An open crossflow cooling tower and its control method

Abstract

This application discloses an open crossflow cooling tower and its control method, relating to the field of cooling tower technology. The tower comprises: an upper tower body and a lower tower body stacked vertically and detachably connected by fasteners; an internal water distribution pipeline is installed within the upper tower body, comprising two sets of parallel inlet branch pipes symmetrically arranged along both sides of the upper tower body; below each set of parallel inlet branch pipes are sequentially arranged an energy consumer, a water distribution tray, spray nozzles, and a packing assembly; wherein, the energy consumer is located below the parallel inlet branch pipes and receives the water flow discharged from the parallel inlet branch pipes; the water distribution tray receives the water overflowing from the periphery of the energy consumer and directs the water flow to the spray nozzles; a water collection tray is provided below the packing assembly, the bottom of the water collection tray is inclined horizontally, and the water outlet of the water collection tray is connected to the water inlet on the external inlet side via a connecting pipe. This application provides a systematic solution that balances transportation compliance, installation efficiency, and operational reliability.

Description

Technical Field

[0001] This application relates to the field of cooling tower technology, and in particular to an open crossflow cooling tower and its control method. Background Technology

[0002] In existing technologies, open crossflow cooling towers typically employ an integral tower structure design. Their height and volume are limited by the dimensions of standard transport vehicles (e.g., road height restrictions of 4.5 meters and width restrictions of 2.55 meters), preventing large cooling towers from being shipped in their complete form. Therefore, traditional solutions often require disassembling the tower into multiple components for on-site assembly and connection. This not only prolongs the installation period but also introduces assembly errors due to complex on-site working environments and insufficient construction precision, affecting the overall sealing and thermal performance of the unit.

[0003] Furthermore, traditional cooling towers generally employ external piping. While core components such as fans, pumps, and water distribution systems can be pre-assembled in the factory, inlet and outlet pipes, valves, and connecting flanges still need to be welded or bolted on-site after the equipment is in place. This method not only increases on-site construction work and labor costs, but also makes exposed piping susceptible to corrosion from the external environment, resulting in lower long-term operational reliability. Simultaneously, because piping work depends on on-site conditions, it is difficult to achieve standardized and modular prefabrication, hindering the rapid deployment and delivery efficiency of cooling tower products.

[0004] Although some manufacturers have attempted to alleviate the above problems through modular transportation or partial pre-assembly, there are still significant technical bottlenecks in areas such as the rationality of tower structure segmentation, the accuracy of rapid on-site docking, and the sealing and maintenance convenience of embedded pipeline systems. A systematic solution that balances transportation compliance, installation efficiency, and operational reliability has not yet been formed. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art and provide a systematic solution that takes into account transportation compliance, installation efficiency and operational reliability, this application provides an open crossflow cooling tower and its control method.

[0006] Firstly, the objective of this invention is achieved through the following technical solution: An open crossflow cooling tower includes an upper tower body and a lower tower body stacked vertically and detachably connected by fasteners. The upper tower body has a built-in water distribution pipeline, which includes two sets of parallel inlet branch pipes symmetrically arranged along both sides of the upper tower body. Below each set of parallel inlet branch pipes are sequentially arranged an energy consumer, a water distribution tray, a spray nozzle, and a packing assembly. The energy consumer is located below the outlet end of the parallel inlet branch pipe and is used to receive the water discharged from the parallel inlet branch pipe. The water distribution tray receives the water overflowing from around the energy consumer and guides the water flow to the spray nozzle. Below the packing assembly is a water collection tray, the bottom of which has an inclined structure with one side higher than the other in the horizontal direction. An outlet hole is provided on the lower side of the water collection tray, and the outlet hole is connected to the inlet hole on the external inlet side via a connecting pipe, forming a closed circulating water circuit.

[0007] By adopting the above technical solution, a layered tower structure with vertically stacked upper and lower layers connected by fasteners allows large cooling towers to be prefabricated in the factory in layers, each layer conforming to standard transport vehicle dimensions (e.g., height limit 4.5 meters, width limit 2.55 meters). On-site installation only requires alignment and fastening to complete the main structure. This fundamentally solves the dilemma between transporting monolithic towers exceeding size limits and the insufficient precision of on-site assembly after disassembly of individual parts, shortening the installation cycle and ensuring the tower's airtightness. Secondly, all water distribution pipes are built into the upper tower body, arranged with symmetrical, same-path inlet branch pipes on both sides. This ensures equal frictional resistance and balanced water flow on both sides, avoiding the drawbacks of traditional external piping, such as susceptibility to corrosion, large on-site welding requirements, and low long-term reliability. Furthermore, the tiered water distribution path—"same-path branch pipe—energy consumer—water distribution tray—sprinkler nozzle—packing assembly"—dissipates the impact energy of the water flow through overflow around the energy consumer, and then the water distribution tray smoothly receives and guides the water to the sprinkler nozzles. This ensures that the water flow completes multi-stage energy attenuation and water redistribution before reaching the packing, improving the uniformity of water distribution on the packing surface. The inclined water collection tray, with one side higher than the other, works in conjunction with the connecting pipes to allow the heat-exchanged water to converge at the outlet and be reinjected into the built-in water distribution pipeline, forming a closed-loop water circuit. This prevents water accumulation in the tray and reduces water replenishment losses, resulting in a comprehensive system solution that balances transportation compliance, installation efficiency, and operational reliability.

[0008] In a preferred embodiment of this application: the spray nozzle is a specially designed spray nozzle with a water storage and pressurization structure. The spray nozzle has a water storage chamber inside for storing water under low flow conditions to increase the spray pressure, so that the cooling tower can maintain a uniform distribution of spray water mist on the surface of the packing assembly within the operating range of 30% to 120% of the nominal flow rate. The water inlet side of the spray nozzle is connected to the lower outlet of the water distribution tray, and the upper part of the water distribution tray corresponds to the peripheral overflow outlet of the energy consumer, forming a gradually decreasing water distribution path.

[0009] By adopting the above technical solution, a water storage and pressurization chamber is added inside the spray nozzle, which can still store a certain water level to maintain the necessary spray pressure under low flow conditions. This allows the cooling tower to maintain a uniform distribution of spray water mist on the packing surface within a wide operating range of 30% to 120% of the nominal flow rate, expanding the effective adjustment range of the equipment and avoiding the problems of local dry areas in the packing and reduced heat exchange efficiency caused by insufficient spray pressure under low load. At the same time, the water inlet side of the spray nozzle is connected to the lower outlet of the water distribution tray, and the upper part of the water distribution tray corresponds to the overflow outlet around the energy consumer, forming a step-by-step deceleration water distribution path of "energy consumer - water distribution tray - spray nozzle". This allows the water flow to complete multi-stage dissipation of kinetic energy before reaching the packing, further improving the uniformity and stability of gas-liquid heat exchange.

[0010] In a preferred embodiment of this application: the packing assembly is composed of several packing units arranged side by side in a horizontal direction. Each packing unit includes two mirror-symmetrical packing plates. The two packing plates are connected and positioned to each other by several support rods that pass through preset holes. After adjacent packing units are assembled together, a honeycomb heat exchange channel is formed inside the packing assembly. Each packing plate is provided with an air inlet section and a water collection section in sequence along the airflow direction. The air inlet section is used to guide external air to enter the honeycomb heat exchange channel evenly, and the water collection section is used to prevent water droplets from splashing after heat exchange. The water flowing out from the lower end of the packing assembly falls into the water collection pan, then flows along the bottom of the inclined pan to the water outlet, and returns to the built-in water distribution pipeline through the connecting pipe.

[0011] By adopting the above technical solution, the packing assembly is constructed as a modular structure consisting of several packing units arranged side by side in a horizontal direction. Each unit consists of two mirror-symmetrical packing plates interlocked by a through rod. This allows the packing assembly to be prefabricated in the factory and quickly assembled on-site, facilitating transportation and single-point maintenance and replacement. The honeycomb-shaped heat exchange channel formed internally after adjacent units are assembled has both a large specific surface area and a stable airflow path. Combined with the air inlet section and water collection section arranged sequentially along the airflow direction for each packing plate, it can ensure that external air enters the heat exchange channel evenly while preventing water droplets from splashing after heat exchange, reducing water drift loss. After heat exchange, the water flows through an inclined water collection tray to the water outlet, and then returns to the built-in water distribution pipeline through a connecting pipe, improving the overall heat exchange efficiency and water recycling rate.

[0012] In a preferred embodiment of this application: each packing unit has several sets of alignment holes spaced apart along the height direction on two packing plates, and four supporting rods are provided, each passing through the corresponding alignment holes, to lock the two packing plates together as one unit; adjacent packing units are connected in series through the extension ends of the supporting rods, thereby forming an integral packing assembly in the upper tower body.

[0013] By adopting the above technical solution, several sets of alignment holes are opened at intervals along the height direction for the two packing plates of each packing unit. Four supporting rods pass through the corresponding holes, locking the two packing plates together. Adjacent packing units are then connected in series through the extensions of the supporting rods to form an integral packing assembly. The relative position between the two packing plates is doubly constrained by the alignment holes and the four supporting rods, avoiding the problems of heat exchange channel misalignment and airflow short-circuiting caused by packing misalignment and deformation under operating vibration and water flow scouring, thus improving structural rigidity and service life. At the same time, the series connection of the rod extensions makes the entire packing assembly a continuous whole in the upper tower body, facilitating one-time hoisting and positioning.

[0014] Secondly, the objective of this invention is achieved through the following technical solution: A control method for an open crossflow cooling tower, applied to an open crossflow cooling tower as described above, includes: Obtain the current environmental conditions and trial operation data of the cooling tower to obtain the initial operating parameters of the cooling tower fan and water pump; During the process of the cooling tower circulating heat dissipation according to the initial operating parameters, the temperature deviation of the actual outlet water temperature of the cooling tower relative to the target set temperature is calculated. The temperature deviation is subjected to energy efficiency optimization and correction processing to obtain the collaborative regression parameters of the cooling tower; Based on the collaborative regression parameters, the energy efficiency optimization adjustment amount under the current environmental conditions is calculated, and the fan speed and water pump frequency of the cooling tower are adjusted synchronously based on the energy efficiency optimization adjustment amount.

[0015] By adopting the above technical solution, the reference heat transfer load corresponding to the current packing heat exchange interface and the deviation cooling load corresponding to the temperature deviation are obtained based on the decoupling of the collaborative regression parameters. The energy efficiency optimization adjustment amount is calculated by combining the two, and then the fan blowing speed is adjusted. The gas-liquid heat balance ratio between the optimal heat dissipation and the corresponding circulating water evaporation at the same time node is used as the adjustment basis to perform synchronous spraying and coordinated treatment of the spray water distribution rate. This application integrates the air volume, water volume and heat exchange load into the gas-liquid heat balance framework for linkage control, avoiding energy waste and water drift loss caused by fan overshoot or excessive spraying, and ensuring that the adjustment on both the air and water sides always maintains thermal matching.

[0016] In a preferred embodiment of this application: the step of calculating the energy efficiency optimization adjustment amount under the current environmental conditions based on the collaborative regression parameters, and synchronously adjusting the fan speed and water pump frequency of the cooling tower based on the energy efficiency optimization adjustment amount, includes: Based on the collaborative regression parameters, obtain the benchmark heat transfer load corresponding to the current packing heat exchange interface and the deviation cooling load corresponding to the temperature deviation. Based on the baseline heat exchange load and the deviation cooling load, the energy efficiency optimization adjustment of the cooling tower under the current environmental conditions is calculated comprehensively. The fan speed of the cooling tower is adjusted according to the energy efficiency optimization adjustment amount, and the gas-liquid heat balance ratio between the optimized heat dissipation and the corresponding circulating water evaporation at the same time node is controlled according to the fan speed. The spray water distribution rate of the cooling tower is adjusted according to the gas-liquid heat balance ratio, and synchronous spraying is performed in conjunction with the blower speed.

[0017] By adopting the above technical solution, this invention first determines the initial operating parameters of the fan and water pump by acquiring the current environmental conditions and trial operation data of the cooling tower; then, during the circulating heat dissipation process, it calculates in real time the direction and magnitude of the deviation between the actual outlet water temperature and the target set temperature, and performs energy efficiency optimization and correction processing to obtain collaborative regression parameters; finally, it synchronously adjusts the fan speed and water pump frequency according to the calculated energy efficiency optimization adjustment amount. This method uses the cooling fan and circulating water pump as coupled objects for joint optimization, avoiding the heat exchange imbalance and ineffective energy consumption caused by the mismatch between air volume and water volume in traditional single-variable regulation, enabling the cooling tower to continuously approach the target outlet water temperature even under fluctuating operating conditions, and significantly reducing the total power consumption of the system.

[0018] In a preferred embodiment of this application: the step of adjusting the spray water distribution rate of the cooling tower according to the gas-liquid heat balance ratio and performing synchronous spraying and coordinated treatment with the blower speed further includes: The wetting deviation angle of the packing layer and the spray coverage diameter of the spray nozzles are obtained after the cooling tower is sprayed with water. The spray water supply pressure for the next spray cycle is calculated based on the wetting deviation angle of the packing layer, and the deviation compensation for the wetting deviation angle of the previous packing layer is performed based on the spray water supply pressure. Adjust the water flow rate of the spray cloth according to the spray coverage diameter and the corresponding spray nozzle outlet outer diameter; The water flow rate of the spray cloth is calculated using formula (1): (1) Where W represents the spray water flow rate, K represents the expansion coefficient of the overflow pressure stabilization-collision water diffusion composite structure, N represents the number of spray nozzles, D1 represents the outer diameter of the spray nozzle outlet, D2 represents the spray coverage diameter of the spray nozzle, and v represents the radial diffusion velocity of the umbrella-shaped atomized water mist. Based on the arrangement of the spray nozzles of the cooling tower, the water distribution coverage parameters of each packing section are obtained, and the water distribution coverage change value is calculated after the wetting deviation angle of the packing layer is compensated. The spray position parameters for the next spray cycle are adjusted based on the change in water coverage, and the spray operation process of the cooling tower is optimized based on the adjusted spray position parameters and the spray water flow rate.

[0019] By adopting the above technical solution, the wetting deviation angle of the packing layer after spraying and the spray coverage diameter of the spray nozzles are obtained. Based on the wetting deviation angle, the spray water supply pressure for the next spraying cycle is calculated to compensate for the deviation of the previous cycle. Furthermore, based on the spray coverage diameter and the outer diameter of the spray nozzle outlet, the spray water flow rate is accurately calculated using the above calculation formula. Then, combined with the water coverage parameters of each packing zone, the water coverage variation value is obtained, and the spray position parameters for the next cycle are adjusted accordingly. This solution achieves closed-loop linkage regulation of spray pressure, flow rate, and position, realizing dynamic and precise compensation for the wetting area of ​​the packing, avoiding localized dry areas and overlapping spraying, and improving water distribution uniformity, heat exchange efficiency, and water resource utilization.

[0020] In a preferred embodiment of this application, the step of calculating the energy efficiency optimization adjustment amount under the current environmental conditions based on the collaborative regression parameters further includes: Obtain the physical structure parameters of the cooling tower; retrieve multi-source operation data of the cooling tower during its historical operation cycle, use the physical structure parameters and the multi-source operation data together as training samples, and perform structure search and hyperparameter optimization on a preset candidate model family based on an automatic machine learning method to construct a digital twin model of equipment performance. The digital twin model of the equipment performance is used to map and output the predicted outlet water temperature and the predicted total power consumption of the system under a given combination of environmental operating parameters and given fan speed and pump frequency. An embedded collaborative optimization controller is constructed with the predicted effluent temperature meeting the target set temperature as a hard constraint and the total power consumption of the prediction system as the objective function. The embedded collaborative optimization controller calls the device performance digital twin model according to a preset optimization cycle to perform traversal calculations on feasible combinations of the fan speed and the water pump frequency, and performs optimization locking on the traversal results to obtain the optimal operating point that minimizes the total power consumption of the system. Based on the optimal operating point, the energy efficiency optimization adjustment amount under the current optimization cycle is calculated, and the energy efficiency optimization adjustment amount is injected back into the collaborative regression parameters for closed-loop correction.

[0021] By adopting the above technical solution, using the physical structural parameters of the cooling tower and historical multi-source operating data as training samples, and employing automatic machine learning methods to perform structure search and hyperparameter optimization on a preset candidate model family, a digital twin model of equipment performance is constructed that can map the predicted outlet water temperature and total system power consumption under a given combination of environmental conditions, fan speed, and pump frequency. An embedded collaborative optimization controller is built with the predicted outlet water temperature meeting the target set temperature as a hard constraint and the predicted total system power consumption minimization as the objective function. This controller traverses feasible combinations according to a preset optimization cycle, locks the optimal operating point, and then back-calculates and injects back into the collaborative regression parameters to form a closed-loop correction. This application upgrades energy efficiency optimization from experience-based adjustment to model-driven optimization based on digital twins, continuously tracking the lowest power consumption condition while ensuring the outlet water temperature meets the target.

[0022] In a preferred embodiment, this application further includes: Obtain meteorological forecast data for the operating area of ​​the cooling tower within a preset time period in the future. The meteorological forecast data includes ambient wet-bulb temperature, ambient dry-bulb temperature, outdoor wind speed, and solar radiation intensity. The meteorological forecast data is input into the digital twin model of the equipment performance, and the predicted water temperature trend and the predicted total power consumption trend of the system are fed forward to obtain the predicted operating condition disturbance result of the cooling tower in the future preset period. Based on the predicted operating condition disturbance, predictive feedforward compensation is performed on the optimal operating point to obtain a predictive operating parameter combination that includes the feedforward compensation amount. The predictive operating parameter combination is used to pre-adjust the fan speed and the water pump frequency before the operating condition disturbance arrives. Based on the current load range and wet-bulb temperature range of the environmental conditions, the spraying strategy corresponding to the optimal operating point is adaptively and dynamically adjusted.

[0023] By adopting the above technical solution, meteorological forecast data including ambient wet-bulb temperature, dry-bulb temperature, outdoor wind speed, and solar radiation intensity for a preset period in the operating area are obtained. This data is then input into a digital twin model of equipment performance to perform feedforward extrapolation of the predicted outlet water temperature trend and the predicted total system power consumption trend for the future period. This yields the predicted operating condition disturbance results, and based on these results, predictive feedforward compensation is performed on the optimal operating point, ensuring that the fan speed and pump frequency are pre-adjusted before the operating condition disturbance arrives. Simultaneously, the spraying strategy is adaptively and linkedly adjusted according to the current load range and wet-bulb temperature range. By combining meteorological feedforward and feedback optimization, transient overshoot of the actual outlet water temperature is effectively avoided, improving the cooling tower's steady-state maintenance capability under drastic weather changes.

[0024] Thirdly, the objective of this invention is achieved through the following technical solution: A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the control method for an open crossflow cooling tower described above.

[0025] In summary, this application includes at least one of the following beneficial technical effects: 1. This application improves transportation compliance and logistics efficiency. By rationally dividing the traditional integrated tower body vertically into upper and lower layers, the height and dimensions of individual modules can be strictly controlled within national road transport limits (e.g., height ≤ 4.2m, width ≤ 2.5m), allowing for the transportation of the entire module using standard flatbed trucks without the need for over-limit approval; this helps to significantly shorten the on-site installation cycle and improve assembly accuracy. All internal components (including packing, spray nozzles, water distribution trays, and built-in piping) of the upper and lower tower bodies have been pre-assembled and functionally tested in the factory. On-site assembly of the main structure is completed by vertically stacking and locking the two tower bodies together with fasteners (e.g., high-strength bolts). 2. Achieve fully integrated and highly reliable operation of the water distribution system. By setting symmetrically arranged parallel water inlet branch pipes in the upper tower body, and cooperating with a multi-stage water flow guiding structure of energy consumer-water distribution tray-spray nozzle, the circulating water is distributed, buffered and evenly sprayed inside the tower body; at the same time, the water collection tray adopts an inclined bottom structure and is designed with low-position water outlet and closed circulation water circuit to effectively prevent water accumulation and enhance drainage smoothness. Attached Figure Description

[0026] Figure 1 This is an overall installation structure diagram of an open crossflow cooling tower according to an embodiment of this application; Figure 2 This is a structural diagram of the same-path water inlet branch pipe in an open crossflow cooling tower according to an embodiment of this application; Figure 3 This is a structural diagram of a packing assembly in an open crossflow cooling tower according to an embodiment of this application; Figure 4 This is a structural diagram of an integral packing assembly in an open crossflow cooling tower according to an embodiment of this application; Figure 5 This is a structural diagram of a spray nozzle in an open crossflow cooling tower according to an embodiment of this application; Figure 6 This is a flowchart of a control method for an open crossflow cooling tower according to an embodiment of this application.

[0027] Explanation of reference numerals in the attached figures: 1. Upper tower body; 2. Lower tower body; 3. Fasteners; 4. Same-path water inlet branch pipe; 5. Spray nozzle; 51. Water inlet pressure stabilizing cylinder; 52. Nozzle water distribution plate; 53. Reduced diameter spray section; 54. Collision water distribution plate; 6. Packing assembly; 61. Alignment through hole; 7. Water collection plate. Detailed Implementation

[0028] The present application will be further described in detail below with reference to the accompanying drawings.

[0029] In one embodiment, such as Figures 1 to 5 As shown, this application discloses an open crossflow cooling tower, which is composed of an upper tower body 1 and a lower tower body 2 stacked vertically. The upper tower body 1 and the lower tower body 2 are detachably connected by fasteners 3. In one specific embodiment, the fastener 3 can be a high-strength bolt-flange assembly, that is, the lower port of the upper tower body 1 and the upper port of the lower tower body 2 are respectively provided with corresponding connecting flanges, and a temperature-resistant and corrosion-resistant rubber sealing ring is sandwiched between the two flanges. Bolts are evenly distributed and locked along the circumference of the flanges to ensure the sealing performance and structural strength of the joint after stacking. In other optional embodiments, the fastener 3 can also be a clamp, pin-lock, or quick-release clamp, or any detachable connection structure that can be pre-assembled in the factory, quickly connected on site, and capable of withstanding the self-weight of the tower body and the operating vibration load. This application does not limit this.

[0030] Through the above-mentioned layered stacked structure, the entire machine can be prefabricated in the factory, with the upper tower body 1 and the lower tower body 2 pre-assembled, and the internal components pre-assembled and tested for sealing. Then, they can be transported to the site independently according to the standard transport vehicle dimensions (such as road height limit of 4.5m and width limit of 2.55m). On-site, only hoisting and alignment and fastener installation 3 are required to complete the main body installation, eliminating the need for complex splicing of scattered parts and shortening the installation cycle.

[0031] An internal water distribution pipeline is installed inside the upper tower body 1. The internal water distribution pipeline includes two sets of parallel water inlet branch pipes 4 symmetrically arranged along both sides of the upper tower body 1. The inlet ends of the two sets of parallel water inlet branch pipes 4 converge inside the upper tower body 1 and extend downward, passing through the lower tower body 2 to form an external water inlet. The external water inlet is located on one side of the bottom of the lower tower body 2, which allows for easy connection of the external circulating water pipe to the external water inlet on site.

[0032] "Same path" refers to the fact that the lengths of the two sets of inlet branch pipes from their confluence point to their respective farthest spray nozzles 5 are equal, their pipe diameters are consistent, and their friction resistance is symmetrical. This ensures that the water flow and pressure at the outlet of the two branch pipes are basically the same, avoiding the problems of "more water on one side and less water on the other side" and uneven heat dissipation caused by traditional non-same path piping with water inlet on one side and outlet on the other side. The same path inlet branch pipes 4 are made of UPVC pipes or 304 stainless steel pipes, and the nominal diameter is selected according to the nominal flow rate of the cooling tower. Each branch pipe extends horizontally along the length of the upper tower body 1, and several outlet ports are opened at intervals on its lower side, corresponding to the positions of the spray nozzles 5 below.

[0033] Below each set of inlet branch pipes 4, there are sequentially arranged an energy dissipator (not shown in the figure), a water distribution tray (not shown in the figure), a spray nozzle 5, and a packing assembly 6. The energy dissipator is located directly below the outlet end of the inlet branch pipe 4 and has a shallow trough-shaped structure with an open top (also called a spillway). In one specific embodiment, the energy dissipator is a long strip of PP plastic trough, with the trough opening extending horizontally, a trough depth of 20~50mm, and a trough width not less than the outer diameter of the branch pipe outlet plus twice the deflection allowance. The water flow discharged from the inlet branch pipe 4 first falls vertically into the energy dissipator, and the vertical impact kinetic energy of the water flow is significantly dissipated on the liquid surface in the trough; when the water level in the trough rises to the edge of the trough, the water flow overflows evenly along the periphery of the energy dissipator at a low speed, thus completing the first stage of deceleration water distribution.

[0034] The water distribution tray receives water overflowing from the periphery of the energy consumer. Located below the energy consumer and above the spray nozzle 5, the water distribution tray has a wide-mouthed, shallow tray shape, with the projection area of ​​the tray opening covering the periphery of the energy consumer's overflow outlet. The bottom of the water distribution tray has a drain hole (or drain pipe) corresponding to the water inlet of the spray nozzle 5 below, used to smoothly guide the water collected in the tray to the spray nozzle 5.

[0035] like Figure 5 As shown, the spray nozzles 5 are located below the discharge holes of the water distribution plate and are evenly distributed along the length of the inlet branch pipe 4. The spray nozzles 5 spray the water discharged from the water distribution plate into a mist or fan shape onto the surface of the packing assembly 6 below. The spray nozzles 5 are specially designed spray nozzles with a water storage and pressurization structure. The spray nozzles 5 are specially designed spray nozzles with an overflow pressure stabilization-collision water dispersion composite structure. Along the water flow direction from top to bottom, they include five functional sections: an inlet pressure stabilizing cylinder 51, an overflow window, a nozzle water distribution plate 52, a narrowing spray section 53, a water dispersion support, and a collision water dispersion plate 54. The whole structure is a multi-stage deceleration atomization structure of "upper cylinder - middle plate - lower funnel - bottom collision plate".

[0036] Specifically, the inlet pressure stabilizing cylinder 51 is a vertically arranged cylindrical cavity with its top opening connected to the lower outlet of the upper inlet branch pipe 4, forming the inlet end of the spray nozzle 5. The side wall of the inlet pressure stabilizing cylinder 51 has a rectangular overflow window that runs through the inside and outside of the cavity. The height of the lower edge of the overflow window from the bottom of the cylinder is the effective water storage height of the spray nozzle 5, which is 20~80mm. When the inlet flow rate is greater than the instantaneous flow capacity of the narrow-diameter spray section 53, the excess water overflows laterally through the overflow window and flows back to the lower water collection circuit, so that the water level in the inlet pressure stabilizing cylinder 51 is always kept constant at the lower edge of the overflow window, thereby forming a constant additional water head at the nozzle inlet that is independent of the flow rate.

[0037] In this embodiment, the nozzle water distribution plate 52 refers to an annular plate integrally formed or fixedly connected to the outer side of the water inlet pressure stabilizing cylinder 51 and the upper edge of the narrow-diameter spray section 53. It is used to receive the water overflowing through the overflow window and distribute it evenly in all directions, while performing axisymmetric rectification on the mainstream water entering the narrow-diameter spray section 53. The narrow-diameter spray section 53 refers to an inverted conical funnel section that gradually narrows downward from the bottom end of the water inlet pressure stabilizing cylinder 51. Its inner wall cone angle is 15°~30°, and a spray hole is provided at the end. The diameter of the spray hole determines the spray speed of the mainstream water in a single nozzle, and is selected in accordance with the nominal flow rate of the cooling tower and the number of spray nozzles 5.

[0038] Specifically, the main stream of water is accelerated and converged downwards from the bottom of the inlet pressure stabilizing cylinder 51 through the narrowing spray section 53, forming an axial jet at the outlet of the spray hole. After leaving the spray hole, the jet impacts the downward-facing collision water distribution plate 54. The collision water distribution plate 54 is fixedly connected to the lower end of the narrowing spray section 53 by several evenly distributed water distribution support pillars. The surface of the collision water distribution plate 54 is perpendicular to the axis of the spray hole and the axial distance from the outlet of the spray hole is 10~30mm. After impacting the collision water distribution plate 54, the axial jet is redirected into a thin film of water that radiates radially outwards along the surface of the plate, and detaches at the outer edge of the collision water distribution plate 54 to form an umbrella-shaped atomized water mist.

[0039] In this embodiment, the spray coverage diameter D2 refers to the maximum outer circle diameter of the effective wetting spot formed when the umbrella-shaped atomized water mist reaches the top surface of the packing assembly 6. This structure achieves a three-stage decoupling: "overflow stabilization ensures a constant additional water head - narrowing jet section 53 ensures a constant jet velocity - collision water distribution plate 54 ensures a constant atomization umbrella angle". This ensures that D2 remains basically stable within a wide operating range of 30% to 120% of the nominal flow rate.

[0040] Specifically, when the cooling tower operates under low flow conditions (below 50% of the nominal flow rate), although the amount of water entering the inlet pressure stabilizing cylinder 51 is small, it is still greater than the instantaneous flow capacity of the spray nozzles. The water level inside the inlet pressure stabilizing cylinder 51 is still stably maintained at the effective water storage height by the overflow window. The jet velocity at the outlet of the spray nozzles is determined by this constant water head. The umbrella-shaped water mist on the collision water distribution plate 54 can still be evenly distributed on the surface of the packing assembly 6 with a coverage diameter D2 close to the design value, avoiding the problem of insufficient water head in conventional straight-through nozzles under low load. The problem of local dry areas in the packing and heat exchange efficiency decay occurs; when the cooling tower is running under high flow conditions (nominal flow rate 100%~120%), the inlet pressure stabilizing cylinder 51 is fully filled under continuous water supply and the excess water is continuously bypassed through the overflow window. The jet velocity at the outlet of the spray hole is still determined by the constant water head corresponding to the lower edge of the overflow window. The umbrella-shaped water mist covering diameter D2 of the collision water distribution plate 54 is still maintained near the design value, thus achieving stable spray atomization pattern in the full operating range of 30%~120% nominal flow rate.

[0041] The packing assembly 6 is located below the spray nozzle 5 and consists of several packing plates or packing units arranged side by side in the horizontal direction. It is used to enable the water sprayed down from above to exchange heat with the external air drawn in from the side of the tower body on the packing interface with a large specific surface area, thereby realizing the evaporation and cooling of the circulating water.

[0042] like Figure 3 and Figure 4 As shown, the packing assembly 6 is composed of several packing units arranged side by side in a horizontal direction. Each packing unit includes two mirror-symmetrical packing plates; four sets of integral packing assemblies are connected in series to form a packing assembly group for use in cooling towers. Figure 4 The diagram shows a set of integral packing components. In one specific embodiment, the packing sheets are made of PVC or PP material and are hot-pressed. Each packing sheet is in the shape of a vertical plate, and the surface facing the other packing sheet is formed with regular undulating corrugations, truncated pyramids, or staggered protrusions, so that when the two packing sheets are mirror-aligned, they form a honeycomb heat exchange channel extending vertically inside.

[0043] Each packing plate is sequentially equipped with an air inlet section and a water collection section along the airflow direction (i.e., the direction in which external air enters from the side of the tower and passes laterally through the packing). The air inlet section is located on the windward side of the packing plate. Its corrugated surface structure is mainly for guiding airflow, and its opening angle is relatively large. This is used to guide external air into the honeycomb heat exchange channel evenly and with low resistance, avoiding airflow deflection and short-circuiting at the inlet. The water collection section is located on the leeward side of the packing plate. Its corrugated surface structure is mainly for blocking water. Its opening angle is relatively small and it has reverse bends to prevent water droplets carried up by the airflow during heat exchange from splashing downstream, thereby preventing water droplets from being drawn away by the tower top fan and causing drift loss.

[0044] After heat exchange, the water flowing out from the lower end of the packing assembly 6 falls into the water collection pan 7 below. It then collects from the outlet holes on the bottom of the inclined pan from the high side to the low side. After being filtered by the filter screen, it returns to the built-in water distribution pipeline through the connecting pipe. Driven by the external water pump, it re-enters the two sets of parallel water inlet branch pipes 4 and enters the next cycle.

[0045] Each packing unit has several sets of alignment through holes 61 spaced apart along the height direction on two packing plates. In one specific embodiment, each packing plate has four sets of alignment through holes 61 spaced apart from top to bottom along the height direction. The spacing between two adjacent sets of through holes is determined according to the effective height of the packing plate (for example, when the effective height of the packing plate is 1000mm, the four sets of through holes are located at 100mm, 400mm, 700mm and 950mm from the top, respectively). The positions of each set of through holes on the two packing plates are mirror images of each other.

[0046] Four support rods (not shown in the figure) pass through the corresponding alignment holes 61, locking the two packing plates together. Both ends of each support rod extend beyond the packing plate and are secured by snap rings, nuts, or limit pins. This ensures that the spacing between the two packing plates is constrained and kept constant by the four rods, preventing misalignment, bulging, or collapse due to water flow erosion or airflow vibration during operation.

[0047] Adjacent packing units are connected in series via the extensions of the support rods. Specifically, each packing unit has extension sections reserved at the ends of the four support rods. When adjacent packing units are assembled side by side in the horizontal direction, the extensions of the support rods of the preceding unit are connected and locked to the corresponding support rods of the following unit through a butt sleeve, threaded joint, or through-rod connection, thereby connecting several packing units into an integral packing assembly 6 in the horizontal direction within the upper tower body 1.

[0048] The water collection tray 7 is located directly below the packing assembly 6 to receive the heat-exchanged water flowing out from the lower end of the packing assembly 6. The bottom of the water collection tray 7 has an inclined structure with one side higher than the other in the horizontal direction. In one specific embodiment, the inclination angle of the tray bottom is 1°, and at least one water outlet is provided on the lower side, with a filter screen covering the water outlet. The filter screen can be made of 304 stainless steel wire mesh or engineering plastic mesh, with a mesh size of 1~5mm, to block leaves, impurities, and packing debris in the water, preventing debris from entering the circulation pipe and clogging the spray nozzles 5. The water outlet is connected to the water inlet on the external inlet side through a connecting pipe, forming a closed circulation water circuit. After heat exchange, the water flows from the higher side to the lower side along the inclined tray bottom under the action of gravity, is filtered by the filter screen, enters the connecting pipe through the water outlet, and finally flows back to the circulation water pipe on the external inlet side. Then, it is driven by an external water pump back to the built-in water distribution pipe to participate in spray heat exchange again. The entire water circuit forms a closed loop.

[0049] It should be noted that in this embodiment, the "water distribution plate" and the "water collection plate 7" are two independent components with different positions and functions: the former is located above the packing assembly 6 and its function is to guide the overflowing water around the energy consumer to the spray nozzle 5; the latter is located below the packing assembly 6 and its function is to collect the heat exchanged water flowing down from the packing assembly 6 and guide it to the water outlet.

[0050] In another embodiment, such as Figure 6 As shown, this application also discloses a control method for an open crossflow cooling tower. This control method is applied to an open crossflow cooling tower as described above. The control method for an open crossflow cooling tower specifically includes the following steps: S1: Obtain the current environmental conditions and trial operation data of the cooling tower to obtain the initial operating parameters of the cooling tower fan and water pump.

[0051] In this embodiment, the current environmental conditions refer to the real-time set of environmental parameters at the location of the cooling tower, including the ambient dry-bulb temperature T_db, ambient wet-bulb temperature T_wb, relative humidity RH, atmospheric pressure P_atm, outdoor wind speed v_w, and solar radiation intensity I_sr; the trial operation data refers to the process parameters collected during short-term operation of the cooling tower at preset speeds before it is formally put into closed-loop control, including the circulating water inlet and outlet temperatures T_in / T_out, circulating water volumetric flow rate Q_w, fan power P_fan, water pump power P_pump, and the steady-state response values ​​of the above parameters at different speeds; the initial operating parameters refer to the tuple (n0, f0) of the fan speed reference value n0 and the water pump frequency reference value f0, which are obtained by matching the current environmental conditions and the trial operation data to start the closed-loop control.

[0052] Specifically, dry-bulb and wet-bulb temperature sensors are installed at the air inlet at the top of the upper tower, while an anemometer and solar radiation sensor are installed on the outside of the lower tower. PT100 inlet and outlet water temperature sensors are installed downstream of the external water inlet and the water collection tray outlet, respectively. Electromagnetic flow meters are installed on the inlet and outlet pipes of the water pump. Power values ​​are read in real time through the communication ports of the fan and water pump frequency converters. The controller collects the above parameters with a sampling period Ts = 1s~10s, and first performs a steady-state test run for no less than 10 minutes at three fan speeds (50%, 75%, and 100%) and three water pump frequencies (50%, 75%, and 100%), obtaining 3×3=9 sets of steady-state (n, f, T_out, Q_w, ...) The controller fits the 9 sets of samples to a binary quadratic response surface T_out = g1(n, f, T_wb) with respect to n and f using the least squares method. =g2(n,f), and among the solutions that satisfy the condition that the outlet water temperature equals the target set temperature T_set under the current environmental conditions, the minimum value is... The (n, f) is taken as the initial operating parameters (n0, f0), and after being sent to the frequency converter, it enters closed-loop heat dissipation operation.

[0053] S2: During the cooling tower's circulation and heat dissipation process according to the initial operating parameters, calculate the temperature deviation of the actual outlet water temperature of the cooling tower relative to the target set temperature.

[0054] In this embodiment, the actual outlet water temperature T_out(k) refers to the circulating water return temperature measured by the downstream temperature sensor of the water collection tray outlet at the kth sampling time; the target set temperature T_set refers to the circulating water return temperature threshold value preset according to the downstream user's process requirements, which is 32℃, 35℃ or 37℃; the temperature deviation ΔT(k) in the direction of temperature deviation refers to the signed difference between the actual outlet water temperature and the target set temperature ΔT(k) = T_out(k) − T_set, where the sign of the sign difference indicates the direction of deviation, with a positive value indicating insufficient heat dissipation and a negative value indicating excessive heat dissipation.

[0055] Specifically, the controller continuously collects T_out(k) at a sampling period of Ts=5s and calculates the instantaneous deviation using the formula ΔT(k)=T_out(k)−T_set. To suppress sensor noise and instantaneous disturbances, the controller applies a moving average filter of length M=6~12 to ΔT(k) to obtain the filtered deviation. Simultaneously set the temperature dead zone ε=0.2℃ and the trigger cycle number N=3, when When the temperature deviation is greater than ε for N consecutive sampling periods, it is determined to be valid, triggering the energy efficiency optimization and correction processing of S3; when When ≤ε, the controller maintains the current (n, f) output unchanged to avoid frequent operation of the frequency converter.

[0056] S3: Perform energy efficiency optimization and correction processing on the temperature deviation to obtain the collaborative regression parameters of the cooling tower.

[0057] In this embodiment, energy efficiency optimization and correction processing refers to... →0 represents a soft constraint, based on the total system power consumption. With the goal of minimization, the process of joint online identification and correction of the two coupled regulation objects, the fan and the water pump, is carried out. The collaborative regression parameter refers to the set of regression coefficients that describe the coupling relationship between the fan speed n and the water pump frequency f under the current operating conditions in order to maintain the gas-liquid thermal balance. It is denoted as Θ=(α, β, γ, δ), where α is the air volume sensitivity coefficient, β is the water volume sensitivity coefficient, γ is the air-water coupling interaction coefficient, and δ is the environmental condition correction term.

[0058] Specifically, the controller continuously stores the most recent L groups (n(k), f(k), T_out(k)) using a sliding window of length L = 100~300. (k), T_wb(k)) data. After each correction is triggered, the loss function is used. The objective is to set λ1 to 1 and λ2 to 0.1~0.3; T_out is expressed as a multivariate linear form of T_out=α×n+β×f+γ×n×f+δ×T_wb+b, and the parameter vector Θ is identified online using the recursive least squares (RLS) method with a forgetting factor λ_f=0.95~0.99; after identification, Θ is written into the co-regression parameter register of the controller for S4 to read and use.

[0059] S4: Based on the collaborative regression parameters, calculate the energy efficiency optimization adjustment amount under the current environmental conditions, and adjust the cooling tower fan speed and water pump frequency synchronously according to the energy efficiency optimization adjustment amount.

[0060] In this embodiment, the energy efficiency optimization adjustment amount refers to the binary vector U=(Δn, Δf) formed by the fan speed increment Δn and the water pump frequency increment Δf required to make the fan and water pump work together to simultaneously eliminate temperature deviation and reduce the total power consumption of the system under the current collaborative regression parameter Θ; synchronous adjustment means that the controller sends Δn and Δf to the fan inverter and water pump inverter respectively at the same timestamp of the same control cycle Tc, so that the execution delay difference between the two is less than 50ms, so as to avoid heat exchange imbalance caused by the asynchrony of air volume and water volume adjustment.

[0061] Specifically, the controller substitutes Θ into the extended linear equation system. ,in (where η is the empirical coefficient for temperature gain and η is the power consumption weight), the optimal (Δn*, Δf*) that makes ΔT'→0 and ∂J / ∂U=0 is obtained; then, rate-of-change limiting |Δn*|≤Δn_max, |Δf*|≤Δf_max are applied to prevent overshoot, Δn_max is taken as 5% / Tc of rated speed, and Δf_max is taken as 2Hz / Tc. The controller uses Tc=10s as the control cycle, and sends the limited n(k+1)=n(k)+Δn and f(k+1)=f(k)+Δf in a package to the frequency converters of the fan and water pump through the CAN bus, and uses T_out(k+1) collected in the next cycle as the control cycle. (k+1) feedback enters the S2~S4 closed loop until... Enter the dead zone ε and remain stable.

[0062] Further, in step S4, based on the collaborative regression parameters, the energy efficiency optimization adjustment amount under the current environmental conditions is calculated, and the cooling tower fan speed and water pump frequency are adjusted synchronously according to the energy efficiency optimization adjustment amount, including: S41: Based on the co-regression parameters, obtain the baseline heat transfer load corresponding to the current packing heat exchange interface and the deviation cooling load corresponding to the temperature deviation.

[0063] In this embodiment, the packing heat exchange interface refers to the contact interface in the packing assembly, which is composed of the sprayed water film and the external air passing through the honeycomb heat exchange channel, and is used to realize the exchange of sensible and latent heat between gas and liquid; the reference heat exchange load Q_base refers to the rated total heat that the packing heat exchange interface can transfer per unit time under the current steady-state condition described by the co-regression parameter Θ, and the unit is kW; the deviation from the cooling load ΔQ_dev refers to the deviation of the current actual heat dissipation from the target heat dissipation, which is equivalently characterized by the temperature deviation ΔT'(k). A positive value indicates that there is a heat dissipation gap, and a negative value indicates that there is a heat dissipation surplus.

[0064] Specifically, the controller calculates the baseline heat exchange load using the following formula: Q_base = c_w × ρ_w × Q_w × (T_in − T_out), where c_w = 4.18 kJ / (kg·℃), ρ_w = 1000 kg / m³, Q_w is the actual circulating water volumetric flow rate measured by the electromagnetic flowmeter, and T_in and T_out are the actual measured inlet and outlet water temperatures, respectively. Simultaneously, the temperature deviation is converted into a deviation from the cooling load in units of heat power using ΔQ_dev = c_w × ρ_w × Q_w × ΔT'(k). Q_base and ΔQ_dev, in kW units, are fed into S42 for comprehensive calculation.

[0065] S42: Based on the baseline heat exchange load and the deviation from the cooling load, calculate the energy efficiency optimization adjustment of the cooling tower under the current environmental conditions.

[0066] In this embodiment, the comprehensive calculation refers to the process of using Q_base as the current matching operating point reference for the fan and water pump, using ΔQ_dev as the load increment that needs to be compensated or released by the coordinated action of the fan and water pump, and combining the coordinated regression parameter Θ to back-calculate the energy efficiency optimization adjustment amount U=(Δn, Δf) for this control cycle; its essence is to upgrade the scalar regulation problem driven by temperature deviation into a binary vector regulation problem based on thermal power balance.

[0067] Specifically, the controller introduces the normalized load ratio η=ΔQ_dev / Q_base as an indicator of adjustment intensity. The larger the absolute value of η, the further it deviates from the current matching operating point and the more aggressive the adjustment is required. The controller calculates the optimization adjustment amount using the following formulas: Δn=η×[α / (α×β−γ²)]×Q_base / k_n, Δf=η×[β / (α×β−γ²)]×Q_base / k_f, where k_n is the empirical coefficient of fan speed-airflow gain, in m³ / (h·rpm); k_f is the empirical coefficient of pump frequency-flow gain, in m³ / (h·Hz). The calculated Δn and Δf are limited by the same rate of change as in S4 and then transmitted to S43 as the energy efficiency optimization adjustment amount for this control cycle.

[0068] S43: Adjust the cooling tower fan speed according to the energy efficiency optimization adjustment amount, and control the gas-liquid heat balance ratio between the optimal heat dissipation and the corresponding circulating water evaporation at the same time node according to the fan speed.

[0069] In this embodiment, the blower velocity v_air refers to the air volume flow rate G_air after the blower draws in external air through the packing assembly, divided by the windward cross-sectional area A_face of the tower, in m / s; the optimal heat dissipation Q_opt refers to the total heat dissipation that the packing heat exchange interface can actually achieve at the same time point when the blower is running at v_air; the circulating water evaporation m_evap refers to the equivalent water mass flow rate of the circulating water carried away by the airflow in the form of evaporation phase change at the same time point, in kg / s; and the gas-liquid heat balance ratio. This refers to the ratio of sensible heat loss Q_sens to latent heat loss Q_lat = m_evap × r (where r is the latent heat of vaporization of water, taken as 2257 kJ / kg) at the same time point. =Q_sens / Q_lat, reflects the matching status of air volume and water volume.

[0070] Specifically, the controller calculates the adjusted oncoming wind speed according to v_air=(n+Δn)×k_v / A_face, where k_v is the fan airflow-speed conversion coefficient; and calculates the optimal heat dissipation Q_opt=ρ_a×G_air×(h_out−h_in) using the enthalpy difference method, where h_in and h_out are the enthalpy values ​​of the air entering and leaving the tower, obtained from the measured T_db and T_wb through the enthalpy-humidity chart; further, Q_opt is decomposed into sensible heat and latent heat components Q_sens=ρ_a×G_air×c_p×(T_out_air−T_in_air) and Q_lat=Q_opt−Q_sens, and obtains the desired heat output based on these components. The controller uses =0.3~0.6 is the target range (0.3~0.4 for summer high wet-bulb temperature, 0.5~0.6 for winter dry conditions), and then... and The deviation ΔR= Pass it to S44.

[0071] S44: Adjust the spray water distribution rate of the cooling tower according to the gas-liquid heat balance ratio, and perform synchronous spraying and coordinated treatment with the blower speed.

[0072] In this embodiment, the spray water distribution rate This refers to the water distribution intensity, measured in m³ / (m²·h), obtained by dividing the volumetric flow rate of circulating water sprayed from all nozzles and evenly covering the surface of the packing assembly per unit time by the water-facing cross-sectional area of ​​the packing. Synchronous spraying and coordinated treatment refers to the controller integrating the adjustment commands of v_air with... The adjustment commands are sent to the fan inverter and the water pump inverter at the same timestamp of the same control cycle Tc, so that the air volume adjustment and the spray water volume adjustment are strictly synchronized at the time of execution.

[0073] Specifically, the controller presses For the current benchmark water distribution rate Perform proportional corrections (increase water distribution to enhance evaporative heat transfer when ΔR>0, and decrease water distribution to suppress drift when ΔR<0), then follow the pump characteristic curve. =h(f) is calculated inversely to obtain the result with The corresponding pump frequency increment Δf' is then superimposed on the Δf obtained in S42 to obtain the final frequency increment Δf_final = Δf + Δf'. The controller, with a control cycle of Tc = 10s, uses the CAN bus time synchronization mechanism to package (Δn, Δf_final) with the same timestamp and send it to the fan inverter and pump inverter, ensuring that the execution delay difference between the two is less than 50ms. After sending, the controller enters the next cycle to re-collect T_out, Q_w, and... The data is executed in a loop from S41 to S44 until... It enters the dead zone ε and maintains a stable position, thereby achieving the lowest energy consumption operation of the air-water synergy while ensuring that the outlet water temperature meets the standard.

[0074] In one embodiment, step S44, adjusting the spray water distribution rate of the cooling tower according to the gas-liquid heat balance ratio and coordinating it with the blower speed for synchronized spraying, further includes: S441: Obtain the wetting deviation angle of the packing layer and the spray coverage diameter of the spray nozzles after the cooling tower has been sprayed with water.

[0075] In this embodiment, the packing layer wetting deviation angle refers to the angle of deviation of the center line of the actual wetted area formed by the spray water mist on the top surface of the packing assembly from the vertical axis of the spray nozzle in the horizontal projection plane. It is used to characterize the degree to which the spray water mist deviates from the ideal vertical spraying direction under the action of gravity, wind disturbance and internal airflow suction. The spray coverage diameter refers to the maximum outer circle diameter of the effective wetting spot (or approximate spot) formed by the umbrella-shaped atomized water mist that detaches from the outer edge of the collision water distribution plate of a single spray nozzle after it diffuses downward and reaches the top surface of the packing assembly. It is denoted as D2 and is essentially determined by the constant additional water head of the inlet pressure stabilizing cylinder, the jet velocity of the narrow-diameter spray section and the umbrella-shaped radiation angle of the collision water distribution plate.

[0076] Specifically, several infrared humidity sensors or capacitive wetting sensors are evenly distributed horizontally on the top surface of the packing assembly in the upper tower body. The sensors cover the theoretical spraying area of ​​each spray nozzle in a matrix array, and continuously read the humidity signal at each point with a sampling period of no more than one spray cycle. The controller performs centroid method calculation on the humidity distribution of the sampling points under the jurisdiction of each spray nozzle to obtain the actual wetting center coordinates of the corresponding spray nozzle. The horizontal angle between the line connecting the actual wetting center coordinates and the projection point of the vertical axis of the spray nozzle on the top surface of the packing and the reference direction is used as the wetting deviation angle of the packing layer. At the same time, the controller performs binarization processing on the humidity signal of each sampling point, divides the wetting and non-wetting areas with a preset humidity threshold, extracts the maximum outer circle diameter of the wetting area as the spray coverage diameter D2 of the spray nozzle, and stores it together with the data of the previous spray cycle in the spray history data cache of the controller.

[0077] S442: Calculate the spray water supply pressure for the next spray cycle based on the wetting deviation angle of the packing layer, and compensate for the deviation angle of the previous packing layer based on the spray water supply pressure.

[0078] In this embodiment, the spray water supply pressure refers to the inlet static pressure required for the built-in water distribution pipeline to drive the circulating water into the inlet pressure stabilizing cylinder and pass through the narrow-diameter spray section and collide with the water distribution plate to form a stable umbrella-shaped atomized water mist; deviation compensation refers to the closed-loop correction process in which the controller uses the spray water supply pressure of the next spray cycle as the execution variable to actively offset the spray direction offset represented by the packing layer wetting deviation angle of the previous spray cycle, so that the actual wetting center of the next cycle returns to the vicinity of the theoretical vertical projection point of the spray nozzle.

[0079] Specifically, the controller calculates the spray water supply pressure for the next spray cycle according to the formula P_next=P_0×(1+k_θ×sinθ_dev), where P_next represents the spray water supply pressure for the next spray cycle, P_0 represents the reference spray water supply pressure for the current spray cycle, θ_dev represents the measured wetting deviation angle of the packing layer obtained in the previous spray cycle, and k_θ represents the pressure-angle compensation gain coefficient, which is taken as 0.05~0.20 and calibrated during the previous trial operation. The controller adds the pump frequency increment corresponding to P_next to the frequency command output by S43 and sends it to the pump frequency converter. This increases (or decreases) the spray water supply pressure, causing the water level response of the inlet pressure stabilizing cylinder and the jet velocity of the narrow-diameter spray section to change accordingly. The direction and diffusion angle of the water mist transformed into umbrella-shaped atomized water mist after colliding with the water distribution plate are actively deflected, which in turn counteracts the directional deviation caused by gravity disturbance and airflow suction, thus achieving closed-loop compensation for the deviation of the previous cycle.

[0080] S443: Adjust the corresponding spray cloth water flow rate according to the spray coverage diameter and the corresponding spray nozzle outlet outer diameter.

[0081] In this embodiment, the outer diameter of the spray nozzle outlet refers to the outer diameter of the outlet pipe of the narrowed spray section at the bottom of the spray nozzle, denoted as D1; ​​the spray coverage diameter, defined by S441 as D2, is the final effective wetting spot diameter formed when the umbrella-shaped water mist reaches the top surface of the packing assembly; the spray water flow rate refers to the circulating water volume flow rate that is sprayed from all the spray nozzles to the packing assembly and forms effective water distribution per unit time under the current spray water supply pressure and spray coverage diameter, and is the target quantity for the frequency linkage adjustment of the water pump.

[0082] Specifically, the controller uses the annular expansion area between the initial projection circle of the umbrella-shaped water mist corresponding to the outer diameter D1 of the spray nozzle outlet and the wetting circle of the top surface of the packing corresponding to the spray coverage diameter D2 as the effective spray projection surface, and the radial diffusion velocity of the umbrella-shaped water mist as the velocity passing through this projection surface. The water flow rate of the spray cloth is calculated by formula (1): (1) Wherein, W represents the water flow rate of the spray cloth, K represents the expansion coefficient of the overflow pressure stabilization-collision water diffusion composite structure, K takes a value of 1.2-1.8, the upper limit is taken when the water level of the inlet pressure stabilization cylinder is still stably limited by the overflow window under low flow conditions and the umbrella-shaped water mist is sufficiently diffused to reflect the pressure stabilization expansion gain, and the lower limit is taken when the overflow window is continuously bypassed and the proportion of mainstream water increases under high flow conditions to avoid overspraying; N represents the number of spray nozzles, D1 represents the outer diameter of the spray nozzle outlet, D2 represents the spray coverage diameter of the spray nozzle, and v represents the radial diffusion velocity of the umbrella-shaped atomized water mist. The controller back-calculates the water flow rate of the spray cloth obtained by equation (1) into the corresponding water pump frequency according to the water pump characteristic curve, and then combines it with the frequency increment corresponding to the spray water supply pressure obtained by S442 to synthesize the water pump frequency command and send it to the water pump frequency converter for execution.

[0083] S444: Based on the spray nozzle arrangement of the cooling tower, obtain the water distribution coverage parameters for each packing zone, and calculate the water distribution coverage change value after deviation compensation of the packing layer wetting deviation angle.

[0084] In this embodiment, the spray nozzle arrangement position refers to the set of two-dimensional coordinates in which each spray nozzle is evenly distributed horizontally below the two sets of parallel water inlet branch pipes; the packing partition refers to the controller dividing the top surface of the packing assembly into several sub-regions corresponding to each spray nozzle, with the horizontal projection point of the spray nozzle on the top surface of the packing assembly as the center and the perpendicular bisector of the line connecting the projection points of adjacent spray nozzles as the boundary; the water distribution coverage parameter refers to the set of parameters consisting of the effective wetting area, wetting center coordinates, and coverage rate formed by the corresponding spray nozzle in each packing partition; the water distribution coverage change value refers to the set of differences between the water distribution coverage parameters of each packing partition and its previous cycle value before and after deviation compensation.

[0085] Specifically, before commissioning, the controller divides the top surface of the packing assembly into Voronoi polygons based on the actual installation coordinates of the spray nozzles inside the upper tower, generating several packing zones and writing the zone number, boundary vertex coordinates, and corresponding spray nozzle number into the zone index table. During operation, the controller calculates the effective wetting area, wetting center coordinates, and coverage rate in each zone according to the humidity array signal collected in S441, obtaining the water distribution coverage parameters for the current cycle. After deviation compensation is completed in S442, the controller drives the spray nozzles to run a complete spray cycle again with the new cycle's spray water supply pressure and new spray water flow rate, and re-collects the water distribution coverage parameters for each zone. The difference between the compensated parameters and the parameters before compensation is calculated item by item to obtain the water distribution coverage change value, which is then stored in the controller's spray collaborative optimization table according to the zone number.

[0086] S445: Adjust the spray position parameters for the next spray cycle based on the change in water distribution coverage, and optimize the cooling tower's spray operation process based on the adjusted spray position parameters and spray water flow rate.

[0087] In this embodiment, the spray position parameters refer to the set of adjustable geometric variables that affect the spray landing point of each spray nozzle, including the inclination angle of the axis of the collision spreader plate relative to the vertical direction (hereinafter referred to as the spray nozzle axis inclination angle), the outlet deflection direction angle, and the effective height of the collision spreader plate relative to the top surface of the packing. The outlet deflection direction angle refers to the azimuth angle of the umbrella-shaped water mist at the outer edge of the collision spreader plate relative to the reference direction in the horizontal projection plane. The spray operation process refers to the overall spray scheduling strategy composed of the spray water supply pressure, the spray water flow rate, and the spray position parameters, which guides the cooling tower to achieve uniform wetting of the packing surface and optimal heat exchange in the next spray cycle.

[0088] Specifically, the controller equips each spray nozzle with an axis tilt adjustment unit and an outlet deflection direction angle adjustment unit that can be driven by a stepper motor or pneumatic actuator. The axis tilt adjustment unit adjusts the installation angle of the water distribution support assembly where the collision water distribution plate is located relative to the narrow-diameter spray section to achieve the overall tilt of the umbrella-shaped water mist. The outlet deflection direction angle adjustment unit rotates the collision water distribution plate assembly around the spray hole axis along the horizontal plane to achieve the deflection of the symmetry axis of the umbrella-shaped water mist in the azimuth angle. When the water distribution coverage change value shows that there is still insufficient wetting or wetting offset in a certain area, the controller calculates the position adjustment increment of the spray nozzle according to the formulas Δα=k_α×δA and Δφ=k_φ×δC, where Δα represents the spray nozzle axis tilt adjustment amount, Δφ represents the outlet deflection direction angle adjustment amount, δA represents the effective wetting area change value of the area, δC represents the azimuth component of the wetting center coordinate change vector of the area, and k_α and k_φ are the position compensation gains calibrated in the previous trial operation. The controller packages and sends the position adjustment increment of each spray nozzle with the water flow rate of the spray cloth obtained by S443, so that the coverage fluctuation of each section of the filler in the next spray cycle is reduced to within the preset threshold (±5%).

[0089] In one embodiment, calculating the energy efficiency optimization adjustment amount under the current environmental conditions based on the cooperative regression parameters further includes: S51: Obtain the physical structural parameters of the cooling tower; retrieve multi-source operating data of the cooling tower during its historical operating cycle, use the physical structural parameters and multi-source operating data together as training samples, and perform structure search and hyperparameter optimization on the preset candidate model family based on automatic machine learning methods to construct a digital twin model of equipment performance.

[0090] In this embodiment, physical structural parameters refer to the set of inherent parameters describing the geometry and material properties of the cooling tower, including the windward cross-section dimensions of the upper tower body, the height of the packing assembly and the corrugation specifications of the packing plates, the number of spray nozzles and the outer diameter of the spray nozzle outlet, the diameter of the fan impeller and the rated air volume, and the rated head and rated flow rate of the water pump. The historical operating cycle refers to the cumulative operating time interval of the cooling tower from its initial commissioning to the present moment, taking no less than three complete calendar years. Multi-source operating data refers to a collection of various heterogeneous data sources continuously collected by the controller within the above-mentioned interval, including environmental condition data, process parameter data, energy consumption data, and fault alarm data. Automatic machine learning methods refer to machine learning modeling methods that can autonomously complete structure search and hyperparameter optimization within a candidate model family without requiring manual specification of the model type and hyperparameters, such as AutoML, Bayesian optimization, and neural architecture search. The pre-set candidate model family refers to a set of alternative machine learning models pre-included in the search space, including multiple linear regression, random forest, gradient boosting tree, support vector regression, and multilayer perceptron.

[0091] Specifically, the controller retrieves multi-source operational data from at least one complete operating season, filters steady-state samples according to a preset sampling period and steady-state criteria, and removes transient disturbance samples and sensor anomaly samples to form the original sample library D_raw. The physical structure parameter set Φ_struct is broadcast and concatenated to the feature vector of each sample in D_raw to form a complete training sample D_train={(x_k,y_k)}, where the input vector x_k=[T_in,n_fan,f_pump,T_db,T_wb,p_atm,v_air,Φ_struct]ᵀ and the output vector y_k=[T_out, ]ᵀ、 =P_fan,act+P_pump,act represents the total power consumption of the system.

[0092] The controller divides D_train into training, validation, and test sets (in a 7:2:1 ratio) chronologically and initiates automatic machine learning optimization on a pre-defined candidate model family: The first layer performs a structure search, performing Bayesian optimization on the structural hyperparameters (tree depth and number of leaf nodes for GBRT / RF, number of hidden layers and neurons for MLP, number of attention heads and encoding layers for Transformer regression networks) for each model type within the candidate model family; the second layer performs hyperparameter optimization, further optimizing training hyperparameters such as learning rate, regularization coefficient, and Dropout rate within the candidate structures identified by the structure search; the root mean square error of T_out prediction on the validation set is used as the benchmark. The optimization objective is to minimize the weighted sum of the root mean square error of the predictions. The optimization is terminated by either a preset maximum number of iterations or an early stopping strategy. After the optimization is terminated, the model with the highest score on the validation set is solidified as the digital twin model M_DT of the equipment performance, and accuracy is verified on the test set.

[0093] S52: Through a digital twin model of equipment performance, the predicted outlet water temperature and the predicted total power consumption of the system are mapped and output under a given combination of environmental operating parameters and given fan speed and pump frequency.

[0094] In this embodiment, the given environmental operating parameters refer to the set of environmental components E=[T_in,T_db,T_wb,p_atm,v_air]ᵀ defined by S81~S82; the given combination of fan speed and pump frequency refers to the two-dimensional decision variables (n_fan,f_pump) fed into the digital twin model group by group by the embedded collaborative optimization controller during the traversal process; the predicted outlet water temperature and the predicted total power consumption of the system refer to the steady-state outlet water temperature estimates output by the digital twin model M_DT under the given E and (n_fan,f_pump) conditions. Estimated total power consumption of the steady-state system .

[0095] Specifically, at the beginning of each optimization cycle, the controller freezes the current environmental condition parameter E_now and concatenates E_now with (n_fan, f_pump) and Φ_struct to form the inference input vector. Input into the M_DT inference engine, according to the formula Complete one mapping output; M_DT supports batch inference, which can receive all generated candidate (n_fan, f_pump) combinations at once within a single optimization cycle and return the corresponding ones in parallel. and .

[0096] S53: An embedded collaborative optimization controller is constructed with the prediction of the outlet water temperature meeting the target set temperature as a hard constraint and the prediction of minimizing the total power consumption of the system as the objective function. The embedded collaborative optimization controller calls the equipment performance digital twin model to perform traversal calculations on feasible combinations of fan speed and pump frequency according to the preset optimization cycle, and performs optimization locking on the traversal results to obtain the optimal operating point that minimizes the total power consumption of the system.

[0097] In this embodiment, the target set temperature refers to the process target value T_out,set that the cooling tower outlet water temperature needs to reach, issued by the upper-level process system according to the production conditions; hard constraints refer to the limiting conditions that must be strictly met during the collaborative optimization process, and candidate solutions that violate the constraints will be directly judged as infeasible; the objective function refers to the scalar evaluation index that needs to be minimized (or maximized) during the collaborative optimization process; the embedded collaborative optimization controller refers to a lightweight optimization solver deployed in the local controller of the cooling tower, using M_DT as the surrogate model, (n_fan,f_pump) as the decision variable, and completing online optimization according to a preset optimization cycle, denoted as OPT_emb.

[0098] Specifically, OPT_emb constructs a collaborative optimization problem according to the following formula (2): min (n_fan,f_pump|E_now,Φ_struct) st (n_fan,f_pump|E_now,Φ_struct)≤T_out,set+ε_T n_fan,min≤n_fan≤n_fan,max f_pump,min≤f_pump≤f_pump,max f_pump≤f_pump,struct_max (2) In equation (2), ε_T represents the allowable engineering margin for the outlet water temperature (taken as 0.1~0.2℃); n_fan,min and n_fan,max are determined by the minimum stable speed and rated speed of the fan inverter, respectively; f_pump,min and f_pump,max are determined by the minimum stable frequency and rated frequency of the water pump inverter, respectively; f_pump,struct_max represents the upper limit of the water pump frequency corresponding to the upper limit of the dual-path flow of "overflow window bypass + jet hole mainstream" obtained by back-calculation of the S81 spray nozzle structural parameters according to f_pump,struct_max=(f_pump,N / Q_pump,N)×[N_sp×C_d×A_ow×√(2g×H_st)+N_sp×(π / 4)×d_jet²×√(2g×H_st)], which is used to ensure that the optimization solution will not drive the inlet pressure stabilizing cylinder to overflow and leave the pressure stabilizing working range of the spray nozzle.

[0099] Specifically, the controller formalizes the hard constraints as the inequality T_pred ≤ T_set + ε_set, and the objective function as min P_pred = P_fan_pred + P_pump_pred, where T_pred represents the predicted outlet water temperature, T_set represents the target set temperature, ε_set represents the outlet water temperature tolerance (taken as 0.2℃), P_pred represents the predicted total power consumption of the system, P_fan_pred represents the predicted power consumption of the fan, and P_pump_pred represents the predicted power consumption of the pump. The controller implements an embedded collaborative optimization controller module in C++ or Python. The module embeds an inference interface for the digital twin model of equipment performance, a constraint discrimination interface, and a target ranking interface, and interfaces with the S822 inference engine and inverter interface via Modbus TCP or OPC UA.

[0100] Furthermore, the preset optimization period refers to the period length (30s~300s, determined by the thermal inertia time constant of the cooling tower) of OPT_emb triggering a complete collaborative optimization calculation at a fixed time interval; the feasible combination refers to the set of candidate solutions obtained by discretizing (n_fan,f_pump) within the two-dimensional feasible domain defined by the four hard constraints in formula (2) at a preset step size; the traversal calculation refers to OPT_emb feeding the feasible combination set into M_DT group by group to obtain the corresponding The batch reasoning process; optimization locking refers to the decision-making process of filtering feasible solutions according to hard constraints in the traversal results, and then selecting the optimal solution according to the objective function; the optimal operating point refers to the point locked within this optimization cycle that satisfies all hard constraints and makes The smallest two-dimensional decision variable value is denoted as (n_fan*, f_pump*).

[0101] Specifically, OPT_emb executes the following process at each preset optimization cycle trigger time: The first step is to discretize the feasible region into candidate grids G={(n_fan,i,f_pump,j)} according to the step size Δn of n_fan (taken as 0.02·n_fan,N) and the step size of f_pump Δf (taken as 0.02·f_pump,N); The second step is to feed G into M_DT in batch inference to obtain the corresponding data for each grid point at once. ,ij, ,ij); The third step is to remove according to hard constraint 1. ,ij>T_out,set+ε_T grid points, to obtain the feasible solution set G_feasible; Fourth step, press within G_feasible Sort in ascending order, and take the first grid point in the sorted order as the optimal running point for this cycle (n_fan*, f_pump*); The fifth step is to perform a local refinement traversal on the neighborhood of (n_fan*, f_pump*) with a finer step size (Δn / 4, Δf / 4) to obtain and lock the refined optimal operating point.

[0102] S54: Calculate the energy efficiency optimization adjustment amount under the current optimization cycle based on the optimal operating point, and inject the energy efficiency optimization adjustment amount back into the co-regression parameters for closed-loop correction.

[0103] In this embodiment, the energy efficiency optimization adjustment amount refers to the two-dimensional adjustment amount set consisting of the difference between the optimal operating point (n_fan*, f_pump*) in the current optimization cycle and the actual operating point (n_fan, act, f_pump, act) issued in the previous control cycle, denoted as Δ_opt=[Δn_fan,Δf_pump]ᵀ; the collaborative regression parameter is the regression coefficient vector θ_reg used to link the fan speed and water pump frequency, which is maintained in real time by the controller according to the multivariate coupled regression algorithm; the closed-loop correction refers to the rolling correction process in which Δ_opt is injected back into θ_reg in a gain superposition manner, so that θ_reg gradually converges towards the global optimal direction indicated by the digital twin optimization in each optimization cycle.

[0104] Specifically, the energy efficiency optimization adjustment amount is calculated using the following formula (3): Δn_fan=n_fan*−n_fan,act Δf_pump=f_pump*−f_pump,act (3) And the energy efficiency optimization adjustment amount is reinjected into the co-regression parameters according to the following formula (4): θ_reg,new=θ_reg,old+η_opt×J_θ×Δ_opt (4) In equation (4), θ_reg,old and θ_reg,new represent the co-regression parameter vectors before and after back-injection, respectively; η_opt represents the optimization back-injection learning rate (taken as 0.05~0.20); and J_θ represents the Jacobian matrix of the co-regression parameter pair (n_fan,f_pump) (calculated and archived incrementally in each control cycle by the above multivariate coupled regression algorithm).

[0105] In one embodiment, a control method for an open crossflow cooling tower further includes: S501: Obtain meteorological forecast data for the cooling tower operating area for a future preset period of time. The meteorological forecast data includes ambient wet-bulb temperature, ambient dry-bulb temperature, outdoor wind speed, and solar radiation intensity.

[0106] In this embodiment, the future preset time period refers to the forecast time window that extends from the current moment into the future and covers one or more optimization cycles, ranging from 15 minutes to 6 hours.

[0107] S502: Input meteorological forecast data into the digital twin model of equipment performance, perform feedforward extrapolation on the predicted water temperature trend and the predicted total power consumption trend of the system within a preset future period, and obtain the predicted operating disturbance results of the cooling tower within the preset future period.

[0108] In this embodiment, feedforward inference refers to the process by which the controller uses meteorological forecast data as the environmental driving variable, the currently issued fan speed and water pump frequency as the decision-side holding variable, and the constructed equipment performance digital twin model M_DT as the mapping core to perform batch inference on the evolution trajectory of cooling tower outlet water temperature and total system power consumption within a future preset time period. The predicted outlet water temperature trend refers to the future outlet water temperature sequence obtained by feedforward inference and arranged in the forecast time sequence. The predicted total system power consumption trend refers to the future total system power consumption sequence obtained by feedforward inference and arranged in the forecast time sequence. The operating condition disturbance prediction result refers to the set of disturbance direction, disturbance amplitude, and disturbance arrival time of the cooling tower operating condition within the future preset time period, jointly characterized by the deviation curves of the predicted outlet water temperature trend and the predicted total system power consumption trend relative to the current steady-state reference value.

[0109] Specifically, the controller splices the ambient wet-bulb temperature, ambient dry-bulb temperature, outdoor wind speed, and solar radiation intensity in the weather forecast buffer according to the forecast time sequence into a feedforward input vector sequence. The feedforward input vector at each time moment, together with the currently operating fan speed, water pump frequency, and defined physical structure parameters, is sent to M_DT. The controller obtains the estimated outlet water temperature and estimated total system power consumption at each time moment in the future preset time period in batches according to the forecast time sequence. After continuous arrangement, the estimated outlet water temperature trend curve and the estimated total system power consumption trend curve are formed.

[0110] The controller further identifies operating condition disturbance characteristics on two trend curves: when the estimated outlet water temperature trend curve exceeds the upper limit of the engineering margin of the target set temperature at a certain time in the future, it is recorded as an "outlet water temperature upward disturbance event" and its arrival time and maximum upward disturbance amplitude are marked; when the estimated total system power consumption trend curve shows a continuous rise or a continuous fall relative to the current benchmark at a certain time in the future, it is recorded as a "power consumption upward disturbance segment" or a "power consumption downward disturbance segment" respectively, and its start and end times and average disturbance amplitude are marked.

[0111] S503: Based on the predicted operating condition disturbance results, predictive feedforward compensation is performed on the optimal operating point to obtain a predictive operating parameter combination that includes the feedforward compensation amount. The predictive operating parameter combination is used to pre-adjust the fan speed and pump frequency before the operating condition disturbance arrives.

[0112] In this embodiment, the optimal operating point is the value of a two-dimensional decision variable consisting of the fan speed and the pump frequency, which has been locked within the current optimization cycle. Predictive feedforward compensation refers to the controller superimposing feedforward adjustment amounts with opposite directions and matching amplitudes on the optimal operating point according to the predicted operating disturbance results before the disturbance arrives, so that the fan and pump can complete the pre-action adjustment process before the disturbance arrives. The feedforward compensation amount refers to the pre-adjustment amount applied to the fan speed and pump frequency of the optimal operating point, which is determined by the amplitude of the disturbance in the outlet water temperature and the direction of the power consumption disturbance. The predictive operating parameter combination refers to the new operating parameter combination formed by superimposing the feedforward compensation amount on the optimal operating point, which is used to issue and execute before the operating disturbance arrives.

[0113] Specifically, the controller generates the feedforward compensation amount according to the following strategy: When the predicted disturbance results show an "outlet water temperature disturbance event", the controller applies a positive feedforward increment to the fan speed and the pump frequency at the optimal operating point one control cycle before the event arrives. This allows the cooling tower to increase the airflow and circulating water flow in advance before the disturbance arrives, thus establishing a heat dissipation margin to avoid transient overshoot of the actual outlet water temperature. The feedforward increment amplitude is given according to the preset linear gain of the disturbance amplitude.

[0114] When the operating condition disturbance prediction results only contain a "power consumption disturbance segment" without an accompanying outlet water temperature disturbance event, the controller will preferentially apply a small negative feedforward increment to the pump frequency during this period, and re-verify the outlet water temperature hard constraint through local refinement traversal. After the verification is passed, the negative increment will be issued as a feedforward compensation amount, so that the system can further reduce the total power consumption while ensuring that the outlet water temperature meets the standard.

[0115] When the power consumption under disturbance prediction results contain a "power consumption under disturbance segment", the controller maintains the optimal operating point unchanged, does not apply feedforward compensation, and only records this period as an energy-saving potential period for historical database statistics.

[0116] The controller superimposes the feedforward compensation amount and the optimal operating point by dimension to form a predictive combination of operating parameters, and reserves an advance of one control cycle according to the arrival time of the operating disturbance, and writes it into the instruction buffer area of ​​the fan inverter and the water pump inverter.

[0117] S504: Based on the current load range and wet-bulb temperature range of the environmental conditions, the spray strategy corresponding to the optimal operating point is adaptively and dynamically adjusted.

[0118] In this embodiment, the current environmental condition refers to the operating condition characterized by the measured / forecasted environmental parameters corresponding to the current moment in the meteorological forecast data stream; the load interval refers to several discrete segments pre-divided by the controller according to the historical operating load distribution of the cooling tower, including at least low load interval, medium load interval, and high load interval, with the interval boundaries obtained from the quantile statistics of the historical load; the wet-bulb temperature interval refers to several discrete segments pre-divided by the controller according to the meteorological data of the operating area, including at least low wet-bulb interval, medium wet-bulb interval, and high wet-bulb interval, with the interval boundaries obtained from the quantile statistics of the local annual wet-bulb temperature; the spray strategy is an overall spray scheduling strategy composed of spray water supply pressure, spray water flow rate, and spray position parameters; the adaptive linkage adjustment refers to the coordinated adjustment process in which the controller selects the corresponding spray strategy adjustment direction from the preset spray strategy scheduling table using the current load interval and wet-bulb temperature interval as a joint index, and synchronously issues it to the optimal operating point for execution.

[0119] For example, before commissioning, the controller generates a sprinkler strategy scheduling table offline according to the following principles and then stores it in the controller strategy library: Low load × low wet bulb conditions: The heat dissipation demand is small and the theoretical heat dissipation margin is sufficient. The spraying strategy is adjusted to "low spray water supply pressure + low spray water flow rate + tightening the spray coverage area" so that the spray water volume is concentrated on the central area of ​​the packing material, avoiding water and power consumption redundancy caused by over-spraying under low load. Low load × high wet bulb conditions: Although the heat dissipation demand is small, the theoretical heat dissipation margin is limited. The spraying strategy is adjusted to "medium spray water supply pressure + medium spray water flow rate + maintain standard spray coverage" to ensure that the surface of the packing is fully wetted and to maximize the approach to the wet bulb temperature limit. Medium load × medium wet bulb condition: As the calibration benchmark condition for the optimal operating point, the spraying strategy remains unchanged from the benchmark spraying scheduling strategy calibrated by S445; High load × low wet bulb condition: The heat dissipation demand is large but the theoretical heat dissipation margin is sufficient. The spraying strategy is adjusted to "medium-high spraying water supply pressure + high spraying water flow rate + maintaining standard spraying coverage" to match the increased circulating water flow rate under high load. High load × high wet bulb conditions: The heat dissipation demand is large and the theoretical heat dissipation margin is limited. The spraying strategy is adjusted to "high spray water supply pressure + high spray water flow rate + expanded spray coverage area" so that the umbrella-shaped atomized water mist forms the largest effective wetting spot on the surface of the packing material. Together with the fan to increase the air volume, the outlet water temperature is brought as close as possible to the wet bulb temperature limit.

[0120] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0121] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the control method for an open crossflow cooling tower as described above.

[0122] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0123] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0124] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. An open-type crossflow cooling tower, characterized in that, include: An upper tower body (1) and a lower tower body (2) are stacked vertically and detachably connected by fasteners (3); the upper tower body (1) is equipped with a built-in water distribution pipeline, which includes two sets of parallel water inlet branch pipes (4) symmetrically arranged on both sides of the upper tower body (1); each set of parallel water inlet branch pipes (4) is provided with an energy consumer, a water distribution tray, a spray nozzle (5) and a packing assembly (6) in sequence below it; wherein, the energy consumer is located at the outlet of the parallel water inlet branch pipe (4). Below the water end, it is used to receive the water flow discharged from the same-path water inlet branch pipe (4); the water distribution plate receives the water flow overflowing from the periphery of the energy consumer and guides the water flow to the spray nozzle (5); a water collection plate (7) is provided below the packing assembly (6), the bottom of the water collection plate (7) is inclined in the horizontal direction with one side higher and the other side lower, and a water outlet is provided on the lower side of the water collection plate (7), the water outlet is connected to the water inlet on the external water inlet side through a connecting pipe to form a closed circulating water circuit.

2. The open crossflow cooling tower according to claim 1, characterized in that, The spray nozzle (5) is a specially designed spray nozzle (5) with a water storage and pressurization structure. The spray nozzle (5) has a water storage chamber inside for storing water level under low flow conditions to increase spray pressure, so that the cooling tower can maintain uniform distribution of spray water mist on the surface of the packing assembly (6) within the working range of 30% to 120% of the nominal flow rate. The water inlet side of the spray nozzle (5) is connected to the lower outlet of the water distribution plate, and the upper part of the water distribution plate corresponds to the peripheral overflow outlet of the energy consumer, forming a water distribution path with progressively decreasing speed.

3. An open crossflow cooling tower according to claim 1, characterized in that, The packing assembly (6) is composed of several packing units arranged side by side in the horizontal direction. Each packing unit includes two mirror-symmetrical packing plates. The two packing plates are connected and positioned by several support rods that pass through preset holes. After adjacent packing units are assembled, a honeycomb heat exchange channel is formed inside the packing assembly (6). Each packing plate is provided with an air inlet section and a water collection section in sequence along the airflow direction. The air inlet section is used to guide the external air to enter the honeycomb heat exchange channel evenly. The water collection section is used to block the splashing of water droplets after heat exchange. The water flowing out from the lower end of the packing assembly (6) falls into the water collection pan (7), and then converges along the bottom of the inclined pan to the water outlet, and returns to the built-in water distribution pipeline through the connecting pipe.

4. An open crossflow cooling tower according to claim 3, characterized in that, Each packing unit has several sets of alignment holes (61) spaced apart along the height direction on two packing plates. There are four support rods that pass through the corresponding alignment holes (61) to lock the two packing plates together. Adjacent packing units are connected in series through the extension ends of the support rods to form an integral packing assembly (6) in the upper tower body (1).

5. A control method for an open crossflow cooling tower, characterized in that, Applied to an open crossflow cooling tower as described in any one of claims 1 to 4, the method comprises: Obtain the current environmental conditions and trial operation data of the cooling tower to obtain the initial operating parameters of the cooling tower fan and water pump; During the process of the cooling tower circulating heat dissipation according to the initial operating parameters, the temperature deviation of the actual outlet water temperature of the cooling tower relative to the target set temperature is calculated. The temperature deviation is subjected to energy efficiency optimization and correction processing to obtain the collaborative regression parameters of the cooling tower; Based on the collaborative regression parameters, the energy efficiency optimization adjustment amount under the current environmental conditions is calculated, and the fan speed and water pump frequency of the cooling tower are adjusted synchronously based on the energy efficiency optimization adjustment amount.

6. The control method for an open crossflow cooling tower according to claim 5, characterized in that, The step of calculating the energy efficiency optimization adjustment amount under the current environmental conditions based on the collaborative regression parameters, and synchronously adjusting the fan speed and water pump frequency of the cooling tower according to the energy efficiency optimization adjustment amount, includes: Based on the collaborative regression parameters, obtain the benchmark heat transfer load corresponding to the current packing heat exchange interface and the deviation cooling load corresponding to the temperature deviation; Based on the baseline heat exchange load and the deviation cooling load, the energy efficiency optimization adjustment of the cooling tower under the current environmental conditions is calculated comprehensively. The fan speed of the cooling tower is adjusted according to the energy efficiency optimization adjustment amount, and the gas-liquid heat balance ratio between the optimized heat dissipation and the corresponding circulating water evaporation at the same time node is controlled according to the fan speed. The spray water distribution rate of the cooling tower is adjusted according to the gas-liquid heat balance ratio, and synchronous spraying is performed in conjunction with the blower speed.

7. The control method for an open crossflow cooling tower according to claim 6, characterized in that, The step of adjusting the spray water distribution rate of the cooling tower according to the gas-liquid heat balance ratio and coordinating it with the blower speed for synchronous spraying also includes: The wetting deviation angle of the packing layer and the spray coverage diameter of the spray nozzles are obtained after the cooling tower is sprayed with water. The spray water supply pressure for the next spray cycle is calculated based on the wetting deviation angle of the packing layer, and the deviation compensation for the wetting deviation angle of the previous packing layer is performed based on the spray water supply pressure. Adjust the water flow rate of the spray cloth according to the spray coverage diameter and the corresponding spray nozzle outlet outer diameter; The water flow rate of the spray cloth is calculated using formula (1): (1) Where W represents the spray water flow rate, K represents the expansion coefficient of the overflow pressure stabilization-collision water diffusion composite structure, N represents the number of spray nozzles, D1 represents the outer diameter of the spray nozzle outlet, D2 represents the spray coverage diameter of the spray nozzle, and v represents the radial diffusion velocity of the umbrella-shaped atomized water mist. Based on the arrangement of the spray nozzles of the cooling tower, the water distribution coverage parameters of each packing section are obtained, and the water distribution coverage change value is calculated after the wetting deviation angle of the packing layer is compensated. The spray position parameters for the next spray cycle are adjusted based on the change in water coverage, and the spray operation process of the cooling tower is optimized based on the adjusted spray position parameters and the spray water flow rate.

8. The control method for an open crossflow cooling tower according to claim 6, characterized in that, The step of calculating the energy efficiency optimization adjustment amount under the current environmental conditions based on the collaborative regression parameters further includes: Obtain the physical structure parameters of the cooling tower; retrieve multi-source operation data of the cooling tower during its historical operation cycle, use the physical structure parameters and the multi-source operation data together as training samples, and perform structure search and hyperparameter optimization on a preset candidate model family based on an automatic machine learning method to construct a digital twin model of equipment performance. The digital twin model of the equipment performance is used to map and output the predicted outlet water temperature and the predicted total power consumption of the system under a given combination of environmental operating parameters and given fan speed and pump frequency. An embedded collaborative optimization controller is constructed with the predicted effluent temperature meeting the target set temperature as a hard constraint and the total power consumption of the prediction system as the objective function. The embedded collaborative optimization controller calls the device performance digital twin model according to a preset optimization cycle to perform traversal calculations on feasible combinations of the fan speed and the water pump frequency, and performs optimization locking on the traversal results to obtain the optimal operating point that minimizes the total power consumption of the system. Based on the optimal operating point, the energy efficiency optimization adjustment amount under the current optimization cycle is calculated, and the energy efficiency optimization adjustment amount is injected back into the collaborative regression parameters for closed-loop correction.

9. The control method for an open crossflow cooling tower according to claim 6, characterized in that, The method also includes: Obtain meteorological forecast data for the operating area of ​​the cooling tower within a preset time period in the future. The meteorological forecast data includes ambient wet-bulb temperature, ambient dry-bulb temperature, outdoor wind speed, and solar radiation intensity. The meteorological forecast data is input into the digital twin model of the equipment performance, and the predicted water temperature trend and the predicted total power consumption trend of the system are fed forward to obtain the predicted operating condition disturbance result of the cooling tower in the future preset period. Based on the predicted operating condition disturbance, predictive feedforward compensation is performed on the optimal operating point to obtain a predictive operating parameter combination that includes the feedforward compensation amount. The predictive operating parameter combination is used to pre-adjust the fan speed and the water pump frequency before the operating condition disturbance arrives. Based on the current load range and wet-bulb temperature range of the environmental conditions, the spraying strategy corresponding to the optimal operating point is adaptively and dynamically adjusted.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the control method for an open crossflow cooling tower as described in any one of claims 5 to 9.

Images