Energy-saving control method for heating, ventilating and air conditioning of building

Abstract

The invention discloses a building heating ventilation air conditioner energy-saving control method which comprises the following steps: acquiring building structure information and building envelope material parameters, and constructing a building thermal inertia characteristic model; collecting indoor and outdoor temperature data of the target building in a plurality of time periods, and generating a thermal response time sequence data set; on the basis of the thermal response time sequence data set, a building thermal inertia characteristic model is combined, and a building thermal buffering capacity prediction function is established; obtaining an electricity price prediction data curve in a future preset time period, wherein the electricity price prediction data curve comprises electricity price fluctuation information corresponding to different time points; according to the building thermal buffering capacity prediction function and the electricity price prediction data curve, an air conditioner operation power scheduling priority strategy is generated; based on the scheduling priority strategy, the optimal start-stop control interval and air supply parameters of each regulation and control time period are calculated in combination with the indoor temperature target range set by the user; and in the optimal start-stop control interval, the operation state of the air conditioning system is regulated and controlled. According to the invention, the purposes of energy conservation and consumption reduction can be achieved.

Description

Technical Field

[0001] This invention relates to the field of air conditioning control technology, and in particular to an energy-saving control method for building heating, ventilation and air conditioning systems. Background Technology

[0002] With the global energy shortage worsening and building carbon emission regulations tightening, the need for energy conservation and emission reduction in the building sector is becoming increasingly urgent. Heating, ventilation, and air conditioning (HVAC) systems, as one of the main sources of building energy consumption, account for over 40% of total building energy consumption. Optimizing the operating efficiency of HVAC systems while ensuring indoor thermal comfort is one of the core issues in the current development of building energy-saving technologies.

[0003] Existing air conditioning control strategies largely rely on simple start-stop logic and fixed temperature settings, failing to dynamically adjust operating status based on building structure thermal inertia or external environmental changes. While some technologies incorporate time-of-use electricity pricing for peak-shifting control, they lack effective integration with the building's thermal buffering capacity for energy consumption scheduling. This results in forced air conditioning operation during periods of high electricity prices, increasing energy costs and reducing overall system efficiency. Furthermore, traditional air supply parameter settings are often based on experience, lacking the ability to finely control the rate of indoor temperature change, making it difficult to ensure indoor thermal comfort while achieving energy savings.

[0004] Therefore, there is an urgent need for an energy-saving control method that can comprehensively sense the thermal response characteristics of buildings, dynamically predict the trend of room temperature changes, and combine electricity price information to schedule start-up and shutdown and control air supply, so as to reduce the overall energy consumption of building air conditioning systems and realize a new building energy consumption optimization control path of "driving energy-saving strategies with thermal buffering capacity". Summary of the Invention

[0005] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for energy-saving control of building heating, ventilation and air conditioning, comprising the following steps: Obtain building structural information and building envelope material parameters to construct a building thermal inertia characteristic model; Collect and process indoor and outdoor temperature data of the target building over multiple time periods to generate a thermal response time series dataset; Based on the aforementioned thermal response time series dataset, and combined with the building thermal inertia characteristic model, a prediction function for building thermal buffering capacity is established. Obtain electricity price forecast data curves for a future preset time period, wherein the electricity price forecast data curves include electricity price fluctuation information corresponding to different time points; Based on the building thermal buffer capacity prediction function and the electricity price prediction data curve, a priority strategy for scheduling the building's air conditioning power in the future time period is generated. Based on the aforementioned scheduling priority strategy and combined with the user-defined indoor temperature target range, the optimal start / stop control interval and air supply parameters for each control period are calculated. Within the optimal start-stop control range, the operating status of the air conditioning system is regulated.

[0007] As a preferred embodiment of the building HVAC energy-saving control method of the present invention, the construction of the building thermal inertia characteristic model includes the following steps: Extract the heat transfer coefficient, heat capacity parameters, and building spatial topology of the building envelope; Based on heat transfer coefficient, heat capacity parameters and building space topology, a multi-physics field coupling model is established that includes wall heat storage effect and air layer convection heat transfer. The thermal inertial characteristics of the building structure are modeled using the finite element method to obtain the thermal time constant distribution matrix; The thermal inertial characteristic model is constructed based on the thermal time constant distribution matrix and the multiphysics coupling model.

[0008] As a preferred embodiment of the building HVAC energy-saving control method of the present invention, the method for establishing the building heat buffer capacity prediction function includes: Fourier transform is performed on the thermal response time series dataset to extract phase lag features that reflect the thermal response delay characteristics; Based on the phase lag characteristics and the historical temperature sequences of the target building collected over multiple time periods, a room temperature change rate prediction model is constructed using a long short-term memory neural network. By combining the thermal resistance parameters in the building thermal inertia characteristic model, a nonlinear mapping function is constructed with outdoor temperature as input and indoor temperature change gradient as output, forming the final thermal buffer capacity prediction function.

[0009] As a preferred embodiment of the energy-saving control method for building heating, ventilation and air conditioning described in this invention, the thermal resistance parameter is calculated from the thermal conductivity of the building envelope material and the thickness of its structural layer.

[0010] As a preferred embodiment of the building HVAC energy-saving control method of the present invention, the expression of the building heat buffer capacity prediction function is:

[0011] in: Indicates the predicted time point The indoor temperature; This is a nonlinear mapping function obtained by training a long short-term memory neural network, which reflects the system's response to input variables; The current indoor temperature; The outdoor temperature at the current moment; This is the air conditioning operation control vector; The phase lag characteristic function is obtained based on the Fourier transform, with the current indoor temperature as the reference. Outdoor temperature Air conditioning control quantity The input variable is used to extract the frequency domain features of the thermal response delay; This is the thermal resistance matrix of the building structure, which is composed of the thermal resistance parameters of each structural unit of the building.

[0012] As a preferred embodiment of the building HVAC energy-saving control method of the present invention, the determination of the air conditioning operating power scheduling priority strategy includes: Based on the electricity price forecast data curve, multiple electricity price ranges are divided, specifically including: high electricity price range, flat electricity price range, and low electricity price range. In high electricity price ranges, the output of the building thermal buffer capacity prediction function is used to determine the building's future short-term time window. If the room has the ability to maintain room temperature, and the result is yes, then the power reduction strategy is implemented, which reduces the compressor operating power or increases the air supply temperature. During periods of low electricity prices, an advance energy storage strategy is implemented, and the operating power of the air conditioners, the trajectory of room temperature changes, and the cumulative energy storage duration during the energy storage phase are recorded to generate a power compensation scheduling table. This power compensation scheduling table is used to dynamically guide the magnitude, duration, and regional distribution of power reductions during subsequent periods of high electricity prices.

[0013] In a preferred embodiment of the building HVAC energy-saving control method of the present invention, the determination of having the ability to maintain room temperature satisfies any of the following conditions: Predict room temperature Still within the set comfortable temperature range ; The current actual temperature is either lower than the user-set target lower temperature limit or higher than the upper temperature limit, with the upper temperature limit being used for heating. The current thermal time constant is greater than the set hysteresis threshold, indicating that the temperature drop / rise changes slowly and has a strong thermal buffering capacity.

[0014] As a preferred embodiment of the building HVAC energy-saving control method of the present invention, the pre-storage energy storage strategy includes: Prediction function based on building thermal buffer capacity The system predicts whether room temperature may exceed the user's set temperature target range during periods of high electricity prices in the future. If the prediction indicates that the building's thermal buffer capacity is insufficient to support the target temperature range, it triggers advance energy storage. Once the advance energy storage is triggered, the system executes the following control measures: During the period from the current moment to the beginning of the high electricity price range, the air supply volume, air supply temperature and compressor operating power are dynamically adjusted to ensure that the room temperature reaches the user's target temperature boundary before the arrival of high electricity prices. Meanwhile, during this energy storage control phase, the system continuously records the air conditioning operating power, room temperature change trajectory, and cumulative energy storage time, and generates a power compensation scheduling table based on the above data. This scheduling table is used to guide energy-saving control strategies in high electricity price ranges.

[0015] As a preferred embodiment of the building HVAC energy-saving control method of the present invention, the calculation method for the optimal start-stop control interval and supply air parameters includes: The upper and lower limits of the user-defined indoor temperature target range are used as comfort boundary conditions, serving as the room temperature control constraint range. Based on the building thermal buffer capacity prediction function, the room temperature change trend under the condition of no active air conditioning control is derived in reverse, and the predicted time point when the room temperature is about to reach the boundary condition is determined as the critical point of temperature change. Combining the current building's thermal inertia state parameters and external load conditions, a model predictive control algorithm is used to solve for the longest sustainable shutdown period for cooling or heating, while ensuring that the room temperature is kept within a comfortable range, and to determine the optimal start-stop control interval. Based on the optimal start-stop control range, the air supply parameters are further calculated to control the rate of change in room temperature.

[0016] As a preferred embodiment of the building HVAC energy-saving control method of the present invention, the operating status of the air conditioning system is regulated within the optimal start-stop control range, wherein the regulation includes dynamically adjusting the compressor power, fan speed and start-stop status based on the generated air supply parameters and control strategy.

[0017] The beneficial effects of this invention are: 1. This invention acquires building structural information and envelope material parameters, and combines finite element analysis with a multiphysics thermal model to construct a complete building thermal inertia characteristic model, which can realistically reflect the heat conduction and heat storage behavior of buildings under unsteady conditions. Furthermore, by combining historical indoor and outdoor temperature data to construct a thermal response time-series dataset, the thermal response patterns of different building areas under different external load conditions can be discovered, thereby improving the dynamic accuracy and foresight of the thermal buffer capacity prediction function.

[0018] 2. This invention, based on a building thermal buffer capacity prediction function, incorporates future electricity price forecast data to construct a priority strategy for air conditioning power scheduling across multiple time periods and scenarios. This strategy allows for flexible adjustment of start-up and shutdown rhythms and operating intensity based on electricity price fluctuations. Especially during peak electricity price periods, the system can determine whether a building has thermal buffer capacity and proactively adopt power reduction or pre-cooling / pre-heating strategies to effectively reduce air conditioning operating power during high-price periods, lower energy costs, and achieve the goals of on-demand scheduling, load shifting, and optimal energy efficiency.

[0019] 3. This invention calculates the longest sustainable downtime within the optimal start-stop control range by combining the user-defined comfort temperature range, the building's current thermal inertia state, and a model predictive control algorithm. Simultaneously, it precisely sets air supply parameters such as air volume, air temperature, and wind speed, achieving fine control over the rate of room temperature change. This method significantly reduces the frequency of air conditioner start-stop operations, extends equipment lifespan, and achieves energy conservation and consumption reduction while ensuring indoor thermal comfort. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is an overall flowchart of a building heating, ventilation and air conditioning energy-saving control method according to the present invention.

[0021] Figure 2 This is a flowchart illustrating the determination of the priority strategy for air conditioning operation power scheduling in a building HVAC energy-saving control method according to the present invention. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth.

[0026] Example 1 Reference Figure 1-2 As an embodiment of the present invention, a building heating, ventilation and air conditioning energy-saving control method is provided, comprising the following steps: S1: Obtain building structure information and building envelope material parameters, and construct a building thermal inertia characteristic model.

[0027] It should be noted that by connecting to a Building Information Modeling (BIM) system or analyzing building structural design drawings, information on the building's spatial structure and thermal parameters of the building envelope can be obtained. This includes the building's spatial topology, thermal conductivity, heat capacity, thickness, material type, and area ratio of various components such as walls, floors, roofs, doors, and windows. The purpose of obtaining building structural information and building envelope material parameters is to provide a foundation for subsequently constructing a building thermal inertia characteristic model. A building's thermal inertia characteristics largely depend on the heat transfer performance and heat storage capacity of its building envelope (including exterior walls, roofs, floors, doors, and windows). Differences in heat conduction and storage among different materials directly affect the dynamic changes in indoor temperature. Therefore, by obtaining building structural information and building envelope material parameters, accurate modeling of the building's thermal inertia characteristic model can be achieved, thereby improving the foresight and accuracy of air conditioning power scheduling priority strategies.

[0028] Specifically, constructing a building thermal inertia characteristic model includes the following steps: First, the heat transfer coefficient, heat capacity parameters, and spatial topology of the building envelope are extracted. Among them, the heat transfer coefficient and heat capacity parameters are commonly used thermal parameters in building thermal simulation. The spatial topology can be obtained from BIM data or architectural drawings, and specifically includes room distribution, connectivity, orientation, floor height, etc., which are used to describe heat conduction paths and heat storage characteristics.

[0029] Then, based on the heat transfer coefficient, heat capacity parameters, and building spatial topology, a multiphysics coupling model is established, incorporating the wall heat storage effect and air-layer convective heat transfer. For example, the steps for establishing the multiphysics coupling model include: First, defining the heat conduction equation of the wall, that is, using thermal conductivity and structural thickness to describe the unsteady heat conduction process in each building envelope, accurately characterizing the heat diffusion behavior over time within solid structures such as walls, providing crucial support for constructing a basic heat transfer model reflecting heat accumulation and heat leakage characteristics; Second, setting the air-layer convective heat transfer model, that is, dynamically characterizing the heat exchange process between the air layer and the solid structure through the relationship between the convective heat transfer coefficient and the inter-room wind speed, thereby reflecting the modulating ability of factors such as supply air, return air, and natural ventilation on changes in the overall thermal environment. The model improves the simulation accuracy of air conditioning response behavior. Third, by combining the building's spatial topology, it connects each room node and its maintenance boundary into a heat transfer network, considering multipath heat propagation and the interactive coupling characteristics of heat between different areas, effectively characterizing the non-uniformity of spatial temperature distribution caused by differences in room layout, orientation, and open structure. Fourth, it uses the finite difference method or finite element method for numerical solution, combining initial temperature conditions with the influence of external boundaries such as solar radiation and outdoor temperature, accurately simulating the building's thermal behavior response process under dynamic boundary conditions, providing a reliable simulation basis for subsequent thermal buffer capacity assessment. This multiphysics coupling model characterizes the "thermal hysteresis" phenomenon of slow temperature change in the building after the air conditioning system stops operating, and simulates the heat exchange process generated by supply air, return air, and natural ventilation, thereby achieving dynamic prediction of heat propagation speed and temperature response sensitivity, providing theoretical support for subsequent scheduling and control.

[0030] Next, the thermal inertia characteristics of the building structure are modeled using the finite element method to obtain a thermal time constant distribution matrix that characterizes the thermal response speed of each space. The thermal time constant measures the building's response speed to heat input and is an important indicator for assessing the regional heating or cooling inertia. Due to significant differences in room structure, orientation, and materials, their thermal responses vary considerably, thus requiring representation in the form of a distribution matrix.

[0031] In the modeling process, the basic idea of ​​the finite element method is to divide the building structure into multiple finite elements. Within each element, a heat conduction model is established based on the extracted thermophysical parameters of the building envelope (including but not limited to thermal conductivity, specific heat capacity, density, etc.). The model comprehensively considers factors such as heat conduction, wall heat storage, air-layer convection heat transfer, and external boundary radiation. The heat conduction process satisfies unsteady-state control differential equations. A numerical solver is used to solve these unsteady-state heat conduction equations, obtaining temperature change curves for each element at different time steps. Furthermore, by fitting these temperature-time curves, the thermal time constant of each element can be extracted and organized into a distribution matrix. Compared to traditional energy-saving control methods that rely on empirical rules, this application uses finite element analysis to model thermal inertia characteristics, enabling accurate energy consumption prediction by zone, structure, and time period, thus improving the overall energy-saving effect and operational stability of the building's HVAC system.

[0032] Finally, a thermal inertial characteristic model for characterizing the overall thermal response of a building was constructed based on the aforementioned thermal time constant distribution matrix and multiphysics coupling model. Based on this model, the building's spatial topology was further integrated to construct a building thermal response simulation model. Multi-field coupling simulations of the heat conduction process were performed under typical boundary conditions to obtain simulation results of temperature changes in different regions. These simulation results were then fitted with residuals to thermal response time-series data, dynamically correcting local thermophysical parameters in the model. This resulted in a highly dynamic and adaptable thermal inertial characteristic model that accurately predicts building temperature response trends, serving as the core foundation for subsequent air conditioning start-stop strategy generation and energy-saving control algorithm optimization. This model not only reveals the building's heat conduction and heat storage mechanisms but also accurately describes the dynamic characteristics of the thermal response, contributing to the optimization of time-based air conditioning power scheduling and the refined control of energy consumption prediction.

[0033] S2: Collect and process indoor and outdoor temperature data of the target building over multiple time periods to generate a thermal response time series dataset.

[0034] It should be noted that the time period can be set to different lengths, such as daily, weekly, monthly, quarterly, or annually. By collecting and processing indoor and outdoor temperature data within these different time periods, the thermal response behavior of the building under different external heat loads, climatic conditions, and usage states can be fully obtained, thus ensuring the comprehensiveness and representativeness of the constructed thermal response time-series dataset. For example, in the hot summer, multiple time periods can be preferably set to noon to evening (e.g., 12:00 to 18:00) every day, when solar radiation is strong, the external ambient temperature is high, and the building's heat load reaches its peak, which helps to capture the upper limit of the heat buffering effect. In the cold winter, it can be set to early morning to night (e.g., 6:00 to 22:00) every day, during which the indoor and outdoor temperature difference is large, and the building's thermal insulation performance has a significant impact on room temperature changes. By collecting and analyzing temperature data in different seasons and typical time periods, the thermal response patterns of the building under various climatic loads can be further refined, providing support for the multi-scenario adaptability of air conditioning power scheduling strategies.

[0035] It should also be noted that indoor and outdoor temperature data can be obtained through temperature sensor systems deployed both inside and outside the target building. Indoor temperature sensors can be placed in the center of a typical room, near exterior walls or windows, and near air supply and return vents. Outdoor temperature sensors can be installed on the sunken and shaded sides of the building facade, the roof, etc. The collected raw temperature data is then cleaned, normalized, and time-series aligned to remove outliers and standardize the time reference.

[0036] Based on the cleaned data, temperature change curves before and after changes in the operating status of the air conditioning system are extracted to construct a "input (e.g., air supply / shutdown) - output (room temperature change)" response sample. Then, each set of air conditioning control behaviors and the corresponding temperature response curves are arranged along the time axis to form a thermal response time series dataset. This time series dataset contains multiple samples of building thermal behavior under different time periods, different external environments, and different air conditioning operating conditions within a time period. These samples serve as inputs for subsequently establishing a building thermal buffer capacity prediction function, used to explore the hysteresis characteristics, stability, and differences in heat transfer rates between different areas of the building's thermal response.

[0037] S3: Based on the aforementioned thermal response time series dataset and combined with the building thermal inertia characteristic model, establish a prediction function for the building's thermal buffering capacity.

[0038] Specifically, the method for establishing the building thermal buffering capacity prediction function includes: Fourier transform is performed on the thermal response time series dataset to extract phase lag features that reflect the thermal response delay characteristics; Based on the phase lag characteristics and the historical temperature sequences of the target building collected over multiple time periods, a room temperature change rate prediction model is constructed using a long short-term memory neural network. By combining the thermal resistance parameters in the building's thermal inertia characteristic model, a nonlinear mapping function is constructed with outdoor temperature as input and indoor temperature change gradient as output, forming the final thermal buffer capacity prediction function. The thermal resistance parameters are calculated from the thermal conductivity of the building envelope materials and the thickness of its structural layers. They are important derived indicators of building structural information and envelope material parameters, used to quantify the heat conduction capacity in each structural unit.

[0039] In some specific implementations, a Fourier transform is performed on the thermal response time-series dataset to extract phase lag features reflecting the thermal response delay characteristics. The specific steps are as follows: First, indoor and outdoor temperature time-series data are collected and organized to form a thermal response sequence, providing basic data for subsequent frequency domain analysis and ensuring the periodicity and comparability of the thermal response behavior. Second, a Fast Fourier Transform is performed on the outdoor and indoor temperature change sequences to calculate their frequency domain amplitude and phase spectra. The phase difference corresponding to the dominant frequency point (usually the daily dominant frequency) is extracted; this phase difference is the thermal lag response. This step transforms the temperature response from the time domain to the frequency domain, effectively revealing the lag response behavior of the building structure to thermal disturbances. Third, the dominant frequency response phase differences of all spatial units are summarized to form a spatially distributed thermal response phase lag feature vector. By obtaining the phase lag features, the degree of lag response of the building structure to changes in heat load can be revealed. This is an important dynamic indicator for measuring thermal buffering capacity, reflecting the heat storage effect and response delay.

[0040] In some specific implementations, the steps for constructing a room temperature change rate prediction model using a Long Short-Term Memory (LSTM) neural network are as follows: First, select historical outdoor temperature sequences, current indoor temperature values, and corresponding phase lag features of the target building over multiple time periods to construct the neural network input dataset, establishing the foundation for modeling. Second, set the prediction target as the room temperature change rate or temperature increment within a future set time window, using this as the network's output label set. This step is guided by temperature change trends, more closely aligning with the actual temperature control prediction targets required for air conditioning regulation. Third, use the input dataset → output label set as training samples, inputting them into a multi-layer LSTM network structure to train the network parameters to minimize prediction errors. After training, a prediction model that can be used to predict the real-time trend of building room temperature changes is obtained. This model can effectively capture the dynamic influence of thermal inertia characteristics on future temperature responses, improving prediction accuracy.

[0041] In some specific implementations, the detailed steps for constructing a nonlinear mapping function with outdoor temperature as input and indoor temperature change gradient as output, based on the thermal resistance parameters in the building thermal inertia characteristic model, are as follows: First, extract the thermal resistance parameters from various building envelopes according to the building thermal inertia characteristic model. ,in For structural thickness, For thermal conductivity, in this step, the resistance parameter is used as an important indicator of the thermal conduction performance of the structure, providing a physical basis for establishing thermal buffering behavior and quantifying the buffering capacity of different structures to temperature changes; secondly, the thermal resistance parameter, phase lag characteristics, and outdoor temperature sequence over a future period are used as input variables, and a mapping function is constructed using a regression neural network or hybrid modeling method to predict the rate of change of indoor temperature or its gradient.

[0042] Compared to traditional predictive modeling methods based on thermal time constants, this invention further introduces frequency domain analysis. By performing Fourier transform on the thermal response data, phase lag is extracted as a dominant feature, and this is combined with an LSTM model to enhance predictive capabilities. Furthermore, by incorporating thermal resistance parameters, a nonlinear mapping function between outdoor and indoor temperature gradients is constructed, improving the model's ability to express the structural thermal conduction characteristics. The thermal buffering capacity prediction function established in this way possesses thermal conduction path awareness capabilities, which is beneficial for improving the foresight and control accuracy of air conditioning system operation scheduling.

[0043] Furthermore, the expression for the building thermal buffering capacity prediction function is as follows:

[0044] in, Indicates the predicted time point The indoor temperature; This is a nonlinear mapping function obtained by training a long short-term memory neural network, which reflects the system's response to input variables; The current indoor temperature; The outdoor temperature at the current moment; For air conditioning operation control vectors, such as control commands for air supply volume, operating status, valve opening, etc.; The phase lag characteristic function is obtained based on the Fourier transform, with the current indoor temperature as the reference. Outdoor temperature Air conditioning control quantity The input variable is used to extract the frequency domain features of the thermal response delay; The thermal resistance matrix of a building structure is composed of the thermal resistance parameters of each structural unit of the building. It reflects the ease or difficulty of heat propagation in the building or is used to characterize the influence of the building envelope on heat conduction.

[0045] For example, suppose a building is operating as follows at 10:00 AM: Current indoor temperature: Current outdoor temperature: ; Current air conditioning control quantity Air supply volume is 800m³ 3 / h, the air conditioner is in the "on" state; Thermal resistance matrix of building structure : Calculated based on the thermal conductivity and thickness of materials such as walls and windows in each room, such as:

[0046] This indicates the distribution of thermal resistance values ​​in three areas of the building.

[0047] Then, Fourier transform was used to perform frequency domain analysis on the air conditioning control input and temperature response data over a week, and the following phase lag feature vector (simplified representation) was extracted:

[0048] This vector represents the degree of response delay of the system on the main frequency components, and can quantify the thermal hysteresis of the building.

[0049] Next, the aforementioned phase lag eigenvectors Compared with the current building thermal resistance matrix These are then fed into a pre-trained long short-term memory neural network model. The model has been trained on a similar building dataset and can fit complex nonlinear thermal response relationships.

[0050] Finally, the LSTM model outputs the result 15 minutes later (i.e., Predicted indoor temperature (minutes): The predicted value indicates that even without adjusting the current air supply volume, the system is expected to reach a room temperature of 25.9°C in 15 minutes, which helps to determine in advance whether it is necessary to turn off the air conditioner, reduce the air supply power, or continue to maintain the current operating status.

[0051] In summary, step S3, by combining the building's thermal inertia characteristic model with the thermal response time-series dataset, establishes a predictive function that dynamically reflects the building's thermal buffering capacity, providing a high-precision decision-making basis for intelligent control of the air conditioning system. Furthermore, during the establishment of the predictive function, thermal resistance parameters calculated from building structural information are introduced and input into the regression model along with phase lag characteristics and future external temperature sequences, constructing a nonlinear mapping function between the gradient of outdoor and indoor temperature changes. This mapping method can identify the moderating effect of various building envelope structures on heat flow, achieving direct prediction from "external thermal disturbance" to "room temperature change trend," significantly enhancing the adaptability of the control model to differences in building structures.

[0052] S4: Obtain the electricity price forecast data curve for a future preset time period, wherein the electricity price forecast data curve includes electricity price fluctuation information corresponding to different time points.

[0053] It should be noted that the electricity price forecast data curve includes the predicted electricity value for each time point. The electricity price forecast data curve is obtained as follows: Electricity price forecast data can be obtained in real time or periodically by accessing the API of the electricity market or electricity operation platform, that is, by connecting to the data interface of the local electricity operator. By combining regional electricity policies and electricity consumption behavior analysis, that is, based on the time-based electricity pricing strategy (such as the summer peak / off-peak electricity pricing system) of the target building's location and the building's past electricity consumption characteristics, a price prediction curve for future periods is generated; Electricity price forecasting models can be introduced to obtain data when authoritative platform data is unavailable. These models can be based on machine learning (such as time series models ARIMA, LSTM, or ensemble regression models) and utilize historical electricity price data, electricity load data, weather and holiday information to generate electricity price forecast curves.

[0054] The electricity price forecast data curve is a time series function, describing the electricity price at each point in time (e.g., every 15 minutes, every hour) within a future period. Its form is as follows:

[0055] in, Indicates a future point in time. This indicates the predicted electricity price for that time period (unit: yuan / kWh).

[0056] It should be noted that by obtaining the electricity price forecast data curve for a future preset time period, the system can identify the off-peak electricity price period (such as nighttime or off-peak hours). The system can then prioritize the operation of air conditioners during periods of lower electricity prices, thereby reducing overall electricity costs.

[0057] S5: Based on the building heat buffer capacity prediction function and the electricity price prediction data curve, generate a priority strategy for air conditioning power scheduling in the building in the future time period.

[0058] Specifically, the determination of the air conditioning operating power scheduling priority strategy includes: S51: Based on the electricity price prediction data curve, multiple electricity price ranges are divided, specifically including: a high electricity price range, a flat electricity price range, and a low electricity price range; wherein, the electricity price in the high electricity price range is higher than a set upper threshold. The electricity price in the flat electricity price range is located at and Between; the electricity price in the low-price range is lower than the set lower threshold. .

[0059] It should be noted that the above thresholds can be set based on historical statistical percentiles (such as the upper 10% and lower 10%) or dynamic pricing strategies published by the demand response platform.

[0060] S52: In high electricity price ranges, based on the output of the building thermal buffer capacity prediction function, determine the building's future short-term time window. If the system has the ability to maintain room temperature, and the result indicates that it does, a power reduction strategy is implemented, either by lowering the compressor's operating power or increasing the supply air temperature. Furthermore, the determination of having the ability to maintain room temperature satisfies any of the following conditions: Predict room temperature Still within the set comfortable temperature range ; The current actual temperature is either lower than the user-set target lower temperature limit or higher than the upper temperature limit, with the upper temperature limit being used for heating. The current thermal time constant is greater than the set hysteresis threshold, indicating that the temperature drop / rise changes slowly and has a strong thermal buffering capacity.

[0061] S53: During periods of low electricity prices, implement an advance energy storage strategy and record the air conditioning operating power, room temperature change trajectory, and cumulative energy storage duration during the energy storage phase. Generate a power compensation scheduling table. The power compensation scheduling table is used to dynamically guide the power reduction magnitude, duration, and regional distribution during subsequent periods of high electricity prices to ensure a smooth transition in system operation.

[0062] Implementing a pre-emptive energy storage strategy can increase the building's thermal buffer space, thereby reducing the operating load during subsequent periods of high electricity prices. Pre-emptive energy storage strategies include: Prediction function based on building thermal buffer capacity The system predicts whether room temperature may exceed the user's set temperature target range during periods of high electricity prices in the future; if the prediction results indicate that the building's thermal buffer capacity is insufficient to support the target temperature range, it will trigger an advance energy storage operation. Once the pre-storage operation is triggered, the system will execute the following control measures: During the period from the current moment to the beginning of the high electricity price range, the air supply volume, air supply temperature and compressor operating power are dynamically adjusted to ensure that the room temperature reaches the user's target temperature boundary before the arrival of high electricity prices. Meanwhile, during this energy storage control phase, the system continuously records the air conditioning operating power, room temperature change trajectory, and cumulative energy storage time, and generates a power compensation scheduling table based on the above data. This scheduling table is used to guide energy-saving control strategies in high electricity price ranges.

[0063] In summary, the priority strategies for air conditioning power scheduling include: During periods of high electricity prices, a power reduction strategy is implemented to reduce energy consumption by utilizing the building's thermal buffering capacity. During periods of low electricity prices, an advance energy storage strategy is implemented, and a power compensation scheduling table is generated based on the energy storage process records.

[0064] S6: Based on the scheduling priority strategy and the indoor temperature target range set by the user, calculate the optimal start / stop control interval and air supply parameters for each control period.

[0065] Specifically, the calculation method for the optimal start-stop control range and air supply parameters includes: The upper and lower limits of the user-defined indoor temperature target range are used as comfort boundary conditions, serving as the room temperature control constraint range. Based on the building thermal buffer capacity prediction function, the room temperature change trend under the condition of no active air conditioning control is derived in reverse, and the predicted time point when the room temperature is about to reach the boundary condition is determined as the critical point of temperature change. Combining the current building's thermal inertia state parameters and external load conditions, the Model Predictive Control (MPC) algorithm is used to solve for the longest sustainable shutdown period for cooling or heating, while ensuring that the room temperature is kept within a comfortable range, and to determine the optimal start-stop control interval. Based on the optimal start-stop control range, air supply parameters such as air volume, air supply temperature, and wind speed are further calculated to control the rate of change in room temperature, achieve the expected energy-saving effect, and improve the efficiency of heat buffer utilization.

[0066] In some specific implementations, the step of determining the predicted time point when the room temperature is about to reach the boundary condition is as follows: based on the thermal buffer capacity prediction function. The simulation predicts the trend of room temperature changes over a future period; during the simulation, the prediction step size is gradually increased to monitor whether the room temperature will change in the future. When the predicted value approaches or reaches the lower or upper temperature limit, record this moment once the predicted value exceeds the temperature boundary. As the critical point of temperature change.

[0067] In some specific implementations, the step of determining the optimal start-stop control interval is as follows: [The text abruptly shifts to a different topic] ...the predicted critical time point... As the upper limit of regulatory constraints, combined with the current moment Constructing the prediction time domain Then, a model predictive control (MPC) algorithm is introduced in the prediction time domain. The system uses maintaining room temperature within a comfortable range as a constraint and minimizing compressor start-stop frequency or energy consumption as the optimization objective to solve for the optimal shutdown time for cooling or heating. The final output is the start-stop time interval, which is the currently acceptable optimal shutdown control interval, used for subsequent calculation of air supply parameters.

[0068] In some specific implementations, the calculation of the "air supply parameters" can be based on the heat balance method, that is, based on the temperature difference between the predicted room temperature and the target temperature, the room volume, the air heat capacity parameters, etc., the heat that needs to be removed or input is estimated; and based on the air supply time window and the set air supply temperature, the required air supply flow rate, wind speed and air supply temperature parameters are derived.

[0069] For example: For example: An office building is operating its central air conditioning system during the summer. The user has set the comfort temperature range to 24℃-26℃, and the current room temperature is 24.2℃.

[0070] First, determine the "optimal start-stop control interval". The determination process is as follows: Based on hot buffer capacity prediction function The study used stepwise simulation to predict the temperature change trend after the air conditioner was turned off. The results show that, under the current condition of turning off the air conditioner, the building's internal thermal buffer can support the room temperature within a comfortable range until the predicted room temperature reaches the critical point of 26℃ at 14:45.

[0071] Therefore, the system determines the predicted time point when the room temperature reaches the boundary condition as 14:45.

[0072] Next, the optimal start-stop control interval is determined: Given that the current time is 14:00 and the critical point is 14:45, the system adopts the Model Predictive Control (MPC) algorithm. Within the prediction time domain of 14:00-14:45, with the constraint that "the room temperature must not exceed 26℃" and the optimization objective of minimizing air conditioning energy consumption, the optimal start-stop control interval is calculated to be: 14:00-14:45 (the acceptable shutdown cooling period).

[0073] Next, the air supply parameters are calculated. The calculation process is as follows: To ensure that the air conditioner can quickly lower the room temperature from 26°C to the lower limit of the comfort range of 24°C after resuming operation at 14:45, the system calculates the following parameters based on the thermal inertia model and the principle of energy conservation: room temperature difference ; Assuming the target cools down within 15 minutes (i.e. ); Assume the room volume V = 300m³ 3 Specific heat capacity of air at constant pressure =1005 J (kg·k); air density =1.2kg / m 3 air supply temperature =18℃.

[0074] The required heat to be removed is:

[0075] The air supply volume meets the energy conservation condition:

[0076] Substitute the parameters to obtain the air supply flow rate. :

[0077] Wind speed is calculated based on the cross-sectional area of ​​the air outlet. If the air outlet is 0.4m... 2 Then the wind speed = / 3600·Cross-sectional area= / 3600·0.4≈0.775m / s.

[0078] S7: Within the optimal start-stop control range, regulate the operating status of the air conditioning system. Regulation includes: dynamically adjusting the compressor power, fan speed and start-stop status based on the generated air supply parameters and control strategy to achieve energy-saving control objectives.

[0079] It should be noted that the control strategy refers to generating control signals to adjust the operating power, air supply intensity, and start-stop rhythm of the air conditioner based on the electricity price level, heat buffer capacity prediction results, and current environmental conditions in the scheduling priority strategy.

[0080] In this step, within the optimal start-stop control range, the system uses the calculated air supply parameters and control strategies (including power reduction strategies, energy storage strategies, etc.) as the basis for execution, and jointly adjusts the operating status of key components in the air conditioning system, such as compressor power adjustment, fan speed control, and start-stop switching, so as to maximize the building's thermal buffering capacity while ensuring indoor temperature comfort, thereby achieving the goal of energy saving and consumption reduction.

[0081] For example: For example, on a summer weekday morning, the system predicts that electricity prices will peak between 2:00 PM and 5:00 PM. Based on the heat buffer capacity prediction function and the electricity price curve, the system determines that the current period is a low-electricity-price energy storage phase and identifies the current time period (1:15 PM to 2:00 PM) as the optimal operating range for air conditioning.

[0082] Based on the scheduling priority strategy and the user-defined comfortable temperature range (24℃–26℃), the system adopts the following control strategy from 13:15 to 14:00: Compressor power regulation: The system dynamically sets the compressor operating frequency based on the indoor and outdoor temperature difference, the heat buffer prediction curve, and the target cooling rate. For example, it initially operates at 70% of the rated power, and automatically adjusts to 85% after 10 minutes based on the heat removal rate to improve cold storage efficiency. Fan speed adjustment: Set the fan speed to 0.75 m / s, corresponding to an air volume of approximately 600 m³ / s. 3 / h), to ensure uniform airflow distribution and avoid discomfort from direct airflow; Supply air temperature control: Maintain the supply air temperature at 18℃ to enhance the heat exchange efficiency per unit supply air volume; Start-stop logic control: The system is preset to shut down the compressor at 14:00, but the fan will continue to run at low speed for 10 minutes to release the residual cooling capacity of the refrigerant in the pipeline, thereby further extending the comfort maintenance time.

[0083] In summary, this invention, by acquiring building structural information and envelope material parameters, and combining finite element analysis with a multiphysics thermal model, constructs a complete building thermal inertia characteristic model that can realistically reflect the heat conduction and heat storage behavior of buildings under unsteady-state conditions. Furthermore, by combining historical indoor and outdoor temperature data to construct a thermal response time-series dataset, the thermal response patterns of different building areas under different external load conditions can be identified, thereby improving the dynamic accuracy and foresight of the thermal buffer capacity prediction function. Based on the established building thermal buffer capacity prediction function, this invention introduces future electricity price prediction data to construct a multi-time period, multi-scenario air conditioning power scheduling priority strategy, which can flexibly adjust the start-up and shutdown rhythm and operating intensity according to electricity price fluctuations. Especially during peak electricity price periods, the system can determine whether a building has thermal buffer capacity and proactively adopt power reduction or pre-cooling / pre-heating strategies to effectively reduce air conditioning operating power during high-price periods, lower energy costs, and achieve the goals of on-demand scheduling, load transfer, and optimal energy efficiency. This invention calculates the longest sustainable downtime within the optimal start-stop control range, combining user-defined comfort temperature ranges, the building's current thermal inertia, and model predictive control algorithms. Simultaneously, it precisely sets air supply parameters such as air volume, air temperature, and wind speed, achieving fine control over the rate of room temperature change. This method significantly reduces the frequency of air conditioner start-stop cycles, extends equipment lifespan, and achieves energy conservation and emission reduction while ensuring indoor thermal comfort.

[0084] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for energy-saving control of building heating, ventilation, and air conditioning systems, characterized in that, Includes the following steps: Obtain building structural information and building envelope material parameters to construct a building thermal inertia characteristic model; Collect and process indoor and outdoor temperature data of the target building over multiple time periods to generate a thermal response time series dataset; Based on the aforementioned thermal response time series dataset, and combined with the building thermal inertia characteristic model, a prediction function for building thermal buffering capacity is established. Obtain electricity price forecast data curves for a future preset time period, wherein the electricity price forecast data curves include electricity price fluctuation information corresponding to different time points; Based on the building thermal buffer capacity prediction function and the electricity price prediction data curve, a priority strategy for scheduling the building's air conditioning power in the future time period is generated. Based on the aforementioned scheduling priority strategy and combined with the user-defined indoor temperature target range, the optimal start / stop control interval and air supply parameters for each control period are calculated. Within the optimal start-stop control range, the operating status of the air conditioning system is regulated.

2. The energy-saving control method for building HVAC as described in claim 1, characterized in that: Constructing the building thermal inertia characteristic model includes the following steps: Extract the heat transfer coefficient, heat capacity parameters, and building spatial topology of the building envelope; Based on heat transfer coefficient, heat capacity parameters and building space topology, a multi-physics field coupling model is established that includes wall heat storage effect and air layer convection heat transfer. The thermal inertial characteristics of the building structure are modeled using the finite element method to obtain the thermal time constant distribution matrix; The thermal inertial characteristic model is constructed based on the thermal time constant distribution matrix and the multiphysics coupling model.

3. The energy-saving control method for building HVAC as described in claim 1, characterized in that: The method for establishing the building thermal buffering capacity prediction function includes: Fourier transform is performed on the thermal response time series dataset to extract phase lag features that reflect the thermal response delay characteristics; Based on the phase lag characteristics and the historical temperature sequences of the target building collected over multiple time periods, a room temperature change rate prediction model is constructed using a long short-term memory neural network. By combining the thermal resistance parameters in the building thermal inertia characteristic model, a nonlinear mapping function is constructed with outdoor temperature as input and indoor temperature change gradient as output, forming the final thermal buffer capacity prediction function.

4. The energy-saving control method for building HVAC as described in claim 3, characterized in that: The thermal resistance parameter is calculated from the thermal conductivity of the building envelope material and the thickness of its structural layer.

5. The energy-saving control method for building HVAC as described in claim 3, characterized in that: The expression for the building thermal buffering capacity prediction function is as follows: in: Indicates the predicted time point The indoor temperature; This is a nonlinear mapping function obtained by training a long short-term memory neural network, which reflects the system's response to input variables; The current indoor temperature; The outdoor temperature at the current moment; This is the air conditioning operation control vector; The phase lag characteristic function is obtained based on the Fourier transform, with the current indoor temperature as the reference. Outdoor temperature Air conditioning control quantity The input variable is used to extract the frequency domain features of the thermal response delay; This is the thermal resistance matrix of the building structure, which is composed of the thermal resistance parameters of each structural unit of the building.

6. The energy-saving control method for building HVAC as described in claim 1, characterized in that: The determination of the air conditioning power scheduling priority strategy includes: Based on the electricity price forecast data curve, multiple electricity price ranges are divided, specifically including: high electricity price range, flat electricity price range, and low electricity price range. In high electricity price ranges, the output of the building thermal buffer capacity prediction function is used to determine the building's future short-term time window. If the room has the ability to maintain room temperature, and the result is yes, then the power reduction strategy is implemented, which reduces the compressor operating power or increases the air supply temperature. During periods of low electricity prices, an advance energy storage strategy is implemented, and the operating power of the air conditioners, the trajectory of room temperature changes, and the cumulative energy storage duration during the energy storage phase are recorded to generate a power compensation scheduling table. This power compensation scheduling table is used to dynamically guide the magnitude, duration, and regional distribution of power reductions during subsequent periods of high electricity prices.

7. The energy-saving control method for building HVAC as described in claim 6, characterized in that: The determination of having the ability to maintain room temperature satisfies any of the following conditions: Predict room temperature Still within the set comfortable temperature range ; The current actual temperature is either lower than the user-set target lower temperature limit or higher than the upper temperature limit, with the upper temperature limit being used for heating. The current thermal time constant is greater than the set hysteresis threshold, indicating that the temperature drop / rise changes slowly and has a strong thermal buffering capacity.

8. The energy-saving control method for building HVAC as described in claim 6, characterized in that: The advance energy storage strategy includes: Prediction function based on building thermal buffer capacity The system predicts whether room temperature may exceed the user's set temperature target range during periods of high electricity prices in the future. If the prediction indicates that the building's thermal buffer capacity is insufficient to support the target temperature range, it triggers advance energy storage. Once the advance energy storage is triggered, the system executes the following control measures: During the period from the current moment to the beginning of the high electricity price range, the air supply volume, air supply temperature and compressor operating power are dynamically adjusted to ensure that the room temperature reaches the user's target temperature boundary before the arrival of high electricity prices. Meanwhile, during this energy storage control phase, the system continuously records the air conditioning operating power, room temperature change trajectory, and cumulative energy storage time, and generates a power compensation scheduling table based on the above data. This scheduling table is used to guide energy-saving control strategies in high electricity price ranges.

9. The energy-saving control method for building HVAC as described in claim 1, characterized in that: The regulation includes: the calculation method for the optimal start-stop control range and air supply parameters includes: The upper and lower limits of the user-defined indoor temperature target range are used as comfort boundary conditions, serving as the room temperature control constraint range. Based on the building thermal buffer capacity prediction function, the room temperature change trend under the condition of no active air conditioning control is derived in reverse, and the predicted time point when the room temperature is about to reach the boundary condition is determined as the critical point of temperature change. Combining the current building's thermal inertia state parameters and external load conditions, a model predictive control algorithm is used to solve for the longest sustainable shutdown period for cooling or heating, while ensuring that the room temperature is kept within a comfortable range, and to determine the optimal start-stop control interval. Based on the optimal start-stop control range, the air supply parameters are further calculated to control the rate of change in room temperature.

10. The energy-saving control method for building HVAC as described in claim 1, characterized in that: Within the optimal start-stop control range, the operating status of the air conditioning system is regulated, wherein the regulation includes dynamically adjusting the compressor power, fan speed and start-stop status based on the generated air supply parameters and control strategy.

Images