Liquid cooling water tower control method and system

Abstract

The invention relates to the technical field of liquid cooling water tower control, in particular to a liquid cooling water tower control method and system.The control method comprises the steps that the water inlet temperature Tin, the water outlet temperature Tout and the environment temperature Tenv of a cooling water tower are collected in real time, and a dynamic adjustment model is established based on the water inlet temperature Tin, the water outlet temperature Tout and the environment temperature Tenv to calculate the target cooling efficiency eta target; and according to the difference value between the eta target and the real-time cooling efficiency eta real, the cooling water flow Q and the fan rotating speed N are dynamically adjusted, and the optimal ratio of Q to N is determined through an energy efficiency optimization algorithm to reduce energy consumption. And when the cooling load is suddenly changed, a feedforward-feedback composite control mode is started. A multi-stage safety control strategy is set, and a thermal runaway prediction model is adopted to prevent risks in advance. The corresponding system comprises a distributed temperature sensing array, a frequency conversion speed regulation module, an edge calculation unit and an emergency cooling execution mechanism. The cooling efficiency can be improved, the energy consumption can be reduced, and the system stability and safety can be enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid-cooled cooling tower control, and specifically to a liquid-cooled cooling tower control method and system thereof. Background Art

[0002] With the rapid development of information technology, the scale and computing power of data centers have been continuously improved; modern data centers integrate a large number of servers, storage devices and network devices, and these devices will generate huge amounts of heat during operation; excessive temperature will seriously affect the performance and lifespan of the devices, and even lead to device failures. Therefore, effective heat dissipation has become a key factor for the stable operation of data centers; the traditional air-cooled heat dissipation method gradually exposes problems of insufficient heat dissipation capacity and excessive energy consumption when facing high-power density devices; as an efficient heat dissipation method, liquid-cooling technology has been more and more widely used in the field of data centers because it can take away more heat and has better energy-saving potential; as an important part of the liquid-cooling system, the performance and control strategy of the liquid-cooled cooling tower directly affect the heat dissipation effect and energy efficiency of the entire data center. Limitations of traditional cooling tower control methods:

[0003] Traditional cooling tower control methods usually adjust based on a single parameter. For example, only the outlet water temperature is used to adjust the flow rate of the cooling water pump or the rotational speed of the fan; this control method is too simple and cannot comprehensively consider various factors during the operation of the cooling tower; in actual operation, the cooling effect of the cooling tower is not only related to the outlet water temperature, but also affected by multiple factors such as the inlet water temperature, ambient temperature, humidity, and cooling load; single-parameter control cannot accurately reflect the actual needs of the system, and it is easy to cause untimely adjustment or over-adjustment, thereby affecting the cooling efficiency and energy consumption.

[0004] Most traditional control methods adopt fixed control strategies and lack dynamic adaptability to changes in the system operating conditions; in the actual operation of data centers, the cooling load will fluctuate frequently with changes in business requirements, and environmental conditions (such as temperature and humidity) will also change with seasons and weather; traditional control methods cannot perceive these changes in real time and adjust control parameters in a timely manner, making it difficult for the cooling tower to maintain the best operating state under different working conditions; for example, when the cooling load suddenly increases, traditional control methods may not be able to respond quickly, resulting in an increase in equipment temperature and affecting the normal operation of the data center; while when the cooling load is small, it may not be able to reduce energy consumption in a timely manner, causing energy waste.

[0005] Therefore, in view of the above problems, a liquid-cooled cooling tower control method and system thereof are proposed. Summary of the Invention

[0006] The purpose of the present invention is to provide a liquid-cooled cooling tower control method and system to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A liquid-cooled cooling tower control method includes the following steps:

[0009] Step S1: Real-time collect the inlet water temperature T in of the cooling tower, the outlet water temperature T out and the ambient temperature T env .

[0010] Step S2: Based on the inlet water temperature T in , the outlet water temperature T out and the ambient temperature T env , establish a dynamic adjustment model and calculate the target cooling efficiency η target .

[0011] Step S3: According to the difference between η target and the real-time cooling efficiency η real , dynamically adjust the cooling water flow Q and the fan speed N;

[0012] Step S4: Calculate the optimal ratio of Q and N through an energy efficiency optimization algorithm to make the system satisfy η real ≥η target and the energy consumption value P≤P max , where P is the real-time energy consumption value of the cooling tower system, and P max is the maximum allowable energy consumption value of the cooling tower system;

[0013] Step S5: When a sudden change in the cooling load is detected, start the feedforward-feedback composite control mode and synchronously adjust the cooperative working parameters of the cooling water circulation pump and the fan.

[0014] Preferably, in step S2, the calculation formula of the dynamic adjustment model is:

[0015] η target =α×(T in -T env )+β×(dT / dt)+γ×Q hist , where α, β, γ

[0016] are dynamic weight coefficients used to adjust the influence degree of different parameters on the calculation of the target cooling efficiency; Q hist is the historical average flow rate; dT / dt is the temperature change rate.

[0017] Preferably, in step S3, the dynamic adjustment includes:

[0018] When Δη = ηtarget -η real > Δη th When Δη th is the set cooling efficiency difference threshold;

[0019] When Δη ≤ Δη th fine-tuning of the fan speed N is adopted, and the adjustment step size does not exceed 2% of the rated speed.

[0020] Preferably, in step S4, the energy efficiency optimization algorithm includes:

[0021] Establish a three-dimensional mapping table to store the correspondence of Q-N-P;

[0022] Use the genetic algorithm to search for the minimum P value combination that satisfies the condition of η real ≥ η target in the mapping table;

[0023] When the ambient temperature T env > 30 °C, the humidity parameter H is introduced for four-dimensional optimization calculation, where H is the real-time humidity of the environment where the cooling water tower is located.

[0024] Preferably, in step S5, the feedforward-feedback composite control mode includes:

[0025] Real-time monitoring of the pressure change in the cooling pipeline through a pressure sensor;

[0026] When the pressure change rate exceeds the set threshold, adjust the frequency conversion parameters of the cooling water circulation pump in advance;

[0027] Synchronously establish a predictive control model for the fan speed to compensate for the adjustment delay caused by the system inertia.

[0028] Preferably, it further includes:

[0029] Set a multi-level safety control strategy. When it is detected that T out > T safe start the emergency cooling circuit;

[0030] The emergency cooling circuit includes a standby heat exchanger and a cold storage device arranged in parallel, and the switching response time < 5 seconds, where T safe is the safety temperature threshold of the cooling water tower outlet.

[0031] Preferably, the safety control strategy includes:

[0032] Construct a thermal runaway prediction model based on the LSTM neural network, and the input parameters include:

[0033] The change rate of the conductivity of the cooling water is the ratio of the change in the conductivity of the cooling water within a set time interval to that time interval.

[0034] The pipeline vibration spectrum characteristics are the characteristic information obtained by performing spectrum analysis on the pipeline vibration signal.

[0035] The motor winding temperature rise curve is the curve of the motor winding temperature changing with time.

[0036] When the predicted risk value > 0.7, start preventive cooling measures in advance.

[0037] Preferably, the liquid-cooled cooling water tower system includes:

[0038] A distributed temperature sensing array arranged on the inlet main pipe, outlet main pipe, and heat dissipation fins at different heights of the cooling water tower.

[0039] A variable frequency speed regulation module that integrally controls the cooling water pump and the axial flow fan.

[0040] An edge computing unit with the dynamic adjustment model and the energy efficiency optimization algorithm built-in.

[0041] An emergency cooling actuator that includes a quick solenoid valve and a cold storage tank linkage device.

[0042] Compared with the prior art, the beneficial effects of the present invention are:

[0043] High cooling efficiency: The dynamic adjustment model accurately calculates the target cooling efficiency, and the multi-parameter collaborative adjustment can quickly respond to different working conditions, greatly improving the cooling efficiency.

[0044] Energy saving and consumption reduction: The energy efficiency optimization algorithm determines the optimal ratio, combined with variable frequency speed regulation, to minimize energy consumption while ensuring the cooling effect.

[0045] Rapid response: The feedforward-feedback composite control mode can quickly respond to sudden changes in the cooling load, compensate for the adjustment delay, and maintain stable cooling performance.

[0046] Safe and reliable: The multi-level safety control strategy and the thermal runaway prediction model can timely respond to overheating conditions and prevent risks in advance to ensure the safety of the system.

[0047] Accurate data: The distributed temperature sensing array comprehensively and highly accurately collects temperature data, providing strong support for the precise control of the system. Description of the Drawings

[0048] Figure 1 It is a schematic flow chart of a liquid-cooled cooling water tower control method of the present invention. Detailed Embodiments

[0049] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0050] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the specification drawings and specific implementation manners.

[0051] Embodiment:

[0052] Please refer to Figure 1 , this embodiment provides a technical solution:

[0053] A liquid-cooled cooling tower control method includes the following steps:

[0054] Step S1: Real-time collect the inlet water temperature T in of the cooling tower, the outlet water temperature T out and the ambient temperature T env ;

[0055] Step S2: Based on the inlet water temperature T in , the outlet water temperature T out and the ambient temperature T env , establish a dynamic adjustment model and calculate the target cooling efficiency η target ;

[0056] Step S3: According to the difference between η target and the real-time cooling efficiency η real , dynamically adjust the cooling water flow Q and the fan speed N;

[0057] Step S4: Calculate the optimal ratio of Q and N through an energy efficiency optimization algorithm to make the system satisfy η real ≥η target and the energy consumption value P≤P max , where P is the real-time energy consumption value of the cooling tower system, and P max is the maximum allowable energy consumption value of the cooling tower system;

[0058] Step S5: When a sudden change in the cooling load is detected, start the feedforward-feedback composite control mode and synchronously adjust the cooperative working parameters of the cooling water circulation pump and the fan.

[0059] In this embodiment, the specific steps of Step S1 include:

[0060] Step S1-1: Sensor selection: For the inlet water temperature T in and the outlet water temperature T outFor the acquisition, a high-precision PT100 temperature sensor is selected. Its measurement accuracy can reach ±0.1°C, which can ensure the accuracy and stability of the collected data and provide a reliable data basis for subsequent control algorithms;

[0061] Step S1-2: The inlet and outlet water temperature sensors are installed at the inlet main pipe of the cooling water tower near the entrance of the cooling tower. The PT100 temperature sensor is installed perpendicular to the water flow direction so that the sensor can fully contact the water flow to accurately measure the actual inlet water temperature; at the position of the outlet main pipe near the outlet of the cooling tower, the PT100 sensor is also installed vertically to accurately collect the outlet water temperature; to ensure data reliability, 3 groups of redundant PT100 sensors are set in the inlet main pipe, and the measurement accuracy can be improved through data fusion or comparison verification;

[0062] Step S1-3: For the installation of the ambient temperature sensor, an ambient temperature sensor with functions of wind protection, rain protection and sun protection is selected to collect the ambient temperature T env ; It is installed at a place 5-10 meters away from the cooling water tower and with good ventilation, and the height from the ground is about 1.5-2 meters to simulate the ambient temperature in the human activity area and reduce the interference of local factors such as the self-cooling of the cooling tower;

[0063] Step S1-4: Set the data acquisition frequency To achieve real-time acquisition, the data acquisition frequencies of the inlet water temperature T in , outlet water temperature T out and ambient temperature T env sensors are all set to not less than 10Hz, so as to capture temperature changes in time, provide timely and accurate data input for the dynamic regulation model, ensure that the system can quickly respond to temperature fluctuations, and achieve efficient and precise control of the cooling water tower.

[0064] In this embodiment, step S2 specifically includes:

[0065] Step S2-1: Define the construction objective of the dynamic regulation model:

[0066] The objective of constructing the dynamic regulation model is based on the real-time collected inlet water temperature T in , outlet water temperature T out and ambient temperature T env to calculate the target cooling efficiency η target ; This target cooling efficiency is used for subsequent comparison with the real-time cooling efficiency η real to realize the dynamic regulation of the cooling water flow Q and the fan speed N, so that the cooling water tower system can operate efficiently under different working conditions;

[0067] Step S2-2: Determine the dynamic regulation model formula:

[0068] The formula η target = α×(Tin -T env ) + β×(dT / dt) + γ×Q hist , where:

[0069] α, β, and γ are dynamic weight coefficients used to adjust the influence degree of different parameters on the calculation of the target cooling efficiency. These three coefficients are used to adjust the influence degree of different parameters on the calculation of the target cooling efficiency; their values are not fixed but are dynamically adjusted according to factors such as the actual operating conditions of the cooling tower, seasonal changes, and equipment aging degree. For example, in a high-temperature environment in summer, the influence of the ambient temperature on the cooling efficiency is relatively large, and the value of α can be appropriately increased; in the initial stage of system operation, the equipment performance is relatively stable, and an initial set of weight coefficients can be set according to historical data and experience.

[0070] T in -T env , which reflects the difference between the inlet water temperature and the ambient temperature of the cooling tower. Generally speaking, the greater the difference between the inlet water temperature and the ambient temperature, the greater the cooling potential of the cooling tower. Therefore, this term occupies an important proportion in the calculation of the target cooling efficiency.

[0071] dT / dt is the temperature change rate, which represents the ratio of the temperature change amount to the time interval within a certain time interval. It can be obtained by performing differential calculation on the data of the inlet water temperature T in or the outlet water temperature T out at consecutive multiple sampling moments. The temperature change rate reflects the thermal dynamic characteristics of the cooling tower system. When the temperature change rate is relatively large, it indicates that the thermal load of the system changes relatively fast, and the cooling strategy needs to be adjusted in a timely manner. Therefore, this term is incorporated into the dynamic adjustment model.

[0072] Q hist is the historical average flow rate, which refers to the average value of the cooling water flow rate within a certain historical time period. It reflects the operating state of the cooling tower in the past period and has certain reference value for the calculation of the current target cooling efficiency. For example, if the historical average flow rate is relatively large, it indicates that the cooling demand of the system was relatively high in the past, and this factor needs to be appropriately considered when calculating the current target cooling efficiency.

[0073] Step S2-3: Determination of dynamic weight coefficients:

[0074] Initial value setting:

[0075] In the initial stage of system startup, according to the design parameters of the cooling tower, the operating experience of previous similar equipment, and theoretical calculations, a set of initial values is set for the dynamic weight coefficients α, β, and γ. For example, the initial value of α can be set to 0.6, the initial value of β can be set to 0.2, and the initial value of γ can be set to 0.2.

[0076] Dynamic adjustment:

[0077] As the system runs, continuously collect the actual operation data, and dynamically adjust the dynamic weight coefficients according to the system's performance indicators (such as cooling efficiency, energy consumption, etc.); an adaptive control algorithm can be adopted, such as a fuzzy control algorithm or a neural network algorithm, to automatically adjust the value of the weight coefficient according to the real-time collected data and the preset performance indicators, so as to make the calculation of the target cooling efficiency more accurate and the system operation more efficient;

[0078] Step S2-4: Calculate the target cooling efficiency:

[0079] After determining the dynamic adjustment model formula and the dynamic weight coefficients, substitute the real-time collected T in , T env , the calculated dT / dt, and Q hist into the formula η target = α×(T in - T env ) + β×(dT / dt) + γ×Q hist to calculate the target cooling efficiency η target . This calculation process can be completed by an edge computing unit or other devices with computing capabilities, and the calculation result will be used for subsequent control strategy formulation.

[0080] In this embodiment, step S3 specifically includes the following steps:

[0081] Step S3-1: Calculate the cooling efficiency difference:

[0082] First, clarify the calculation method of the cooling efficiency difference Δη, that is, Δη = η target - η real , where η target is the target cooling efficiency calculated based on the dynamic adjustment model in step S2, and η real is the current actual cooling efficiency of the cooling water tower; the actual cooling efficiency η real can be calculated through the heat exchange principle of the cooling water tower. For example where Q c is the actual heat exchange amount, which can be calculated according to the flow rate of the cooling water, the temperature difference between the inlet and outlet, and the specific heat capacity of water, that is, Q c = c×m×(T in - T out ), c is the specific heat capacity of water, and m is the mass flow rate of the cooling water; Q max is the theoretical maximum heat exchange amount, which is related to factors such as the ambient temperature and the flow rate of the cooling water;

[0083] Step S3-2: Set the cooling efficiency difference threshold Δη th :

[0084] Set a threshold value Δη for the cooling efficiency difference according to the design parameters, operating experience of the cooling water tower, and the performance requirements of the system. th This threshold value is used to determine whether to preferentially adjust the cooling water flow rate Q or adopt a refined adjustment of the fan speed N; for example, for a cooling water tower of a specific specification, after multiple tests and optimizations, Δη th is set to 5%.

[0085] Step S3-3: Adjustment strategy when Δη ≤ Δη th :

[0086] Preferentially adjust the cooling water flow rate Q:

[0087] When Δη = η target - η real > Δη th , it indicates that the gap between the actual cooling efficiency and the target cooling efficiency is large. At this time, preferentially adjust the cooling water flow rate Q because adjusting the cooling water flow rate can quickly change the heat exchange capacity of the cooling water tower and has a relatively significant impact on improving the cooling efficiency;

[0088] The adjustment amplitude is non-linearly positively correlated with Δη;

[0089] The adjustment amplitude does not have a simple linear relationship with Δη, but is adjusted in a non-linearly positive correlation manner; for example, a non-linear function f(Δη) can be established to determine the adjustment amplitude Δη, such as ΔQ = k × Δη n , where k is the proportionality coefficient and n is an exponent greater than 1, which can be adjusted according to the actual situation. The purpose of this design is to increase the adjustment amplitude when Δη is large to quickly narrow the gap with the target cooling efficiency; when Δη is small, appropriately reduce the adjustment amplitude to avoid over-adjustment causing system instability;

[0090] Feedback control during the adjustment process:

[0091] During the process of adjusting the cooling water flow rate, continuously monitor the change of the cooling efficiency difference Δη; if Δη gradually decreases and approaches Δη th as the flow rate is adjusted, then appropriately reduce the adjustment amplitude; if Δη does not change significantly or continues to increase, it is necessary to re-evaluate the adjustment strategy or increase the adjustment amplitude;

[0092] Step S3-4: Adjustment strategy when Δη ≤ Δη th :

[0093] Adopt a refined adjustment of the fan speed N:

[0094] When Δη ≤ Δη thWhen it indicates that the actual cooling efficiency is relatively close to the target cooling efficiency, fine-tuning of the fan speed N is adopted at this time; because the adjustment of the fan speed has a relatively slow and precise impact on the cooling efficiency, it is suitable for fine-tuning when approaching the target cooling efficiency to achieve precise control;

[0095] The adjustment step size does not exceed 2% of the rated speed. To avoid excessive adjustment amplitude causing system instability or increased energy consumption, it is stipulated that the adjustment step size of the fan speed does not exceed 2% of the rated speed; for example, if the rated speed of the fan is 1500 r / min, the maximum step size for each adjustment is 1500×2% = 30 r / min; during the adjustment process, it is also necessary to continuously monitor the change of the cooling efficiency difference Δη, and decide whether to continue the adjustment and the adjustment direction according to the change situation;

[0096] Step S3-5: Safety monitoring during the adjustment process:

[0097] During the adjustment of the cooling water flow Q and the fan speed N, various parameters of the system are monitored in real time, such as the cooling water pressure, the motor current, etc., to ensure that the adjustment process will not affect the safe operation of the system; if abnormal situations occur during the adjustment process, such as too high pressure, too large current, etc., the adjustment should be stopped immediately and corresponding protection measures should be taken.

[0098] In this embodiment, step S4 specifically includes the following steps:

[0099] Step S4-1: Establish a three-dimensional mapping table:

[0100] Data collection and preparation:

[0101] Under different operating conditions of the cooling water tower system, a large amount of data on the cooling water flow Q, the fan speed N, and the system energy consumption value P are collected; specifically, the cooling water flow Q is changed by adjusting the frequency of the cooling water circulation pump, the fan speed N is changed by adjusting the frequency converter of the fan, and at the same time, the energy consumption value P of the system is monitored in real time by using a power sensor; when collecting data, various typical operating conditions that the system may encounter should be covered, such as different environmental temperatures, different cooling loads, etc.;

[0102] Mapping table construction: The collected data of Q, N, and P are sorted out to construct a three-dimensional mapping table; in this mapping table, each combination of Q and N corresponds to a specific P value; Q can be used as one coordinate axis, N as another coordinate axis, and P as the corresponding value stored in the corresponding coordinate position; this mapping table reflects the internal relationship between the cooling water flow, the fan speed, and the system energy consumption, providing basic data for subsequent optimization calculations;

[0103] Step S4-2: Use the genetic algorithm to search for the combination with the minimum P value:

[0104] Determine the fitness function. The fitness function is the basis for the genetic algorithm to search. Its goal is to minimize the system energy consumption value P under the condition of satisfying η real ≥η target ; Therefore, the fitness function can be defined as (when η real ≥η target ), when the condition η real ≥η target is not satisfied, F = 0; In this way, the genetic algorithm will preferentially select those Q-N combinations that can not only meet the cooling efficiency requirements but also minimize the energy consumption;

[0105] Initialize the population. Randomly generate a certain number of Q-N combinations as the initial population; Each Q-N combination can be regarded as an individual in the genetic algorithm, and the size of the population can be adjusted according to the actual situation, generally between dozens and hundreds of individuals;

[0106] Genetic operations:

[0107] Selection operation: Select individuals in the population according to the value of the fitness function; The higher the fitness value of an individual, the greater the probability of being selected, which can ensure that excellent individuals have more opportunities to participate in subsequent genetic operations;

[0108] Crossover operation: Randomly select two individuals from the selected individuals, and exchange or combine their Q and N values to generate new individuals; The crossover operation can increase the diversity of the population and help search for better solutions;

[0109] Mutation operation: Randomly modify the Q or N value of an individual with a certain probability to prevent the algorithm from falling into a local optimal solution;

[0110] Iterative search. Continuously repeat the selection, crossover, and mutation operations until the termination condition is met; The termination condition can be reaching a preset number of iterations, or the optimal solution found does not change significantly within a certain number of iterations; Finally, the genetic algorithm will search for the combination of the minimum P value that satisfies η real ≥η target in the three-dimensional mapping table, that is, the optimal ratio of Q to N;

[0111] Step S4-3: When T env > 30°C, introduce the humidity parameter H for four-dimensional optimization calculation:

[0112] Humidity data acquisition. When the ambient temperature T env > 30°C, humidity has a significant impact on the cooling efficiency and energy consumption of the cooling tower; Therefore, it is necessary to increase the acquisition of the humidity parameter H; A humidity sensor can be used to monitor the humidity of the environment where the cooling tower is located in real time;

[0113] Four - dimensional mapping table expansion: On the basis of the original three - dimensional mapping table, the humidity dimension is added to construct a four - dimensional mapping table; the new mapping table stores the correspondence of Q - N - P - H, that is, each combination of Q, N, and H corresponds to a specific P value;

[0114] Four - dimensional optimization calculation: Expand the genetic algorithm by incorporating the humidity parameter H into the fitness function and search range; the fitness function is modified to consider the minimization of energy consumption under the influence of humidity while meeting the cooling efficiency requirements; by searching in the four - dimensional mapping table, find the optimal combination of Q, N, and H in a high - temperature and high - humidity environment to further optimize the energy efficiency of the system;

[0115] Step S4 - 4: Apply the optimal ratio for system adjustment:

[0116] After obtaining the optimal ratio of Q and N, adjust the cooling water pump and axial - flow fan through the variable - frequency speed - regulation module to make the cooling water flow rate and fan speed reach the optimal values; during the adjustment process, continuously monitor the real - time cooling efficiency η real and the system energy consumption value P to ensure that the system meets η real ≥η target and P≤P max requirements; if there are deviations during the adjustment process, fine - tune the optimal ratio in a timely manner according to the actual situation.

[0117] In this embodiment, the specific operation steps of step S5 are as follows:

[0118] Step S5 - 1: Cooling load mutation detection:

[0119] Pressure sensor installation and data acquisition:

[0120] Install pressure sensors at key positions in the cooling water pipeline. These key positions include the inlets and outlets of the cooling water circulation pump, the inlets and outlets of the cooling water tower, etc.; the pressure sensors need to have high precision and fast response characteristics to accurately monitor the pressure changes in the cooling pipeline in real - time; set a reasonable data acquisition frequency, for example, collect pressure data once per second, to ensure that the subtle changes in pressure can be captured in a timely manner;

[0121] Set the pressure change rate threshold:

[0122] According to the design parameters of the cooling water tower system, historical operation data, and actual working conditions, set a pressure change rate threshold; this threshold is used to judge whether the cooling load has mutated; for example, after multiple tests and analyses, it is determined that when the pressure change rate exceeds 5% within a short period (such as 10 seconds), it is judged that the cooling load has mutated;

[0123] Step S5 - 2: Feed - forward control - Adjust the variable - frequency parameters of the cooling water circulation pump in advance:

[0124] Pressure change rate judgment and triggering:

[0125] Calculate the pressure change rate in real time and compare it with the set threshold; when the pressure change rate exceeds the set threshold, it indicates that a sudden change in the cooling load has occurred, and the feedforward control mechanism is immediately activated;

[0126] Variable frequency parameter adjustment strategy:

[0127] Adjust the variable frequency parameters of the cooling water circulation pump in advance according to the amplitude and direction of the pressure change; if the pressure suddenly increases, it means that the cooling load may increase. At this time, it is necessary to increase the frequency of the cooling water circulation pump to increase the flow rate of the cooling water to meet the higher cooling demand; conversely, if the pressure suddenly decreases, the frequency of the pump can be appropriately reduced to reduce unnecessary energy consumption; for example, establish a mapping relationship table between the pressure change rate and the pump frequency adjustment amount, and find the corresponding pump frequency adjustment amount from the table according to the actual pressure change rate;

[0128] Step S5-3: Feedback control - Establish a fan speed prediction control model:

[0129] Prediction control model construction:

[0130] Based on the historical operation data of the system, thermodynamic principles, and control theory, establish a prediction control model for the fan speed; this model uses the pressure change rate, ambient temperature, real-time cooling efficiency, etc. as input parameters, and predicts the adjustment amount that the fan speed needs to make to compensate for the adjustment delay caused by the system inertia through mathematical algorithms; for example, a neural network algorithm can be used to train a large amount of historical data so that the model can accurately predict the optimal adjustment value of the fan speed under different working conditions;

[0131] Model real-time update and optimization:

[0132] As the system runs, continuously collect new operation data and perform real-time update and optimization on the prediction control model; because the operating conditions of the system may change with factors such as time and environment, real-time updating of the model can improve the accuracy and adaptability of the model, ensuring that it can better compensate for the adjustment delay caused by the system inertia;

[0133] Step S5-4: Coordinated working parameter synchronous adjustment:

[0134] Control signal sending:

[0135] When a sudden change in the cooling load is detected and the feedforward-feedback composite control mode is activated, the edge computing unit sends a control signal to the variable frequency speed regulation module according to the adjustment result of the variable frequency parameters of the cooling water circulation pump in the feedforward control and the adjustment amount obtained from the fan speed prediction control model in the feedback control;

[0136] Coordinated adjustment implementation:

[0137] After receiving the control signal, the variable frequency speed regulation module synchronously adjusts the operating parameters of the cooling water pump and the axial flow fan; ensures that the flow rate adjustment of the cooling water pump and the fan speed adjustment cooperate with each other, enabling the cooling water tower system to quickly and stably adapt to sudden changes in the cooling load, and maintaining good cooling effect and energy efficiency level;

[0138] Step S5-5: Monitoring and evaluation of the adjustment effect:

[0139] Parameter monitoring:

[0140] During the adjustment process, continuously monitor parameters such as the pressure of the cooling pipeline, the inlet and outlet temperatures of the cooling water, the real-time cooling efficiency, and the system energy consumption; evaluate the adjustment effect of the feedforward-feedback compound control mode based on the changes of these parameters.

[0141] Dynamic adjustment:

[0142] If the monitored parameters indicate that the adjustment effect is not ideal, for example, the cooling efficiency has not increased significantly or the system energy consumption is too high, analyze the reasons in a timely manner and dynamically adjust the cooperative working parameters of the cooling water circulation pump and the fan; the adjustment process can be optimized by adjusting the parameters of the predictive control model, modifying the adjustment strategy of the feedforward control, etc., so that the system reaches the optimal operating state.

[0143] In this embodiment, the method further includes:

[0144] Set a multi-level safety control strategy. When it is detected that T out >T safe start the emergency cooling circuit;

[0145] The emergency cooling circuit includes a standby heat exchanger and a cold storage device arranged in parallel, and the switching response time < 5 seconds, where T safe is the safety temperature threshold of the cooling water tower effluent;

[0146] Furthermore, the safety control strategy includes:

[0147] Construct a thermal runaway prediction model based on the LSTM neural network, and the input parameters include:

[0148] The change rate of the cooling water conductivity, which is the ratio of the change amount of the cooling water conductivity within a set time interval to the time interval;

[0149] The pipeline vibration spectrum characteristic, which is the characteristic information obtained by performing spectrum analysis on the pipeline vibration signal;

[0150] The motor winding temperature rise curve, which is the curve of the motor winding temperature changing with time;

[0151] When the predicted risk value > 0.7, start preventive cooling measures in advance.

[0152] The liquid-cooled cooling water tower system includes:

[0153] A distributed temperature sensing array arranged on the inlet main pipe, outlet main pipe and heat dissipation fins at different heights of the cooling water tower;

[0154] A variable frequency speed regulation module for integrally controlling the cooling water pump and the axial flow fan;

[0155] An edge computing unit with the dynamic regulation model and the energy efficiency optimization algorithm built therein;

[0156] An emergency cooling actuator including a quick solenoid valve and a cold storage tank linkage device.

[0157] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A control method for a liquid-cooled cooling water tower, characterized in that: It includes the following steps: Step S1: Collect the inlet water temperature T of the cooling water tower in real time in , the outlet water temperature T out and the ambient temperature T env ; Step S2: Based on the inlet water temperature T in , the outlet water temperature T out and the ambient temperature T env to establish a dynamic adjustment model and calculate the target cooling efficiency η target ; Step S3: Dynamically adjust the cooling water flow rate Q and the fan speed N according to the difference between η target and the real-time cooling efficiency η real ; Step S4: Calculate the optimal ratio of Q and N through the energy efficiency optimization algorithm to make the system satisfy η real ≥η target and the energy consumption value P ≤ P max , where P is the real-time energy consumption value of the cooling tower system, and P max is the maximum allowable energy consumption value of the cooling tower system; Step S5: When a sudden change in cooling load is detected, start the feedforward-feedback composite control mode and synchronously adjust the cooperative working parameters of the cooling water circulation pump and the fan.

2. The liquid-cooled cooling water tower control method according to claim 1, characterized in that: In the said step S2, the calculation formula of the dynamic adjustment model is: η target = α×(T in - T env ) + β×(dT / dt) + γ×Q hist , where α, β, γ is a dynamic weight coefficient used to adjust the influence degree of different parameters on the calculation of the target cooling efficiency; Q hist is the historical average flow rate; dT / dt is the temperature change rate.

3. The liquid-cooled cooling tower control method according to claim 1, wherein: In the said step S3, the dynamic adjustment includes: When Δη = η target - η real > Δη th , the cooling water flow rate Q is preferentially adjusted, and the adjustment amplitude is non-linearly positively correlated with Δη, where Δη is the difference between the target cooling efficiency and the real-time cooling efficiency; Δη th is the set threshold value of the cooling efficiency difference; When Δη ≤ Δη th a refined adjustment of the fan speed N is adopted, and the adjustment step size does not exceed 2% of the rated speed.

4. The liquid-cooled cooling tower control method according to claim 1, wherein: In the said step S4, the energy efficiency optimization algorithm includes: Establish a three-dimensional mapping table to store the correspondence of Q-N-P; Use the genetic algorithm to search for the combination of the minimum P value in the mapping table that satisfies η real ≥ η target under the condition; When the ambient temperature T env > 30 °C, the humidity parameter H is introduced for four-dimensional optimization calculation, where H is the real-time humidity of the environment where the cooling water tower is located.

5. The liquid-cooled cooling tower control method according to claim 1, characterized in that: In the said step S5, the feedforward-feedback composite control mode includes: Real-time monitor the pressure change of the cooling pipeline through a pressure sensor; When the pressure change rate exceeds the set threshold, adjust the frequency conversion parameters of the cooling water circulation pump in advance; Synchronously establish a predictive control model for the fan speed to compensate for the adjustment delay caused by the system inertia.

6. The liquid-cooled cooling tower control method according to claim 1, characterized in that: It also includes: Set a multi-level security control strategy. When it is detected that T out >T safe is reached, start the emergency cooling circuit; The emergency cooling circuit includes a standby heat exchanger and a cold storage device arranged in parallel, with a switching response time < 5 s, where T safe is the safety temperature threshold of the cooling water tower outlet water.

7. The liquid-cooled cooling water tower control method according to claim 6, wherein: The said safety control strategy includes: Construct a thermal runaway prediction model based on the LSTM neural network, and the input parameters include: The change rate of the cooling water conductivity, which is the ratio of the change amount of the cooling water conductivity within a set time interval to this time interval; The pipeline vibration spectrum characteristics, which are the characteristic information obtained by performing spectrum analysis on the pipeline vibration signal; The motor winding temperature rise curve, which is the curve of the motor winding temperature changing with time; When the predicted risk value > 0.7, start the preventive cooling measures in advance.

8. A liquid-cooled cooling water tower system for implementing the control method according to any one of claims 1-7, characterized in that, It includes: A distributed temperature sensing array arranged on the inlet main pipe, outlet main pipe and cooling fins at different heights of the cooling water tower; A variable frequency speed regulation module that integrally controls the cooling water pump and the axial flow fan; An edge computing unit with the said dynamic adjustment model and energy efficiency optimization algorithm built in; An emergency cooling actuator, which includes a quick solenoid valve and a cold storage tank linkage device.

Images