Self-adaptive active-passive combined air supplement control method for cooling tower

Abstract

The invention discloses a self-adaptive active-passive combined air supplement control method for a cooling tower. The self-adaptive active-passive combined air supplement control method comprises the following steps: S1, data acquisition and index calculation; s2, carrying out digital twinning-CFD parameter representation and working condition identification; s3, target setting and adaptive scheduling are carried out; and S4, active-passive combined air supplement adjustment is carried out. Compared with a traditional constant air adding or single air guiding scheme, under the working conditions of crosswind and hot backflow, the air inlet uniformity is remarkably improved, UI can be stably smaller than or equal to 0.15 (within the preferable range), and a tower circumference reverse pressure / backflow strip is remarkably weakened; the performance of the cold end is improved: delta Tcool is greater than or equal to 0.8-2.0 DEG C, and the unit vacuum is improved; the energy consumption is controlled: the passive domain is preferentially utilized on the premise of reaching the target, and the energy consumption of the fan does not exceed the set Plim; frequent switching and abnormal vibration are avoided through hysteresis + constraint + self-protection logic, and engineering landing is facilitated.

Description

Technical Field

[0001] This application relates to an adaptive active-passive combined air supply control method for cooling towers. Background Technology

[0002] Natural draft cooling towers are prone to uneven airflow around the tower, localized back pressure, and "hot air strip" recirculation under conditions such as crosswinds, heat recirculation, and non-uniform perimeter airflow. This leads to uneven packing wetting, deteriorated heat exchange, decreased unit vacuum, and increased outlet water temperature. Existing solutions mostly adopt single air guides or constant air supply schemes, lacking linkage and adaptive and active-passive coordination based on multiple variables such as wind speed, wind direction, tower internal pressure difference, and outlet water temperature. Especially in scenarios with sudden changes in wind direction and heat recirculation from adjacent towers, control is lagging, energy consumption is high, and stability is insufficient. Therefore, it is necessary to propose a closed-loop method and system that can automatically switch and allocate between "passive air supply (air guide / direction)" and "active air supply (fan-assisted air supply)" according to operating conditions, so as to stabilize the air volume and uniformity around the tower and reduce the outlet water temperature while ensuring controllable energy consumption. Summary of the Invention

[0003] To address the aforementioned issues, this application proposes an adaptive active-passive combined air supply control method for cooling towers, comprising the following steps:

[0004] S1. Data collection and indicator calculation;

[0005] S2. Digital Twin - CFD Parameter Characterization and Operating Condition Identification;

[0006] S3. Goal setting and adaptive scheduling;

[0007] S4. Perform active-passive combined air supply adjustment. Compared with traditional constant air supply or single air guide schemes, this application achieves the following under crosswind and hot recirculation conditions: significantly improved air intake uniformity: UI can be stably ≤0.15 (within the preferred range), and significantly reduced back pressure / recirculation strips around the tower; improved cold end performance: ΔT cool Achieve ≥0.8–2.0℃, improve unit vacuum; control energy consumption: prioritize the use of passive domains while meeting targets, and ensure that fan energy consumption does not exceed the set P. lim Improved stability: Hysteresis, constraint, and self-protection logic avoid frequent switching and abnormal vibrations, making it easier to implement in engineering projects.

[0008] Preferably, it also includes processes for constraint and fault self-protection, as well as online evaluation and self-optimization.

[0009] Preferably, the data collection and indicator calculation are performed in the following manner:

[0010] S1.1 acquires the following data via a sensor array: inflow velocity u, wind direction θ, relative humidity RH, and dry / wet bulb temperature T. db / T wb The static pressure p at each measuring point along the circumference of the tower skirt i Speed ​​v i Pressure difference Δp in the lower part of the tower in-out and circulating water outlet temperature T out Inlet water temperature T in ;

[0011] S1.2 Calculate the circumferential airflow non-uniformity coefficient UI and the airflow volume Q in :

[0012] Where UI = σ(v) i) / μ(v i ), i∈{1…N};

[0013] σ(v i —Standard deviation of wind speed at various measuring points around the cooling tower;

[0014] μ(v i — The average wind speed at each measuring point around the cooling tower;

[0015] i — represents the i-th measurement point;

[0016] N represents the number of circumferential measurement points;

[0017] It reflects the relative dispersion of wind speed: if the wind speeds at all measuring points are similar, the standard deviation σ is small, and the UI value approaches 0; conversely, when the differences are large, the UI value increases.

[0018] Q in =∑(v i ·A i );

[0019] Q in —Total air volume at the cooling tower inlet, in m³ 3 / s;

[0020] v i —The average wind speed at the i-th measuring point or air inlet unit, in m / s;

[0021] A i —The area of ​​the i-th measuring point (or unit), in m² 2 ;

[0022] Σ — represents the summation of the areas of all measuring points or zones;

[0023] This formula is the basic expression for the conservation of airflow and is used to calculate the total airflow around the cooling tower. If multiple air inlets or zones are arranged around the tower, the total airflow at the entire cooling tower inlet is obtained by multiplying the wind speed of each zone by the area of ​​that zone and then summing all the zones. In actual systems, Qin is one of the control objectives to ensure that the airflow into the cooling tower is sufficient and uniform.

[0024] S1.3 calculates the water temperature drop ΔT cool =T in -T out And estimate the heat reflux criterion R. rec =(T rim -T amb ) / T amb The threshold value of turbulent kinetic energy k* in the near-tower region is k*>0.

[0025] Preferably, N≥8.

[0026] Preferably, the digital twin-CFD parameter characterization and operating condition identification are performed in the following manner:

[0027] S2.1 Utilizing offline CFD-model calibration, a parameter characterization set Π={α(θ),β(u),γ(k),…} is obtained, which is used to estimate the effect of different guide angles / openings and makeup air volumes on UI and ΔT. cool Response sensitivity S UI S T ;

[0028] Π represents the control law or set of control parameters. It is a mathematical abstraction of the entire adaptive supplementary air control method and encompasses multiple control factors.

[0029] α(θ) is the active / passive air supply ratio function, which depends on the wind direction angle θ; meaning: when the external wind direction changes, the opening ratio of the active fan and the passive guide valve is adjusted by α(θ) to maintain uniform air intake;

[0030] β(u) is the make-up air volume adjustment function, which depends on the wind speed u;

[0031] Meaning: The need for active wind supply and the intensity of such supply are determined based on the measured wind speed.

[0032] γ(k) is the control gain function, which depends on the control gain parameter k; its meaning is to adjust the stability of the system control and prevent oscillations caused by excessively fast or strong feedback.

[0033] S2.2 determines the current operating condition category based on the θ-uk-Rrec combination: downwind, crosswind, side-backwind, and high risk of heat recirculation.

[0034] Preferably, S2.2 is performed according to the following steps: Based on the environmental crosswind direction θ, determine the downwind zone, crosswind zone, and side-backwind zone of the tower's air inlet; and determine whether R exists under the crosswind speed u. rec A value >0 and k*>0 indicate the presence of hot reflow, and the larger the values ​​of Rec and k*, the higher the risk of hot reflow.

[0035] Preferably, the target setting and adaptive scheduling are performed in the following manner:

[0036] S3.1 Set the target based on load and seasonal operating conditions: UI≤UI*, preferably UI*∈[0.10,0.18]; ΔT cool ≥ΔT*, preferably ΔT*∈[0.8℃,2.0℃]; and set an upper limit P for energy consumption. lim (The sum of the power of the fan and the valve drive);

[0037] S3.2 Establishing a multi-objective cost function

[0038]

[0039] J – the control objective function (dimensionless), used to quantify the degree of deviation between the system's operating state and the target value; the smaller J is, the closer the system is to the optimization objective.

[0040] ω1, ω2, ω3 — weighting coefficients, dimensionless, representing the importance of each control objective in the comprehensive evaluation;

[0041] UI—Measured airflow unevenness coefficient;

[0042] UI*—Target air intake non-uniformity coefficient, set value, usually ≤0.3;

[0043] ΔT* — Target cooling temperature drop, set value, unit: °C;

[0044] ΔT cool —Actual cooling temperature drop, in °C;

[0045] P—Actual power of the make-up air fan, in kW;

[0046] P lim — Maximum allowable power of the wind turbine, in kW.

[0047] Preferably, the active-passive combined air supply adjustment is performed according to the following method:

[0048] S4.1 Passive Priority: First, optimize the angle of attack α of the air guide vane within the passive domain. d Elevation angle β d Circumferential opening φ d Given the partition baffle state, solve for {α} that minimizes J. d ,βd , φ d} and execute;

[0049] S4.2 Trigger active air supply: If at the passive limit (α d , β d , φ d reaches the safety / structural boundary) and still satisfies UI > UI* or ΔT cool < ΔT*, then allocate the active air supply flow rate Q fan and the frequency f j of each partition fan according to MPC / feedforward-feedback composite control;

[0050] S4.3 Proportional ratio and switching: Use the hysteresis intervals H1 / H2 to prevent frequent switching:

[0051] If UI > UI* + H1 or R rec > R*: Increase the active proportion η = Q fan / (Q fan + Q passive );

[0052] If UI < UI* - H2 and ΔT cool ≥ ΔT*: Decrease η and rely more on passive;

[0053] Where H1 ∈ [0.02, 0.05], H2 ∈ [0.01, 0.03], R* ∈ [0.006, 0.015];

[0054] UI: Measured inlet air non-uniformity coefficient;

[0055] UI*: Target uniformity coefficient, ≤ 0.3;

[0056] H1, H2: Upper and lower limit hysteresis thresholds to avoid frequent switching;

[0057] R rec : Return flow intensity index, calculated based on CFD or measured pressure difference / temperature field;

[0058] R*: Return flow intensity threshold;

[0059] ΔT cool : Actual cooling temperature drop;

[0060] ΔT*: Target cooling temperature drop;

[0061] η: Proportion of active air supply, defined as the ratio of the fan air volume to the total air supply volume;

[0062] Q fan : Active fan flow rate;

[0063] Q passive : Passive inlet air flow rate.

[0064] Preferably, the constraints and fault self-protection are implemented in the following manner:

[0065] S5.1 constraint: The negative pressure at any point in the circumference of the tower skirt shall not exceed the structural / waterproof limit P. min The air guide plate's attitude is in [α] min ,α max ]、[β min ,β max Within, the fan frequency f j ∈[f min ,f max And the noise / vibration does not exceed the threshold;

[0066] S5.2 Fault Self-Protection: When the sensor fails / drifts, it enters a steady-state conservative state, and the fan frequency is limited. safe Fixed air guide angle α safe And trigger an alarm.

[0067] Preferably, the online evaluation and self-optimization are performed in the following manner:

[0068] Real-time recording of UI, ΔT cool P and vacuum parameters form a rolling window KPI key performance indicator; if J does not improve within M consecutive control cycles, the weights and targets are automatically reset or the spare parameter Π is called.

[0069] This application offers the following beneficial effects: Compared to traditional constant air supply or single airflow guidance schemes, under crosswind and hot recirculation conditions, this application achieves: significantly improved airflow uniformity: UI can be stably ≤0.15 (within the preferred range), and significantly reduced back pressure / recirculation stripes around the tower; improved cold end performance: ΔT cool Achieve ≥0.8–2.0℃, improve unit vacuum; control energy consumption: prioritize the use of passive domains while meeting targets, and ensure that fan energy consumption does not exceed the set P. lim Improved stability: Hysteresis, constraint, and self-protection logic avoid frequent switching and abnormal vibrations, making it easier to implement in engineering projects. Attached Figure Description

[0070] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0071] Figure 1 This is a flowchart of this application.

[0072] Figure 2 A schematic diagram of the arrangement of the tower's circumferential sensors and the air supply / guiding components.

[0073] Figure 3This is a schematic diagram of the overall structure of this application.

[0074] Figure 4 This is a schematic diagram of the response curve and hysteresis interval of the active-passive ratio η-UI / ΔT.

[0075] Figure 5 For operating condition identification criteria (θ-ukR) rec (Illustration of switching between control domains) Detailed Implementation

[0076] To clearly illustrate the technical features of this solution, the following detailed description, in conjunction with the accompanying drawings, will explain this application in detail.

[0077] An adaptive active-passive combined makeup air control method for cooling towers, such as Figure 1-5 As shown, it includes the following steps:

[0078] S1. Data collection and indicator calculation;

[0079] S1.1 acquires the following data via a sensor array: inflow velocity u, wind direction θ, relative humidity RH, and dry / wet bulb temperature T. db / T wb The static pressure p at each measuring point along the circumference of the tower skirt i Speed ​​v i Pressure difference Δp in the lower part of the tower in-out and circulating water outlet temperature T out Inlet water temperature T in ;

[0080] S1.2 Calculate the circumferential airflow non-uniformity coefficient UI and the airflow volume Q in :

[0081] Where UI = σ(v) i) / μ(v i ), i∈{1…N};

[0082] σ(v i —Standard deviation of wind speed at various measuring points around the cooling tower;

[0083] μ(v i — The average wind speed at each measuring point around the cooling tower;

[0084] i — represents the i-th measurement point;

[0085] N represents the number of circumferential measurement points, N≥8;

[0086] Q in =∑(v i ·A i );

[0087] Qin —Total air volume at the cooling tower inlet, in m³ 3 / s;

[0088] v i —The average wind speed at the i-th measuring point or air inlet unit, in m / s;

[0089] A i —The area of ​​the i-th measuring point (or unit), in m² 2 ;

[0090] Σ — represents the summation of the areas of all measuring points or zones;

[0091] S1.3 calculates the water temperature drop ΔT cool =T in -T out And estimate the heat reflux criterion R. rec =(T rim -T amb ) / T amb The threshold value of turbulent kinetic energy k* in the near-tower region is k*>0.

[0092] S2. Digital Twin - CFD Parameter Characterization and Operating Condition Identification;

[0093] S2.1 Utilizing offline CFD-model calibration, a parameter characterization set Π={α(θ),β(u),γ(k),…} is obtained, which is used to estimate the effect of different guide angles / openings and makeup air volumes on UI and ΔT. cool Response sensitivity S UI S T ;

[0094] Π represents the control law or set of control parameters. It is a mathematical abstraction of the entire adaptive supplementary air control method and encompasses multiple control factors.

[0095] α(θ) is the active / passive wind supply ratio function, which depends on the wind direction angle θ;

[0096] β(u) is the make-up air volume adjustment function, which depends on the wind speed u;

[0097] γ(k) is the control gain function, which depends on the control gain parameter k;

[0098] S2.2 determines the current operating condition category based on the θ-uk-Rrec combination: downwind, crosswind, side-backwind, and high risk of heat recirculation.

[0099] Based on the environmental crosswind direction θ, determine the downwind zone, crosswind zone, and side-backwind zone of the tower's air inlet. Then, determine whether R exists under the crosswind speed u. recA value >0 and k*>0 indicate the presence of hot reflow, and the larger the values ​​of Rec and k*, the higher the risk of hot reflow.

[0100] S3. Goal setting and adaptive scheduling;

[0101] S3.1 Set the target based on load and seasonal operating conditions: UI≤UI*, preferably UI*∈[0.10,0.18]; ΔT cool ≥ΔT*, preferably ΔT*∈[0.8℃,2.0℃]; and set an upper limit P for energy consumption. lim (The sum of the power of the fan and the valve drive);

[0102] S3.2 Establishing a multi-objective cost function

[0103]

[0104] J – the control objective function (dimensionless), used to quantify the degree of deviation between the system's operating state and the target value; the smaller J is, the closer the system is to the optimization objective.

[0105] ω1, ω2, ω3 — weighting coefficients, dimensionless, representing the importance of each control objective in the comprehensive evaluation;

[0106] UI—Measured airflow unevenness coefficient;

[0107] UI*—Target air intake non-uniformity coefficient, set value, usually ≤0.3;

[0108] ΔT* — Target cooling temperature drop, set value, unit: °C;

[0109] ΔT cool —Actual cooling temperature drop, in °C;

[0110] P—Actual power of the make-up air fan, in kW;

[0111] P lim — Maximum allowable power of the wind turbine, in kW.

[0112] S4. Perform active-passive combined air supply adjustment.

[0113] S4.1 Passive Priority: First, optimize the angle of attack α of the air guide vane within the passive domain. d Elevation angle β d Circumferential opening φ d Given the partition baffle state, solve for {α} that minimizes J. d ,β d ,φ d} and execute;

[0114] S4.2 Triggering active air supply: If in the passive limit (α) d βd , φ d Still satisfy UI > UI* or ΔT when reaching the safety / structural boundary cool < ΔT*, then allocate the active air supply flow rate Q according to MPC / feedforward-feedback composite control fan And the frequency f of each partition fan j ;

[0115] S4.3 Proportion ratio and switching: Use the hysteresis intervals H1 / H2 to prevent frequent switching:

[0116] If UI > UI* + H1 or R rec > R*: Increase the active proportion η = Q fan / (Q fan + Q passive );

[0117] If UI < UI* - H2 and ΔT cool ≥ ΔT*: Decrease η and rely more on passive;

[0118] Where H1 ∈ [0.02, 0.05], H2 ∈ [0.01, 0.03], R* ∈ [0.006, 0.015];

[0119] UI: Measured inlet air non-uniformity coefficient;

[0120] UI*: Target uniformity coefficient, ≤ 0.3;

[0121] H1, H2: Upper and lower limit hysteresis thresholds to avoid frequent switching;

[0122] R rec : Return flow intensity index, calculated based on CFD or measured pressure difference / temperature field;

[0123] R*: Return flow intensity threshold;

[0124] ΔT cool : Actual cooling temperature drop; ​​​​​​​​​​​​​​​​​​​​​​The air guide plate's attitude is in [α] min ,α max ]、[β min ,β max Within, the fan frequency f j ∈[f min ,f max And the noise / vibration does not exceed the threshold;

[0131] S5.2 Fault Self-Protection: When the sensor fails / drifts, it enters a steady-state conservative state, and the fan frequency is limited. safe Fixed air guide angle α safe And trigger an alarm.

[0132] S6. Online evaluation and self-optimization process.

[0133] Real-time recording of UI, ΔT cool P and vacuum parameters form a rolling window KPI key performance indicator; if J does not improve within M consecutive control cycles, the weights and targets are automatically reset or the spare parameter Π is called.

[0134] In a natural ventilation tower with an inner diameter D of 160m:

[0135] Sensor arrangement: A set of measuring points is arranged at 45° intervals around the tower, for a total of 8 sets. Each set includes v, p, and T. db T wb R H ; 2-4 differential pressure Δp probes are installed at the lower part of the tower; 2 sets of paired sensors for the inlet and outlet water temperatures of the circulating water;

[0136] Target: During the summer high-humidity windy season, set UI* = 0.15, ΔT* = 1.2℃, and Plim = 55kW;

[0137] Control: Prioritize adjusting the angle of attack α of the air guide vane in the passive domain. d ∈[5°, 25°], elevation angle β d ∈[0°,15°] and circumferential opening φd; if UI>0.17 or R rec When UI > 0.01, active air supply is triggered, and η is allocated from 0 to 0.4; when UI < 0.13 and ΔT cool At ≥1.2℃, η decreases according to the hysteresis H2=0.02.

[0138] Results: Under the conditions of θ≈70° crosswind and u=6–8m / s, UI decreased from 0.28 to 0.14–0.16; ΔT cool Temperature rises by approximately 1.0–1.6°C; average fan power ≤42kW.

[0139] Establish the prediction time domain T with Π as a constraint p The cost function for MPC with a time step of 120s and a control step size of Δt = 5s is as shown above.

[0140] Risk of hot reflow R rec When the value is greater than 0.012, the weight of w1 is increased and the maximum angle change rate of the local guide vane is limited to ≤2° / step to suppress induced vortices.

[0141] Actual measurements show that when the wind direction oscillates rapidly (±30° / 5min), the UI peak suppression rate is ≥40%, and ΔT cool The volatility decreased by approximately 35%.

[0142] If the wind speed sensor for a certain circumferential segment loses connection for more than 30 seconds, the system uses interpolation compensation with adjacent segment values ​​and Π, and the UI uses dimensionality reduction estimation; if the connection is continuously lost for more than 10 minutes, the system enters the conservative table: the wind turbine frequency is limited to f. safe =25Hz, air guide fixed α safe =12°, and an alarm will be triggered.

[0143] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for adaptive active-passive combined air supply control of a cooling tower, characterized in that: Includes the following steps: S1. Data collection and indicator calculation; S2. Digital Twin - CFD Parameter Characterization and Operating Condition Identification; S3. Goal setting and adaptive scheduling; S4. Perform active-passive combined air supply adjustment.

2. The adaptive active-passive combined air supply control method for cooling towers as described in claim 1, characterized in that: It also includes constraints and fault self-protection, as well as online evaluation and self-optimization processes.

3. The adaptive active-passive combined air supply control method for cooling towers as described in claim 1, characterized in that: The data collection and indicator calculation are performed in the following manner: S1.1 acquires the following data via a sensor array: inflow velocity u, wind direction θ, relative humidity RH, and dry / wet bulb temperature T. db / T wb The static pressure p at each measuring point along the circumference of the tower skirt i Speed ​​v i Pressure difference Δp in the lower part of the tower in-out and circulating water outlet temperature T out Inlet water temperature T in ; S1.2 Calculate the circumferential airflow non-uniformity coefficient UI and the airflow volume Q in : where UI = σ(v i) / μ(v i ), i ∈ {1…N}; σ(v i —Standard deviation of wind speed at various measuring points around the cooling tower; μ(v i — The average wind speed at each measuring point around the cooling tower; i — represents the i-th measurement point; N represents the number of circumferential measurement points; Q in =∑(v i ·A i ); Q in —Total air volume at the cooling tower inlet, in m³ 3 / s; v i —The average wind speed at the i-th measuring point or air inlet unit, in m / s; A i —The area of ​​the i-th measuring point or unit, in m² 2 ; Σ — represents the summation of the areas of all measuring points or zones; S1.3 calculates the water temperature drop ΔT cool =T in -T out And estimate the heat reflux criterion R. rec =(T rim -T amb ) / T amb The threshold value of turbulent kinetic energy k* in the near-tower region is k*>0.

4. The adaptive active-passive combined air supply control method for cooling towers as described in claim 3, characterized in that: N≥8。 5. The adaptive active-passive combined air supply control method for cooling towers as described in claim 2, characterized in that: The digital twin-CFD parameter characterization and operating condition identification are performed in the following manner: S2.1 Utilizing offline CFD-model calibration, a parameter characterization set Π={α(θ),β(u),γ(k),…} is obtained, which is used to estimate the effect of different guide angles / openings and makeup air volumes on UI and ΔT. cool Response sensitivity S UI S T ; Π represents the control law or set of control parameters. It is the mathematical abstraction of the entire adaptive supplementary air control method and encompasses multiple control factors. α(θ) is the active / passive wind supply ratio function, which depends on the wind direction angle θ; β(u) is the make-up air volume adjustment function, which depends on the wind speed u; γ(k) is the control gain function, which depends on the control gain parameter k; S2.2 determines the current operating condition category based on the θ-uk-Rrec combination: downwind, crosswind, side-backwind, and high risk of heat recirculation.

6. The adaptive active-passive combined air supply control method for cooling towers as described in claim 5, characterized in that: S2.2 is carried out according to the following steps: Based on the environmental crosswind direction θ, determine the downwind zone, crosswind zone, and side-backwind zone of the tower's air inlet, and determine whether R exists under the crosswind speed u. rec A value >0 and k*>0 indicate the presence of hot reflow, and the larger the values ​​of Rec and k*, the higher the risk of hot reflow.

7. The adaptive active-passive combined air supply control method for cooling towers as described in claim 1, characterized in that: The target setting and adaptive scheduling are performed in the following manner: S3.1 Set the target based on load and seasonal operating conditions: UI≤UI*, UI*∈[0.10,0.18]; ΔT cool ≥ΔT*, ΔT*∈[0.8℃,2.0℃]; and set an upper limit P for energy consumption. lim The sum of the driving power of the fan and the valve; S3.2 Establishing a multi-objective cost function J—the control objective function, dimensionless, used to quantify the degree of deviation between the system's operating state and the target value; the smaller J is, the closer the system is to the optimization objective; ω1, ω2, ω3 — weighting coefficients, dimensionless, representing the importance of each control objective in the comprehensive evaluation; UI—Measured airflow unevenness coefficient; UI*—Target air intake non-uniformity coefficient, set value, usually ≤0.3; ΔT* — Target cooling temperature drop, set value, unit: °C; ΔT cool —Actual cooling temperature drop, in °C; P—Actual power of the make-up air fan, in kW; P lim — Maximum allowable power of the wind turbine, in kW.

8. The adaptive active-passive combined air supply control method for cooling towers as described in claim 1, characterized in that: Adjust the active-passive combined air supply as follows: S4.1 Passive Priority: First, optimize the angle of attack α of the air guide vane within the passive domain. d Elevation angle β d Circumferential opening φ d Given the partition baffle state, solve for {α} that minimizes J. d ,β d ,φ d } and execute; S4.2 Triggering active air supply: If in the passive limit, α d β d φ d Even when the safety / structural boundary is reached, UI > UI* or ΔT must still be satisfied. cool If <ΔT*, then the active makeup air flow rate Q is allocated according to MPC / feedforward-feedback composite control. fan With the frequency f of each zone's fan j ; S4.3 Proportional Mixing and Switching: Hysteresis intervals H1 / H2 are used to prevent frequent switching. If UI > UI*+H1 or R rec R*: Increase the proportion of active components η = Q fan / (Q fan +Q passive ); If UI < UI* - H2 and ΔT cool ≥ ΔT*: Reduce η and rely more on passive components; Where H1∈[0.02,0.05], H2∈[0.01,0.03], R*∈[0.006,0.015]; UI: Actual measured airflow unevenness coefficient; UI*: Target uniformity coefficient, ≤0.3; H1, H2: Upper and lower limit hysteresis thresholds to avoid frequent switching; R rec : Reflux intensity index, calculated based on CFD or measured pressure difference / temperature field; R*: Reflux intensity threshold; ΔT cool Actual cooling temperature drop; ΔT*: Target cooling temperature drop; η: Active air supply ratio, defined as the ratio of fan air volume to total air supply volume; Q fan Active fan flow rate; Q passive Passive air intake flow rate.

9. The adaptive active-passive combined air supply control method for cooling towers as described in claim 2, characterized in that: The constraints and fault self-protection are implemented in the following manner: S5.1 constraint: The negative pressure at any point in the circumference of the tower skirt shall not exceed the structural / waterproof limit P. min The air guide plate's attitude is in [α] min ,α max ]、[β min ,β max Within, the fan frequency f j ∈[f min ,f max And the noise / vibration does not exceed the threshold; S5.2 Fault Self-Protection: When the sensor fails / drifts, it enters a steady-state conservative state, and the fan frequency is limited. safe Fixed air guide angle α safe And trigger an alarm.

10. The adaptive active-passive combined air supply control method for cooling towers as described in claim 2, characterized in that: The online evaluation and self-optimization are performed in the following manner: Real-time recording of UI, ΔT cool P and vacuum parameters form a rolling window KPI key performance indicator; if J does not improve within M consecutive control cycles, the weights and targets are automatically reset or the spare parameter Π is called.

Images