Intelligent control method, device and medium for building equipment

Abstract

The application belongs to the technical field of building control, and particularly relates to an intelligent control method, equipment and medium for a building device, which comprises the following steps: obtaining a target area heating state vector of a target building, a target device state vector of the target building and a target external environment state vector of the target building according to an original environment data set of the target building, an original device operation data set of the target building and an original external environment data set of the target building; and generating a priority of a target area state of the target building according to the target area heating state vector of the target building, the target device state vector of the target building and the target external environment state vector of the target building, so as to perform control. The intelligent control method for heating and ventilation devices can fuse multi-source dynamic data, accurately depict the thermal environment state of the building, perceive the health level of the device, and perform global collaboration and forward-looking regulation and control according to the above, so as to solve problems such as high energy consumption, poor environmental stability, short service life of the device and the like.

Description

Technical Field

[0001] This invention relates to the field of building control technology, specifically to an intelligent control method, device, and medium for building equipment. Background Technology

[0002] In industrial buildings and multi-functional commercial buildings, environmental control systems play a crucial role in maintaining a suitable internal thermal and humidity environment and air quality. However, existing control methods still face multiple challenges in actual operation. Buildings contain multiple functional areas, each with dynamically changing and unevenly distributed internal thermal disturbances due to differences in equipment heat dissipation, personnel flow, lighting, and the heat storage characteristics of the building envelope. Simultaneously, outdoor meteorological conditions such as temperature, humidity, solar radiation, and wind speed fluctuate dramatically in real time, creating strong nonlinear external disturbances through infiltration into the building envelope and the introduction of fresh air. This deep coupling of internal and external thermal disturbances results in a building thermal environment characterized by strong time-varying, large lag, and multivariate coupling. Traditional control methods based on fixed setpoints or simple feedback (such as PID) struggle to accurately respond to rapidly changing load distributions, easily leading to areas becoming overcooled or overheated, temperature oscillations, and control delays, severely impacting environmental stability.

[0003] On the other hand, raw data from multiple sources, such as temperature, humidity, pressure, flow rate, and power, collected by field sensors often contain measurement noise, communication packet loss, sensor drift, and outliers. Directly using such low-quality data in the control loop can easily lead to command distortion and system oscillation, reducing control reliability and robustness. Existing control strategies mostly focus on local closed-loop regulation of single equipment (such as compressors, fans, and pumps) or single thermal areas, lacking the ability to model and characterize the overall thermal coupling relationships within a building (such as inter-area heat transfer, airflow organization, and energy migration processes). Therefore, it is difficult to achieve cross-area energy coordination and load balancing. A typical manifestation is that some areas receive excessive cooling or dehumidification while adjacent areas receive insufficient supply, forcing the system to operate at high energy consumption to compensate for local imbalances, resulting in low overall energy efficiency.

[0004] Furthermore, traditional control methods typically treat equipment as an ideal response unit, ignoring its real-time health status, such as load rate, motor winding temperature deviation, heat exchanger surface fouling, and bearing vibration amplitude. Continuing to operate under conventional commands when equipment is in a sub-healthy or high-loss state not only exacerbates energy consumption degradation but also accelerates component fatigue and fault accumulation, significantly increasing the risk of unplanned downtime and shortening the equipment's lifespan.

[0005] In summary, existing environmental control technologies have significant shortcomings in dynamic thermal disturbance coupling characterization, multi-source data quality assurance, inter-regional collaborative optimization, and equipment health perception fusion, making it difficult to meet the comprehensive requirements of modern industrial buildings and multi-functional commercial buildings for low energy consumption, high environmental stability, and high equipment reliability. Therefore, there is an urgent need to develop an intelligent control method that can integrate multi-source dynamic data, accurately characterize the building's thermal environment, perceive equipment health levels in real time, and possess global collaborative and forward-looking adjustment capabilities. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent control method, device, and medium for building equipment to solve the problems mentioned in the background art.

[0007] This invention protects an intelligent control method for building equipment, comprising the following steps: Based on the original environmental dataset, the original equipment operation dataset, and the original external environment dataset of the target building, obtain the target area state vector, the target equipment state vector, and the target external environment state vector of the target building. Based on the target area state vector, target device state vector, and target external environment state vector of the target building, a priority of the target area state of the target building is generated. This priority is used to generate intelligent control commands for the target building. The priority Y of the target area state of the target building satisfies the following condition: Where φ1 is the regional state weight, φ2 is the equipment state weight, and φ3 is the external environment state weight.

[0008] Specifically, the original environmental dataset of the target building includes: N1 subsets of original environmental data corresponding to monitoring areas, wherein any subset of original environmental data corresponding to the monitoring area includes a list of original humidity and a list of original temperature corresponding to the monitoring area. The list of original humidity corresponding to the monitoring area includes several collected original humidity values, and the list of original temperature corresponding to the monitoring area includes several real-time collected original temperatures.

[0009] Specifically, the raw equipment operation dataset of the target building includes: N2 subsets of raw equipment operation data corresponding to target operating equipment, wherein any subset of raw equipment operation data corresponding to the target operating equipment includes a list of raw equipment temperatures, a list of raw equipment operating power, and a list of raw equipment heat generation power. The list of raw equipment operating power includes several collected raw equipment temperatures, the list of raw equipment operating power includes several collected raw equipment operating powers, and the list of raw equipment heat generation power includes several collected raw equipment heat generation powers.

[0010] Specifically, the original external environment dataset of the target building includes a list of original outdoor temperatures, a list of original outdoor humidity, and a list of original light intensity corresponding to the target building.

[0011] Specifically, the steps of obtaining the target area state vector, target equipment state vector, and target external environment state vector of the target building based on the target building's original environmental dataset, original equipment operation dataset, and original external environment dataset also include the following steps: Based on the original environmental dataset of the target building, generate a spatial distribution feature vector of the target building; Based on the original environmental dataset of the target building, generate a spatial correlation feature vector of the target building; Based on the original environmental dataset of the target building, generate a temporal evolution feature vector of the target building; Based on the original environmental dataset of the target building, generate a thermal comfort feature vector of the target building; The spatial distribution feature vector, spatial correlation feature vector, temporal evolution feature vector, and thermal comfort feature vector of the target building are concatenated to obtain the target area environmental state vector of the target building.

[0012] Specifically, the steps of obtaining the target area state vector, target equipment state vector, and target external environment state vector of the target building based on the target building's original environmental dataset, original equipment operation dataset, and original external environment dataset also include the following steps: Based on the original equipment operation dataset of the target building, generate the equipment operation feature vector of the target building; Based on the aforementioned equipment operation feature vector and the equipment historical operation dataset, generate the equipment mode feature vector of the target building; Based on the original equipment operation dataset of the target building, generate the equipment collaboration feature vector of the target building; The target equipment state vector of the target building is obtained by weighted concatenation of the equipment operation feature vector, the equipment mode feature vector, and the equipment coordination feature vector of the target building.

[0013] Specifically, the steps of obtaining the target area state vector, target equipment state vector, and target external environment state vector of the target building based on the target building's original environmental dataset, original equipment operation dataset, and original external environment dataset also include the following steps: Based on the original external environment dataset of the target building, extract the basic feature vector of the external environment of the target building; Based on the original external environment dataset of the target building, construct the external environment evolution feature vector of the target building; Based on the original external environment dataset of the target building, generate the external environment influence weight vector of the target building; The target building's external environment basic feature vector, external environment evolution feature vector, and external environment influence weight vector are fused together to obtain the target building's external environment state vector.

[0014] The present invention also protects a storage medium storing program instructions, which, when executed, implement the above-described intelligent control method for building equipment.

[0015] The present invention also protects an electronic device, including a processor and a memory, wherein the memory stores program instructions, and the processor executes the program instructions to implement the above-described intelligent control method for building equipment.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention utilizes multi-source data fusion and feature-based modeling to collaboratively optimize the control efficiency of building HVAC systems. First, the method systematically collects and preprocesses data on the building's internal environment, equipment operating status, and external climate, providing a high-quality data foundation for accurate decision-making. By extracting and fusing multi-dimensional features such as spatial distribution, correlations, temporal evolution, and thermal comfort of the target area, a regional environmental state vector is constructed that comprehensively reflects the spatiotemporal dynamics of the building's thermal environment. Simultaneously, by analyzing equipment operating efficiency, energy consumption, health status, operating modes, and inter-equipment coordination indicators, an equipment state vector is generated that quantitatively characterizes the real-time performance and health level of the equipment. Furthermore, by analyzing the fundamental characteristics and evolution patterns of external environmental data and calculating their dynamic influence weights, an environmental state vector is generated that can proactively quantify and compensate for external climate disturbances. Finally, by fusing the above three types of state vectors and defining a target area state priority function, global collaborative regulation is achieved, simultaneously optimizing equipment energy efficiency and healthy operation while proactively offsetting external disturbances, all while ensuring regional thermal comfort. This invention effectively improves the accuracy, overall energy efficiency, and operational reliability of equipment system control in complex building environments. Attached Figure Description

[0017] Figure 1 This is a flowchart of an intelligent control method for building equipment according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 This invention provides an intelligent control method for building equipment, comprising the following steps: S1, obtain the original environmental dataset, the original equipment operation dataset, and the original external environment dataset of the target building.

[0020] Specifically, the target building is a building that requires user-defined optimization control; for example, the target building is an office building.

[0021] Specifically, the target building has N1 monitored spatial areas; for example, the monitored area is a room.

[0022] Furthermore, the original environmental dataset of the target building includes: N1 subsets of original environmental data corresponding to the monitored areas, wherein any subset of original environmental data corresponding to the monitored area includes an original humidity list and an original temperature list corresponding to the monitored area. The original humidity list corresponding to the monitored area includes several collected original humidity values, and the original temperature list corresponding to the monitored area includes several real-time collected original temperatures. Further, it can be: the original environmental dataset of the target building A = {A1, ..., A...} i , ..., A N1}, the original environmental data subset A corresponding to the i-th monitored area i ={A 1 i A 2 i}, the original humidity list A corresponding to the area to be monitored 1 i ={A 1 i1 , ..., A 1 ix1 , ..., A 1 iP1}, the original temperature list A corresponding to the area to be monitored 2 i ={A 2 i1 , ..., A 2 ix2 , ..., A 2 iP2}, A 1 i1x It is A i The corresponding x1th original humidity in the original humidity list for the area to be monitored, A 2 ix2 Yes, it is A. i The x2th original temperature in the original temperature list corresponding to the area to be monitored, where i ranges from 1 to N1, x1 ranges from 1 to P1, P1 is the original humidity value, and x2 ranges from 1 to P2, P2 is the original temperature value.

[0023] Specifically, the target building has N2 target devices, for example, the target device is an air conditioning unit.

[0024] Specifically, the raw equipment operation dataset of the target building includes: N2 subsets of raw equipment operation data corresponding to the target equipment, wherein any subset of raw equipment operation data corresponding to the target equipment includes a list of raw equipment temperatures, a list of raw equipment operating power, and a list of raw equipment heat generation power corresponding to the target equipment. The list of raw equipment operating power includes several collected raw equipment temperatures, the list of raw equipment operating power includes several collected raw equipment operating powers, and the list of raw equipment heat generation power includes several collected raw equipment heat generation powers. Further, the raw equipment operation dataset of the target building can be B = {B1, ..., B...} j , ..., B N2}, the subset B of the original equipment operation data corresponding to the j-th target device j ={B 1 j B 2 j B 3 j}, the original equipment temperature list B corresponding to the target equipment 1 j ={B 1 j1 , ..., B 1 j(y1) , ..., B 1 j(Q1)}, the original equipment operating power list B corresponding to the target device 2 j ={B 2 j1 , ..., B 2 j(y2) , ..., B 2 j(Q2)}, the original equipment heat dissipation power list B corresponding to the target device 3 j ={B 3 j1 , ..., B 3 j(y3) , ..., B 3 j(Q3)}, B 1 j(y1) It is B j The corresponding y1th original device temperature of the target device, B 2 j(y2) It is B j The corresponding y2th original device operating power of the target device, B 3j(y3) It is B j The corresponding target device's y3th original device heating power; j ranges from 1 to N2, y1 ranges from 1 to Q1, Q1 is the original device temperature quantity; y2 ranges from 1 to Q2, Q2 is the original device operating power quantity; y3 ranges from 1 to Q3, Q3 is the original device heating power quantity. For example, the target device is a target heating device.

[0025] Specifically, the original external environment dataset of the target building includes a list of original outdoor temperatures corresponding to the target building, a list of original outdoor humidity corresponding to the target building, and a list of original light intensity corresponding to the target building; further, it can be: the original external environment dataset of the target building C={C1, C2, C3}, and the original outdoor temperature list corresponding to the target building C1={C 11 , ..., C 1r1 , ..., C 1S1}, the original outdoor humidity list C2={C 21 , ..., C 2r2 , ..., C 2S2}, the original illuminance list C3 corresponding to the target building = {C 31 , ..., C 3r3 , ..., C 3S3}, C 1r1 This is the r1th original outdoor temperature, where r1 ranges from 1 to S1, and S1 is the number of original outdoor temperatures. 2r2 This is the r2th original outdoor humidity value, where r2 ranges from 1 to S2, and S2 is the original outdoor humidity value. 3r3 It is the r3rd original light intensity, where r3 ranges from 1 to S3, and S3 is the number of original light intensities.

[0026] Compared to existing technologies that only require building temperature and humidity to control heating equipment, this invention can also collect real-time data of the heating equipment and real-time data of the outdoor environment, which is conducive to more precise control of the heating equipment.

[0027] S2, Based on the original environmental dataset of the target building, generate the target area environmental state vector of the target building.

[0028] Specifically, step S2 also includes the following steps: S21, Based on the original environmental dataset of the target building, generate the spatial distribution feature vector of the target building; further understood as: the spatial distribution feature vector of the target building V1=(V 1 1, V 2 1, V 3 1, V4 1), V 1 1 represents the standard deviation of the current average humidity across all monitored areas, V 2 1 is the standard deviation of the current average temperature across all monitored areas, V 3 1 is the coefficient of variation of humidity distribution, V 4 1 is the coefficient of variation of the temperature distribution.

[0029] Preferred, V 1 1. Meets the following conditions: .

[0030] Preferred, V 2 1. Meets the following conditions: .

[0031] Preferred, V 3 1. Meets the following conditions: .

[0032] Preferred, V 4 1. Meets the following conditions: .

[0033] S22, Based on the original environmental dataset of the target building, generate a spatial association feature vector of the target building; further understood as: the spatial association feature vector of the target building V2=(V 1 2, V 2 2), V 1 2 is the average Pearson coefficient of the temperature data list of all adjacent monitoring areas, V 2 2 is the maximum Pearson coefficient for the temperature data list of all adjacent monitoring areas, V 3 2 is the minimum Pearson coefficient for the temperature data list of all adjacent monitoring areas.

[0034] Preferred, V 1 2. Meets the following conditions: Where U is the edge set in the correlation graph of adjacent monitoring areas, representing the pairs of adjacent monitoring areas; corr() is the function to calculate the correlation coefficient between two original temperature data lists, A 0 m Yes, A 0 n yes S23, Based on the original environmental dataset of the target building, generate the temporal evolution feature vector of the target building; further understood as: the temporal evolution feature vector of the target building V3 = (V 1 3, V 2 3), V1 3 represents the intensity of the trend in all the raw temperature data, V 2 3 represents the intensity of the changing trend of all raw humidity data.

[0035] Preferred, V 1 3. Meets the following conditions: .

[0036] Preferred, V 2 3. Meets the following conditions: .

[0037] S24, Based on the original environmental dataset of the target building, generate the thermal comfort feature vector of the target building; further understood as: the thermal comfort feature vector of the target building V4 = (V 1 4, V 2 4, V 3 4), V 1 4 is the mean of the prediction and evaluation voting indicators for all monitored areas, V 2 4 represents the maximum percentage of predicted areas that do not meet the forecast. 3 4 represents the average absolute deviation between the temperature values ​​of all monitored areas and the predicted comfort temperature setpoint.

[0038] Preferred, V 1 4. Meets the following conditions: Among them, PMV i It is the PMV value of the i-th monitoring area, where the PMV value is calculated by combining the mean of the original temperature data list and the mean of the original humidity data list with the standard metabolic rate and clothing thermal resistance; those skilled in the art know that the PMV value is obtained by a pre-set PMV calculation model, for example, by inputting the known mean of the original temperature data list and the mean of the original humidity data list, the standard metabolic rate and the clothing thermal resistance into the pre-set PMV calculation model.

[0039] Preferred, V 2 4. Meets the following conditions: Wherein, PMV0 is the average PMV value of all monitored areas, ε1 is the first parameter, and ε2 is the second parameter; preferably, ε1=0.03353 and ε2=0.2179.

[0040] Preferred, V 3 4. Meets the following conditions: Where △A is the predicted comfort temperature setpoint, A 0i It is the mean of the original temperature data list for the i-th monitoring area.

[0041] S24, the spatial distribution feature vector, spatial correlation feature vector, temporal evolution feature vector, and thermal comfort feature vector of the target building are concatenated to obtain the target area environmental state vector of the target building; further understood as: the target area environmental state vector V0 of the target building = (V 1 1, V 2 1, V 3 1, V 4 1, V 1 2, V 2 2, V 1 3, V 2 3, V 1 4, V 2 4, V 3 4).

[0042] By extracting multi-dimensional features (spatial distribution, correlation, temporal evolution, and thermal comfort) from environmental data (temperature and humidity) of multiple monitored areas, a high-dimensional, structured environmental state representation with spatiotemporal awareness capabilities was constructed. This enables the system to not only reflect the current temperature and humidity status of each area, but also capture the heat transfer relationships between areas, historical trends, and human comfort needs, providing precise input for subsequent regionalized and differentiated regulation.

[0043] S3, Generate the target device state vector of the target building based on the original device operation dataset of the target building.

[0044] Specifically, step S3 also includes the following steps: S31, Based on the original equipment operation dataset of the target building, generate the equipment operation feature vector of the target building; further understood as: the equipment operation feature vector E1 of the target building = (E eff E energy H health ), where E eff It is a characteristic of equipment operating efficiency, E energy It is the energy consumption characteristic of the equipment, H health It is a characteristic of equipment health.

[0045] Preferred, E eff The following conditions must be met: .

[0046] Preferred, E energy The following conditions must be met: TOP is the cumulative running time of the target device.

[0047] Preferably, H health The following conditions must be met: , where △Q 1 j It refers to the number of intermediate device temperatures that are greater than the preset overheating threshold in the intermediate device temperature list corresponding to the target device within the j-th subset of the original device operation data set of the target building.

[0048] S32, Based on the equipment operation feature vector of the target building, generate the equipment mode feature vector of the target building; further understood as: the equipment mode feature vector of the target building E2 = (E 1 2, E 2 2, E 3 2), E 1 2 is the eigenvalue of the efficient mode, E 2 2 is the eigenvalue of the inefficient mode, E 3 2 represents the characteristic value of the abnormal mode; among them, the characteristic value of the efficient mode is 0 or 1, where 1 indicates an efficient mode and 0 indicates an inefficient mode; the characteristic value of the inefficient mode is 0 or 1, where 1 indicates an inefficient mode and 0 indicates a non-inefficient mode; the characteristic value of the abnormal mode is 0 or 1, where 1 indicates an abnormal mode and 0 indicates a non-abnormal mode; E eff ≥ηhigh and H health When the value is ≥0.9, it is in high-efficiency mode; E eff <ηlow and H health When the value is less than 0.7, it is an abnormal mode; otherwise, when the aforementioned conditions are not met, it is an inefficient mode; ηhigh is the preset upper limit threshold for efficiency, and ηlow is the preset lower limit threshold for efficiency.

[0049] S33, Based on the original equipment operation dataset of the target building, generate the equipment coordination feature vector of the target building; further understood as: the equipment coordination feature vector of the target building E3 = (E 1 3, E 2 3, E 3 3), E 1 3 represents load balancing, E 2 3 represents temperature uniformity, E 3 3 represents the power synchronization rate.

[0050] Preferred, E 1 3. Meets the following conditions: , where max() is the maximum value function and std() is the standard deviation function.

[0051] Preferred, E 2 3. Meets the following conditions: , where max(B 1 j ) refers to B 1 j The maximum value.

[0052] Preferred, E 3 3. Meets the following conditions: , where corr() is the Pearson coefficient.

[0053] S34, the equipment operation feature vector, the equipment mode feature vector, and the equipment coordination feature vector of the target building are weighted and concatenated to obtain the target equipment state vector of the target building; further understood as: the target equipment state vector E of the target building. 0 =((α1 / 3)×E eff (α1 / 3)×E energy (α1 / 3)×H health (α2 / 3)×E 1 2, (α² / 3) × E 2 2, (α² / 3) × E 3 2, (α³ / 3) × E 1 3, (α3 / 3) × E 2 3, (α3 / 3) × E 3 3), where α1+α2+α3=1, α1 is the equipment operation weight, α2 is the equipment mode weight, and α3 is the equipment collaboration weight.

[0054] Preferably, α1 is 0.5, α2 is 0.3, and α3 is 0.2.

[0055] By analyzing equipment operating data (temperature, power, and heat generation), a multi-dimensional equipment state vector is constructed, encompassing equipment operating efficiency, energy consumption characteristics, health status, operating modes, and inter-equipment collaboration. This enables the system to assess the health status and energy efficiency level of each piece of equipment in real time, identify abnormal or inefficient operating modes, and provide a basis for load balancing and collaborative control among equipment, thereby extending equipment lifespan and improving overall system energy efficiency.

[0056] S4. Generate the target external environment state vector of the target building based on the original external environment dataset of the target building.

[0057] Specifically, step S4 also includes the following steps: S41, based on the original external environment dataset of the target building, extract the basic feature vector of the target building's external environment; further understood as: the basic feature vector of the target building's external environment F1 = (F 11, F 2 1, F 3 1, F 4 1), F 1 1 represents the outdoor temperature characteristic, F 2 1 represents outdoor humidity characteristics, F 3 1 represents the light intensity characteristic, F 4 1 represents the outdoor environmental parameter value.

[0058] Preferred, F 1 1. Meets the following conditions: .

[0059] Preferred, F 2 1. Meets the following conditions: .

[0060] Preferred, F 3 1. Meets the following conditions: .

[0061] Preferred, F 4 1. Meets the following conditions: Where γ1 is the outdoor temperature weight, γ2 is the outdoor humidity weight, and γ3 is the light intensity weight.

[0062] S42, based on the original external environment dataset of the target building, construct the external environment evolution feature vector of the target building; further understood as: F2 = (F 11 2, F 12 2, F 13 2, F 21 2, F 22 2, F 23 2, F 31 2, F 32 2, F 33 2), F 11 2 represents the intensity of the trend change in outdoor temperature, F 12 2 represents the intensity of the periodic change in outdoor temperature, F 13 2 represents the intensity of outdoor temperature fluctuations; F 21 2 represents the intensity of the trend change in outdoor humidity, F 22 2 represents the intensity of the periodic variation in outdoor humidity, F 23 2 represents the intensity of outdoor humidity fluctuations, F 31 2 represents the trend change intensity of light intensity, F 32 2 represents the periodic variation intensity of light intensity, F. 33 2 represents the fluctuation intensity of light intensity.

[0063] Preferred, F11 2. Meets the following conditions: , where t 1 r1 It is C 1r1 The corresponding timestamp.

[0064] Preferred, F 12 2. Meets the following conditions: Where K1 is the number of outdoor temperature samples taken daily.

[0065] Preferred, F 13 2. Meets the following conditions: , where t 1 0 is the average of the sampling time corresponding to C1.

[0066] Preferred, F 21 2. Meets the following conditions: , where t r2 It is C 2r2 The corresponding timestamp.

[0067] Preferred, F 22 2. Meets the following conditions: Where K2 is the number of outdoor humidity samples taken daily.

[0068] Preferred, F 23 2. Meets the following conditions: , where t 2 0 represents the average sampling time corresponding to C2.

[0069] Preferred, F 31 2. Meets the following conditions: , where t r3 It is C 3r3 The corresponding timestamp.

[0070] Preferred, F 32 2. Meets the following conditions: Where K3 is the number of natural sunlight intensity samples.

[0071] Preferred, F 33 2. Meets the following conditions: , where t 3 0 represents the average sampling time corresponding to C3.

[0072] S43, Based on the original external environment dataset of the target building, generate the external environment influence weight vector of the target building; further understood as: the external environment influence weight vector of the target building F3 = (F 1 3, F 2 3, F 3 3), F 1 3 represents the weight of the influence of outdoor temperature, F 2 3 represents the weighting of outdoor humidity, F 3 3 is the weight of light intensity.

[0073] Preferred, F 1 3. Meets the following conditions: .

[0074] Preferred, F 2 3. Meets the following conditions: .

[0075] Preferred, F 3 3. Meets the following conditions: .

[0076] Among them, G T This is the influence value of outdoor temperature, G H This is the value influenced by outdoor humidity, G. R It is the influence value of light intensity.

[0077] Furthermore, G T The following conditions must be met: .

[0078] Furthermore, G H The following conditions must be met: .

[0079] Furthermore, G R The following conditions must be met: .

[0080] Wherein, β1 is the preset outdoor temperature weighting value, β2 is the preset outdoor humidity weighting value, and β3 is the preset light intensity weighting value.

[0081] Furthermore, β1+β2+β3=1; preferably, β1=0.4, β2=0.3, β3=0.3.

[0082] Furthermore, D T It is the current deviation of the outdoor temperature, D H It is the current deviation of outdoor humidity, D RThis refers to the current deviation of light intensity; methods known to those skilled in the art for determining the current deviation of any outdoor temperature, outdoor humidity, and light intensity in the prior art will not be elaborated here. For example, the current deviation is a variance value.

[0083] Furthermore, R T It is the rate of change of outdoor temperature, R H It is the rate of change of outdoor humidity, R R It is the rate of change of light intensity. Those skilled in the art are familiar with methods for understanding the rates of change of any outdoor temperature, outdoor humidity, and light intensity in the prior art, which will not be elaborated here; for example, the rate of change is a slope.

[0084] Furthermore, L T Historical correlation of outdoor temperature, L H Historical correlation of outdoor humidity, L R It is the historical correlation of light intensity; the methods by which those skilled in the art know the historical correlation of any outdoor temperature, outdoor humidity and light intensity in the prior art will not be elaborated here; for example, the historical correlation is the Pearson coefficient.

[0085] S44, the basic feature vector of the external environment of the target building, the evolution feature vector of the external environment of the target building, and the influence weight vector of the external environment of the target building are fused to obtain the target external environment state vector of the target building; further, the target external environment state vector F of the target building... 0 =((θ1 / 4)×F 1 1, (θ1 / 4)×F 2 1, (θ1 / 4)×F 3 1, (θ1 / 4)×F 4 1, (θ² / 9) × F 11 2, (θ² / 9) × F 12 2, (θ² / 9) × F 13 2, (θ² / 9) × F 21 2, (θ² / 9) × F 22 2, (θ² / 9) × F 23 2, (θ² / 9) × F 31 2, (θ² / 9) × F 32 2, (θ² / 9) × F 33 2, (θ³ / 3) × F 1 3, (θ³ / 3) × F 2 3, (θ³ / 3) × F 33), where θ1+θ2+θ3=1, θ1 is the weight of the outdoor temperature feature, θ2 is the weight of the outdoor humidity feature, and θ3 is the weight of the outdoor environmental influence feature.

[0086] Preferably, θ1 is 0.4, θ2 is 0.3, and θ1 is 0.3.

[0087] By extracting features and performing weight analysis on external environmental data (outdoor temperature and humidity, light intensity), a state vector quantifying the impact of external climate on building heat load was constructed. This enables the system to proactively sense and predict the impact of external environmental changes on the indoor environment, achieving forward-looking compensation control and enhancing the system's adaptability and stability to climate disturbances.

[0088] S5, based on the target area heating state vector of the target building, the target equipment state vector of the target building, and the target external environment state vector of the target building, generate the priority of the target area state of the target building, so as to generate control commands for the target building based on the priority of the target area state of the target building; further, the priority Y of the target area state of the target building meets the following conditions: Wherein, φ1 is the heating state weight, φ2 is the equipment state weight, and φ3 is the external environment state weight. Furthermore, those skilled in the art know that for each feature value in the target area heating state vector, the target equipment state vector, and the target external environment state vector of the target building input in Y that is not within the range [0, 1], normalization processing is required to ensure that the feature value falls within the range [0, 1]. Those skilled in the art are familiar with any normalization algorithm, which will not be elaborated upon here.

[0089] This embodiment utilizes multi-source data fusion and feature-based modeling to collaboratively optimize the control efficiency of building HVAC systems. First, the method systematically collects and preprocesses data on the building's internal environment, equipment operating status, and external climate, providing a high-quality data foundation for accurate decision-making. By extracting and fusing multi-dimensional features such as spatial distribution, correlations, temporal evolution, and thermal comfort of the target area, a regional environmental state vector is constructed that comprehensively reflects the spatiotemporal dynamics of the building's thermal environment. Simultaneously, by analyzing equipment operating efficiency, energy consumption, health status, operating modes, and inter-equipment coordination indicators, an equipment state vector is generated that quantitatively characterizes the real-time performance and health level of the equipment. Furthermore, by analyzing the fundamental characteristics and evolution patterns of external environmental data and calculating their dynamic influence weights, an environmental state vector is generated that can proactively quantify and compensate for external climate disturbances. Finally, by fusing the above three types of state vectors and defining a priority function for the target area's state, global collaborative control is achieved, simultaneously optimizing equipment energy efficiency and healthy operation while proactively offsetting external disturbances, all while ensuring regional thermal comfort. This invention effectively improves the accuracy, overall energy efficiency, and operational reliability of HVAC system control in complex building environments.

[0090] Embodiments of the present invention also provide a non-transitory computer-readable storage medium, which can be disposed in an electronic device to store at least one instruction or at least one program related to implementing a method in the method embodiments, wherein the at least one instruction or the at least one program is loaded and executed by the processor to implement the intelligent control method for building equipment provided in the above embodiments.

[0091] Embodiments of the present invention also provide an electronic device, including a processor and the aforementioned non-transitory computer-readable storage medium.

[0092] While specific embodiments of the invention have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of the invention. It should also be understood that various modifications can be made to the embodiments without departing from the scope and spirit of the invention. The scope of the invention is defined by the appended claims.

Claims

1. A method for intelligent control of building equipment, characterized in that, Includes the following steps: Based on the original environmental dataset, the original equipment operation dataset, and the original external environment dataset of the target building, obtain the target area state vector, the target equipment state vector, and the target external environment state vector of the target building. Based on the target area state vector, target device state vector, and target external environment state vector of the target building, a priority of the target area state of the target building is generated. This priority is used to generate intelligent control commands for the target building. The priority Y of the target area state of the target building satisfies the following condition: Where φ1 is the regional state weight, φ2 is the equipment state weight, and φ3 is the external environment state weight.

2. The intelligent control method for building equipment according to claim 1, characterized in that, The original environmental dataset of the target building includes: N1 subsets of original environmental data corresponding to the monitored areas, wherein any subset of original environmental data corresponding to the monitored area includes a list of original humidity and a list of original temperature corresponding to the monitored area. The list of original humidity corresponding to the monitored area includes several collected original humidity values, and the list of original temperature corresponding to the monitored area includes several real-time collected original temperatures.

3. The intelligent control method for building equipment according to claim 1, characterized in that, The raw equipment operation dataset of the target building includes: N2 subsets of raw equipment operation data corresponding to target operating equipment. Each subset of raw equipment operation data corresponding to a target operating equipment includes a list of raw equipment temperatures, a list of raw equipment operating power, and a list of raw equipment heat generation power. The list of raw equipment operating power includes several collected raw equipment temperatures, the list of raw equipment operating power includes several collected raw equipment operating powers, and the list of raw equipment heat generation power includes several collected raw equipment heat generation powers.

4. The intelligent control method for building equipment according to claim 1, characterized in that, The original external environment dataset of the target building includes a list of original outdoor temperatures, a list of original outdoor humidity, and a list of original light intensity corresponding to the target building.

5. The intelligent control method for building equipment according to claim 1, characterized in that, The steps of obtaining the target area state vector, target device state vector, and target external environment state vector of the target building based on the target building's original environmental dataset, original equipment operation dataset, and original external environment dataset also include the following steps: Based on the original environmental dataset of the target building, generate a spatial distribution feature vector of the target building; Based on the original environmental dataset of the target building, generate a spatial correlation feature vector of the target building; Based on the original environmental dataset of the target building, generate a temporal evolution feature vector of the target building; Based on the original environmental dataset of the target building, generate a thermal comfort feature vector of the target building; The spatial distribution feature vector, spatial correlation feature vector, temporal evolution feature vector, and thermal comfort feature vector of the target building are concatenated to obtain the target area environmental state vector of the target building.

6. The intelligent control method for building equipment according to claim 1, characterized in that, The steps of obtaining the target area state vector, target device state vector, and target external environment state vector of the target building based on the target building's original environmental dataset, original equipment operation dataset, and original external environment dataset also include the following steps: Based on the original equipment operation dataset of the target building, generate the equipment operation feature vector of the target building; Based on the aforementioned equipment operation feature vector and the equipment historical operation dataset, generate the equipment mode feature vector of the target building; Based on the original equipment operation dataset of the target building, generate the equipment collaboration feature vector of the target building; The target equipment state vector of the target building is obtained by weighted concatenation of the equipment operation feature vector, the equipment mode feature vector, and the equipment coordination feature vector of the target building.

7. The intelligent control method for building equipment according to claim 1, characterized in that, The steps of obtaining the target area state vector, target device state vector, and target external environment state vector of the target building based on the target building's original environmental dataset, original equipment operation dataset, and original external environment dataset also include the following steps: Based on the original external environment dataset of the target building, extract the basic feature vector of the external environment of the target building; Based on the original external environment dataset of the target building, construct the external environment evolution feature vector of the target building; Based on the original external environment dataset of the target building, generate the external environment influence weight vector of the target building; The target building's external environment basic feature vector, external environment evolution feature vector, and external environment influence weight vector are fused together to obtain the target building's external environment state vector.

8. A storage medium storing program instructions, which, when executed, implement the intelligent control method for building equipment as described in any one of claims 1 to 7.

9. An electronic device, comprising a processor and a memory, wherein the memory stores program instructions, and the processor executes the program instructions to implement the intelligent control method for building equipment as described in any one of claims 1 to 7.

Images