Air conditioning control method fusing outdoor temperature feedforward and indoor heat source distribution perception

Abstract

The present application relates to air conditioner control technical field, especially to a kind of air conditioner control method of fusing outdoor temperature feedforward and indoor heat source distribution perception, the method includes the following steps: based on building outer wall temperature sensor continuous acquisition external environment temperature, carry out within day time period temperature variation prediction, obtain environment temperature prediction curve;Based on environment temperature prediction curve, carry out refrigeration power demand calculation, drive indoor air conditioner to complete initial refrigeration processing;Based on indoor infrared sensor, carry out indoor personnel real-time perception, and construct personnel moving track distribution data;Personnel moving track distribution data are fitted with heat source intensity spatial distribution, and construct indoor heat source distribution diagram;Based on indoor heat source distribution diagram, carry out regional heat load deviation evaluation, obtain the refrigeration deviation of different regions.The present application effectively improves indoor temperature regulation precision, comfort balance and energy utilization efficiency, while reducing building operation energy consumption and operation and maintenance cost.

Description

Technical Field

[0001] This invention relates to the field of air conditioning control technology, and in particular to an air conditioning control method that integrates outdoor temperature feedforward and indoor heat source distribution sensing. Background Technology

[0002] With the continuous increase in demand for building energy management and the ever-improving requirements for indoor environmental comfort, intelligent control of air conditioning systems has become an important research direction in the field of building energy conservation. Traditional air conditioning control systems generally adopt a feedback regulation strategy based on indoor temperature sensors. That is, a cooling command is triggered when the indoor temperature exceeds a preset threshold, driving the compressor to start and deliver cooling capacity into the room. This control method is a typical passive response mechanism. The system can only compensate and adjust after the temperature deviation has already occurred, resulting in a significant lag in indoor temperature during the rapid increase of heat load, making it difficult to maintain a stable thermal comfort environment.

[0003] Existing air conditioning control systems typically only collect the current outdoor temperature value as an auxiliary reference, lacking the ability to continuously track and predict outdoor temperature trends. When the outdoor ambient temperature continues to rise due to factors such as solar radiation and seasonal changes, the heat transfer of the building envelope increases accordingly, and the indoor heat load accumulates rapidly in a short period of time. Because traditional control systems cannot anticipate the upcoming increase in heat load, the compressor's cooling capacity configuration often lags behind actual demand, causing periodic overshooting of indoor temperature, making it difficult to effectively coordinate the contradiction between energy consumption and comfort. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes an air conditioning control method that integrates outdoor temperature feedforward and indoor heat source distribution sensing, thereby solving at least one of the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides an air conditioning control method that integrates outdoor temperature feedforward and indoor heat source distribution sensing, comprising the following steps:

[0006] Step S1: Based on the continuous collection of external ambient temperature by the building exterior wall temperature sensor, perform intraday temperature change prediction to obtain the ambient temperature prediction curve;

[0007] Step S2: Calculate the cooling power demand based on the ambient temperature prediction curve and drive the indoor air conditioner to complete the initial cooling process;

[0008] Step S3: Real-time perception of indoor personnel based on indoor infrared sensors to construct personnel movement trajectory distribution data;

[0009] Step S4: Fit the spatial distribution of heat source intensity to the personnel movement trajectory distribution data to construct an indoor heat source distribution map;

[0010] Step S5: Based on the indoor heat source distribution map, assess the regional heat load deviation to obtain the cooling deviation in different regions;

[0011] Step S6: Adjust the cooling parameters of each air outlet differently based on the cooling deviation, and send the adjustment to the execution unit of each air outlet to execute indoor sensing control.

[0012] In this invention, step S1 specifically involves the following steps:

[0013] Based on the continuous collection of external ambient temperature by building exterior wall temperature sensors, a real-time outdoor temperature sequence is constructed.

[0014] Anomaly filtering is performed on the outdoor real-time temperature sequence to obtain a standard temperature sequence;

[0015] A sliding window statistical analysis is performed on a standard temperature sequence to generate a temperature change feature sequence; the temperature change feature sequence includes temperature change rate, fluctuation amplitude, and trend slope characteristics.

[0016] The temperature change feature sequence is segmented and extracted, and intraday temperature change prediction is performed to obtain the ambient temperature prediction curve.

[0017] In this invention, step S2 specifically involves the following steps:

[0018] Set the indoor comfort temperature value; estimate the initial indoor heat load based on the ambient temperature prediction curve to obtain the original indoor heat load parameters;

[0019] The basic cooling power is obtained by calculating the cooling power requirement based on the original indoor heat load parameters according to the indoor comfort temperature value.

[0020] The operating power of the air conditioner compressor is set based on the basic cooling power to drive the indoor air conditioner to complete the initial cooling process.

[0021] In this invention, step S3 specifically involves the following steps:

[0022] Real-time detection of indoor occupants is achieved using indoor infrared sensors, and all indoor occupants are identified.

[0023] The spatial coordinates of the people inside the room are obtained by calculating their positions.

[0024] Continuous inter-frame difference calculation is performed on the spatial coordinates of personnel to generate a dynamic position change sequence;

[0025] Based on the dynamic change sequence of location, the distribution of personnel movement trajectories is analyzed to construct personnel movement trajectory distribution data.

[0026] In this invention, step S4 specifically involves the following steps:

[0027] Set the dynamic and static human body heat dissipation power calibration values;

[0028] Based on the dynamic and static human body heat dissipation power calibration values, the individual heat source intensity is assigned to the personnel movement trajectory distribution data, and spatial grid interpolation is performed to output the indoor heat source coordinate system.

[0029] Based on the indoor heat source coordinate system, the spatial distribution of heat source intensity is fitted to construct an indoor heat source distribution map.

[0030] In this invention, the dynamic and static human body heat dissipation power calibration values ​​are specifically: the dynamic heat dissipation power calibration value is 120W, and the static heat dissipation power calibration value is 80W.

[0031] In this invention, step S5 specifically involves the following steps:

[0032] The indoor heat source distribution map is divided into grids, and the heat source influence of each grid node is analyzed to obtain the expected temperature value of each node.

[0033] The expected temperature difference is calculated based on the indoor comfort temperature value to obtain the expected temperature difference value for all nodes.

[0034] Construct a temperature gradient distribution map based on the expected temperature difference; identify all indoor air outlets;

[0035] The heat load deviation of the corresponding area of ​​the air outlet is evaluated based on the temperature difference gradient distribution map to obtain the cooling deviation of different areas.

[0036] In this invention, step S6 specifically involves the following steps:

[0037] Based on the cooling deviation, the cooling parameters of each air outlet are adjusted differently to generate a set of cooling control parameters for each air outlet.

[0038] Generate packaging control instructions based on the aforementioned dynamic cooling regulation parameter set;

[0039] Based on the encapsulation control command, it is sent to the execution unit of each air outlet to perform indoor sensing control.

[0040] In this invention, the dynamic cooling control parameter set includes target supply air temperature, air volume level, and running time.

[0041] The specific benefits of this invention are as follows: By continuously collecting external ambient temperature data and combining it with intraday time-of-day characteristics for temperature change prediction, the trend of external temperature rise or fall can be predicted in advance. This transforms the air conditioning system from traditional passive response control to active predictive control, reducing the impact of sudden changes in ambient temperature on indoor comfort and improving the foresight and stability of air conditioning regulation. By pre-calculating the cooling power demand based on the ambient temperature prediction curve and driving the air conditioner to perform initial cooling, indoor temperature regulation can be synchronized with changes in external heat load, reducing frequent start-stop and large power adjustments of the air conditioner, improving cooling efficiency, and reducing energy consumption. By using infrared sensors to perceive the number, location, and activity trajectory of people indoors in real time, the spatial distribution and flow patterns of people can be accurately reflected, providing a reliable basis for indoor heat source identification and avoiding the perception blind spots caused by traditional air conditioning control based solely on fixed temperature measurement points. By spatially fitting the heat source intensity of people's movement trajectories and constructing a heat source distribution map, the indoor heat accumulation areas and heat source changes can be intuitively reflected, improving the accuracy of dynamic heat load identification and making air conditioning control more closely reflect the actual distribution of people and heat demand. By assessing regional heat load deviations based on heat source distribution maps, the difference between actual heat and target comfort levels in each region can be accurately identified. This enables quantitative analysis of cooling demands in different regions and improves the air conditioning system's ability to regulate localized high-temperature areas and uneven heating / cooling. By adjusting the cooling parameters of each air outlet according to the cooling deviation and issuing control commands to the corresponding execution units, zoned, directional, and precise airflow control can be achieved. This improves cooling efficiency in densely populated areas while reducing energy waste in low-load areas, thus enhancing overall comfort and energy efficiency. Attached Figure Description

[0042] Figure 1 This is a flowchart illustrating the steps of an air conditioning control method that integrates outdoor temperature feedforward and indoor heat source distribution sensing according to the present invention.

[0043] Figure 2 This is a detailed flowchart illustrating the implementation steps of step S1.

[0044] Figure 3 This is a detailed flowchart illustrating the implementation steps of step S2;

[0045] Figure 4 This is a schematic diagram illustrating the distribution of indoor heat sources. Detailed Implementation

[0046] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0047] In the embodiments of the present invention, see Figure 1This is a flowchart illustrating the steps of an air conditioning control method integrating outdoor temperature feedforward and indoor heat source distribution sensing according to the present invention. In this example, the steps of the air conditioning control method integrating outdoor temperature feedforward and indoor heat source distribution sensing include:

[0048] Step S1: Based on the continuous collection of external ambient temperature by the building exterior wall temperature sensor, perform intraday temperature change prediction to obtain the ambient temperature prediction curve;

[0049] Step S2: Calculate the cooling power demand based on the ambient temperature prediction curve and drive the indoor air conditioner to complete the initial cooling process;

[0050] Step S3: Real-time perception of indoor personnel based on indoor infrared sensors to construct personnel movement trajectory distribution data;

[0051] Step S4: Fit the spatial distribution of heat source intensity to the personnel movement trajectory distribution data to construct an indoor heat source distribution map;

[0052] Step S5: Based on the indoor heat source distribution map, assess the regional heat load deviation to obtain the cooling deviation in different regions;

[0053] Step S6: Adjust the cooling parameters of each air outlet differently based on the cooling deviation, and send the adjustment to the execution unit of each air outlet to execute indoor sensing control.

[0054] In this embodiment, multiple temperature sensors are deployed on the sun-facing, shaded, and side walls of the building's exterior to continuously collect outdoor ambient temperature data. The sampling period is set to 30 seconds, and each sampled data point is accompanied by a timestamp to construct a time series relationship. Data collected from multiple measuring points are fused at the same time point, for example, using a weighted average method to obtain a comprehensive ambient temperature value, in order to eliminate biases caused by direct sunlight or partial shading at a single measuring point. The fused temperature data is arranged in chronological order to form a real-time outdoor temperature sequence, while simultaneously recording the intraday time period and seasonal information, such as the warming phase from 8:00 AM to 12:00 PM in summer, the high-temperature maintenance phase in the afternoon, and the cooling phase at night. Trend analysis is performed on the temperature sequence, calculating the rate of temperature change and trend slope at adjacent time points, and comparing it with historical temperature change patterns for the same time period. Using a 30-minute forecast window, the system projects future ambient temperatures hourly. For example, if the current temperature is 32.5℃, based on the morning's continuous warming trend, the forecast is 33.8℃ in 30 minutes, 35.0℃ in 60 minutes, and 36.2℃ in 120 minutes. A seasonal correction factor is also introduced to adjust for the warming or cooling rates in different seasons, making the forecast curve more closely reflect actual environmental changes. The temperature values ​​at each forecast time point are connected chronologically to form a continuous ambient temperature forecast curve.

[0055] The indoor comfort temperature is set at 24℃, and the current average indoor temperature is obtained, for example, 26℃, forming a base temperature difference of 2℃. Based on this, and considering the predicted upward trend of future outdoor temperatures, the heat transfer increment of the building envelope is estimated. Factors such as the exterior wall area, the proportion of glass curtain walls, and the heat transfer coefficient are comprehensively considered to convert future external heat input into equivalent indoor heat load. The heat load within the next 120-minute time window is further accumulated. For example, if the outdoor temperature rises from 32.5℃ to 36.2℃, the corresponding external heat input increases continuously, thus indicating that additional cooling capacity is needed to offset the future heat increment. The current temperature difference load and the future heat increment load are superimposed to obtain the total cooling demand, which is then converted into a base cooling power parameter. Subsequently, based on this base cooling power, the compressor operating frequency, supply air temperature, and air volume configuration are adjusted so that the air conditioner enters pre-cooling mode before the indoor temperature rises significantly, allowing the indoor temperature to gradually converge towards a stable 24℃.

[0056] Infrared sensors were deployed on the indoor ceiling and in key areas to continuously monitor personnel activity. The sampling period was maintained at 30 seconds to ensure consistency with the outdoor temperature data collection timescale. The infrared sensors identified personnel presence based on differences in human infrared radiation and assigned a unique identifier to each person. When multiple personnel targets were detected, spatial fusion positioning was performed by combining the coverage areas of multiple sensors to determine the specific coordinates of each person indoors, and the location was mapped using a 1m×1m grid. The position coordinates of the same person at consecutive time points were then continuously recorded. For example, if person A passed through coordinates (8,5), (9,5), and (10,6) sequentially, a continuous movement path was formed by linking these points with timestamps. By calculating the position difference between adjacent time points, the movement distance and direction were obtained, and the states of stationary, slow movement, and rapid movement were categorized. A stationary state corresponds to a position change close to 0, a moving state corresponds to continuous position changes, and a high-frequency movement state corresponds to frequent directional changes. All continuous position change information of personnel was integrated in chronological order to form personnel movement trajectory distribution data.

[0057] Indoor heat sources are modeled and analyzed based on the distribution data of human movement trajectories. Human body heat dissipation power is calibrated, with 120W for dynamic states and 80W for static states. The activity status of individuals is determined based on changes in their trajectories, and a corresponding heat dissipation power value is assigned. For example, continuously moving individuals are marked as 120W heat sources, and stationary individuals as 80W heat sources. The indoor space is then divided into 1m×1m grids, and the heat source intensity of each individual is mapped to their respective grid node. Heat sources from multiple individuals within the same grid are accumulated. Considering the heat diffusion characteristics of the human body to the surrounding space, attenuation diffusion calculations are performed on adjacent grids, expanding the heat source distribution from discrete points to a continuous thermal field. Spatial interpolation is performed on all grid heat sources to form a continuous heat source intensity distribution result. Weighting is then applied to areas where individuals spend a long time, resulting in a more pronounced heat source aggregation effect in high-density activity areas. An indoor heat source distribution map is then constructed.

[0058] The heat source intensity of each grid node is superimposed with the initial indoor baseline temperature field to calculate the expected temperature for each node. For example, areas with high population density may reach 26℃ to 28℃, ordinary areas approximately 24.5℃ to 25.5℃, and unoccupied areas maintained at around 24℃. Then, using the comfortable indoor temperature of 24℃ as a benchmark, the temperature difference for each node is calculated to obtain the expected temperature difference value. Further analysis of temperature difference changes between adjacent nodes is conducted to construct a temperature difference gradient distribution structure, identifying high-temperature deviation areas and low-temperature areas. The difference in cooling demand is calculated based on the magnitude of the temperature difference in different areas; for example, high-temperature areas correspond to a 3℃ temperature difference, medium-temperature areas to a 1℃ to 2℃ temperature difference, and low-temperature areas close to 0℃. The average temperature difference of each area is used as a heat load deviation index to obtain the cooling deviation amount for different areas, which characterizes the degree of insufficient cooling in that area relative to the comfortable temperature.

[0059] A mapping relationship is established between air outlets and spatial grid areas to determine the main service area covered by each air outlet and to calculate the average cooling deviation for that area. For example, an area with a deviation of 3℃ corresponds to air outlet A, and an area with a deviation of 1℃ corresponds to air outlet B. Air outlet parameters are adjusted based on the cooling deviation, including target supply air temperature, airflow level, and operating time. High deviation areas correspond to target supply air temperatures of 20℃ to 22℃, high airflow levels, and longer operating times; medium deviation areas correspond to supply air temperatures of 23℃ to 24℃ and medium airflow; low deviation areas maintain a supply air temperature of 24℃ and low airflow. Subsequently, the parameters of each air outlet are structured and encapsulated to generate control commands, which are then sent to the execution units of each air outlet via a control link. Each air outlet independently executes adjustment actions according to the received control commands, achieving differentiated air supply control by region. This allows high heat load areas to obtain stronger cooling capacity, while low heat load areas maintain energy-saving operation, thereby realizing coordinated air conditioning control based on outdoor temperature feedforward and indoor heat source distribution sensing.

[0060] In one specific embodiment, multiple temperature sensors are installed on the building's exterior wall, with a sampling period of 30 seconds. Data from three measuring points at the same time are weighted and fused (0.4 / 0.3 / 0.3). For example, if the data at a certain time are 32.8℃, 32.3℃, and 32.6℃ respectively, the overall temperature is calculated as follows:

[0061] ;

[0062] Continuous sampling yielded the following temperature sequence: 32.6℃ → 33.2℃ → 34.1℃ → 35.0℃ → 36.0℃. The temperature rise rate was calculated based on the linear trend.

[0063] ;

[0064] By combining the time dimension to extrapolate future temperatures, the predicted values ​​are: 37.7℃ at 13:00 and 39.4℃ at 14:00, forming an ambient temperature prediction curve to characterize the trend of continuous increase in external heat load.

[0065] Let the indoor comfortable temperature be Current indoor temperature The basic temperature difference is:

[0066] ;

[0067] The indoor heat load model is defined as follows:

[0068] ;

[0069] In this invention, the building is an office building with medium thermal insulation performance. The building envelope adopts double-glazed windows and insulated walls. After conversion based on the building's thermal parameters, the equivalent overall heat transfer coefficient is approximately 120 W / ℃, and the building's thermal inertia compensation is preferably 300 W.

[0070] in , Substituting the current outside temperature of 32.6℃ into the calculation:

[0071] ;

[0072] Considering the impact of a predicted temperature increase of 2℃:

[0073] ;

[0074] Therefore, the total load is:

[0075] ;

[0076] Corresponding basic cooling capacity:

[0077] ;

[0078] The compressor operating power is set to approximately 1.3kW to achieve initial cooling pre-conditioning, causing the room temperature to converge towards 24℃.

[0079] Three individuals were identified using infrared sensors, with their states as follows: A is stationary, B is moving, and C is stationary, corresponding to heat dissipation power ratings of 80W, 120W, and 80W, respectively. The individuals' positions are A(8,5), B(12,9), and C(15,6). Trajectories were calculated using inter-frame difference; for example, B's displacement was 1.2m within 30s, while A's displacement was <0.1m.

[0080] As is known to those skilled in the art, the natural micro-movement amplitude of the human body in a completely static state (such as sitting upright at an office or standing while waiting) is usually no more than 0.1m / 30s, which is considered postural maintenance micro-movement and has minimal impact on the body's metabolic heat production rate, with heat dissipation power maintained at a basal resting metabolic level of approximately 80W. However, when the human body undergoes significant displacement behavior (such as walking or moving to another workstation), its walking speed is usually no less than 0.8m / s, and the corresponding displacement within a 30s sampling period usually exceeds 1.0m, significantly increasing the metabolic heat production rate and achieving a heat dissipation power of 120W or more. There is a certain transition range between these two typical states; that is, when a person is in a state of slight activity such as standing up, turning around, or retrieving objects over a short distance, the displacement within 30s is approximately 0.3m to 0.8m. In this embodiment, 0.5m / 30s is used as the threshold for distinguishing between dynamic and static states, which falls within the middle of the aforementioned transition range.

[0081] The preferred dynamic determination rule in this case is: movement > 0.5m / 30s is determined to be dynamic (120W), otherwise it is static (80W). A heat source set is formed: A=80W, B=120W, C=80W, and it is bound to coordinates to form trajectory distribution data.

[0082] The space is divided into 1m×1m grids, and heat source superposition calculations are performed on each grid. For example, if grid G(12,9) contains person B, then the direct heat source is 120W.

[0083] In thermodynamics, heat transfer in space follows Fourier's law of heat conduction and Newton's law of cooling. The intensity of heat dissipation from a heat source to the surrounding medium decreases with increasing distance, and this decrease is closely related to the properties of the heat transfer medium, the intensity of the heat source, and ambient airflow conditions. It is well known to those skilled in the art that, under typical indoor environmental conditions, air, as a heat transfer medium, has a thermal conductivity of approximately 0.026 W / (m·℃), and its convective heat transfer coefficient is typically between 5 and 25 W / (m·℃). 2The range is ·℃). On a spatial scale of 1 meter, the ratio of heat transferred from human body thermal radiation to adjacent areas via thermal convection is typically estimated to be between 20% and 40% using engineering simplification. This means the attenuation coefficient ranges from approximately 0.2 to 0.4. In this study, 0.3 is preferred as the neighborhood diffusion coefficient for calculation.

[0084] ;

[0085] Therefore, the total heat source is:

[0086] ;

[0087] The temperature mapping relationship is defined as follows:

[0088] ;

[0089] The temperature of the grid is then:

[0090] ;

[0091] Multiple grids are superimposed to form a continuous thermal field. For example, in high-density areas, 500W → 25.5℃ and 900W → 26.7℃ are used to construct an indoor heat source distribution map.

[0092] Using a comfortable temperature of 24℃ as a baseline, the temperature difference is calculated for each grid. For example, if area A is 26.7℃, area B is 25.5℃, and area C is 24.5℃, then the temperature difference is: A=2.7℃, B=1.5℃, and C=0.5℃.

[0093] The expression for the temperature gradient is:

[0094] ;

[0095] For example, the distance from A to B is 2m:

[0096] ;

[0097] The refrigeration deviation is defined as: ;

[0098] Therefore: Zone A requires 270W, Zone B requires 150W, and Zone C requires 50W, resulting in differences in regional cooling demand.

[0099] Map the air outlets O1 / O2 / O3 to areas A / B / C respectively. The target supply air temperature is calculated as follows:

[0100] ;

[0101] The results show that: Area A is 22.65℃, Area B is 23.25℃, and Area C is 23.75℃.

[0102] Airflow settings are divided according to deviation: Zone A 100%, Zone B 70%, Zone C 40%.

[0103] The runtime model is as follows:

[0104] ;

[0105] Results: Zone A 23.5 min, Zone B 17.5 min, Zone C 12.5 min.

[0106] The system encapsulates control commands and sends them to the execution units at each air outlet to achieve strong cooling in high heat load areas and energy-saving operation in low load areas, thereby stabilizing the indoor temperature to around 24°C and completing the control of outdoor feedforward and indoor heat source sensing.

[0107] In this embodiment, see Figure 2 The diagram below illustrates the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0108] Based on the continuous collection of external ambient temperature by building exterior wall temperature sensors, a real-time outdoor temperature sequence is constructed.

[0109] Anomaly filtering is performed on the outdoor real-time temperature sequence to obtain a standard temperature sequence;

[0110] A sliding window statistical analysis is performed on a standard temperature sequence to generate a temperature change feature sequence; the temperature change feature sequence includes temperature change rate, fluctuation amplitude, and trend slope characteristics.

[0111] The temperature change feature sequence is segmented and extracted, and intraday temperature change prediction is performed to obtain the ambient temperature prediction curve.

[0112] In this embodiment, multiple temperature sensors are deployed at different locations on the building's exterior walls, including the sun-facing side, the shaded side, and the side walls. Environmental temperature data is collected synchronously from multiple measuring points. Each temperature sensor collects temperature data at a fixed sampling period, for example, every 30 seconds, and simultaneously records the corresponding timestamp. The collected temperature data is arranged in chronological order, and the temperature values ​​obtained from multiple measuring points at the same time are fused to obtain the comprehensive environmental temperature value for that time. For example, if the sun-facing temperature is 33.2℃, the shaded side temperature is 31.8℃, and the side wall temperature is 32.5℃, the current comprehensive environmental temperature of 32.5℃ is obtained through a weighted average. The corresponding date, month, and season information are also recorded, and the current time period is marked, such as the warming period from 8:00 AM to 12:00 PM in summer or the cooling period from 3:00 PM to 6:00 PM. As temperature data is continuously collected, the comprehensive environmental temperature values ​​corresponding to each time point are associated and stored in chronological order, forming a continuously updated real-time outdoor temperature sequence.

[0113] The temperature change between adjacent sampling points is calculated to analyze whether the temperature change within a unit sampling period is within a reasonable range. For example, if the sampling period is 30 seconds, and the temperature changes instantaneously from 32.5℃ to 38.0℃, the 5.5℃ temperature change in a short time significantly exceeds the normal variation pattern of the natural environment, thus it is marked as an abnormal candidate data. An analysis window is then constructed using the 10 most recent sampling points, and the mean temperature, median temperature, and dispersion parameters within the window are calculated to analyze the deviation between the current temperature value and the overall distribution characteristics of the window. When the current temperature value exceeds a preset deviation threshold for the window's average value, it is determined to be abnormal temperature data. For identified abnormal values, compensation correction is performed using adjacent valid temperature data to maintain continuous and smooth temperature changes. For example, if the valid temperatures before and after an abnormal point are 32.4℃ and 32.8℃ respectively, a corrected temperature value of 32.6℃ is generated to replace the abnormal data. After processing all abnormal points, temperature data that conforms to the actual environmental variation pattern is retained, and the temperature sequence is reconstructed according to the original time sequence to obtain a standard temperature sequence, thereby reducing the impact of sensor jitter, local heat source interference, and communication anomalies on environmental temperature analysis.

[0114] A sliding window analysis method was used to extract the pattern of ambient temperature changes. Standard temperature data from the most recent 30 minutes were selected as the statistical window, and the window content was continuously updated over time. The temperature difference between the beginning and end of the window and the corresponding time interval were calculated to obtain the rate of temperature change. For example, if the window started at 31.0℃ and ended at 33.0℃, the temperature rose by 2.0℃ in 30 minutes, corresponding to a temperature rise rate of 4.0℃ per hour. Subsequently, the difference between the highest and lowest temperatures within the window was calculated to obtain the temperature fluctuation amplitude, which reflects the stability of the ambient temperature. A small fluctuation amplitude indicates a relatively stable ambient temperature, while a continuously increasing fluctuation amplitude indicates a faster rate of temperature change. Further trend fitting was performed on all temperature data within the window to calculate the slope of the temperature change trend, which characterizes the direction and intensity of temperature change. A trend slope greater than zero indicates a warming phase, a trend slope less than zero indicates a cooling phase, and a trend slope close to zero indicates a small change in ambient temperature. The temperature change rate, fluctuation amplitude, and trend slope corresponding to each time window are associated and stored in chronological order to form a temperature change feature sequence, thereby realizing a quantitative characterization of the dynamic change law of ambient temperature.

[0115] Intraday temperature variation patterns are analyzed based on temperature change characteristic sequences and timestamp information. The day is divided into time periods according to temperature change characteristics, including a stable period in the early morning, a warming period in the morning, a high-temperature period at noon, a gradual cooling period in the afternoon, and a cooling period at night. Seasonal temperature change characteristic templates for different seasons are established in conjunction with the current date's seasonal information. For example, the period from 8:00 AM to 12:00 PM in summer typically shows a continuous warming phase, while the warming rate is relatively slower during the same period in winter. The temperature change rate, fluctuation amplitude, and trend slope characteristics corresponding to the current time period are then extracted and matched with historical temperature change characteristics for the same period. When the current ambient temperature continues to rise and the trend slope maintains a positive growth, the current environment is determined to be in a typical warming state, and future ambient temperature change trends are predicted based on the current change patterns. Hourly predictions of the ambient temperature for the next 30 minutes, 60 minutes, and 120 minutes are made by combining the current temperature value, time period characteristics, seasonal characteristics, and historical change patterns for the same period. For example, if the current ambient temperature is 32.5℃, it is predicted that the temperature will reach 33.8℃ in 30 minutes, 35.0℃ in 60 minutes, and 36.2℃ in 120 minutes. Connecting the predicted temperature values ​​at each future time point in chronological order forms a continuous curve, resulting in the ambient temperature prediction curve, which reflects the future trend of heat load changes outside the building.

[0116] In this embodiment, see Figure 3 The diagram below illustrates the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0117] Set the indoor comfort temperature value; estimate the initial indoor heat load based on the ambient temperature prediction curve to obtain the original indoor heat load parameters;

[0118] The basic cooling power is obtained by calculating the cooling power requirement based on the original indoor heat load parameters according to the indoor comfort temperature value.

[0119] The operating power of the air conditioner compressor is set based on the basic cooling power to drive the indoor air conditioner to complete the initial cooling process.

[0120] In this embodiment, an indoor comfort temperature value is set according to the comfort requirements of office buildings, commercial spaces, or public buildings, preferably 24℃, and this temperature value is used as the target control temperature of the air conditioning system. Then, the obtained ambient temperature prediction curve is retrieved to analyze the trend of ambient temperature changes in the building's surrounding environment over a future period. For example, if the current outdoor ambient temperature is 32.5℃, it is predicted to reach 33.8℃ in 30 minutes, 35.0℃ in 60 minutes, and 36.2℃ in 120 minutes. Based on the predicted temperature changes, the potential heat transfer from the building envelope is estimated. Specifically, the heat transfer caused by the future increase in ambient temperature is calculated by combining the building's exterior wall area, glass curtain wall area, building orientation, and heat transfer performance parameters of the envelope. As the ambient temperature continues to rise, the heat transfer from the building's exterior walls and glass areas increases synchronously, leading to a gradual increase in the building's internal heat load. Subsequently, the heat transfer values ​​at each moment within the prediction period are cumulatively analyzed to obtain the total heat load demand that the building may bear within the future time window, forming the original indoor heat load parameters. The original heat load parameter is used to characterize the degree of impact of changes in the external environment on the building's internal temperature. This allows the air conditioning equipment to detect the upcoming heat load growth trend before the indoor temperature rises significantly, providing a basis for subsequent cooling capacity configuration, thereby realizing the transformation from traditional passive response to active pre-regulation control.

[0121] After obtaining the original indoor heat load parameters, a cooling demand analysis is further conducted in conjunction with the set indoor comfort temperature value of 24℃. The current average indoor temperature is obtained, and the temperature difference between the current indoor temperature and the target comfort temperature is calculated. For example, if the current average indoor temperature is 26℃, the corresponding temperature difference is 2℃. Simultaneously, the indoor heat growth trend over a future period is analyzed based on the obtained original indoor heat load parameters. When the ambient temperature prediction curve shows that the outdoor temperature will rise from 32.5℃ to 36.2℃ within the next 120 minutes, it indicates that the building's internal heat load will continue to increase. Therefore, it is necessary not only to eliminate the existing temperature difference but also to reserve corresponding cooling capacity in advance to offset the upcoming increase in external heat input. During the calculation process, the building space volume, indoor air specific heat capacity, and original heat load parameters are comprehensively correlated to determine the total amount of heat to be removed to maintain a stable indoor temperature of 24℃. Subsequently, the heat demand is converted into the corresponding cooling power demand according to the preset control cycle. For example, the greater the total heat to be removed within the preset time window, the higher the corresponding basic cooling power; when the predicted temperature rise trend is significantly enhanced, the basic cooling power configuration ratio is appropriately increased. The basic cooling power parameter is calculated, which is used to characterize the minimum cooling capacity required to maintain a comfortable temperature in the current building environment.

[0122] A mapping relationship is established between the basic cooling capacity and the compressor's operating capacity. Based on the current basic cooling capacity demand, the target output power and corresponding operating frequency of the compressor are determined. When the basic cooling capacity demand is in the low-load range (≤5 kW), the compressor operates at a low frequency of 30 Hz; when the basic cooling capacity demand is in the medium-load range (5 kW~15 kW), the compressor operates at a medium frequency of 40 Hz; when the basic cooling capacity demand is in the high-load range (>15 kW), the compressor operates at a high frequency of 50 Hz, to match the current cooling demand while also considering energy consumption control. For example, the current average indoor temperature is 26℃, the target comfortable temperature is 24℃, and the temperature difference is 2℃; combined with a building space volume of 800 m³, an indoor air specific heat capacity of 1.005 kJ / (kg·℃), and an air density of 1.2 kg / m³... 3 The calculated heat removal required to eliminate the current temperature difference is approximately 1932 kJ. Further analysis of the ambient temperature forecast curve, which indicates an increase in outdoor temperature from 32.5°C to 36.2°C within the next 120 minutes, suggests an additional heat transfer to the building envelope of approximately 3200 kJ, bringing the total heat demand to approximately 5132 kJ. Based on a 60-minute control cycle, this corresponds to a basic cooling capacity of approximately 1.42 kW, falling within the low-load range. The compressor operates at a low frequency of 30 Hz, with the supply air temperature set at 16°C and the supply air volume set at 60% of the rated air volume. When the ambient temperature forecast curve indicates a continued warming trend and a total heat demand exceeding 15 kW, the compressor operating frequency is increased to 50 Hz, the supply air temperature is adjusted to 13°C, and the supply air volume is increased to 90% of the rated air volume, allowing the cooling capacity to be released earlier and offsetting the impact of increased external heat load. After the compressor's operating power is determined, each air outlet performs initial air supply according to the principle of uniform distribution, and the indoor temperature gradually converges from 26℃ to the target comfortable temperature of 24℃, forming a stable initial temperature field distribution in the indoor space.

[0123] In this embodiment, step S3 includes the following steps:

[0124] Real-time detection of indoor occupants is achieved using indoor infrared sensors, and all indoor occupants are identified.

[0125] The spatial coordinates of the people inside the room are obtained by calculating their positions.

[0126] Continuous inter-frame difference calculation is performed on the spatial coordinates of personnel to generate a dynamic position change sequence;

[0127] Based on the dynamic change sequence of location, the distribution of personnel movement trajectories is analyzed to construct personnel movement trajectory distribution data.

[0128] In this embodiment, multiple infrared sensors are deployed in the indoor ceiling, wall corners, and main activity areas to provide full coverage monitoring of the indoor space. The infrared sensors continuously detect at a fixed sampling period, preferably consistent with the aforementioned ambient temperature acquisition period of 30 seconds, to ensure consistency in subsequent time-series data analysis. The infrared sensors employ a combination of passive pyroelectric infrared (PIR) sensors and infrared array sensors. The PIR sensors detect human movement-triggered events, while the infrared array sensors acquire information about the distribution of human heat sources. Each infrared sensor is installed on the ceiling at a height of 2.6–3.2 m above the ground. The spacing between adjacent sensors is preferably 4–6 m, and adjacent monitoring areas maintain a 10%–20% overlap to reduce the impact of blind spots on personnel positioning. Each infrared sensor corresponds to preset installation coordinates, and a unified indoor planar coordinate system is established, allowing the detection results of each sensor to be mapped to the same spatial coordinate system.

[0129] Within each sampling period, each infrared sensor outputs information on the human body's heat source status within its corresponding monitoring area, including the center location of the heat source, the area covered by the heat source, and the heat source intensity. When two or more adjacent infrared sensors simultaneously detect the same human body heat source, spatial cross-location is performed based on the installation location of each sensor and the corresponding heat source center location to calculate the real-time spatial coordinates of the human body in the indoor planar coordinate system. When the human body is located within the monitoring area of ​​a single sensor, the center location of the heat source in the monitoring area corresponding to that sensor is used as the current spatial coordinates of the person. Through this method, each person has a unique spatial coordinate in each sampling period.

[0130] Infrared sensors identify the presence of people by detecting differences in infrared radiation between the human body and the surrounding environment, and output corresponding personnel detection information. To avoid blind spots in single sensors, data from multiple infrared sensors are fused and analyzed to identify all personnel targets in the current space. When multiple heat source targets are detected, each heat source target is independently numbered to generate a corresponding personnel identification number. For example, personnel identified in the current space are marked as Person A, Person B, Person C, etc., as unique identifiers. For continuously existing personnel targets, the numbering is maintained continuously; for newly added personnel targets, a new identification number is automatically assigned; for personnel targets leaving the monitoring area, the departure time is recorded and numbering tracking is terminated. At the same time, the appearance time, dwell time, and activity area of ​​each personnel target are correlated and recorded to form an indoor personnel status dataset. This step can accurately obtain the current number and distribution of indoor personnel, providing basic data support for subsequent personnel location and heat source distribution analysis, enabling the air conditioning control process to gradually shift from traditional environmental temperature control to personnel perception-driven control.

[0131] After personnel identification, spatial positioning calculations are performed on each individual. An indoor spatial coordinate model is established, using the room's floor plan as a reference plane to construct a two-dimensional coordinate system, and assigning a corresponding spatial coordinate range to each area. Subsequently, cross-positioning analysis is performed based on the detection results of multiple infrared sensors, determining the actual location of the personnel by the overlap relationship between the detection areas of different sensors. When a personnel target is detected by multiple infrared sensors simultaneously, the personnel's center position is calculated based on the coverage area and detection intensity information of each sensor; when the personnel is located at the boundary of adjacent areas, the position is corrected by combining multiple detection results. For larger office areas, meeting areas, or open office spaces, they can be further divided into multiple grid units, for example, using a 1 m × 1 m spatial scale, and the personnel positions are mapped to the corresponding grid areas. The spatial position coordinates of each personnel at the current moment are obtained through continuous calculation, for example, personnel A is located at coordinates (8, 5), personnel B is located at coordinates (12, 9), personnel C is located at coordinates (15, 6), etc. All personnel position coordinates are associated with and stored with corresponding timestamps, forming a real-time personnel spatial position dataset. This step allows for an accurate understanding of the distribution and spatial aggregation of people indoors, providing a locational basis for subsequent analysis of personnel movement behavior and calculation of heat source spatial distribution, thereby improving the targeting and accuracy of air conditioning cooling resource allocation.

[0132] The system extracts the location coordinates of the same person at different sampling times in chronological order. For example, if person A is located at coordinates (8, 5) at the current sampling time and at coordinates (9, 5) at the next sampling time, it indicates that the person's location has changed between the two sampling periods. The position difference between adjacent time points is then calculated to obtain the person's movement direction and distance. The position change results from multiple consecutive sampling periods are correlated and statistically analyzed to obtain the movement change pattern of the person within a certain time range. When the position change is close to zero for multiple consecutive sampling periods, the person is considered stationary; when the position change continuously increases, the person is considered to be moving; when the direction of position change changes frequently, the person is considered to be in a high-frequency activity state. Simultaneously, the position change, movement direction, and movement speed corresponding to each time point are stored in a time-correlated manner to generate corresponding dynamic change data. As the monitoring time continues, each person forms a continuous record of position changes, and a dynamic position change sequence is constructed in chronological order. This sequence reflects the process of a person changing from stationary to slow movement, rapid movement, and other different activity states.

[0133] The system extracts location change records of the same person over a continuous time period and connects them in chronological order to form a complete movement path. For example, if person A passes through coordinates (8, 5), (9, 5), (10, 6), and (11, 6) sequentially at multiple consecutive moments, a corresponding movement trajectory is formed. The trajectory length, movement frequency, and activity range per unit time are then statistically analyzed to assess the intensity of human activity. When a person stays in a certain area for a long time, that area is considered a stable residence area; when a person frequently moves between multiple areas, it is considered a high-activity area. Further, all human trajectories are overlaid and statistically analyzed to assess the degree of human concentration and passage frequency in different areas. For example, a meeting room area with many people gathering for a long time constitutes a high-density activity area; an office corridor area, although with shorter dwell times, has a higher passage frequency, thus constituting a high-flow activity area. Spatial distribution weights are then generated based on the number of times the trajectory is covered and the dwell time, and human movement trajectory distribution data is constructed. This distribution data not only reflects the current spatial distribution of people indoors but also reveals potential activity trends in the near future. For example, if there is a continuous gathering of people in a certain area, the probability of heat accumulation in that area is relatively high; if the activity of people in a certain area gradually decreases, the heat growth trend is relatively weak. This step can provide dynamic behavioral reference for subsequent human heat source modeling, heat source distribution map construction, and regional air conditioning cooling control, making air conditioning control more consistent with real human activity scenarios.

[0134] In this embodiment, see Figure 4 Step S4 includes the following steps:

[0135] The specific steps of step S4 are as follows:

[0136] Set dynamic and static human body heat dissipation power calibration values; the specific dynamic and static human body heat dissipation power calibration values ​​are: dynamic heat dissipation power calibration value is 120W, and static heat dissipation power calibration value is 80W.

[0137] Based on the dynamic and static human body heat dissipation power calibration values, the individual heat source intensity is assigned to the personnel movement trajectory distribution data, and spatial grid interpolation is performed to output the indoor heat source coordinate system.

[0138] Based on the indoor heat source coordinate system, the spatial distribution of heat source intensity is fitted to construct an indoor heat source distribution map.

[0139] In this embodiment, based on the metabolic heat production characteristics of the human body under different activity states, dynamic heat dissipation power calibration values ​​and static heat dissipation power calibration values ​​are preset, wherein the dynamic heat dissipation power calibration value is set to 120W and the static heat dissipation power calibration value is set to 80W. Dynamic states mainly correspond to behaviors such as continuous walking, frequent movement, area switching, and sustained activity within the office area; static states mainly correspond to behaviors such as prolonged sitting at work, staying in meetings, reading, or maintaining a relatively fixed position. Combining the position dynamic change sequence generated in step S10, the position change of each person within a preset analysis time window is statistically analyzed. When the position change exceeds a preset activity threshold for multiple consecutive sampling periods, the person is identified as a dynamic person and assigned a heat dissipation power value of 120W; when the position change is close to zero or remains at a low range for multiple consecutive sampling periods, the person is identified as a static person and assigned a heat dissipation power value of 80W. For personnel transitioning from static to dynamic or vice versa, the corresponding heat dissipation power calibration value is switched in real time according to the current activity state. Subsequently, the personnel identification number, spatial location coordinates, activity status, and corresponding heat dissipation power are associated and stored to form an individual heat source dataset. This step transforms abstract personnel activity behaviors into quantifiable heat output parameters, ensuring that each person corresponds to a specific heat source contribution value.

[0140] The indoor space is divided into 1m×1m grids to construct a unified spatial grid model. The coordinates of each person's current location and corresponding heat dissipation power are then extracted and mapped to the corresponding grid cell. For example, if person A is located at coordinates (8, 5) and is in a dynamic state, a heat source intensity of 120W is assigned to the corresponding grid cell; if person B is located at coordinates (12, 9) and is in a static state, a heat source intensity of 80W is assigned to the corresponding grid cell. When multiple people are in the same grid cell, the heat source intensities of each person are accumulated. For example, if two people are in a dynamic state and one person is in a static state in a grid area, the corresponding heat source intensity is 320W. Since human body heat dissipation diffuses into the surrounding space, heat diffusion processing is applied not only to the grid cell containing the heat source but also to neighboring grid cells. Based on spatial distance, the central heat source gradually attenuates and expands towards the surrounding grid cells; the closer the distance, the greater the heat contribution, and the farther the distance, the smaller the heat contribution. After spatial diffusion calculations for all people's heat sources, continuous heat source intensity distribution data is generated for each grid cell. By associating the spatial coordinates of each grid with the heat source intensity, an indoor heat source coordinate system is established. This coordinate system can intuitively reflect the heat distribution of the human body at different locations indoors.

[0141] The spatial coordinates and heat source intensity values ​​corresponding to each grid node are extracted, and a heat source distribution matrix is ​​established according to the spatial position relationship. Since actual human body heat dissipation is not limited to a single location but forms a continuous heat field in the surrounding area, spatial fitting processing is performed on the distance from the heat dissipation source point. By analyzing the heat source intensity variation relationship between adjacent grids, the heat source data is smoothly expanded to make the local heat distribution more consistent with the actual indoor heat diffusion law. For example, densely populated areas may form multiple adjacent high heat source grids, so a continuous high-temperature heat zone is formed through fitting processing; sparsely populated areas form low heat source areas; and uninhabited areas correspond to heat source intensities close to zero. At the same time, combined with the distribution data of personnel movement trajectories, higher heat accumulation weights are given to areas where people stay for a long time, and lower heat accumulation weights are given to areas where people stay for a short time, so that the heat source distribution results can simultaneously reflect the current heat status and heat accumulation trend. After completing all spatial fitting, continuous heat source intensity distribution results are generated in the indoor planar coordinate model, and corresponding heat distribution areas are formed according to the heat source intensity. Among them, high heat source areas mainly correspond to conference rooms, densely populated areas, and high-activity areas; medium heat source areas correspond to ordinary office areas; and low heat source areas correspond to corridors and uninhabited areas. An indoor heat source distribution map is constructed to visualize the indoor heat sources and heat accumulation status, providing a core basis for subsequent temperature gradient calculation, air conditioning supply strategy generation, and regional cooling control.

[0142] In this embodiment, step S5 includes the following steps:

[0143] The indoor heat source distribution map is divided into grids, and the heat source influence of each grid node is analyzed to obtain the expected temperature value of each node.

[0144] The expected temperature difference is calculated based on the indoor comfort temperature value to obtain the expected temperature difference value for all nodes.

[0145] Construct a temperature gradient distribution map based on the expected temperature difference; identify all indoor air outlets;

[0146] The heat load deviation of the corresponding area of ​​the air outlet is evaluated based on the temperature difference gradient distribution map to obtain the cooling deviation of different areas.

[0147] In this embodiment, the entire indoor space is uniformly divided using a 1m×1m grid, ensuring each grid corresponds to a unique spatial coordinate node. Since human heat sources continuously release heat into the surrounding space, the analysis considers not only the node's own heat source intensity but also the cumulative effect of heat sources from adjacent grids on the current node. In densely populated areas, multiple high-heat-source nodes in the vicinity create thermal coupling, causing a continuous increase in local temperature; in less populated areas, the thermal impact is relatively weaker. During the analysis, a heat attenuation relationship is established based on the spatial distance between the node and the heat source; the closer the distance, the greater the heat source impact, and the farther the distance, the weaker the impact. For example, when multiple 120W dynamic heat sources exist around a node, its expected temperature rise is significantly higher than in areas with only an 80W static heat source. Subsequently, the node's own heat source contribution, the diffusion contribution of neighboring heat sources, and the current indoor base temperature field are comprehensively superimposed to calculate the expected temperature value corresponding to each grid node. For example, when the initial indoor temperature is 24℃, the expected temperature of nodes in densely populated areas may reach 26℃ to 28℃, the expected temperature of nodes in ordinary office areas may reach 24.5℃ to 25.5℃, while unoccupied areas will remain around 24℃. By analyzing all grid nodes one by one, an expected temperature field covering the entire indoor space is formed.

[0148] For each grid node, its corresponding expected temperature value is compared and analyzed with 24℃ to calculate the degree of temperature deviation. For example, if the expected temperature of a node is 27℃, the corresponding expected temperature difference is 3℃; if the expected temperature of a node is 25.5℃, the corresponding expected temperature difference is 1.5℃; and if the expected temperature of a node remains at 24℃, the corresponding expected temperature difference is 0℃. This method quantifies the degree of deviation of different areas from the comfort temperature target. Simultaneously, the temperature difference results of all nodes are statistically analyzed to identify high-temperature deviation areas, medium-temperature deviation areas, and low-temperature deviation areas. High-temperature deviation areas typically correspond to areas with long-term population gatherings, high-frequency activity areas, and meeting areas; medium-temperature deviation areas typically correspond to ordinary office areas; and low-temperature deviation areas mainly correspond to areas with fewer people or no one. After completing the calculation for all nodes, the spatial coordinates corresponding to each node are associated with and stored with the expected temperature difference value, forming an indoor temperature difference data matrix. This matrix not only reflects the current uneven distribution of heat but also clearly indicates which areas have significant cooling demand gaps.

[0149] The temperature difference between adjacent grid nodes is analyzed, and the direction and magnitude of temperature difference changes between different nodes are calculated. A large temperature difference between adjacent areas indicates significant heat accumulation; a small temperature difference indicates a relatively uniform temperature distribution. By continuously analyzing all grid nodes, a temperature gradient field covering the entire indoor space is established, accurately representing the direction of heat accumulation, the direction of heat diffusion, and the intensity of temperature difference changes. For example, a significant temperature gradient between the conference room area and the surrounding office areas indicates that heat is diffusing to the surrounding areas. Simultaneously, the structural information of the indoor air conditioning equipment is read, identifying the location coordinates of all air outlets. This includes ceiling outlets, side wall outlets, and independent air outlets, and each outlet is mapped to its corresponding spatial coordinate system. For example, outlet A is located at coordinates (5, 4), outlet B at coordinates (10, 8), and outlet C at coordinates (15, 6). The spatial coverage relationship between the outlets and the surrounding grid areas is established, clarifying the main air supply area covered by each outlet.

[0150] The expected temperature difference values ​​of all grid nodes within the coverage area of ​​each air outlet are extracted, and the average temperature difference, maximum temperature difference, and heat accumulation degree within the area are statistically analyzed. For example, if the average temperature difference in the area covered by air outlet A reaches 3℃, while the average temperature difference in the area covered by air outlet B is only 1℃, it indicates that the area corresponding to air outlet A has a higher cooling demand. Subsequently, the deviation between the actual heat load and the target comfort heat load of the area is analyzed by combining the heat source intensity distribution and temperature difference gradient trend within the area. When the temperature difference in the area is large and the heat source continues to increase, it is determined that there is a large cooling gap in the area; when the temperature difference in the area is small and the heat source is relatively stable, it is determined that the cooling capacity basically meets the demand. Further, the corresponding cooling deviation is calculated based on the degree of heat load deviation, which is used to characterize the cooling capacity that still needs to be increased or decreased in the area. For example, high heat load areas correspond to large cooling deviations, medium heat load areas correspond to medium cooling deviations, and low heat load areas correspond to small cooling deviations. After completing the analysis of all air outlet areas, the distribution results of cooling deviations covering the entire indoor space are generated. It accurately reflects the differences in actual cooling needs in different areas, providing a direct basis for dynamically generating control parameters such as air supply temperature, air volume, and air direction at each air outlet, thereby achieving refined zoned cooling control based on the distribution of heat sources for people.

[0151] In this embodiment, the specific steps of step S6 are as follows:

[0152] Based on the cooling deviation, the cooling parameters of each air outlet are adjusted differently to generate a set of cooling control parameters for each air outlet.

[0153] Generate packaging control instructions based on the aforementioned dynamic cooling regulation parameter set;

[0154] Based on the encapsulation control command, it is sent to the execution unit of each air outlet to perform indoor sensing control.

[0155] The dynamic cooling control parameter set includes target supply air temperature, air volume level, and operating time.

[0156] In this embodiment, a one-to-one mapping relationship is established between the spatial coverage area corresponding to each air outlet and the cooling deviation. The cooling deviation for the area corresponding to air outlet A is 3℃, for air outlet B it is 1℃, and for air outlet C it is 0.5℃. The cooling capacity of each air outlet is adjusted in stages according to the magnitude of the cooling deviation: high deviation areas correspond to enhanced cooling strategies, medium deviation areas to standard cooling strategies, and low deviation areas to weak cooling or maintenance strategies. Subsequently, three types of core control parameters are determined for each air outlet, including target supply air temperature, airflow level, and operating time.

[0157] Regarding the target supply air temperature setting, when the cooling deviation in a region is large, the target supply air temperature at the corresponding air outlet is appropriately reduced, for example, adjusted from a base supply air temperature of 24℃ to the range of 20℃ to 22℃, to enhance local cooling capacity. When the cooling deviation is small, the target supply air temperature is maintained between 23℃ and 24℃ to avoid over-cooling. Regarding the airflow setting, the airflow is graded according to the size of the cooling deviation, for example, divided into low, medium, and high airflow levels. High deviation areas correspond to high airflow output, medium deviation areas to medium airflow output, and low deviation areas to low airflow output. Regarding the operating time, it is dynamically set according to the continuity of the regional heat load. For example, areas with continuously increasing heat load correspond to longer operating cycles, while areas with stable or decreasing heat load correspond to shorter operating cycles or intermittent operation modes.

[0158] The target supply air temperature, air volume level, and operating time for each air outlet are standardized, and the data is encoded according to a preset communication protocol to ensure compatibility with the input interface of the air conditioning control execution unit. During the control command encapsulation process, to guarantee independent control capabilities between different air outlets, an air outlet identifier field, a timestamp field, and a priority field are introduced into the data structure. The priority field is sorted and assigned values ​​based on the magnitude of the cooling deviation, with air outlets exhibiting larger deviations having higher priority.

[0159] For example, air outlet A, with a cooling deviation of 3℃, is assigned priority 1; air outlet B, with a cooling deviation of 1℃, is assigned priority 2; and air outlet C, with a cooling deviation of 0.5℃, is assigned priority 3. Subsequently, all air outlet control parameters are uniformly encapsulated into a complete control instruction set, giving it parsable, executable, and schedulable characteristics. A verification field is added during the encapsulation process to ensure the integrity and consistency of the instructions during transmission, preventing control parameter deviations due to communication errors. The resulting encapsulated control instructions not only contain the specific cooling parameters of each air outlet but also the execution order and priority control logic, enabling the air conditioning control to have both unified scheduling capabilities and differentiated execution capabilities.

[0160] The generated encapsulated control commands are sent to each air outlet execution unit via the air conditioning control bus or wireless communication link. Each execution unit parses the corresponding control parameters based on its own identification information and executes the corresponding adjustment actions. Specific actions include adjusting the supply air temperature, adjusting the airflow level, and controlling the operating duration. When an air outlet receives a high-priority control command, it prioritizes operations to enhance cooling capacity, such as lowering the supply air temperature and increasing the airflow output. When it receives a medium-priority command, it maintains a medium airflow and moderate temperature output. When it receives a low-priority command, it executes an energy-saving operation mode to reduce energy consumption.

[0161] During execution, each air outlet provides real-time feedback on its current operating status, including actual supply air temperature, air volume output, and runtime progress, and dynamically matches and verifies this against control commands. When a deviation between the actual output and the target parameters is detected, adaptive correction is performed to gradually bring the outlet output closer to the target cooling control parameters. This method achieves precise cooling control of different indoor areas based on heat source distribution and temperature gradients, enabling high-heat-load areas to achieve stronger cooling capacity while low-heat-load areas maintain energy-efficient operation, thereby achieving balanced and comfortable control and energy optimization of the overall indoor environment.

[0162] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0163] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein are implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. An air conditioning control method integrating outdoor temperature feedforward and indoor heat source distribution sensing, characterized in that, Includes the following steps: Step S1: Based on the continuous collection of external ambient temperature by the building exterior wall temperature sensor, perform intraday temperature change prediction to obtain the ambient temperature prediction curve; Step S2: Calculate the cooling power demand based on the ambient temperature prediction curve and drive the indoor air conditioner to complete the initial cooling process; Step S3: Real-time perception of indoor personnel based on indoor infrared sensors to construct personnel movement trajectory distribution data; Step S4: Fit the spatial distribution of heat source intensity to the personnel movement trajectory distribution data to construct an indoor heat source distribution map; Step S5: Based on the indoor heat source distribution map, assess the regional heat load deviation to obtain the cooling deviation in different regions; Step S6: Adjust the cooling parameters of each air outlet differently based on the cooling deviation, and send the adjustment to the execution unit of each air outlet to execute indoor sensing control.

2. The air conditioning control method according to claim 1, which integrates outdoor temperature feedforward and indoor heat source distribution sensing, is characterized in that, The specific steps of step S1 are as follows: Based on the continuous collection of external ambient temperature by building exterior wall temperature sensors, a real-time outdoor temperature sequence is constructed. Anomaly filtering is performed on the outdoor real-time temperature sequence to obtain a standard temperature sequence; A sliding window statistical analysis is performed on a standard temperature sequence to generate a temperature change feature sequence; the temperature change feature sequence includes temperature change rate, fluctuation amplitude, and trend slope characteristics. The temperature change feature sequence is segmented and extracted, and intraday temperature change prediction is performed to obtain the ambient temperature prediction curve.

3. The air conditioning control method integrating outdoor temperature feedforward and indoor heat source distribution sensing according to claim 2, characterized in that, The specific steps of step S2 are as follows: Set the indoor comfort temperature value; estimate the initial indoor heat load based on the ambient temperature prediction curve to obtain the original indoor heat load parameters; The basic cooling power is obtained by calculating the cooling power requirement based on the original indoor heat load parameters according to the indoor comfort temperature value. The operating power of the air conditioner compressor is set based on the basic cooling power to drive the indoor air conditioner to complete the initial cooling process.

4. The air conditioning control method integrating outdoor temperature feedforward and indoor heat source distribution sensing according to claim 3, characterized in that, Step S3 is as follows: Real-time detection of indoor occupants is achieved using indoor infrared sensors, and all indoor occupants are identified. The spatial coordinates of the people inside the room are obtained by calculating their positions. Continuous inter-frame difference calculation is performed on the spatial coordinates of personnel to generate a dynamic position change sequence; Based on the dynamic change sequence of location, the distribution of personnel movement trajectories is analyzed to construct personnel movement trajectory distribution data.

5. The air conditioning control method integrating outdoor temperature feedforward and indoor heat source distribution sensing according to claim 4, characterized in that, The specific steps of step S4 are as follows: Set the dynamic and static human body heat dissipation power calibration values; Based on the dynamic and static human body heat dissipation power calibration values, the individual heat source intensity is assigned to the personnel movement trajectory distribution data, and spatial grid interpolation is performed to output the indoor heat source coordinate system. Based on the indoor heat source coordinate system, the spatial distribution of heat source intensity is fitted to construct an indoor heat source distribution map.

6. The air conditioning control method according to claim 5, which integrates outdoor temperature feedforward and indoor heat source distribution sensing, is characterized in that... The specific dynamic and static human body heat dissipation power calibration values ​​are as follows: the dynamic heat dissipation power calibration value is 120W, and the static heat dissipation power calibration value is 80W.

7. The air conditioning control method according to claim 6, which integrates outdoor temperature feedforward and indoor heat source distribution sensing, is characterized in that, The specific steps of step S5 are as follows: The indoor heat source distribution map is divided into grids, and the heat source influence of each grid node is analyzed to obtain the expected temperature value of each node. The expected temperature difference is calculated based on the indoor comfort temperature value to obtain the expected temperature difference value for all nodes. Construct a temperature gradient distribution map based on the expected temperature difference; identify all indoor air outlets; The heat load deviation of the corresponding area of ​​the air outlet is evaluated based on the temperature difference gradient distribution map to obtain the cooling deviation of different areas.

8. The air conditioning control method according to claim 7, which integrates outdoor temperature feedforward and indoor heat source distribution sensing, is characterized in that, The specific steps of step S6 are as follows: Based on the cooling deviation, the cooling parameters of each air outlet are adjusted differently to generate a set of cooling control parameters for each air outlet. Generate packaging control instructions based on the aforementioned dynamic cooling regulation parameter set; Based on the encapsulation control command, it is sent to the execution unit of each air outlet to perform indoor sensing control.

9. The air conditioning control method according to claim 8, which integrates outdoor temperature feedforward and indoor heat source distribution sensing, is characterized in that... The dynamic cooling control parameter set includes target supply air temperature, air volume level, and operating time.

Images