Air volume control method for multiple air blowers connected in parallel

Abstract

The invention relates to the technical field of non-variable-capacity pump set parallel control, and discloses a multi-blower parallel air volume control method, which comprises the following steps of: acquiring the real-time rotating speed and the real-time input power of a driving frequency converter of each single blower in a parallel system, and superposing a periodic rotating speed perturbation signal at a common pressure given value to obtain a multi-blower parallel air volume control result; power response amplitudes of all single blowers are extracted synchronously, the change rate of real-time power relative to the rotating speed is calculated to generate a power response gradient representing the energy conversion characteristic of the unit, the unbalanced blowers are recognized according to deviation parameters of the power response gradient relative to the global average value, the rotating speed instruction gain of the blowers is reduced, and the unit energy conversion characteristic is obtained. According to the method, load distribution is converted into an active adjustment mechanism based on the physical characteristics of the impeller, accurate identification of the offset of the physical operation sites of the parallel unit is achieved, and the accuracy of the offset of the physical operation sites of the parallel unit is improved on the premise that branch flow sensing hardware is not needed. And the recessive surge risk caused by single-machine performance discretization is inhibited.

Description

Technical Field

[0001] This invention relates to a method for controlling the air volume of multiple blowers in parallel, belonging to the field of parallel control technology for non-variable displacement pump sets. Background Technology

[0002] Currently, operating multiple blowers in parallel is a common method to maintain stable pressure and air volume in large-scale systems such as industrial aeration and municipal sewage treatment. Existing technologies mostly adopt closed-loop feedback regulation logic based on the pressure of the main pipe, and send a unified speed command to each frequency converter in the parallel group through the controller to achieve proportional distribution of air volume or static deviation compensation.

[0003] However, non-variable displacement pump sets are highly sensitive to fluid kinetic energy conversion. The operational stability of parallel systems is constrained by the discrete physical characteristics of individual units. In actual engineering, the resistance of each branch pipeline is asymmetrical due to installation errors. Due to different operating times, each unit experiences physical performance degradation such as impeller scaling and mechanical wear, resulting in performance differences within the parallel array. Traditional pressure feedback mechanisms only capture the energy balance at the system level and cannot reflect the flow fluctuations within the branches. When the system load decreases, the degraded units are often the first to enter the rotating stall or surge critical region, causing high-frequency pulsation and mechanical fatigue. Moreover, this process is often masked by stable main pipe pressure data. Even if branch flow meters or dynamic pressure sensors are added, they will be limited by the measurement drift of the sensors in high-temperature, dusty, and humid media environments, increasing the risk of control logic failure.

[0004] Therefore, how to utilize the original electrical signals of the generating units to monitor the characteristic gradient response of each unit, and to achieve adaptive load distribution based on the actual physical carrying capacity of each unit, is the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A method for controlling the air volume of multiple blowers in parallel, the method comprising the following steps:

[0006] Step S1: Collect the real-time speed and real-time input power of each blower drive frequency converter in the parallel system;

[0007] Step S2: A periodic speed micro-perturbation signal is superimposed on the common pressure setpoint of the parallel system. The frequency of the speed micro-perturbation signal is lower than the mechanical resonance frequency of the parallel system and its amplitude satisfies that the transient fluctuation of the main pipe pressure is lower than the preset pressure fluctuation threshold.

[0008] Step S3: Simultaneously extract the power response amplitude of each individual blower under the action of speed micro-disturbance signal. By calculating the rate of change of real-time input power relative to real-time speed, generate the power response gradient characterizing the energy conversion characteristics of each individual blower under the current operating condition. The power response gradient corresponds to the first derivative of the impeller power characteristic curve of the non-variable displacement pump.

[0009] Step S4: Calculate the arithmetic mean of the power response gradient of each blower operating online, and calculate the difference between each power response gradient and the arithmetic mean to obtain the relative deviation parameter of each blower.

[0010] Step S5: Identify the single blower whose relative deviation parameter exceeds the preset consistent working range as an unbalanced blower, reduce the speed command gain of the unbalanced blower, so as to transfer the load demand to the other blowers whose relative deviation parameter is within the consistent working range, so that the working point of each single blower converges to the consistent working range.

[0011] Preferably, the speed micro-perturbation signal of each individual blower is triggered by time-division polling; step S3 specifically includes: each individual blower sequentially performs speed micro-perturbation and synchronously calculates the corresponding power response gradient within a preset independent detection window; during non-detection window periods, the value of the corresponding power response gradient is locked to isolate the hydrodynamic coupling interference between parallel branches.

[0012] Preferably, step S4 further includes: determining that when the power response gradients of all online blowers increase synchronously and the relative deviation parameters are all within the consistent operating range, the current operating condition change originates from the change in pipeline resistance, and maintaining the current load distribution weight.

[0013] Preferably, the method further includes the following steps: during the application of the speed micro-perturbation signal, the real-time torque ripple signal of the drive motor is acquired synchronously; the spectral energy distribution information entropy of the real-time torque ripple signal is calculated to identify the rotational stall characteristics at the corresponding blower impeller; when the information entropy is lower than a preset critical threshold, a safety margin correction parameter is generated and the corresponding load distribution weight is reduced, so that the operating position of the corresponding single blower moves away from the surge zone.

[0014] Preferably, the method further includes the following steps: detecting the phase lag angle of the real-time input power response to the speed micro-disturbance signal; comparing the response phase lag angle with a preset flow direction discrimination threshold; when the response phase lag angle exceeds the flow direction discrimination threshold, determining that the corresponding single blower has a fluid backflow trend, and increasing the lower limit of the blower's speed output to generate endogenous static pressure resistance against the main pipe pressure.

[0015] Preferably, the method further includes the following steps: synchronously collecting the real-time active power fluctuation and real-time reactive power fluctuation of each individual blower drive motor; calculating the ratio between the real-time active power fluctuation and the real-time reactive power fluctuation. : ,in, This represents the real-time fluctuation of active power. This represents the real-time reactive power fluctuation; when the proportional relationship is... When the fluid load response range is within the preset range determined by the motor torque angle characteristics, step S5 is executed.

[0016] Preferably, the method further includes the following steps: extracting the envelope attenuation rate of the power response amplitude after the speed micro-disturbance signal stops; comparing the envelope attenuation rate with a preset single-machine health benchmark value; and generating an impeller fouling degradation warning for the corresponding blower when the deviation of the envelope attenuation rate relative to the single-machine health benchmark value exceeds a preset safety deviation.

[0017] Preferably, the frequency of the speed micro-perturbation signal is set to Hz to Hz, and the amplitude of the speed perturbation signal is set to the rated speed of the corresponding single blower. to .

[0018] Preferably, the method further includes the following steps: real-time detection of the fluctuation variance of the main pipe pressure of the parallel system; when the fluctuation variance exceeds a preset stability threshold, reducing the amplitude of the speed micro-disturbance signal and extending the triggering period of the detection window until the fluctuation variance recovers to within the stability threshold.

[0019] Preferably, the method further includes the following steps: recording the offset trajectory of the power response gradient of each individual blower relative to the operating time, and establishing a single-blower efficiency evolution model based on the offset trajectory; and using the single-blower efficiency evolution model to pre-correct the initial load allocation weight of each individual blower under different total load demands.

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] 1. In the parallel air volume control of multiple blowers, by applying periodic micro-disturbance signals to the common control target and combining real-time speed and power response to calculate the characteristic gradient factor, the traditional passive load distribution based on the main pipe pressure is transformed into an active distribution method based on the energy conversion efficiency characteristics of a single unit. This mechanism directly maps the physical characteristics of non-variable displacement pumps, enabling the controller to sense the differences in work capacity of each unit caused by impeller wear, scaling, and other factors, and guide the operating point to converge towards the consistent safe range, thus avoiding the hidden surge risk caused by the traditional proportional distribution when the performance of a single unit deteriorates.

[0022] 2. This invention utilizes the synergistic effect of time-domain orthogonal arrangement and response phase lag monitoring to eliminate hydrodynamic coupling interference caused by multiple units in parallel. The non-overlapping time-domain disturbance triggering ensures the purity of feature gradient extraction and solves the signal intermodulation problem in large and medium-sized unit arrays. At the same time, through the instantaneous determination of fluid backflow trend by phase lag, the system can generate endogenous static pressure resistance at the moment of main pipe load fluctuation, realizing millisecond-level flow direction protection without physical check valve feedback, and enhancing the system's operational safety under transient conditions.

[0023] 3. By comparing the global mean with the envelope attenuation analysis of the power feedback signal, the system achieves decoupled diagnosis of systemic resistance drift and individual unit degradation. The system compares the characteristic gradient factor of a single unit with the global average value, distinguishes between external environmental disturbances such as changes in the opening degree of pipeline valves and the decline in the physical performance of a single unit. Combined with the energy dissipation mode characterized by the envelope attenuation rate, the system achieves online assessment of the health of branch hardware without stopping the production process, reducing unnecessary regulation oscillations and subsequent maintenance costs. Attached Figure Description

[0024] Figure 1 This is a flowchart of the parallel blower air volume control based on the power response gradient of the present invention;

[0025] Figure 2 This is a graph showing the evolution of power response gradient and deviation parameters under different impeller fouling conditions according to the present invention. Detailed Implementation

[0026] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0027] This invention provides a method for controlling the airflow of multiple blowers operating in parallel. During the operation of a parallel system, the fluid kinetic energy conversion of a non-variable displacement pump is constrained by the discretization of the impeller's physical characteristics. Traditional main pipe pressure feedback regulation logic cannot reflect the asymmetry of branch resistance caused by installation errors or operational wear. The solution of this invention involves collecting the real-time speed and real-time input power of the frequency converters driving each individual blower in the parallel system, where the real-time speed is denoted as... The unit is r / min, and the real-time input power is denoted as... The unit is kW; by establishing a stable electrical signal acquisition link, the system acquires raw physical samples reflecting the impeller's working state, eliminating the dependence on branch flow hardware sensing devices; static operating data is affected by fluctuations in pipeline resistance, making it difficult to directly extract characteristic quantities representing the working characteristics of a single unit; to address this challenge, the system superimposes periodic speed micro-perturbation signals onto the common pressure setpoint of the parallel system, with the frequency of the speed micro-perturbation signals set at... Hz to Within the Hz range, and its amplitude is set to the rated speed of the corresponding single blower. to The frequency of the speed micro-disturbance signal is lower than the mechanical resonance frequency of the parallel system, and its amplitude satisfies the requirement that the transient fluctuation of the main pipe pressure is lower than the preset pressure fluctuation threshold. This allows for the generation of a controlled power characteristic response without affecting the stability of the production process. Fluid dynamic coupling in a multi-machine parallel system can cause interference between disturbance signals in different branches. The system employs a time-domain orthogonal allocation method, enabling time-division polling triggering of the speed micro-disturbance signal for each individual blower. Each blower sequentially performs speed micro-disturbance within a preset independent detection window and locks the current power response gradient value during non-detection windows. By arranging the detection behavior in a non-overlapping time domain, fluid dynamic coupling interference between parallel branches is isolated, ensuring the physical purity of the extracted features. Under a defined pressure boundary, impellers in different physical states exhibit varying sensitivities to speed changes. The system synchronously extracts the power response amplitude of each individual blower under the action of the speed micro-disturbance signal and calculates the real-time input power. Relative to real-time speed The rate of change of , generating the power response gradient Power response gradient The calculation satisfies the relation Power response gradient Corresponding to the first derivative of the impeller power characteristic curve of a non-variable displacement pump, it is used to characterize the load response sensitivity of a single blower under current operating conditions.

[0028] External environmental disturbances such as changes in pipeline resistance can cause synchronous shifts in the characteristic gradients of each unit in a parallel system. To address this challenge, the system calculates the power response gradient of each individual blower operating online. arithmetic mean and each power response gradient Compared with the arithmetic mean respectively Perform the difference calculation to obtain the relative deviation parameter. Relative deviation parameter Satisfying the relation When determining the power response gradient of all online operating blowers Synchronous increase and relative deviation parameter When all loads are within the preset consistent operating range, the system determines that the current operating condition change originates from a change in pipeline resistance and maintains the current load allocation weight. Units with degraded physical performance are at risk of being the first to slip into the surge zone under low load conditions. Therefore, the system identifies the relative deviation parameter. A single blower exceeding the preset consistent operating range is considered an unbalanced blower, and the speed command gain of the unbalanced blower is reduced; this is achieved by transferring load demand to the relative deviation parameter. For the remaining blowers operating within the consistent operating range, the system drives the operating points of each individual blower to converge towards the consistent operating range. This adjustment method achieves active load distribution based on the energy conversion efficiency characteristics of individual blowers, suppressing the risk of latent surge. The electromagnetic environment in industrial settings can affect the accuracy of power signal measurements; therefore, the system synchronously collects the real-time active power fluctuations of the drive motors of each individual blower. With real-time reactive power fluctuation ; Calculate real-time active power fluctuations With real-time reactive power fluctuation proportional relationship ,in ,in, This represents the real-time active power fluctuation, in kW. This represents the real-time reactive power fluctuation, measured in kvar; when the proportional relationship... When the system is within the preset fluid load response range determined by the motor torque angle characteristics, it allows the system to perform load distribution weight adjustment steps, thereby filtering out non-physical electrical noise interference.

[0029] The tendency for fluid backflow in parallel branches can lead to instability in the unit's operating state, and the system will detect the real-time input power. Response phase lag angle to speed micro-disturbance signals ; The response phase lag angle The response phase lag angle is compared with the preset flow direction discrimination threshold. When the flow direction discrimination threshold is exceeded, it is determined that the corresponding single blower has a fluid backflow trend. The system increases the lower limit of the blower's speed output to generate static pressure resistance against the main pipe pressure. Impeller scaling or wear will cause changes in mechanical energy dissipation characteristics. The system extracts the envelope attenuation rate of the power response amplitude after the speed micro-disturbance signal stops. ; envelope attenuation rate Compare with the preset single-machine health benchmark value, when the envelope attenuation rate When the deviation from the single-unit health baseline value exceeds the preset safety deviation, the system generates an early warning for impeller scaling and degradation of the corresponding blower, realizing online assessment of the single-unit hardware health; and monitors the fluctuation variance of the main pipe pressure in the parallel system in real time. When the variance of the fluctuation When the preset stability threshold is exceeded, the amplitude of the speed micro-perturbation signal is reduced, and the trigger period of the detection window is extended until the fluctuation variance is reduced. It has recovered to within the stability threshold.

[0030] Example 1: In a large-scale wastewater treatment aeration system comprising four centrifugal blowers operating in parallel, the resistance of each branch pipeline is asymmetrical due to installation errors, and the impellers of each blower exhibit varying degrees of scaling or wear due to differences in operating time. When the aeration load demand decreases, if the controller sends a uniform speed command to all units, the units with degraded physical performance tend to be the first to slip into the rotational stall zone, leading to high-frequency surge in a single unit and causing mechanical fatigue of the blades. To address this challenge, the system superimposes a frequency of [frequency value missing] into the common pressure setpoint. Hz and amplitude at rated speed The speed micro-perturbation signal is used; the speed micro-perturbation signal is orthogonally triggered in the time domain among the four units through perturbation tokens allocated by the controller, so that each blower only superimposes the perturbation within its own independent detection window, and locks the current characteristic value during non-detection windows; the system synchronously collects real-time speed and real-time input power, and calculates the power response gradient of a single blower. The power response gradient satisfies the calculation formula. ,in, This is the power response gradient, which corresponds to the first derivative of the impeller power characteristic curve of a non-variable displacement pump. This is the real-time input power, in kW. The speed is the real-time rotational speed, expressed in r / min; to isolate the interference of pipeline resistance drift on the single-unit state determination, the system calculates the arithmetic mean of the power response gradients of all online blowers in the parallel group in real time, denoted as . For each blower, the system calculates the relative deviation parameter of its power response gradient relative to the global average. That is, satisfying the relation ,in, This is a relative deviation parameter used to quantify the degree of deviation between the performance of a single unit and the overall performance of the parallel group. The arithmetic mean; when the relative deviation parameter of an unbalanced blower is identified. Exceeding the consistent operating range, and real-time input power Response phase lag angle to speed micro-disturbance signals When the flow direction threshold is exceeded, it is determined that the branch has a tendency for fluid backflow. The phase lag angle, measured in rad, reflects the degree of reverse impact of fluid potential energy on the impeller.

[0031] To resolve the conflict between operational safety and load balance, the controller automatically reduces the speed command gain of the unbalanced blower, dynamically transferring load demand to other blowers whose relative deviation parameters are within the consistent operating range. Simultaneously, the system calculates real-time active power fluctuations while adjusting the weights. With real-time reactive power fluctuation proportional relationship It satisfies the relation ,in, For proportional relationships, This represents the real-time active power fluctuation, in kW. This represents the real-time reactive power fluctuation, measured in kvar; when the proportional relationship... When the system is within the physical load response range determined by the motor torque angle characteristics, it confirms that the current gradient change originates from the impeller fluid response rather than grid harmonic interference. Through this adaptive allocation mechanism, the operating points of each unit in the parallel group converge towards the consistency safety range, achieving instant arbitration of the flow stability of each branch without adding external physical check valve status feedback and branch flow meters. The parallel system, which was originally limited by the discrete performance of individual units, generates endogenous static pressure resistance by adjusting the lower limit of the speed output at the moment of load fluctuation, blocking the backflow of fluid to the performance degradation branch.

[0032] Example 2: To verify the accuracy of identifying performance degradation caused by scale buildup on a single impeller in a parallel system, the experiment was conducted on a physical test platform containing four centrifugal blowers. The physical test platform was equipped with a 200kW variable frequency drive motor, and the accompanying data acquisition system had a power resolution of 0.1kW and a speed measurement accuracy of 1r / min, with a sampling frequency set to 100Hz. This experiment addressed the asymmetric branch resistance caused by installation errors, simulating different network resistance coefficients by adjusting the opening of the manual butterfly valves in each branch. The sampling period was set with the technical consideration of balancing the real-time performance of response feature extraction with the execution frequency of the control algorithm. To capture the dynamic response triggered by minor speed disturbances, the sampling period should be shorter than the speed disturbance. One-tenth of the signal period; under the experimental conditions, the frequency of the speed micro-disturbance signal was set to 0.5Hz, so the sampling period was set to 20ms; the amplitude setting of the speed micro-disturbance signal needs to balance the signal-to-noise ratio of the extracted signal with the stability of the main pipe pressure; when the amplitude is lower than 0.5% of the rated speed of the corresponding single blower, the fluctuation variance of the power response gradient will increase due to the sensor measurement noise and fluid pulsation interference; after multiple calibrations, 1% of the rated speed was selected as the amplitude of the speed micro-disturbance signal in this experiment; after the test started, periodic speed micro-disturbance signals were applied to four blowers under different maintenance conditions; the test group simulated four physical state gradients of impeller without scaling, slight scaling, moderate scaling and heavy scaling; the system synchronously collected the real-time speed of each unit. With real-time input power The power response gradient of each unit is calculated according to the gradient solution procedure described above. To simulate signal interference in a real industrial environment, a Gaussian white noise with a signal-to-noise ratio of 20dB was actively superimposed on the test signal source.

[0033] Table 1: Power characteristic response data under different impeller fouling conditions

[0034]

[0035] Referring to Table 1, the power response gradient increases with the degree of impeller fouling. The trend shows a monotonically decreasing trend, reflecting the decline in the impeller's work capacity; when the relative deviation parameters of heavily scaled units... When the value reaches -0.40, it exceeds the lower limit of the preset consistency working range; the system then detects that the response phase lag angle of the unit exceeds the preset flow direction discrimination threshold, triggering the load weight correction procedure; in order to verify the anti-interference of the deep coupling verification procedure of electromagnetic and flow characteristics, the experiment introduces the fifth harmonic interference with a frequency of 250Hz at the power input end; without the processing of the present invention, the pseudo power fluctuation caused by the interference causes the gradient calculation value to shift by more than 15%.

[0036] Table 2: Comparison of Judgment Data Before and After Introducing Active-Reactive Power Ratio Verification

[0037]

[0038] Referring to Table 2, the present invention utilizes real-time active power fluctuations. With real-time reactive power fluctuation proportional relationship As a criterion for physical properties; the calculated proportional relationship when subjected to the fifth harmonic interference of the power grid. The system deviates from the preset fluid load response range determined by the motor torque angle characteristics, and locks the load distribution weight to avoid malfunctions caused by non-physical fluctuations. Further testing shows that when the amplitude of the speed micro-disturbance signal increases to 3% of the rated speed, the transient fluctuation of the main pipe pressure exceeds the process safety threshold of 5 kPa.

[0039] Example 3: This example combines Figures 1 to 2 This document describes a method for controlling the air volume of multiple blowers connected in parallel, such as... Figure 1 As shown, the process includes: Step S1: Collect the real-time speed and real-time input power of the frequency converter driven by each blower in the parallel system, and establish an electrical signal acquisition link to obtain physical samples reflecting the working state of the impeller; Step S2: Superimpose periodic speed micro-disturbance signals with a frequency lower than the mechanical resonance frequency and controlled amplitude into the common pressure setpoint of the parallel system, and excite the power characteristic response under the premise of maintaining the stability of the main pipe pressure; Step S3: Synchronously extract the power response amplitude of each blower under micro-disturbance and calculate the rate of change of real-time power relative to speed to generate a power response gradient characterizing the energy conversion characteristics of the unit; Step S4: Calculate the arithmetic mean of the power response gradient of the blowers in online operation and perform a difference operation between the individual machine gradient and the mean to obtain the relative deviation parameter that quantifies the degree of deviation of the individual machine performance; Step S5: Identify the unbalanced blowers whose relative deviation parameters exceed the preset consistent working range and reduce their speed command gain to drive the working point of each individual blower to converge to the consistent working range to suppress latent surge.

[0040] like Figure 2 As shown, the horizontal axis, from left to right, represents five physical evolution stages: no scaling, slight scaling, moderate scaling, heavy scaling, and extreme scaling. The left vertical axis corresponds to the power response gradient G, and the right vertical axis corresponds to the relative deviation parameter D. The legend clearly marks the three types of data trajectories, with the solid line representing the power response gradient G and the dashed line representing the arithmetic mean. The dotted lines represent the relative deviation parameter D. Observing the curve trend, it can be seen that as the impeller condition deteriorates from no scaling to extreme scaling, the solid line representing the power response gradient G shows a significant monotonically decreasing trend, while the line representing the arithmetic mean... The dashed line remains at a relatively stable level. The difference between the two causes the dotted line representing the relative deviation parameter D to drop rapidly from the positive range to the negative range. Thus, the health status of the equipment and the load distribution requirements can be clearly defined through the numerical boundary of parameter D.

[0041] Example 4: In an aeration system of a municipal wastewater treatment plant containing three centrifugal blowers operating in parallel, the energy conversion characteristics of blower No. 3 deviate asymmetrically from the other blowers due to high-order harmonic interference from the motor and uneven scale buildup on the impeller surface. To identify the unbalanced blower, the system collects data on each blower at a period of... Real-time speed within the micro-disturbance period of s With real-time input power The controller is used to calculate the power response gradient of each unit. arithmetic mean and sample standard deviation Sample standard deviation The calculation satisfies the first relation. ,in, For the first The power response gradient of the blower. This refers to the total number of blowers operating online, and in this embodiment... for The system defines the consistent working interval as... Centered on twice the sample standard deviation The numerical domain of the radius, i.e., the interval boundary satisfies When the relative deviation parameter of blower No. 3 Continuous When the data acquisition cycle exceeds the consistent operating range, the system executes a load weight correction procedure; the reduction in speed command gain is determined by the deviation ratio, i.e., by calculating the gain correction operator. The command to adjust the speed of blower No. 3, and the gain correction operator. Satisfying the second relation ,in, The preset adjustment coefficient is set to [value] in this embodiment. This is used to smoothly transfer part of the load demand of blower No. 3 to the relative deviation parameter. Blower No. 1 and Blower No. 2 are within the consistent operating range.

[0042] To detect the flow stability of blower No. 3 under low-load conditions, the system synchronously acquires the real-time torque ripple signal of the drive motor through the inverter control bus while applying a speed micro-disturbance signal, and calculates the spectral energy distribution entropy using a spectrum analysis algorithm. The system sets the duration of the sliding observation window to [duration]. s, sampling frequency is Hz, perform a Fast Fourier Transform on the acquired torque fluctuation sequence to obtain the frequency range in the range of Hz. Hz to Power spectrum distribution within Hz; the system calculates the energy proportion of each frequency component. The logarithmic weighted sum yields the information entropy of the spectral energy distribution. Spectral energy distribution information entropy The calculation satisfies the third relation. ,in, For the first The energy percentage at each frequency point This represents the total number of sampling points in the spectrum; when the spectral energy distribution tends to concentrate at a specific stall frequency and the spectral energy distribution information entropy... Below the preset critical threshold At that time, the system determined that there was a rotational stall characteristic at the corresponding blower impeller; the system then generated a safety margin correction parameter, which further reduced the speed command gain. This forces the operating point of blower No. 3 to shift away from the surge zone; through this adaptive adjustment method, the power response gradient of blower No. 3 is... Finally returning to Within the numerical domain, the global energy efficiency balance of the parallel system was maintained under the condition of single-machine performance degradation.

[0043] Example 5: In a system containing three units with a rated speed of For newly built aeration stations with blower units operating at r / min, pre-deployment calibration is performed on-site. The system performs frequency range calibration for each individual unit within the specified range. Hz to Hz and step frequency is Hz sweep frequency detection, by acquiring the high-frequency pulsation component of the drive motor bus current, calculates the information entropy of the torque ripple spectrum energy distribution under different speed micro-disturbance frequencies. , To identify the entropy value used for identifying impeller rotational stall characteristics, the frequency point where no rotational stall cluster characteristics appear and the torque response fluctuation is minimized is established as the frequency of the speed micro-disturbance signal under this field environment, which will then be used to generate the power response gradient. Provides an excitation load with a high signal-to-noise ratio, wherein, It characterizes the energy conversion characteristics of each individual blower under the current operating conditions, and realizes the physical matching between the working characteristics of the individual blower and the characteristics of the on-site pipeline network.

[0044] To determine the dynamic boundaries of the consistent operating range, the system performs baseline calibration within the steady-state operating window of the parallel system, and the controller... ms is the sampling interval for continuous recording. Group power response gradient The calculation sample is obtained by calculating the arithmetic mean of the sample set. with sample standard deviation Simultaneously monitor the transient fluctuations in the main pipe pressure. , This represents the average power response gradient level of the online operating units. Indicates the degree of dispersion of a sample from the mean. This represents the variance of the pressure fluctuation in the main pipe, where the transient fluctuation of the main pipe pressure is... The measured variance exceeds kPa will reduce the amplitude of the speed perturbation signal from the rated speed. Downgraded to And repeat the baseline acquisition process until the sample standard deviation is reached. The values ​​tend to converge, and the resulting range of values ​​twice the standard deviation is used as the boundary for identifying unbalanced blowers.

[0045] Example 6: When performing load commissioning on a parallel system consisting of four centrifugal blowers, in order to determine the flow direction discrimination threshold in the phase lag monitoring procedure... The system maintains the pressure in the public pipeline network at the rated pressure. In a steady-state environment, the backflow point simulation during the deceleration process of a single blower is performed by superimposing a frequency of [frequency value] before the blower speed decreases to the closing point of the branch check valve. The Hz speed micro-perturbation signal is recorded and the phase offset value of the real-time input power is recorded, where, The phase angle limit used to determine the tendency of fluid to flow back into the branch is expressed in rad. The system is set with a rated main pipe pressure value in kPa. The system extracts the angular displacement sequence where the power response phase lags behind the speed disturbance command, and establishes the arithmetic mean of the phase angles of the last three sampling periods before the check valve actuates as the flow direction discrimination threshold. This provides a quantitative criterion for generating static pressure resistance against the main pipe pressure by increasing the lower limit of the output speed.

[0046] During the pre-calibration process of performing feature gradient extraction in the parallel system, a stability threshold for the variance of the main pipe pressure fluctuation is calibrated. The system continuously collects data within a window when all online blowers are operating at their rated conditions and free from external interference. The real-time feedback value of the main pipe pressure is used to calculate the standard deviation of the pressure sample set, and its squared value is defined as the background pressure pulsation variance. To determine whether the system pressure is within the critical variance limit for abnormal fluctuations, the unit is... The system will calculate the variance of background pressure fluctuations. Multiple values ​​as stability thresholds The configuration table of the controller is written to record the variance of the main pipe pressure fluctuation during the operation of the parallel system. The stability threshold was exceeded for five consecutive sampling periods. At that time, the system triggers an amplitude reduction command to reduce the amplitude of the speed micro-disturbance signal until the fluctuation variance of the main pipe pressure is reduced. Restored to stability threshold Within.

[0047] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for controlling the air volume of multiple blowers connected in parallel, characterized in that, The method includes the following steps: Step S1: Collect the real-time speed and real-time input power of each blower drive frequency converter in the parallel system; Step S2: A periodic speed micro-perturbation signal is superimposed on the common pressure setpoint of the parallel system. The frequency of the speed micro-perturbation signal is lower than the mechanical resonance frequency of the parallel system and its amplitude satisfies that the transient fluctuation of the main pipe pressure is lower than the preset pressure fluctuation threshold. Step S3: Simultaneously extract the power response amplitude of each individual blower under the action of speed micro-disturbance signal. By calculating the rate of change of real-time input power relative to real-time speed, generate the power response gradient characterizing the energy conversion characteristics of each individual blower under the current operating condition. The power response gradient corresponds to the first derivative of the impeller power characteristic curve of the non-variable displacement pump. Step S4: Calculate the arithmetic mean of the power response gradient of each blower operating online, and calculate the difference between each power response gradient and the arithmetic mean to obtain the relative deviation parameter of each blower. Step S5: Identify the single blower whose relative deviation parameter exceeds the preset consistent working range as an unbalanced blower, reduce the speed command gain of the unbalanced blower, so as to transfer the load demand to the other blowers whose relative deviation parameter is within the consistent working range, so that the working point of each single blower converges to the consistent working range.

2. The method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, The speed micro-perturbation signal of each individual blower is triggered by time-division polling; step S3 specifically includes: each individual blower sequentially performs speed micro-perturbation and synchronously calculates the corresponding power response gradient within a preset independent detection window; during non-detection window periods, the value of the corresponding power response gradient is locked.

3. The method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, Step S4 also includes: determining that when the power response gradient of all online blowers increases synchronously and the relative deviation parameters are all in the consistent operating range, the current operating condition change is due to the change in pipeline resistance, and maintaining the current load distribution weight.

4. The method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, The method also includes the following steps: during the application of the speed micro-perturbation signal, the real-time torque ripple signal of the drive motor is acquired synchronously; the spectral energy distribution information entropy of the real-time torque ripple signal is calculated to identify the rotational stall characteristics at the corresponding blower impeller; when the information entropy is lower than the preset critical threshold, a safety margin correction parameter is generated and the corresponding load distribution weight is reduced, so that the operating position of the corresponding single blower moves away from the surge zone.

5. The method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, The method also includes the following steps: detecting the phase lag angle of the real-time input power response to the speed micro-disturbance signal; comparing the response phase lag angle with a preset flow direction discrimination threshold; when the response phase lag angle exceeds the flow direction discrimination threshold, determining that the corresponding single blower has a fluid backflow trend, and increasing the lower limit of the blower's speed output to generate endogenous static pressure resistance against the main pipe pressure.

6. The method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, The method also includes the following steps: synchronously collecting the real-time active power fluctuation and real-time reactive power fluctuation of each individual blower drive motor; and calculating the ratio between the real-time active power fluctuation and the real-time reactive power fluctuation. : ,in, This represents the real-time fluctuation of active power. This represents the real-time reactive power fluctuation; when the proportional relationship is... When the fluid load response range is within the preset range determined by the motor torque angle characteristics, step S5 is executed.

7. The method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, The method also includes the following steps: extracting the envelope attenuation rate of the power response amplitude after the speed micro-disturbance signal stops; comparing the envelope attenuation rate with a preset single-machine health benchmark value; and generating an impeller fouling degradation warning for the corresponding blower when the offset of the envelope attenuation rate relative to the single-machine health benchmark value exceeds a preset safety deviation.

8. The method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, The frequency of the speed micro-disturbance signal is set to Hz to Hz, and the amplitude of the speed perturbation signal is set to the rated speed of the corresponding single blower. to .

9. The method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, The method also includes the following steps: real-time detection of the fluctuation variance of the main pipe pressure of the parallel system; when the fluctuation variance exceeds the preset stability threshold, reducing the amplitude of the speed micro-disturbance signal and extending the triggering period of the detection window until the fluctuation variance recovers to within the stability threshold.

10. A method for controlling the air volume of multiple blowers in parallel according to claim 1, characterized in that, The method also includes the following steps: recording the offset trajectory of the power response gradient of each individual blower relative to the operating time, and establishing a single-unit efficiency evolution model based on the offset trajectory; using the single-unit efficiency evolution model to pre-correct the initial load allocation weight of each individual blower under different total load demands.

Images