Self-adaptive adjusting method and system of liquid cooling air conditioning system

Abstract

The invention discloses a self-adaptive adjustment method and system for a liquid cooling air conditioning system, and the method comprises the steps: collecting data during the operation of the liquid cooling air conditioning system, and carrying out the preprocessing of the data, so as to obtain sample data; obtaining the sample data, and calculating a thermal load demand of the liquid cooling air conditioning system to obtain thermal load demand data; the thermal load demand data and the temperature data set are obtained and input into a load prediction model, the load of the liquid cooling air conditioning system is predicted, and prediction data are obtained; a cooling liquid flow and temperature adjusting instruction is generated according to the prediction data, and the cooling liquid flow and temperature are dynamically adjusted through an adjusting valve and a cooling liquid temperature adjusting device; according to the method, the environment temperature, the load change and the system operation state are monitored in real time, the flow and the temperature of the cooling liquid are dynamically adjusted, so that self-adaptive adjustment of the liquid cooling air conditioner system is achieved, and the system operates efficiently and stably.

Description

Technical Field

[0001] The present invention relates to the field of liquid-cooled air conditioner control, and particularly to an adaptive adjustment method and system for a liquid-cooled air conditioner system. Background Art

[0002] A liquid-cooled air conditioner is an air conditioner system that uses a liquid as a cooling medium. Different from traditional air-cooled air conditioners, a liquid-cooled air conditioner absorbs and dissipates heat through liquid circulation, has higher heat conduction efficiency and lower noise level. Liquid-cooled air conditioners are widely used in data centers, industrial equipment, high-end commercial buildings, and fields that require efficient cooling. The core working principle of a liquid-cooled air conditioner is to transfer heat from the heat source to the heat dissipation device through liquid circulation to achieve the purpose of cooling.

[0003] Traditional liquid-cooled air conditioner systems usually operate with fixed parameters, that is, the flow rate and temperature settings of the coolant are static and cannot be dynamically adjusted according to the actual load and environmental changes. Although this fixed-parameter operation mode can meet the basic cooling requirements under design conditions, there are many problems in actual applications. The energy efficiency of the system is low. Since the coolant flow rate and temperature are fixed, when the load is low or the environmental temperature is cold, the system still operates with high energy consumption, resulting in energy waste. The stability of the system is insufficient. When the load suddenly changes or the environmental temperature changes drastically, the fixed-parameter operation cannot respond quickly, which may cause the equipment to overheat or overcool, affecting the reliability of the system and the service life of the equipment. In addition, the maintenance cost of the traditional system is high. Fixed-parameter operation may cause the equipment to run at high load for a long time, accelerating equipment wear and increasing maintenance and replacement costs. Summary of the Invention

[0004] The purpose of the present invention is to solve the above problems and design an adaptive adjustment method and system for a liquid-cooled air conditioner system.

[0005] The first aspect of the present invention provides an adaptive adjustment method for a liquid-cooled air conditioner system, and the adaptive adjustment method for the liquid-cooled air conditioner system includes the following steps:

[0006] Collect data during the operation of the liquid-cooled air conditioner system, preprocess the data, and obtain sample data;

[0007] Obtain the sample data, calculate the heat load demand of the liquid-cooled air conditioner system, and obtain heat load demand data;

[0008] Obtain the heat load demand data and the temperature data set, input them into the load prediction model, predict the load of the liquid-cooled air conditioner system, and obtain prediction data;

[0009] Generate adjustment instructions for the coolant flow rate and temperature according to the prediction data, and dynamically adjust the coolant flow rate and temperature through the regulating valve and the coolant temperature regulating device;

[0010] Monitor the operating status of the adjusted system in real time, obtain feedback data, and perform adaptive adjustment of the operating status parameters.

[0011] Furthermore, in the above-mentioned adaptive adjustment method of a liquid-cooled air-conditioning system, collecting the data during the operation of the liquid-cooled air-conditioning system, preprocessing the data, and obtaining sample data, including:

[0012] Call different sensors to collect the ambient temperature, the inlet temperature of the coolant, the outlet temperature of the coolant, the coolant flow rate, and the system load data during the operation of the liquid-cooled air-conditioning system;

[0013] Read the collected data for data cleaning, and perform operations such as filling missing values with the average value and processing outliers;

[0014] Align the timestamps of the data collected by different sensors, fuse the multi-source data into a unified data set, and obtain sample data.

[0015] Furthermore, in the above-mentioned adaptive adjustment method of a liquid-cooled air-conditioning system, obtaining the sample data, calculating the heat load demand of the liquid-cooled air-conditioning system, and obtaining the heat load demand data, including:

[0016] Obtain the sample data, determine the physical properties of the coolant, and obtain the specific heat capacity of the coolant and the density of the coolant;

[0017] Calculate the mass flow rate of the coolant according to the density of the coolant and the coolant flow rate;

[0018] Calculate the temperature difference between the inlet and outlet of the coolant according to the inlet temperature of the coolant and the outlet temperature of the coolant;

[0019] The heat load demand is calculated by the following formula:

[0020] Q = m * c * ΔT,

[0021] where Q represents the heat load demand, unit: kW; m represents the mass flow rate of the coolant; c represents the specific heat capacity of the coolant; ΔT represents the temperature difference between the inlet and outlet of the coolant.

[0022] Furthermore, in the above-mentioned adaptive adjustment method of a liquid-cooled air-conditioning system, inputting the heat load demand data and the temperature data set into a load prediction model to predict the load of the liquid-cooled air-conditioning system, and obtaining prediction data, including:

[0023] Obtain the data of the air-conditioning load demand at the current moment, the air-conditioning heat load demand at the previous moment, the dry-bulb temperature at the current moment, the relative humidity at the current moment, the dry-bulb temperature at the previous moment, the relative humidity at the previous moment, the predicted value of the daily maximum temperature, and the predicted value of the daily minimum temperature as input factors;

[0024] Build an LSTM model, including an input layer: receiving time series data; an LSTM layer: capturing long-term dependencies in the time series; a fully connected layer: mapping the LSTM output to the target variable; an output layer: outputting the predicted value of the load of the liquid cooling air conditioning system;

[0025] Optimize the LSTM model. Introduce Dropout in the third fully connected layer of the LSTM model. By randomly discarding a part of the neurons in each training iteration, the model cannot overly rely on specific neurons, thereby improving the generalization performance of the model;

[0026] Set the model parameters, specifically the time step, the number of LSTM units, the activation function, and the optimizer, to obtain the load prediction model;

[0027] Input the input factors into the load prediction model for training and output the prediction data.

[0028] Furthermore, in the above-mentioned adaptive adjustment method for a liquid cooling air conditioning system, the generating of adjustment instructions for the coolant flow rate and temperature according to the prediction data, and dynamically adjusting the coolant flow rate and temperature through a control valve and a coolant temperature adjustment device includes:

[0029] Generate adjustment instructions according to the heat load demand of the prediction data in combination with preset rules. When the heat load demand increases, increase the coolant flow rate and decrease the coolant temperature; when the heat load demand decreases, decrease the coolant flow rate and increase the coolant temperature;

[0030] Obtain the adjustment instructions, control the coolant flow rate by calling the opening of the control valve, and use a flow sensor to monitor the coolant flow rate in real time;

[0031] Obtain the adjustment instructions, adjust the coolant temperature by calling the coolant temperature adjustment device, using a heater or a cooling tower, and use a temperature sensor to monitor the coolant temperature in real time.

[0032] Furthermore, in the above-mentioned adaptive adjustment method for a liquid cooling air conditioning system, the real-time monitoring of the operating state of the adjusted system, obtaining feedback data, and performing adaptive adjustment of the operating state parameters further includes:

[0033] Use a flow sensor to monitor the coolant flow rate in real time, use a temperature sensor to monitor the coolant temperature in real time, use a power meter to monitor the total energy consumption of the system, and monitor external factors such as the ambient temperature and humidity to obtain monitoring data;

[0034] Transmit the obtained monitoring data back to the input load prediction model at a frequency of once every 5 minutes for load prediction of the liquid cooling air conditioning system, and perform adaptive adjustment of the coolant flow rate and temperature parameters for the current air conditioning operating state.

[0035] The second aspect of the present invention provides an adaptive regulation system for a liquid-cooled air-conditioning system, characterized in that the adaptive regulation system of the liquid-cooled air-conditioning system includes a data acquisition module, a heat load calculation module, a demand prediction module, an instruction regulation module, and an adaptive regulation module, where:

[0036] The data acquisition module is used to collect data during the operation of the liquid-cooled air-conditioning system, preprocess the data, and obtain sample data;

[0037] The heat load calculation module is used to obtain the sample data, calculate the heat load demand of the liquid-cooled air-conditioning system, and obtain heat load demand data;

[0038] The demand prediction module is used to obtain the heat load demand data and the temperature data set, input them into the load prediction model, predict the load of the liquid-cooled air-conditioning system, and obtain prediction data;

[0039] The instruction regulation module is used to generate adjustment instructions for the coolant flow rate and temperature according to the prediction data, and dynamically adjust the coolant flow rate and temperature through a regulating valve and a coolant temperature regulating device;

[0040] The adaptive regulation module is used to monitor the operation state of the adjusted system in real time, obtain feedback data, and perform adaptive regulation of the operation state parameters.

[0041] Further, in the above-mentioned adaptive regulation system of a liquid-cooled air-conditioning system, the demand prediction module includes an acquisition sub-module, a construction sub-module, an optimization sub-module, a setting sub-module, and a training sub-module, where:

[0042] The acquisition sub-module is used to obtain data such as the air-conditioning load demand at the current moment, the air-conditioning heat load demand at the previous moment, the dry-bulb temperature at the current moment, the relative humidity at the current moment, the dry-bulb temperature at the previous moment, the relative humidity at the previous moment, the predicted value of the highest temperature of the day, and the predicted value of the lowest temperature of the day as input factors;

[0043] The construction sub-module is used to construct an LSTM model, including an input layer: receiving time series data; an LSTM layer: capturing long-term dependencies in the time series; a fully connected layer: mapping the LSTM output to the target variable; an output layer: outputting the predicted value of the load of the liquid-cooled air-conditioning system;

[0044] The optimization sub-module is used to optimize the LSTM model, introduce Dropout in the third fully connected layer of the LSTM model, and randomly discard a part of neurons in each training iteration, so that the model cannot overly rely on specific neurons and improve the generalization performance of the model;

[0045] A setting sub-module is used to set model parameters, specifically the time step, the number of LSTM units, the activation function, and the optimizer, to obtain a load prediction model;

[0046] A training sub-module is used to input the input factors into the load prediction model for training and output prediction data.

[0047] Its beneficial effects are as follows: by collecting the data during the operation of the liquid-cooled air-conditioning system, preprocessing the data to obtain sample data; acquiring the sample data, calculating the heat load demand of the liquid-cooled air-conditioning system to obtain heat load demand data; acquiring the heat load demand data and the temperature dataset, inputting them into the load prediction model to predict the load of the liquid-cooled air-conditioning system and obtain prediction data; generating adjustment instructions for the coolant flow rate and temperature according to the prediction data, and dynamically adjusting the coolant flow rate and temperature through the regulating valve and the coolant temperature regulating device; monitoring the operation state of the adjusted system in real time, acquiring feedback data, and performing adaptive adjustment of the operation state parameters. This method dynamically adjusts the coolant flow rate and temperature by monitoring the environmental temperature, load changes, and system operation state in real time to achieve efficient and stable operation. The present invention has the advantages of high energy efficiency, high stability, and high intelligence, and can be widely applied to liquid-cooled air-conditioning systems in fields such as data centers, industrial equipment, and commercial buildings. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention.

[0049] Figure 1 It is a schematic diagram of an embodiment of an adaptive adjustment method for a liquid-cooled air-conditioning system in an embodiment of the present invention;

[0050] Figure 2 It is the first schematic diagram of an adaptive adjustment system for a liquid-cooled air-conditioning system in an embodiment of the present invention;

[0051] Figure 3 It is the second schematic diagram of an adaptive adjustment system for a liquid-cooled air-conditioning system in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] In order to make the objectives, technical solutions, and advantages of the present invention clearer, 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.

[0053] Those skilled in the art can understand that, unless specifically stated, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "including" used in the specification of the present invention means the presence of the described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or their groups.

[0054] The present invention will be specifically described below in conjunction with the accompanying drawings. As Figure 1 shown, an adaptive adjustment method for a liquid-cooled air-conditioning system, the adaptive adjustment method for the liquid-cooled air-conditioning system includes the following steps:

[0055] Step 101: Collect data during the operation of the liquid-cooled air-conditioning system, preprocess the data, and obtain sample data;

[0056] Specifically, in this embodiment, collecting data during the operation of the liquid-cooled air-conditioning system, preprocessing the data, and obtaining sample data includes: calling different sensors to collect the ambient temperature, coolant inlet temperature, coolant outlet temperature, coolant flow rate, and system load data during the operation of the liquid-cooled air-conditioning system; reading the collected data for data cleaning, performing operations on filling the missing values with the average value and processing the outliers; aligning the time stamps of the data collected by different sensors, and fusing the multi-source data into a unified data set to obtain sample data.

[0057] Step 102: Obtain the sample data, calculate the heat load demand of the liquid-cooled air-conditioning system, and obtain the heat load demand data;

[0058] Specifically, in this embodiment, obtaining the sample data, calculating the heat load demand of the liquid-cooled air-conditioning system, and obtaining the heat load demand data includes: obtaining the sample data, determining the physical properties of the coolant, and obtaining the specific heat capacity of the coolant and the density of the coolant; calculating the mass flow rate of the coolant according to the density of the coolant and the coolant flow rate; calculating the temperature difference between the inlet and outlet of the coolant according to the coolant inlet temperature and the coolant outlet temperature.

[0059] The heat load demand is calculated by the following formula:

[0060] Q = m * c * ΔT,

[0061] where Q represents the heat load demand, unit: kW; m represents the mass flow rate of the coolant; c represents the specific heat capacity of the coolant; ΔT represents the temperature difference between the inlet and outlet of the coolant.

[0062] Step 103: Obtain the heat load demand data and the temperature data set, input them into the load prediction model, predict the load of the liquid-cooled air-conditioning system, and obtain the prediction data;

[0063] Specifically, in this embodiment, the heat load demand data and the temperature data set are obtained and input into the load prediction model to predict the load of the liquid-cooled air-conditioning system, and the prediction data is obtained, including:

[0064] Obtain the data of the current air-conditioning load demand, the previous air-conditioning heat load demand, the current dry-bulb temperature, the current relative humidity, the previous dry-bulb temperature, the previous relative humidity, the predicted value of the daily maximum temperature and the predicted value of the daily minimum temperature as input factors;

[0065] Build an LSTM model, including an input layer: receiving time series data; an LSTM layer: capturing long-term dependencies in the time series; a fully connected layer: mapping the LSTM output to the target variable; an output layer: outputting the predicted value of the liquid-cooled air-conditioning system load;

[0066] The long short-term memory network LST is a special recurrent neural network (RNN) that can effectively handle long-term dependencies in time series data. In a liquid-cooled air-conditioning system, LSTM can be used to train historical heat load demand data, predict future heat load demands, and optimize the adjustment strategies for coolant flow rate and temperature accordingly. The heat load demand of a liquid-cooled air-conditioning system changes over time and is affected by factors such as ambient temperature and system load. Through the LSTM model, historical data can be used to predict future heat load demands and generate corresponding adjustment strategies for coolant flow rate and temperature to achieve efficient and energy-saving operation.

[0067] Optimize the LSTM model by introducing Dropout in the third fully connected layer of the LSTM model. By randomly discarding a part of the neurons in each training iteration, the model cannot overly rely on specific neurons, thus improving the generalization performance of the model;

[0068] Set the model parameters, specifically the time step, the number of LSTM cells, the activation function, and the optimizer, to obtain the load prediction model;

[0069] Input the input factors into the load prediction model for training and output the prediction data.

[0070] Step 104, generate adjustment instructions for the coolant flow rate and temperature according to the prediction data, and dynamically adjust the coolant flow rate and temperature through the regulating valve and the coolant temperature regulating device;

[0071] Specifically, in this embodiment, adjustment instructions for the coolant flow rate and temperature are generated based on the prediction data, and the coolant flow rate and temperature are dynamically adjusted through a regulating valve and a coolant temperature adjustment device, including: generating adjustment instructions according to the heat load demand of the prediction data in combination with preset rules. When the heat load demand increases, the coolant flow rate is increased and the coolant temperature is decreased; when the heat load demand decreases, the coolant flow rate is decreased and the coolant temperature is increased; obtaining the adjustment instructions, controlling the coolant flow rate by calling the opening degree of the regulating valve, and using a flow sensor to monitor the coolant flow rate in real time; obtaining the adjustment instructions, adjusting the coolant temperature by calling the coolant temperature adjustment device, using a heater or a cooling tower, and using a temperature sensor to monitor the coolant temperature in real time.

[0072] Step 105, monitor the running state of the system after adjustment in real time, obtain feedback data, and perform adaptive adjustment of the running state parameters;

[0073] Specifically, in this embodiment, the running state of the system after adjustment is monitored in real time, feedback data is obtained, and adaptive adjustment of the running state parameters is performed, further including: using a flow sensor to monitor the coolant flow rate in real time, using a temperature sensor to monitor the coolant temperature in real time, using a power meter to monitor the total energy consumption of the system, monitoring external factors such as the ambient temperature and humidity, and obtaining monitoring data; transmitting the obtained monitoring data back to the input load prediction model at a frequency of once every 5 minutes to perform load prediction for the liquid-cooled air-conditioning system, and performing adaptive adjustment of the coolant flow rate and temperature parameters for the current air-conditioning running state.

[0074] High-precision prediction: The improved LSTM model can capture complex time series patterns.

[0075] Dynamic adjustment: Adjust the coolant flow rate and temperature in real time according to the prediction results.

[0076] Energy-saving and efficient: Reduce the system energy consumption by optimizing the adjustment strategy.

[0077] As Figure 2 shown, it is the first schematic diagram of an adaptive adjustment system for a liquid-cooled air-conditioning system. The adaptive adjustment system for the liquid-cooled air-conditioning system includes a data acquisition module, a heat load calculation module, a demand prediction module, an instruction adjustment module, and an adaptive adjustment module, where:

[0078] The data acquisition module is used to collect the data during the operation of the liquid-cooled air-conditioning system, preprocess the data, and obtain sample data;

[0079] The heat load calculation module is used to obtain the sample data, calculate the heat load demand of the liquid-cooled air-conditioning system, and obtain the heat load demand data;

[0080] A demand forecasting module, which is used to obtain heat load demand data and temperature datasets, input them into a load forecasting model, forecast the load of the liquid-cooled air-conditioning system, and obtain forecast data;

[0081] An instruction adjustment module, which is used to generate adjustment instructions for the coolant flow rate and temperature according to the forecast data, and dynamically adjust the coolant flow rate and temperature through a regulating valve and a coolant temperature adjustment device;

[0082] An adaptive adjustment module, which is used to monitor the operating state of the adjusted system in real time, obtain feedback data, and perform adaptive adjustment of the operating state parameters.

[0083] As Figure 3 shown, it is the second schematic diagram of an adaptive adjustment system for a liquid-cooled air-conditioning system.

[0084] In this embodiment, the demand forecasting module includes an acquisition sub-module, a construction sub-module, an optimization sub-module, a setting sub-module, and a training sub-module, where:

[0085] The acquisition sub-module is used to acquire data such as the air-conditioning load demand at the current moment, the air-conditioning heat load demand at the previous moment, the dry-bulb temperature at the current moment, the relative humidity at the current moment, the dry-bulb temperature at the previous moment, the relative humidity at the previous moment, the forecast value of the daily maximum temperature, and the forecast value of the daily minimum temperature, as input factors;

[0086] The construction sub-module is used to construct an LSTM model, including an input layer: receiving time series data; an LSTM layer: capturing long-term dependencies in the time series; a fully connected layer: mapping the LSTM output to the target variable; an output layer: outputting the predicted value of the load of the liquid-cooled air-conditioning system;

[0087] The optimization sub-module is used to optimize the LSTM model, introduce Dropout in the third fully connected layer of the LSTM model, and randomly discard a part of the neurons in each training iteration, so that the model cannot overly rely on specific neurons, and improve the generalization performance of the model;

[0088] The setting sub-module is used to set model parameters, specifically the time step, the number of LSTM units, the activation function, and the optimizer, to obtain a load forecasting model;

[0089] The training sub-module is used to input the input factors into the load forecasting model for training and output forecast data.

[0090] Through the implementation of the above solution, data during the operation of the liquid cooling air conditioning system is collected, preprocessed to obtain sample data; the sample data is acquired, the heat load demand of the liquid cooling air conditioning system is calculated to obtain heat load demand data; the heat load demand data and the temperature data set are acquired and input into the load prediction model to predict the load of the liquid cooling air conditioning system, obtaining prediction data; adjustment instructions for the coolant flow rate and temperature are generated based on the prediction data, and the coolant flow rate and temperature are dynamically adjusted through the regulating valve and the coolant temperature regulating device; the operation state of the system after adjustment is monitored in real time, feedback data is acquired, and self-adaptive adjustment of the operation state parameters is performed. This method dynamically adjusts the coolant flow rate and temperature by monitoring the ambient temperature, load changes, and system operation state in real time, so as to achieve self-adaptive adjustment of the liquid cooling air conditioning system, enabling the system to operate efficiently and stably.

[0091] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. An adaptive adjustment method for a liquid-cooled air-conditioning system, characterized in that The adaptive adjustment method of the liquid cooling air conditioning system includes the following steps: Collect data during the operation of the liquid cooling air conditioning system, preprocess the data, and obtain sample data; Obtain the sample data, calculate the heat load demand of the liquid cooling air conditioning system, and obtain heat load demand data; Obtain the heat load demand data and the temperature data set, input them into the load prediction model, predict the load of the liquid cooling air conditioning system, and obtain prediction data; Generate adjustment instructions for the coolant flow rate and temperature according to the prediction data, and dynamically adjust the coolant flow rate and temperature through the regulating valve and the coolant temperature regulating device; Real-time monitor the operating state of the adjusted system, obtain feedback data, and perform adaptive adjustment of the operating state parameters.

2. The adaptive adjustment method of a liquid-cooled air-conditioning system according to claim 1, wherein, The step of collecting data during the operation of the liquid cooling air conditioning system, preprocessing the data, and obtaining sample data includes: Call different sensors to collect the ambient temperature, coolant inlet temperature, coolant outlet temperature, coolant flow rate, and system load data during the operation of the liquid cooling air conditioning system; Read the collected data for data cleaning, and perform operations such as filling the missing values with the average value and processing the outliers; Align the timestamps of the data collected by different sensors, fuse the multi-source data into a unified data set, and obtain sample data.

3. The adaptive adjustment method of a liquid-cooled air-conditioning system according to claim 1, wherein, The step of obtaining the sample data, calculating the heat load demand of the liquid cooling air conditioning system, and obtaining heat load demand data includes: Obtain the sample data, determine the physical properties of the coolant, and obtain the specific heat capacity of the coolant and the density of the coolant; Calculate the mass flow rate of the coolant according to the density of the coolant and the coolant flow rate; Calculate the temperature difference between the inlet and outlet of the coolant according to the coolant inlet temperature and the coolant outlet temperature; The heat load demand is calculated by the following formula: Q = m * c * ΔT, where Q represents the heat load demand, unit: kW; m represents the mass flow rate of the coolant; c represents the specific heat capacity of the coolant; ΔT represents the temperature difference between the inlet and outlet of the coolant.

4. The adaptive adjustment method of a liquid-cooled air-conditioning system according to claim 1, characterized in that, The step of obtaining the heat load demand data and the temperature data set, inputting them into the load prediction model, predicting the load of the liquid cooling air conditioning system, and obtaining prediction data includes: Obtain the data of the air conditioning load demand at the current moment, the air conditioning heat load demand at the previous moment, the dry bulb temperature at the current moment, the relative humidity at the current moment, the dry bulb temperature at the previous moment, the relative humidity at the previous moment, the predicted value of the highest temperature of the day, and the predicted value of the lowest temperature of the day as input factors; Build an LSTM model, including an input layer: receiving time series data; an LSTM layer: capturing long-term dependencies in the time series; a fully connected layer: mapping the LSTM output to the target variable; an output layer: outputting the predicted value of the load of the liquid cooling air conditioning system; Optimize the LSTM model, introduce Dropout in the third fully connected layer of the LSTM model, and randomly discard a part of the neurons in each training iteration, so that the model cannot overly rely on specific neurons and improve the generalization performance of the model; Set the model parameters, specifically the time step, the number of LSTM units, the activation function, and the optimizer, to obtain the load prediction model; Input the input factors into the load prediction model for training and output prediction data.

5. The adaptive adjustment method of a liquid-cooled air-conditioning system according to claim 4, wherein, Generating adjustment instructions for the coolant flow rate and temperature based on the prediction data, and dynamically adjusting the coolant flow rate and temperature through a regulating valve and a coolant temperature adjustment device, including: Generating adjustment instructions according to the heat load demand of the prediction data and combining preset rules. When the heat load demand increases, increase the coolant flow rate and decrease the coolant temperature; when the heat load demand decreases, decrease the coolant flow rate and increase the coolant temperature; Obtaining the adjustment instructions, controlling the coolant flow rate by calling the opening degree of the regulating valve, and using a flow sensor to monitor the coolant flow rate in real time; Obtaining the adjustment instructions, adjusting the coolant temperature by calling the coolant temperature adjustment device, using a heater or a cooling tower, and using a temperature sensor to monitor the coolant temperature in real time.

6. The adaptive adjustment method of a liquid-cooled air-conditioning system according to claim 1, characterized in that The real-time monitoring of the adjusted system operation state, obtaining feedback data, and performing adaptive adjustment of the operation state parameters further includes: Using a flow sensor to monitor the coolant flow rate in real time, using a temperature sensor to monitor the coolant temperature in real time, using a power meter to monitor the total energy consumption of the system, and monitoring external factors such as ambient temperature and humidity to obtain monitoring data; Transmitting the obtained monitoring data back to the input load prediction model at a frequency of once every 5 minutes for load prediction of the liquid-cooled air-conditioning system, and performing adaptive adjustment of the coolant flow rate and temperature parameters for the current air-conditioning operation state.

7. An adaptive regulation system for a liquid-cooled air-conditioning system, characterized in that, The adaptive adjustment system of the liquid-cooled air-conditioning system includes a data acquisition module, a heat load calculation module, a demand prediction module, an instruction adjustment module, and an adaptive adjustment module, where: The data acquisition module is used to collect data during the operation of the liquid-cooled air-conditioning system, preprocess the data, and obtain sample data; The heat load calculation module is used to obtain the sample data, calculate the heat load demand of the liquid-cooled air-conditioning system, and obtain heat load demand data; The demand prediction module is used to obtain the heat load demand data and the temperature data set, input them into the load prediction model, predict the load of the liquid-cooled air-conditioning system, and obtain prediction data; The instruction adjustment module is used to generate adjustment instructions for the coolant flow rate and temperature based on the prediction data, and dynamically adjust the coolant flow rate and temperature through a regulating valve and a coolant temperature adjustment device; The adaptive adjustment module is used to monitor the adjusted system operation state in real time, obtain feedback data, and perform adaptive adjustment of the operation state parameters.

8. The adaptive regulation system of a liquid-cooled air-conditioning system according to claim 7, wherein, The demand prediction module includes an acquisition sub-module, a construction sub-module, an optimization sub-module, a setting sub-module, and a training sub-module, where: The acquisition sub-module is used to obtain data such as the current air-conditioning load demand, the previous air-conditioning heat load demand, the current dry bulb temperature, the current relative humidity, the previous dry bulb temperature, the previous relative humidity, the predicted value of the daily maximum temperature, and the predicted value of the daily minimum temperature as input factors; The construction sub-module is used to construct an LSTM model, including an input layer: receiving time series data; an LSTM layer: capturing long-term dependencies in the time series; a fully connected layer: mapping the LSTM output to the target variable; an output layer: outputting the predicted value of the load of the liquid-cooled air-conditioning system; An optimization sub-module, which is used to optimize the LSTM model. Dropout is introduced into the fully connected layer of the third layer of the LSTM model. By randomly discarding a part of neurons in each training iteration, the model cannot overly rely on specific neurons, thereby improving the generalization performance of the model; A setting sub-module, which is used to set model parameters, specifically the time step, the number of LSTM cells, the activation function, and the optimizer, to obtain a load prediction model; A training sub-module, which is used to input the input factors into the load prediction model for training and output prediction data.

Images