 0,2 is the square wave reference wind speed, A2 is the square wave amplitude, ω2 is the square wave period frequency, φ2 is the square wave phase angle, and the wind speed of the time-varying airflow with a square wave waveform change curve jumps between the maximum and minimum values, representing the sudden wind speed disturbance; when the time-varying airflow waveform change curve is a triangular wave, the target wind speed function is Where V 0,3 is the triangular wave reference wind speed, A3 is the triangular wave amplitude, T is the triangular wave period, and modT represents the position of the time variable within one period.

[0038] The target wind speed function is sampled at a uniform time interval Δt to obtain the target wind speed time series Where k is the sampling point number, n is the total length of the sequence, k = 0, 1, 2, ..., n, V k is the target wind speed at the sampling point, V k =V(kΔt), the time interval Δt selected in this embodiment is 10ms, that is, the target wind speed function is sampled once every 10ms; the target wind speed time series is brought into the control signal prediction model to obtain the predicted control signal time series Among them, PWM 0，k =f i -1 (V k ), by establishing the target wind speed function and generating the predicted control signal time series, it is possible to predict the control signal of any complex time-varying airflow waveform, thereby generating any complex time-varying airflow waveform.

[0039] S3: Use the TCP protocol to send the predicted control signal time series to each fan controller, and each fan controller caches the predicted control signal time series as a local control signal time series;

[0040] Specifically, when sending down the predicted control signal time series, the predicted control signal time series is first split into multiple predicted control signal time segments, and the TCP transmission protocol is used to send the predicted control signal time segments to each fan controller in batches, while ensuring that the predicted control signal time segments accurately reach each fan controller, avoiding the loss or delay of the predicted control signal time series due to network congestion; each fan controller caches the predicted control signal time segments locally, and then reorganizes them into local control signal time series in sequence; each fan controller generates pre-execution data based on the local control signal time series, waiting for the synchronization trigger instruction. By dividing the control instruction into pre-execution data and synchronization trigger instruction, each fan controller caches the pre-execution data in advance, and the synchronization trigger instruction is decoupled from the pre-execution data, does not rely on the real-time network, can reduce the impact of real-time communication jitter in the process of multi-fan airflow control, and does not interrupt the control of each fan controller when the communication is momentarily interrupted.

[0041] S4: Synchronize the global clock through the NTP protocol, and based on the local control signal time sequence, output the initial multi-point control signal synchronously at the preset time point through the local timer;

[0042] Specifically, the global clock is aligned through the network using the NTP protocol to establish a unified time base for each fan controller. Based on the unified time base, each fan controller triggers a synchronization trigger instruction at a preset time point through a local timer. The NTP protocol is a network time synchronization protocol. The use of the NTP protocol can ensure the time alignment of the coordinated changes of the airflow; each fan controller executes the pre-execution data and outputs the initial multi-point control signal. The initial multi-point control signal is output in the form of a predicted control signal time series. The initial multi-point control signal is f i -1 (V k ); Each fan controller can achieve millisecond-level synchronous control to adapt to the high-frequency wind speed changes required by time-varying airflow. The output error of each fan controller is within ±1%, meeting the high consistency output requirements of multi-fan airflow control.

[0043] S5: Measure the actual wind speed in real time, compare it with the target wind speed time series, generate a compensation multi-point control signal based on the PI / PID feedback adjustment algorithm, and superimpose the initial multi-point control signal to output a dynamic multi-point control signal.

[0044] Specifically, a wind speed sensor is installed in each fan or area to measure the actual wind speed V of each fan and each area in real time. meas , calculate the actual wind speed error ΔV=V between the actual wind speed and the target wind speed at the sampling point k -V measBased on the PI / PID feedback regulation algorithm, a compensation multi-point control signal ΔPWM is generated. After the compensation multi-point control signal is fed back to each fan controller, each fan controller superimposes the compensation multi-point control signal with the initial multi-point control signal to output a dynamic multi-point control signal PWM. a , namely PWM a =f i -1 (V k )+ΔPWM; based on the error pattern between actual and target wind speeds, real-time feedback is provided to each fan controller, outputting dynamically adjustable multi-point control signals to achieve dynamic and precise compensation for multi-fan airflow control, thereby combating performance degradation caused by fan motor aging or environmental disturbances and ensuring long-term consistent output from multiple fans.

[0045] The present invention also provides a multi-fan airflow control system based on multi-point modeling and synchronous triggering, and applies the multi-fan airflow control method based on multi-point modeling and synchronous triggering, such as Figure 2 As shown, the system includes:

[0046] Multi-point modeling module: used to sample the multi-point control signals and wind speed responses of each fan, and generate a control signal wind speed fitting model and a control signal prediction model;

[0047] Control signal prediction module: used to generate target wind speed time series and predict control signal time series;

[0048] Fan Multi-Point Control Module: This module includes multiple fan controllers. Each fan controller caches the predicted control signal time series issued in batches and outputs the initial multi-point control signal. The fan multi-point control module can be deployed in different scales according to the needs of various scenarios such as large wind walls, experimental wind tunnels, or indoor wind environment simulations, and has large-scale scalability.

[0049] Local synchronization execution module: including a local timer, which is used to trigger the synchronization trigger instruction at a preset time point under the time base unified by the NTP protocol. The local timer selected in this embodiment is a hardware counter;

[0050] Wind speed feedback adjustment module: includes multiple wind speed sensors, and multiple wind speed sensors are set in each fan area to measure the actual wind speed in real time and feed it back to the fan multi-point control module. The fan multi-point control module outputs dynamic multi-point control signals; through the wind speed feedback adjustment module, in the face of fan performance degradation caused by motor aging and environmental disturbances, long-term consistency output is ensured to perform closed-loop adaptive compensation for multi-fan airflow control in real time.

[0051] Furthermore, the multi-fan airflow control system described in this embodiment adopts a control architecture that centrally and uniformly plans multi-point control signals, locally distributed caches multi-point control signals, aligns global clocks with the network, and executes local synchronous executions. This can support precise coordinated control of thousands of fans and reduce dependence on real-time communications.

[0052] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A multi-fan airflow control method based on multi-point modeling and synchronous triggering, characterized in that: The steps include: S1: Sample the multi-point control signals and wind speed responses of each fan, establish a control signal wind speed fitting model, and reversely infer the model to obtain a control signal prediction model; S2: Establish a target wind speed function based on the time-varying airflow characteristics to be generated, discretely generate a target wind speed time series, and generate a predicted control signal time series based on the control signal prediction model; S3: Use the TCP protocol to send the predicted control signal time series to each fan controller, and each fan controller caches the predicted control signal time series as a local control signal time series; S4: Synchronize the global clock through the NTP protocol, and based on the local control signal time sequence, output the initial multi-point control signal synchronously at the preset time point through the local timer; S5: Measure the actual wind speed in real time, compare it with the target wind speed time series, generate a compensation control signal based on the PI / PID feedback adjustment algorithm, and superimpose the initial multi-point control signal to output a dynamic multi-point control signal.

2. The multi-fan airflow control method based on multi-point modeling and synchronous triggering according to claim 1, characterized in that: The specific steps of establishing the control signal wind speed fitting model in S1 are: performing interpolation fitting on the sampled data to obtain the control signal wind speed fitting model, and the control signal wind speed fitting model is specifically V=f i (PWM), where V is the fan speed, PWM is the multi-point control signal, i is the fan number, and f i The mapping relationship between the multi-point control signal of each fan and the fan speed, the control signal prediction model is PWM=f i -1 (V).

3. The multi-fan airflow control method based on multi-point modeling and synchronous triggering according to claim 2, characterized in that: The specific steps of establishing the target wind speed function in S2 are: according to the time-varying airflow characteristics to be generated, confirm the time-varying airflow waveform change curve, the time-varying airflow waveform change curve includes sine wave, square wave and triangle wave, and establish the target wind speed function according to the time-varying airflow waveform change curve.

4. The multi-fan airflow control method based on multi-point modeling and synchronous triggering according to claim 3 is characterized in that: When the time-varying airflow waveform curve is a sine wave, the target wind speed function of the continuous time variable t is V(t) = V 0,1 +A1sin(2πω1t+φ1), where V 0,1 is the sinusoidal reference wind speed, A1 is the sinusoidal amplitude, ω1 is the sinusoidal period frequency, and φ1 is the sinusoidal phase angle. When the time-varying airflow waveform curve is a square wave, the target wind speed function is V(t) = V 0,2 +A2sgn[sin(2πω2t+φ2)], where sgn(x) is the sign function, when x>0, sgn(x)=1, when x=0, sgn(x)=0, when x<0, sgn(x)=-1, V 0,2 is the square wave reference wind speed, A2 is the square wave amplitude, ω2 is the square wave period frequency, and φ2 is the square wave phase angle. When the time-varying airflow waveform curve is a triangular wave, the target wind speed function is: Where V 0,3 is the triangular wave reference wind speed, A3 is the triangular wave amplitude, T is the triangular wave period, and modT represents the position of the time variable within one period.

5. The multi-fan airflow control method based on multi-point modeling and synchronous triggering according to claim 4 is characterized in that: The specific steps of generating the prediction control signal time series in S2 are: sampling the target wind speed function at a uniform time interval Δt to obtain the target wind speed time series Where k is the sampling point number, n is the total length of the sequence, k = 0, 1, 2, ..., n, V k is the target wind speed at the sampling point, V k =V(kΔt); bring the target wind speed time series into the control signal prediction model to obtain the predicted control signal time series Among them, PWM 0，k =f i -1 (V k ).

6. The multi-fan airflow control method based on multi-point modeling and synchronous triggering according to claim 5, characterized in that: The specific steps of S3 are: splitting the predicted control signal time series into multiple predicted control signal time segments, and using the TCP transmission protocol to send the predicted control signal time segments to each fan controller in batches; after each fan controller caches the predicted control signal time segments locally, it is reorganized in sequence into a local control signal time series.

7. The multi-fan airflow control method based on multi-point modeling and synchronous triggering according to claim 6, characterized in that: The specific steps of S4 are as follows: each fan controller generates pre-execution data based on the local control signal time series and waits for the synchronization trigger instruction; aligns the global clock using the NTP protocol through the network to establish a unified time base for each fan controller; each fan controller triggers the synchronization trigger instruction at a preset time point through a local timer based on the unified time base; each fan controller executes the pre-execution data and synchronously outputs the initial multi-point control signal, and the airflow speed of the multiple fans changes synchronously with the output of the initial multi-point control signal. The initial multi-point control signal is f i -1 (V k ).

8. The multi-fan airflow control method based on multi-point modeling and synchronous triggering according to claim 7, characterized in that: The specific steps of S5 are: measuring the actual wind speed V of each fan and each area in real time meas , calculate the actual wind speed error ΔV=V k -V meas Based on the PI / PID feedback regulation algorithm, a compensation multi-point control signal ΔPWM is generated, and the initial multi-point control signal is superimposed to generate a dynamic multi-point control signal PWM. a , PWM a =f i -1 (V k )+ΔPWM.

9. A multi-fan airflow control system based on multi-point modeling and synchronous triggering, applied to a multi-fan airflow control method based on multi-point modeling and synchronous triggering according to any one of claims 1 to 8, characterized in that: include: Multi-point modeling module: used to sample the multi-point control signals and wind speed responses of each fan, and generate a control signal wind speed fitting model and a control signal prediction model; Control signal prediction module: used to generate target wind speed time series and predict control signal time series; Fan multi-point control module: includes multiple fan controllers, each of which caches the predicted control signal time series issued in batches and outputs the initial multi-point control signal; Local synchronous execution module: includes a local timer, which is used to trigger the synchronous trigger instruction at a preset time point under a unified time base; Wind speed feedback adjustment module: includes a wind speed sensor, which is used to measure the actual wind speed in real time and feed it back to the fan multi-point control module. The fan multi-point control module outputs a dynamic multi-point control signal.

Images