Heating ventilation air conditioner control method and device based on heat exchange amount prediction, equipment and medium

Abstract

The invention discloses a heating ventilation air conditioner control method, device and equipment based on heat exchange amount prediction and a medium, and relates to the technical field of air conditioner control. The method comprises the steps that according to historical time sequence data and weather forecast data in a future set time period, a time sequence prediction model is adopted to predict the heat exchange amount in the future set time period in a rolling mode; the circulating water amount in the future set time period is calculated according to the heat exchange amount in the future set time period and the water supply and return temperature difference set interval; the set water supply temperature in the future set time period is predicted according to the heat exchange amount, the circulating water amount and the weather forecast data in the future set time period; based on the set water supply temperature in the future set time period, an optimization algorithm is adopted, and an air conditioning unit control scheme in the future set time period is determined with the purpose of total energy consumption minimization; and when the air conditioner main unit is controlled in real time, the current air conditioner unit control scheme is adjusted through real-time self-correction. According to the invention, the energy-saving efficiency and the control precision are improved.

Description

Technical Field

[0001] The present application relates to the technical field of air-conditioning control, and particularly to a heating, ventilation and air-conditioning (HVAC) control method, device, equipment and medium based on heat transfer prediction. Background Art

[0002] Most of the related HVAC systems are controlled based on fixed parameters, such as a constant water flow rate or a fixed temperature set point. This method has the following problems: 1. Energy waste: The parameters cannot be dynamically adjusted according to the actual demand, resulting in excessive heating or cooling; 2. Low operating efficiency: The equipment operates at a high load for a long time, which is likely to cause component wear and excessive energy consumption; 3. Poor adaptability: It cannot meet the flexible requirements in different time periods and different scenarios. Summary of the Invention

[0003] The purpose of the present application is to provide a heating, ventilation and air-conditioning (HVAC) control method, device, equipment and medium based on heat transfer prediction, which improves the energy-saving efficiency and control accuracy.

[0004] To achieve the above purpose, the present application provides the following solutions:

[0005] In the first aspect, the present application provides a heating, ventilation and air-conditioning (HVAC) control method based on heat transfer prediction. The HVAC control method based on heat transfer prediction includes:

[0006] Collect the historical time series data of the air-conditioning host and the weather forecast data within a future set time period;

[0007] According to the historical time series data and the weather forecast data within the future set time period, use a time series prediction model to roll and predict the heat transfer within the future set time period;

[0008] According to the relationship between the supply and return water temperature difference, the circulating water volume and the heat transfer, use the heat transfer within the future set time period and the set interval of the supply and return water temperature difference to calculate the circulating water volume within the future set time period;

[0009] Input the heat transfer, the circulating water volume and the weather forecast data within the future set time period into a set supply water temperature prediction model to obtain the set supply water temperature within the future set time period;

[0010] Based on the set supply water temperature within the future set time period, use an optimization algorithm to output the control scheme of the air-conditioning unit within the future set time period with the goal of minimizing the total energy consumption; the control scheme of the air-conditioning unit includes the enabling state of the air-conditioning unit and the operating parameters of the variable frequency compressor;

[0011] When controlling the air-conditioning main unit using the described air-conditioning unit control scheme, real-time self-correction calculations are performed on the circulating water volume, set supply water temperature, air-conditioning unit enabling status, and variable-frequency compressor operating parameters of the air-conditioning main unit based on the real-time transmitted data and the weather interface data for the current period. The current air-conditioning unit control scheme is adjusted according to the calculation results.

[0012] Optionally, the time series prediction model and the set supply water temperature prediction model are continuously iteratively trained using a reinforcement learning algorithm based on the historical operation data of the air-conditioning main unit, where the historical operation data includes predicted heat transfer, actual heat transfer, total energy consumption, and the operating parameters of the air-conditioning unit and the HVAC water pump; the actions during the training process include the circulating water volume, set return water temperature, and the operating parameters of the air-conditioning unit enabling status and the variable-frequency compressor; the goal of the reward function during the training process is to minimize the total energy consumption.

[0013] Optionally, collecting the historical time series data of the air-conditioning main unit specifically includes:

[0014] Collecting the historical data of the supply and return water temperatures and the circulating water volume of the air-conditioning main unit;

[0015] Calculating the heat transfer based on the product of the supply and return water temperature difference and the circulating water volume;

[0016] Statistically aligning the calculated heat transfer and the historical weather data according to the time dimension at a unified interval, and cleaning the data using a data cleaning method to obtain the historical time series data marked with the time dimension; the time series characteristics of the historical time series data include hour, week, month, quarter, and whether it is a holiday.

[0017] Optionally, the heat transfer within the future set time period is a probability interval of the heat transfer within the future set time period.

[0018] Optionally, according to the relationship between the supply and return water temperature difference, the circulating water volume, and the heat transfer, using the heat transfer within the future set time period and the set interval of the supply and return water temperature difference, calculating the circulating water volume within the future set time period, specifically including:

[0019] According to the relationship between the supply and return water temperature difference, the circulating water volume, and the heat transfer, using the probability interval of the heat transfer within the future set time period and the set interval of the supply and return water temperature difference, obtaining the probability interval of the circulating water volume for each time period within the future set time period;

[0020] Generating a circulating water volume probability curve based on the probability interval of the circulating water volume for each time period within the future set time period;

[0021] Based on the probability curve of the circulating water volume, through a linear programming optimization algorithm, while meeting the set heat exchange requirements and the constraints of the supply water temperature, the energy consumption of the HVAC pump is minimized to obtain the circulating water volume within a future set time period.

[0022] Optionally, the time series prediction model is a long short-term memory network, a Prophet model, or an autoregressive moving average model; the set supply water temperature prediction model is a backpropagation neural network.

[0023] Optionally, the optimization algorithm is a particle swarm optimization, a genetic annealing algorithm, a mixed integer programming, or a decision tree algorithm.

[0024] In a second aspect, the present application provides an HVAC control device based on heat transfer prediction. The HVAC control device based on heat transfer prediction applies the HVAC control method based on heat transfer prediction. The HVAC control device based on heat transfer prediction includes:

[0025] A data acquisition module, configured to collect historical time series data of the air conditioner host and weather forecast data within a future set time period;

[0026] A heat transfer prediction module, configured to use a time series prediction model to roll and predict the heat transfer within a future set time period according to the historical time series data and the weather forecast data within the future set time period;

[0027] A circulating water volume calculation module, configured to calculate the circulating water volume within a future set time period according to the relationship between the supply and return water temperature difference, the circulating water volume, and the heat transfer, and by using the heat transfer within the future set time period and the set interval of the supply and return water temperature difference;

[0028] A set supply water temperature prediction module, configured to input the heat transfer, the circulating water volume, and the weather forecast data within a future set time period into a set supply water temperature prediction model to obtain the set supply water temperature within the future set time period;

[0029] An air conditioner unit control scheme optimization module, configured to output an air conditioner unit control scheme within a future set time period based on the set supply water temperature within the future set time period, by using an optimization algorithm with the goal of minimizing the total energy consumption; the air conditioner unit control scheme includes the enabled state of the air conditioner unit and the operating parameters of the variable frequency compressor;

[0030] A deviation correction module, configured to perform real-time self-deviation correction calculations on the circulating water volume, the set supply water temperature, the enabled state of the air conditioner unit, and the operating parameters of the variable frequency compressor of the air conditioner host according to the real-time feedback data and the weather interface data when controlling the air conditioner host by using the air conditioner unit control scheme, and adjust the current air conditioner unit control scheme according to the calculation results.

[0031] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the steps of the heat transfer amount prediction-based HVAC control method described in any one of the above.

[0032] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the heat transfer amount prediction-based HVAC control method described in any one of the above are implemented.

[0033] According to the specific embodiments provided by the present application, the following technical effects are disclosed in the present application:

[0034] The present application provides a heat transfer amount prediction-based HVAC control method, device, equipment, and medium. According to the historical time series data and weather forecast data within a future set time period, a time series prediction model is used to roll-predict the heat transfer amount within the future set time period, and the set water supply temperature is predicted based on the rolled heat transfer amount. With the goal of minimizing the total energy consumption, an optimization algorithm is used to output the control scheme of the air conditioning unit within the future set time period, thereby reducing energy waste and improving the energy-saving efficiency. In addition, real-time self-correction calculation is performed on the currently running control scheme according to the real-time feedback data, and the control strategy is adjusted in a timely manner, thereby improving the control accuracy. Description of the Drawings

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0036] Figure 1 It is a schematic flowchart of a heat transfer amount prediction-based HVAC control method provided by an embodiment of the present application.

[0037] Figure 2 It is a schematic structural diagram of a computer device provided by an embodiment of the present application. Detailed Embodiments

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0039] To make the above objects, features, and advantages of the present application more apparent and understandable, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] The present application provides a control method for a heating, ventilation, and air conditioning (HVAC) system based on heat transfer prediction, as Figure 1 shown. The control method for the HVAC system based on heat transfer prediction includes:

[0041] Step 101: Collect historical time series data of the air conditioning main unit and weather forecast data for a future set time period.

[0042] Step 102: According to the historical time series data and the weather forecast data for the future set time period, use a time series prediction model to predict the heat transfer amount for the future set time period in a rolling manner.

[0043] Step 103: According to the relationship between the supply and return water temperature difference, the circulating water volume, and the heat transfer amount, use the heat transfer amount for the future set time period and the set interval of the supply and return water temperature difference to calculate the circulating water volume for the future set time period.

[0044] Step 104: Input the heat transfer amount, the circulating water volume, and the weather forecast data for the future set time period into a set supply water temperature prediction model to obtain the set supply water temperature for the future set time period.

[0045] Step 105: Based on the set supply water temperature for the future set time period, use an optimization algorithm to output a control plan for the air conditioning unit for the future set time period with the goal of minimizing the total energy consumption; the control plan for the air conditioning unit includes the operating state of the air conditioning unit and the operating parameters of the variable frequency compressor.

[0046] Step 106: When using the control plan for the air conditioning unit to control the air conditioning main unit, perform real-time self-correction calculations on the circulating water volume, the set supply water temperature, the operating state of the air conditioning unit, and the operating parameters of the variable frequency compressor of the air conditioning main unit according to the real-time feedback data and the weather interface data for the current time period, and adjust the current control plan for the air conditioning unit according to the calculation results.

[0047] The time series prediction model and the set water supply temperature prediction model are continuously iteratively trained using a reinforcement learning algorithm based on the historical operation data of the air conditioning host. The historical operation data includes predicted heat transfer, actual heat transfer, total energy consumption, and the operation parameters of the air conditioning unit and the HVAC water pump. The adjustment actions of the control strategy during the training process include the circulating water volume, the set return water temperature, and the operation parameters of the air conditioning unit startup status and the variable frequency compressor. The goal of the reward function during the training process is to minimize the total energy consumption. In this application, the model is continuously iteratively trained through the reinforcement learning algorithm, the energy consumption of the control scheme is evaluated, and the circulating water volume prediction and the control strategy of the air conditioning unit are optimized. Through continuous reinforcement learning, the control strategy is optimized, and the energy efficiency of the air conditioning system is continuously optimized.

[0048] The objective function of the total energy consumption is:

[0049] where c i is the comprehensive energy consumption of the i-th HVAC pump, and x i is the coefficient of the i-th HVAC pump. The objective function of the total energy consumption includes constraints such as flow rate, water pressure, pump efficiency, pump startup energy consumption, pump startup duration, and maintenance. N is the number of pumps.

[0050] ① Flow rate constraint factor:

[0051] where Q i is the flow rate of the i-th HVAC pump, Q pre is the predicted total water supply, and Q max is the maximum allowable water volume.

[0052] ② Water pressure constraint factor:

[0053] where P i is the water pressure of the i-th HVAC pump, P pre is the predicted water pressure, and P max is the maximum allowable water pressure.

[0054] ③ Optimal efficiency factor:

[0055] where η i is the efficiency of the i-th pump, and η min is the expected minimum efficiency.

[0056] ④ Pump startup energy consumption factor:

[0057] where E start_i is the pump startup energy consumption of the i-th pump, and E start_max is the maximum pump startup energy consumption limit.

[0058] ⑤ Pump startup duration factor:

[0059] Among them, T start_i is the starting pump duration of the i-th pump, and T start_min is the minimum starting pump duration per time, and T start_max is the maximum starting pump duration per time.

[0060] ⑥ Maintenance factor:

[0061] Among them, F i is the operation duration within a maintenance cycle of the i-th pump, and F max is the maximum maintenance cycle.

[0062] It is necessary to calculate the minimum value of the sum of the comprehensive energy consumption of the pumps according to the above.

[0063] This application uses technologies such as time series prediction algorithms, optimization algorithms, deep learning, and reinforcement learning to model and predict the operating parameters of HVAC units and HVAC pumps and perform optimal control.

[0064] In an exemplary embodiment, the future set time period is the next 24 hours. This application first collects the supply and return water temperatures of the air-conditioning main unit, the circulating water volume, and weather forecast data through sensors, and uses a time series prediction algorithm to build a model to predict the heat transfer amount in the next 24 hours. Then, according to the prediction result of the heat transfer amount, the circulating water volume of the HVAC pump in the next 24 hours is dynamically planned to ensure that the system energy consumption is reduced while meeting the cooling / heating demand. In this application, the HVAC pump refers to the HVAC water pump. In addition, because the HVAC system is a typical non-linear time-varying system, in addition to the prediction result, an algorithm real-time self-correction module is also required to combine the current data to perform self-correction adjustment on the circulating water volume, optimize the system operating parameters, realize dynamic correction and optimal control of the circulating water volume of the HVAC pump and the set return water temperature of the main unit, and provide further matching of the actual cooling / heating demand. The system supports switching between multiple scenario modes and meets the requirements of different usage scenarios through specific control strategies. At the same time, the reinforcement learning algorithm is used to continuously iterate and train the model to optimize the control strategy and achieve higher energy efficiency.

[0065] In an exemplary embodiment, in step 101, the historical time series data of the air-conditioning main unit is collected, specifically including:

[0066] Collect the historical data of the supply and return water temperatures and the circulating water volume of the air-conditioning main unit, specifically by collecting the supply and return water temperatures and the circulating water volume of the air-conditioning main unit through sensors. The supply and return water temperatures include the supply water temperature and the return water temperature.

[0067] Calculate the heat transfer amount according to the product of the supply and return water temperature difference and the circulating water volume. The supply and return water temperature difference is the temperature difference between the supply water temperature and the return water temperature.

[0068] Using an interface to crawl historical weather data, which includes weather phenomena, outdoor temperature, wind force, ultraviolet intensity, relative humidity, etc. Weather phenomena are weather characterization parameters such as overcast, sunny, rainy, or snowy. In fact, the text descriptions of each weather characterization parameter correspond to one of a set of numbers. For example, sunny is 1, overcast turning to cloudy is 20, and sleet is 106.

[0069] Align the calculated heat exchange amount and historical weather data statistically at a unified time interval dimension, and clean the data using data cleaning methods to obtain the historical time-series data marked with the time dimension, that is, the standard time-series dataset; the time-series characteristics of the historical time-series data include hour, week, month, quarter, and whether it is a holiday.

[0070] More specifically, step 101 includes: The central control system computer of the HVAC system collects and stores the key data required for system operation through sensors and network interfaces. The key data includes the supply water temperature T of the air-conditioning host in and the return water temperature T out , the circulating water volume Q of the HVAC pump l , the pump frequency F of the HVAC pump, the opening degree L of the expansion valve, the energy consumption Q of the pump motor w , the total energy consumption Q 0 . In addition, there is also data collected through the weather interface, such as the outdoor temperature T w , the weather phenomenon W h , the ultraviolet index level UV, the wind force W D , the relative humidity f e , etc. And the stored original data is statistically processed according to different rules in dimensions such as minutes, hours, days, etc. The statistical rules include maximum value Max, minimum value Min, average value Avg, difference Sub, etc. The statistical data and the original data are stored together in the database of the central control system computer server.

[0071] Align the statistically processed multi-factor data in ascending order according to the continuous time series. Using the data of the supply water temperature T s (t) and the return water temperature T r (t) of the host and the circulating water volume Q l (t) of the HVAC pump in each time period t, calculate the actual heat exchange amount Q c (t).

[0072] The calculation formula for the actual heat exchange amount is: Q c (t) = ΔT(t)·Q l (t)·ρ·c p = (T s (t) - T r (t))·Q l (t)·ρ·c p .

[0073] where ΔT(t) is the temperature difference between the supply and return water, ρ is the density of the heat transfer fluid, and c p is the specific heat capacity of the heat transfer fluid.

[0074] Then, abnormal noise data is cleaned by a data cleaning method. The data cleaning method includes data interpolation, data deduplication, and data conversion. The data interpolation methods include median interpolation, normal mean filling, Bessel interpolation, regression prediction filling, etc. The data deduplication methods include clustering deduplication, logical deduplication, weighted average merging deduplication, etc. The data conversion methods include field standardization, merging synonymous terms, data normalization, etc. After the data cleaning method, the time dimension of the time series data is marked. The marked time series features include hour, week, month, quarter, whether it is a holiday, etc., to obtain a standard time series data sample set.

[0075] In an exemplary embodiment, the time series prediction model is a Long Short-Term Memory (LSTM) network, a Prophet model, or an AutoRegressive Moving Average (ARMA) model. The time series prediction model rolls and predicts the heat transfer amount Q c (t) pred in the next 24 hours based on the time series feature information in the next 24 hours and combined with the weather forecast data in the next 24 hours. The predicted heat transfer amount obtained is a probability interval.

[0076] In an exemplary embodiment, step 102 specifically includes: According to the historical time series data and the weather forecast data within the next 24 hours, a time series prediction model is used to roll and predict the heat transfer amount within the next 24 hours. The heat transfer amount within the future set time period is a probability interval of the heat transfer amount within the future set time period, that is, the predicted heat transfer amount is a probability interval. The heat transfer amount within the future set time period is time series data composed of the heat transfer amounts of multiple time periods.

[0077] In an exemplary embodiment, step 103 specifically includes:

[0078] According to the relationship between the temperature difference between the supply and return water, the circulating water volume, and the heat transfer amount, a probability interval of the circulating water volume for each time period within the future set time period is obtained by using the probability interval of the heat transfer amount within the future set time period and the set interval of the temperature difference between the supply and return water.

[0079] A circulating water volume probability curve is generated according to the probability interval of the circulating water volume for each time period within the future set time period.

[0080] According to the circulating water volume probability curve, through the linear programming optimization algorithm, on the basis of meeting the set heat exchange demand and the supply water temperature constraint, the energy consumption of the HVAC water pump is minimized to obtain the circulating water volume within the future set time period, reduce the unnecessary circulating water volume demand, and thus reduce the system energy consumption. The supply water temperature constraint is that the circulating water volume is not lower than the minimum limit value.

[0081] More specifically, according to the predicted heat exchange data for the next 24 hours, the median 95% confidence interval is the lowest value of the confidence interval, is the highest value of the confidence interval. Then, the set interval of the supply and return water temperature difference is designed as a typical empirical interval, usually 5 - 10 °C. According to the theoretical formula of the supply and return water temperature difference, the circulating water volume, and the heat exchange amount, the circulating water volume Q l (t) pred for each time period t within the next 24 hours is calculated to predict the probability interval, and a probability curve of the HVAC pump circulating water volume for the next 24 hours is generated. Through the linear programming optimization algorithm, on the basis of meeting the heat exchange demand and ensuring that the supply water temperature is not lower than the minimum limit value, the energy consumption Q p (t) of the HVAC water pump is minimized to reduce the unnecessary circulating water volume demand, and thus reduce the system energy consumption, where k is a constant.

[0082] In an exemplary embodiment, a neural network black box model, i.e., a set supply water temperature prediction model, is established with the heat exchange amount, the circulating water volume, and weather factors as inputs and the set supply water temperature as the output. According to the predicted heat exchange amount and circulating water volume data for the next 24 hours and the weather forecast data for the next 24 hours, the set supply water temperature of the air conditioner unit for the next 24 hours is obtained. The neural network black box model is a Back Propagation (BP) neural network.

[0083] In an exemplary embodiment, step 105 specifically includes: summarizing the historical operation data of the air conditioner unit, analyzing which units in the air conditioner unit have higher operation efficiency and which have lower efficiency under the same working conditions through the unit operation power and the unit energy consumption, and setting the activation priority of each machine according to the operation efficiency of the air conditioner unit.

[0084] Based on the set supply water temperature within the future set time period, an optimization algorithm is used to output the control scheme of the air conditioner unit within the future set time period with the goal of minimizing the total energy consumption; the control scheme of the air conditioner unit includes the activation state of the air conditioner unit and the operation parameters of the variable frequency compressor. The operation parameters of the variable frequency compressor are the frequency or speed of the variable frequency compressor.

[0085] The control scheme of the air-conditioning units within a set future time period, that is, which units should be activated and which units should remain off within the set future time period, and how the frequency or rotational speed of the variable-frequency compressors should be adjusted to achieve the goal of minimizing the overall energy consumption and realizing the coordinated energy conservation between the HVAC water pumps and the air-conditioning units.

[0086] In an exemplary embodiment, the optimization algorithm is a particle swarm optimization, a genetic annealing algorithm, a mixed integer programming, or a decision tree algorithm.

[0087] This application uses historical operation data and corresponding energy consumption records, and through an optimization algorithm, deeply learns and analyzes these data to summarize the relationship between the system energy consumption and the set water supply temperature, the activation status of the air-conditioning units, and the frequency of the variable-frequency compressors. The optimization model is based on the predicted data of the set water supply temperature for the next 24 hours, the energy consumption performance of the pump sets, and weather factors (weather, outdoor temperature, wind force, ultraviolet intensity, relative humidity, etc.), and automatically selects the optimal control decision for the air-conditioning units. Its output result is how the activation status, operation duration of the air-conditioning units, and the frequency or rotational speed of the variable-frequency compressors should be adjusted within the next 24 hours, generating an optimal control scheme for the air-conditioning units within the next 24 hours to achieve the goal of minimizing the overall energy consumption and realizing the coordinated energy conservation between the circulating water pumps and the units. For example, in the case of low usage at night, the system scheme will reduce the operation frequency of the circulating water pumps and turn off some of the air-conditioning main units, only maintaining the minimum load demand in the core area, thus significantly reducing the energy consumption.

[0088] In an exemplary embodiment, in step 106, the real-time feedback data includes the circulating water volume of the HVAC water pump, the temperature difference between the supply and return water, the set temperature of the terminal panel, and the energy consumption of the air-conditioning unit. The weather interface data includes weather phenomena, outdoor temperature, wind force, ultraviolet intensity, and relative humidity. By automatically correcting in real time to adjust the current and future 24-hour operation plans, and re-controlling the operation parameters of the entire system according to the calculation results, the dynamic matching and optimization of the heat exchange capacity of the HVAC system are realized, optimizing the system performance and reducing the system energy consumption.

[0089] The technical effects of this application are as follows:

[0090] 1. Through the regulation of the circulating water volume and the optimization of the unit, this application significantly improves the energy-saving efficiency of the HVAC system. Traditional systems usually operate with a constant cooling or heating capacity and cannot be flexibly adjusted according to actual needs, which easily leads to energy waste. In contrast, this application combines a time series prediction model, utilizes historical operation data and real-time weather information to accurately predict the heat exchange demand within the next 24 hours, and dynamically adjusts operation parameters such as the circulating water volume of the HVAC pump. For example, during low-demand periods, the system automatically reduces the circulating water volume and heat exchange capacity to avoid excessive cooling or heating; during high-demand periods, it optimizes the system operation strategy in advance to ensure efficient heating or cooling. By adjusting the circulating water volume and unit power, unnecessary energy losses are reduced. This demand-driven control method minimizes energy waste while meeting the demand, and the energy-saving efficiency is increased by more than 20% compared to traditional systems.

[0091] 2. Through real-time self-correcting regulation and reinforcement learning, this application achieves a significant improvement in control accuracy. The real-time self-correcting regulation module is based on the real-time operation data of the HVAC pump, combines variables such as weather phenomena and outdoor temperature, and dynamically controls and adjusts the circulating water volume of the HVAC pump. This closed-loop control mechanism can quickly respond to environmental and load changes, ensuring that the operation parameters are always highly matched with the actual needs. In addition, the reinforcement learning model optimizes the control strategy of the air-conditioning host through continuous training and learning of historical operation data, enabling the system to adapt to complex operation scenarios and find the optimal regulation plan. Compared with the traditional fixed-rule control method, this application not only improves the control accuracy but also effectively reduces energy consumption, enhancing the reliability and stability of the system.

[0092] 3. This application incorporates the prediction of heat exchange volume with weather and holiday information. Traditional HVAC systems often adopt a single operation mode and are difficult to flexibly respond to demand changes in different time periods or special scenarios. This application can adapt to various scenarios, including weekdays, holidays, nights, rainy and snowy days, etc., and provides control strategies for different scenarios. For example, during holidays, the system preferentially reduces the load in non-essential areas and only maintains the basic operation requirements; during peak demand periods, the system quickly increases the circulating water volume and heat exchange capacity to meet large-scale centralized cooling / heating demands. Ensure that the air-conditioning system can adjust operation parameters according to the actual usage scenario, thereby more efficiently and flexibly meeting the demand and enhancing the user experience.

[0093] 4. On the premise of meeting the heat exchange requirements, this application effectively reduces unnecessary energy consumption through accurate prediction and dynamic regulation. For example, the system continuously learns and evaluates the energy consumption of the control scheme. In the case of low usage at night, it reduces the operating frequency of the circulating water pump and shuts down some air-conditioning main units, only maintaining the minimum load demand in the core area, thus significantly reducing energy consumption. At the same time, the reinforcement learning module optimizes the system operation parameters, reduces the unnecessary operation time of the equipment, reduces the wear of mechanical components, and extends the service life of the equipment. Compared with the traditional system, this application can not only significantly reduce the operating cost, but also reduce the maintenance cost, bringing higher economic benefits to users.

[0094] Based on the same inventive concept, the embodiment of this application also provides a heat exchange quantity prediction-based HVAC control device for implementing the above-mentioned heat exchange quantity prediction-based HVAC control method. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more of the following embodiments of the heat exchange quantity prediction-based HVAC control device can refer to the limitations on the heat exchange quantity prediction-based HVAC control method in the above text, and will not be elaborated here.

[0095] In an exemplary embodiment, this application provides a heat exchange quantity prediction-based HVAC control device. The heat exchange quantity prediction-based HVAC control device applies the above-mentioned heat exchange quantity prediction-based HVAC control method. The heat exchange quantity prediction-based HVAC control device includes:

[0096] A data acquisition module for collecting the historical time series data of the air-conditioning main unit and the weather forecast data within a future set time period.

[0097] A heat exchange quantity prediction module for rollingly predicting the heat exchange quantity within a future set time period by using a time series prediction model based on the historical time series data and the weather forecast data within a future set time period.

[0098] A circulating water volume calculation module for calculating the circulating water volume within a future set time period by using the heat exchange quantity within a future set time period and the set range of the supply and return water temperature difference according to the relationship between the supply and return water temperature difference, the circulating water volume and the heat exchange quantity.

[0099] A set supply water temperature prediction module for inputting the heat exchange quantity, the circulating water volume and the weather forecast data within a future set time period into a set supply water temperature prediction model to obtain the set supply water temperature within a future set time period.

[0100] An air conditioner unit control scheme optimization module, which is used to output the air conditioner unit control scheme within a future set time period based on the set water supply temperature within the future set time period, adopt an optimization algorithm, and aim at minimizing the total energy consumption; the air conditioner unit control scheme includes the air conditioner unit startup state and the variable frequency compressor operation parameters.

[0101] A deviation correction module, which is used to perform real-time self-correction calculation on the circulating water volume, set water supply temperature, air conditioner unit startup state and variable frequency compressor operation parameters of the air conditioner main unit according to the real-time feedback data and the weather interface data of the current time period when controlling the air conditioner main unit by using the air conditioner unit control scheme, and adjust the current air conditioner unit control scheme according to the calculation result.

[0102] In an exemplary embodiment, a computer device is provided. The computer device can be a server or a terminal, and its internal structure diagram can be as Figure 2 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store heating, ventilation, and air conditioning control data based on heat transfer prediction. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it realizes a heating, ventilation, and air conditioning control method based on heat transfer prediction.

[0103] Those skilled in the art can understand that Figure 2 the structure shown in

[0104] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are realized.

[0105] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0106] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the various embodiments provided in this application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0107] The databases involved in the various embodiments provided in this application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., and are not limited thereto. The processors involved in the various embodiments provided in this application can be general-purpose processors, central processors, graphics processors, digital signal processors, data processing logics of programmable logics, etc., and are not limited thereto.

[0108] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0109] In this article, specific examples are used to elaborate on the principles and implementation manners of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A HVAC control method based on heat exchange prediction, characterized in that: The HVAC control method based on heat exchange amount prediction includes: Collect historical time series data of the air conditioner host and weather forecast data within a set time period in the future; Based on the historical time series data and the weather forecast data within the future set time period, a time series prediction model is used to make a rolling prediction of the heat exchange amount within the future set time period; According to the relationship between the supply and return water temperature difference, the circulating water volume and the heat exchange, the circulating water volume in the future set time period is calculated by using the heat exchange in the future set time period and the supply and return water temperature difference setting interval; The heat exchange amount, circulating water volume and weather forecast data within a future set time period are input into a set water supply temperature prediction model to obtain a set water supply temperature within a future set time period; Based on the set water supply temperature in the future set time period, an optimization algorithm is used to minimize the total energy consumption and output the air conditioning unit control plan in the future set time period; the air conditioning unit control plan includes the air conditioning unit activation state and the variable frequency compressor operation parameters; When the air-conditioning unit control scheme is used to control the air-conditioning host, real-time self-correction calculations are performed on the circulating water volume, set water supply temperature, air-conditioning unit activation status and variable frequency compressor operating parameters of the air-conditioning host based on real-time feedback data and weather interface data of the current period, and the current air-conditioning unit control scheme is adjusted based on the calculation results.

2. The HVAC control method based on heat exchange prediction according to claim 1, characterized in that: The time series prediction model and the set water supply temperature prediction model are iteratively trained based on the historical operation data of the air-conditioning host using a reinforcement learning algorithm, wherein the historical operation data includes the predicted heat exchange, the actual heat exchange, the total energy consumption, and the operating parameters of the air-conditioning unit and the HVAC water pump; the actions in the training process include the circulating water volume, the set return water temperature, the activation status of the air-conditioning unit and the operating parameters of the variable frequency compressor; the goal of the reward function in the training process is to minimize the total energy consumption.

3. The HVAC control method based on heat exchange prediction according to claim 1, characterized in that: Collect historical time series data of the air conditioner host, including: Collecting historical data of the supply and return water temperature and circulating water volume of the air conditioner host; The heat exchange is calculated based on the product of the supply and return water temperature difference and the circulating water volume; The calculated heat exchange amount and the historical weather data are statistically aligned according to a uniformly spaced time dimension, and the data is cleaned using a data cleaning method to obtain the historical time series data marked with a time dimension; the time series features of the historical time series data include hours, weeks, months, quarters, and whether it is a holiday.

4. The HVAC control method based on heat exchange prediction according to claim 1, characterized in that: The heat exchange amount within the future set time period is a probability interval of the heat exchange amount within the future set time period.

5. The HVAC control method based on heat exchange prediction according to claim 4, characterized in that: According to the relationship between the supply and return water temperature difference, the circulating water volume and the heat exchange, the heat exchange in the future set time period and the supply and return water temperature difference setting interval are used to calculate the circulating water volume in the future set time period, including: According to the relationship between the supply and return water temperature difference, the circulating water volume and the heat exchange, the probability interval of the heat exchange in the future set time period and the supply and return water temperature difference setting interval are used to obtain the probability interval of the circulating water volume in each period in the future set time period; Generate a circulating water volume probability curve according to the probability interval of the circulating water volume in each period within the future set time period; According to the circulating water volume probability curve, a linear programming optimization algorithm is used to minimize the HVAC water pump energy consumption on the basis of meeting the set heat exchange demand and water supply temperature constraints, and obtain the circulating water volume within the future set time period.

6. The HVAC control method based on heat exchange prediction according to claim 1, characterized in that: The time series prediction model is a long short-term memory network, a Prophet model or an autoregressive moving average model; the set water supply temperature prediction model is a back propagation neural network.

7. The HVAC control method based on heat exchange prediction according to claim 1, characterized in that: The optimization algorithm is particle swarm optimization, genetic annealing algorithm, mixed integer programming or decision tree algorithm.

8. A HVAC control device based on heat exchange prediction, characterized in that: The HVAC control device based on heat exchange prediction applies the HVAC control method based on heat exchange prediction according to any one of claims 1 to 7, and the HVAC control device based on heat exchange prediction includes: The data collection module is used to collect the historical time series data of the air conditioner host and the weather forecast data within the future set time period; A heat exchange amount prediction module is used to use a time series prediction model to make a rolling prediction of the heat exchange amount within a future set time period based on the historical time series data and the weather forecast data within a future set time period; The circulating water volume calculation module is used to calculate the circulating water volume in the future set time period according to the relationship between the supply and return water temperature difference, the circulating water volume and the heat exchange, using the heat exchange in the future set time period and the supply and return water temperature difference setting interval; A set water supply temperature prediction module is used to input the heat exchange, circulating water volume and weather forecast data within a future set time period into a set water supply temperature prediction model to obtain a set water supply temperature within a future set time period; An air conditioning unit control scheme optimization module is used to output an air conditioning unit control scheme within a future set time period based on a set water supply temperature within a future set time period, using an optimization algorithm and taking total energy consumption minimization as a goal; the air conditioning unit control scheme includes an air conditioning unit activation state and variable frequency compressor operating parameters; The correction module is used to control the air-conditioning host by adopting the air-conditioning unit control scheme, and to perform real-time self-correction calculations on the circulating water volume, set water supply temperature, air-conditioning unit activation status and variable frequency compressor operating parameters of the air-conditioning host according to real-time feedback data and weather interface data, and adjust the current air-conditioning unit control scheme according to the calculation results.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the HVAC control method based on heat exchange prediction described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the HVAC control method based on heat exchange prediction described in any one of claims 1 to 7 is implemented.

Images