Intelligent building integrated control method and system based on cloud computing

Abstract

The invention discloses an intelligent building integrated control method and system based on cloud computing, relates to the technical field of building integrated control, and adopts an environment sensing module to collect building environment data in real time, ensure the accuracy and timeliness of the data and provide basic data support for intelligent regulation and control. And through the energy prediction and scheduling module, the future power load Ppr and the personnel density Dus are dynamically predicted based on the time sequence prediction model and the regression model, the energy supply is optimized, and the unnecessary energy consumption is reduced. And a user behavior analysis module is adopted to accurately identify the peak period, the equipment use frequency and the energy consumption mode in the building based on the personnel density and the equipment use data, so that the energy waste is reduced. The operation mode of the air conditioner heating and ventilating system is dynamically adjusted, continuous operation of high-power refrigerating or heating equipment in the low-load period is avoided, and therefore unnecessary energy consumption is reduced. Intelligent light adjustment is realized, illumination brightness is adjusted according to personnel activity conditions, and long-term invalid operation of the illumination system is avoided.

Description

Technical Field

[0001] This invention relates to the field of building integrated control technology, specifically to a cloud computing-based intelligent building integrated control method and system. Background Technology

[0002] Energy optimization is a core component of intelligent building management systems, encompassing electricity consumption control, heating and cooling management, intelligent lighting control, and water resource optimization. These systems utilize cloud computing and AI technologies to process data, dynamically adjust equipment operation, and improve energy efficiency. Intelligent energy management optimization is a key module of this system, primarily used to predict energy consumption trends, optimize equipment scheduling, reduce energy waste, and ultimately achieve the goals of intelligent and energy-efficient building operation.

[0003] Currently, most buildings rely on fixed rules or manual adjustments for energy management, lacking dynamic and intelligent control methods. Many commercial and office buildings' air conditioning, heating, and lighting systems often operate according to fixed schedules rather than adjusting to real-time demand, resulting in high energy consumption and low energy efficiency even when there is less foot traffic.

[0004] The primary cause of these problems lies in the lack of real-time data analysis capabilities in traditional building management systems, leading to lagging or ineffective energy dispatch strategies. Because peak electricity consumption times cannot be accurately predicted, buildings continue to operate at high power during off-peak hours, resulting in significant energy waste. Simultaneously, because user habit data is not effectively utilized, the system cannot optimize the usage patterns of equipment such as heating, lighting, and elevators based on actual needs, leading to low equipment operating efficiency and even causing excessive energy consumption or energy shortages. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a cloud computing-based intelligent building integrated control method and system, which solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution: a cloud computing-based intelligent building integrated control system, comprising an environmental perception module, an energy prediction and scheduling module, a user behavior analysis module, an adaptive energy consumption control module, a feedback and self-optimization module, and an energy efficiency assessment and reporting module;

[0007] The environmental perception module collects environmental data of the smart building in real time through sensors, including atmospheric radiation intensity Ra, temperature gradient ∇T, energy consumption density Ed, and wind speed Wou, fits them into the original data set Q, and performs preprocessing to obtain the environmental dataset QH.

[0008] The energy forecasting and scheduling module uses the acquired environmental dataset QH to predict energy demand for different time periods in the future through time series forecasting models and regression models, and obtains the predicted power load value Ppre and the predicted population density value Dus.

[0009] The user behavior analysis module analyzes peak periods of pedestrian traffic and equipment usage habits inside the building based on the predicted power load value Ppre and the predicted personnel density value Dus, and obtains user behavior patterns.

[0010] The adaptive energy consumption control module adjusts the operating mode of the equipment based on user behavior patterns and energy consumption density Ed, and obtains the energy regulation consumption density nEd.

[0011] The feedback and self-optimization module obtains the energy saving rate ηeff by comparing the energy regulation consumption density nEd with the energy consumption density Ed.

[0012] The energy efficiency assessment and reporting module summarizes the parameters and energy saving rate ηeff of the adjusted equipment operating mode and generates a visual report.

[0013] Preferably, the environmental perception module includes a data acquisition unit and a data preprocessing unit;

[0014] The data acquisition unit collects environmental data of the smart building through radiation sensors, temperature sensors, smart meters, and wind speed sensors. Specifically, it collects atmospheric radiation intensity Ra through radiation sensors, temperature gradient ∇T through temperature sensors, energy consumption density Ed through smart meters, and wind speed Wou through wind speed sensors, and fits them into the original data set Q.

[0015] The temperature gradient ∇T is obtained using the following formula:

[0016] ;

[0017] In the formula, Tin represents the indoor temperature, Tou represents the outdoor temperature, η represents the radiation absorptivity of the building's exterior walls, and Cp represents the specific heat capacity. Indicates air density;

[0018] Energy consumption density Ed is obtained using the following formula;

[0019] ;

[0020] In the formula, Pe represents the power consumption of lighting equipment, Pha represents the energy consumption of air conditioning, heating and ventilation, Pw represents the power consumption of elevator system, and A represents the total building area.

[0021] The data preprocessing unit cleans and standardizes the acquired raw data set Q to obtain the environmental dataset QH;

[0022] Cleaning includes outlier detection and missing value imputation. Outlier detection is performed by processing the original data set Q using an outlier detection algorithm to remove outliers from Q. Missing value imputation is performed by filling in missing data in the original data set Q using temporal interpolation.

[0023] Standardization is achieved by using Min-Max normalization to unify the dimensions of the data in the original dataset Q to the same dimension.

[0024] Preferably, the energy forecasting and scheduling module uses time series forecasting models and regression models to predict the power load forecast Ppre and the population density forecast Dus for different future periods based on the environmental dataset QH;

[0025] First, the acquired environmental dataset QH is fitted with historical power load data and historical population density data to form a training set, which is then used to train the time series prediction model and the regression model.

[0026] The power load forecast value Ppre is obtained by using a time series forecasting model to predict the power load.

[0027] The electricity load forecast value Ppre is obtained using the following formula:

[0028] ;

[0029] In the formula, Ppre(t) represents the predicted power load at time t, QHi(td) represents the value of the i-th data in the environmental dataset QH at time (td), αi represents the time series prediction model coefficient of the i-th data, C represents the error compensation term, and n represents the total number of data.

[0030] The predicted population density, Dus, is obtained by using a regression model to predict population density.

[0031] The predicted population density value Dus is obtained using the following formula:

[0032] ;

[0033] In the formula, Dus(t) represents the predicted population density at time t, m represents the total number of time points, and Dus(td) represents the population density at time (td). This represents the regression weight value at the j-th time point. Let B represent the regression model coefficient for the i-th data point, and let B represent the error term.

[0034] Preferably, the user behavior analysis module includes a peak period identification unit and a device usage unit;

[0035] The peak period identification unit calculates the comprehensive load index Iode inside the building based on the obtained power load forecast value Ppre and personnel density forecast value Dus.

[0036] The composite load index Iode is obtained using the following formula:

[0037] ;

[0038] In the formula, These represent the preset weight values ​​for the predicted power load Ppre and the predicted population density Dus, respectively.

[0039] By using a time series clustering algorithm, the load index change rate ΔIod between adjacent time points is obtained, and the peak index IPE is calculated and compared with the preset peak threshold TIP to identify peak times.

[0040] The load index change rate ΔIod is obtained by the difference between the comprehensive load index Iode(t) at time t and the comprehensive load index Iode(t-1) at time t-1.

[0041] The Peak Index (IPE) is obtained using the following formula:

[0042] ;

[0043] In the formula, max(ΔIod) represents the peak value of the load index change rate ΔIod;

[0044] Peak hours are obtained through the following methods:

[0045] When the peak index IPE is greater than the peak threshold TIP, it indicates that the peak time has arrived.

[0046] Preferably, the equipment usage unit obtains user behavior patterns, including peak hours, elevator usage frequency, and lighting usage, based on the acquired peak index IPE and the predicted personnel density Dus.

[0047] Peak periods are obtained by combining peak time and dwell time Ts;

[0048] The residence time Ts is obtained using the following formula:

[0049] ;

[0050] In the formula, Dus(t, (ai, Tavg)) represents the population density at the ai-th sampling point at time t, which is corrected based on the average dwell time Tavg, where Tavg represents the average dwell time, and Na represents the total number of sampling points.

[0051] The peak time period is obtained by comparing the obtained residence time Ts with the time threshold HTs.

[0052] When the dwell time Ts > the time threshold HTs, it indicates that the period is at its peak.

[0053] By fitting peak time periods with peak hours, peak periods can be obtained;

[0054] When the dwell time Ts > the time threshold HTs and the peak index IPE > the peak threshold TIP, it indicates that the period is in a peak period.

[0055] Elevator usage frequency is obtained by the ratio of elevator usage count DNE to the fixed cycle TG; elevator usage count DNE is obtained by the following formula:

[0056] ;

[0057] In the formula, HfL(a) represents the floor height of the a-th floor, N represents the total number of floors, and Cee represents the elevator capacity;

[0058] Lighting usage Lon is obtained by comparing the current lighting power with the maximum lighting power in the building. The correlation between lighting usage Lon and the predicted population density Dus (Corr) is used to determine the correlation between lighting and population distribution.

[0059] Correlation (Corr) is obtained using the following formula:

[0060] ;

[0061] In the formula, pDus represents the average value of the predicted population density, and pLon represents the average proportion of light usage;

[0062] The correlation between lighting and personnel distribution was obtained through the following methods:

[0063] When 0 < Correlation (Corr) < 0.3, it indicates that there is no correlation between the matching of lighting and personnel distribution;

[0064] When 0.3 ≤ Correlation < 1, it indicates that the matching between lighting and personnel distribution is correlated.

[0065] Preferably, the adaptive energy consumption control module adjusts the operating mode of the equipment based on the acquired user behavior pattern and energy consumption density Ed, including air conditioning, heating and ventilation adjustment, intelligent lighting adjustment and elevator operation adjustment, and acquires the adjusted air conditioning, heating and ventilation energy consumption nPha, the adjusted lighting energy consumption nPe and the adjusted elevator operation energy consumption nPele.

[0066] Air conditioning, heating, and ventilation regulation includes temperature adjustment strategies and adaptive fan speed control.

[0067] Temperature adjustment strategies are obtained through the following methods:

[0068] When it is during peak hours and the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, the ventilation volume and carbon dioxide concentration are adjusted. After the peak hours, the fan speed is reduced to reduce energy consumption.

[0069] Adaptive wind speed adjustment is obtained through the following methods:

[0070] When the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, and the temperature gradient ∇T is greater than the preset temperature threshold H∇T, then adjust the wind speed Wou to improve air circulation efficiency.

[0071] The adjusted energy consumption (nPha) for air conditioning, heating, and ventilation can be obtained using the following formula.

[0072] ;

[0073] In the formula, C1, C2, and C3 represent the load adjustment coefficients of the air conditioning, heating, and ventilation system;

[0074] Intelligent lighting adjustment is obtained through the following methods:

[0075] If the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, and the lighting usage Lon is less than the preset lighting usage threshold TLon, then increase the lighting brightness.

[0076] When the predicted population density Dus(t) at time t is less than the preset density threshold TDu, the lighting brightness is reduced.

[0077] The adjusted light energy consumption nPe is obtained using the following formula:

[0078] ;

[0079] In the formula, D1 and D2 represent the lighting adjustment coefficients;

[0080] Elevator operation adjustments are obtained through the following methods:

[0081] When the elevator usage frequency exceeds the preset frequency threshold during peak hours, a priority target scheduling method is adopted to control the elevator to run empty and adjust the passenger carrying efficiency.

[0082] When not in peak hours, the elevator is adjusted to an energy-saving mode to reduce unnecessary starts.

[0083] Elevator operating energy consumption nPele is obtained using the following formula:

[0084] ;

[0085] In the formula, HfL(a) represents the floor height of the a-th floor, N represents the total number of floors, Cee represents the elevator capacity, and Efl represents the energy consumption of a single floor elevator operation.

[0086] The energy regulation consumption density nEd is calculated by fitting the adjusted air conditioning heating and ventilation energy consumption nPha, the adjusted lighting energy consumption nPe, and the adjusted elevator operation energy consumption nPele.

[0087] Energy regulation consumption density nEd is obtained using the following formula:

[0088] .

[0089] Preferably, the feedback and self-optimization module includes an execution effect evaluation unit and a model adaptive optimization unit;

[0090] The performance evaluation unit calculates the energy saving rate ηeff based on the obtained energy regulation consumption density nEd and energy consumption density Ed, and judges the effectiveness of the equipment operation mode through the energy saving rate ηeff;

[0091] The energy saving rate ηeff is obtained using the following formula:

[0092] ;

[0093] The validity of the device's operating mode is obtained through the following methods:

[0094] When the energy saving rate ηeff > 20%, it indicates that it is in the first effective level, the adjustment effect of the equipment operation mode is very effective, and it should continue to be used.

[0095] When 20% ≥ energy saving rate ηeff ≥ 5%, it indicates that it is in the second effective level, and the adjustment of the equipment operation mode has a certain effect, but there is still room for improvement.

[0096] When 5% > energy saving rate ηeff > 0, it indicates that the equipment is in the third effective level, the adjustment effect of the equipment operation mode is poor, and the equipment operation mode needs to be readjusted.

[0097] Preferably, the model adaptive optimization unit adjusts the parameters of the time series prediction model and the regression model according to the level of the energy saving rate ηeff;

[0098] When the energy saving rate ηeff is at the third effective level, the gradient descent algorithm is used to adjust the parameters of the time series prediction model and the regression model, including adjusting the time parameter α of the time series prediction model and the regression parameter β of the regression model; obtaining the new time parameter nα and the new regression parameter nβ, and returning the new time parameter nα and the new regression parameter nβ to the time series prediction model and the regression model to recalculate the predicted value of electricity load Ppre and the predicted value of population density Dus;

[0099] The new time parameter nα is obtained using the following formula:

[0100] ;

[0101] In the formula, θ represents the learning rate, and ∂ represents the partial derivative;

[0102] The new regression parameter nβ is obtained using the following formula:

[0103] .

[0104] Preferably, the energy efficiency assessment and reporting module summarizes the parameters of the adjusted equipment operation mode, including energy regulation consumption density nEd, new time parameter nα, and new regression parameter nβ, with the energy saving rate ηeff, and generates a visual report.

[0105] Visualized reports include line charts and pie charts;

[0106] The line graph shows energy consumption over different time periods;

[0107] The pie chart illustrates the contribution of adjusting equipment operating modes to energy conservation.

[0108] A cloud computing-based integrated control method for smart buildings includes the following steps:

[0109] Step 1: The environmental perception module collects environmental data of the smart building in real time through sensors, including atmospheric radiation intensity Ra, temperature gradient ∇T, energy consumption density Ed, and wind speed Wou, fits them into the original data set Q, and performs preprocessing to obtain the environmental dataset QH.

[0110] Step 2: The energy forecasting and scheduling module uses the acquired environmental dataset QH to predict energy demand for different time periods in the future through time series forecasting models and regression models, and obtains the predicted power load value Ppre and the predicted population density value Dus.

[0111] Step 3: The user behavior analysis module analyzes peak periods of pedestrian traffic and equipment usage habits inside the building based on the predicted power load value Ppre and the predicted personnel density value Dus, and obtains user behavior patterns.

[0112] Step 4: The adaptive energy consumption control module adjusts the operating mode of the equipment based on the user's behavior pattern and the energy consumption density Ed, and obtains the energy regulation consumption density nEd.

[0113] Step 5: The feedback and self-optimization module obtains the energy saving rate ηeff by comparing the energy regulation consumption density nEd with the energy consumption density Ed.

[0114] Step Six: The Energy Efficiency Assessment and Reporting module summarizes the parameters and energy saving rate ηeff of the adjusted equipment operating mode and generates a visual report.

[0115] This invention provides a cloud computing-based intelligent building integrated control method and system, which has the following beneficial effects:

[0116] (1) During system operation, the environmental perception module collects building environment data in real time to ensure the accuracy and timeliness of the data, providing basic data support for intelligent control. Through the energy prediction and scheduling module, the future power load Ppr and personnel density Dus are dynamically predicted based on time series prediction models and regression models to optimize energy supply and reduce unnecessary energy consumption. The user behavior analysis module is used to accurately identify peak periods, equipment usage frequency and energy consumption patterns inside the building based on personnel density and equipment usage data, optimize equipment scheduling strategies and reduce energy waste.

[0117] Dynamically adjust the operating mode of the air conditioning, heating, and ventilation systems to avoid continuous operation of high-power cooling or heating equipment during low-load periods, thereby reducing unnecessary energy consumption. Intelligent lighting adjustment adjusts brightness based on occupant activity, ensuring lights turn on when people are present and off when they leave, preventing prolonged ineffective operation of the lighting system. Elevator operation optimization intelligently schedules elevator routes and stops based on passenger flow and peak-hour predictions, reducing empty runs and improving transportation efficiency.

[0118] (2) By combining the comprehensive load index Iode calculation formula with the power load forecast Ppre and the personnel density forecast Dus, a building overall load assessment model is constructed to improve the accuracy of peak period identification. A time series clustering algorithm is used to calculate the load index change rate ΔIod and the peak index IPE to ensure more accurate peak period identification and avoid the misjudgment problem caused by fixed peak period settings in traditional systems. By comparing the peak index IPE with the peak threshold TIP, the specific time point when the building is in the peak period is dynamically determined, making equipment scheduling more flexible.

[0119] A peak-time identification mechanism is employed, combining the Peak Performance Index (IPE) and dwell time (Ts) to dual-determine the true peak periods, avoiding misjudgments caused by relying on a single indicator. This method can accurately identify the occupancy status of people in different building functional areas such as conference rooms, office areas, and shopping malls, optimizing regional energy dispatching and improving energy efficiency.

[0120] (3) Intelligent ventilation is based on the predicted population density Dus. During peak hours, if the population density exceeds the density threshold TDu, the system automatically adjusts the ventilation volume and CO2 concentration to ensure air quality. After the peak period ends, the fan speed is reduced to reduce unnecessary energy consumption.

[0121] The system features adaptive wind speed adjustment. When the predicted personnel density (Dus) exceeds the density threshold (TDu) and the temperature gradient (∇T) exceeds the temperature threshold (H∇T), the system dynamically adjusts the wind speed (Wou) to ensure a balance between comfort and energy efficiency. By dynamically optimizing temperature and wind speed adjustment strategies, it reduces cooling / heating energy consumption by 10%–20%, improving building energy efficiency. If the predicted personnel density (Dus) exceeds the density threshold (TDu) and the lighting usage (Lon) is less than the preset lighting usage threshold (TLon), the system automatically increases lighting brightness to ensure a comfortable working environment. If the predicted personnel density (Dus) exceeds the density threshold (TDu), the system automatically reduces lighting brightness to decrease lighting energy consumption.

[0122] (4) Using the energy saving rate ηeff as the optimization evaluation standard, automatically determine whether the current equipment operation mode has achieved the energy saving target: If the energy saving rate ηeff is at the third effective level, it means that the prediction model has failed to accurately match the actual energy consumption trend and the parameters need to be optimized. If the energy saving rate ηeff is at the second or first level, it means that the existing prediction model is usable and does not need to be significantly adjusted, only continuous monitoring is required.

[0123] The gradient descent algorithm is used to optimize the parameters α of the time series prediction model and β of the regression model, thereby improving the model's predictive ability. Through parameter optimization, the system becomes more adaptable to changes in the building operating environment, and the self-learning ability of the prediction model is enhanced. Attached Figure Description

[0124] Figure 1 This is a schematic diagram of the block diagram of an intelligent building integrated control system based on cloud computing according to the present invention.

[0125] Figure 2 This is a schematic diagram illustrating the steps of a cloud computing-based intelligent building integrated control method according to the present invention.

[0126] Figure 3 This is a schematic diagram of the model parameter calculation process of the present invention;

[0127] Figure 4 This is a line graph showing the predicted population density of the present invention.

[0128] Figure 5 Line graph for user dwell time analysis and peak period identification in this invention. Detailed Implementation

[0129] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0130] Example 1

[0131] This invention provides a cloud-based intelligent building integrated control system. Please refer to [link / reference]. Figure 1 - Figure 5 It includes an environmental perception module, an energy prediction and scheduling module, a user behavior analysis module, an adaptive energy consumption control module, a feedback and self-optimization module, and an energy efficiency assessment and reporting module;

[0132] The environmental perception module collects environmental data of the smart building in real time through sensors, including atmospheric radiation intensity Ra, temperature gradient ∇T, energy consumption density Ed, and wind speed Wou, fits them into the original data set Q, and performs preprocessing to obtain the environmental dataset QH.

[0133] The energy forecasting and scheduling module uses the acquired environmental dataset QH to predict energy demand for different time periods in the future through time series forecasting models and regression models, and obtains the predicted power load value Ppre and the predicted population density value Dus.

[0134] The user behavior analysis module analyzes peak periods of pedestrian traffic and equipment usage habits inside the building based on the predicted power load value Ppre and the predicted personnel density value Dus, and obtains user behavior patterns.

[0135] The adaptive energy consumption control module adjusts the operating mode of the equipment based on user behavior patterns and energy consumption density Ed, and obtains the energy regulation consumption density nEd.

[0136] The feedback and self-optimization module obtains the energy saving rate ηeff by comparing the energy regulation consumption density nEd with the energy consumption density Ed.

[0137] The energy efficiency assessment and reporting module summarizes the parameters and energy saving rate ηeff of the adjusted equipment operating mode and generates a visual report.

[0138] In this embodiment, an environmental sensing module collects building environment data in real time to ensure data accuracy and timeliness, providing fundamental data support for intelligent control. An energy forecasting and scheduling module dynamically predicts future electricity load (Ppr) and occupancy density (Dus) based on time series forecasting and regression models, optimizing energy supply and reducing unnecessary energy consumption. A user behavior analysis module accurately identifies peak periods, equipment usage frequency, and energy consumption patterns within the building based on occupancy density and equipment usage data, optimizing equipment scheduling strategies and reducing energy waste.

[0139] Dynamically adjust the operating mode of the air conditioning, heating, and ventilation systems to avoid continuous operation of high-power cooling or heating equipment during low-load periods, thereby reducing unnecessary energy consumption. Intelligent lighting adjustment adjusts brightness based on occupant activity, ensuring lights turn on when people are present and off when they leave, preventing prolonged ineffective operation of the lighting system. Elevator operation optimization intelligently schedules elevator routes and stops based on passenger flow and peak-hour predictions, reducing empty runs and improving transportation efficiency.

[0140] The system employs a feedback and self-optimization module to calculate the energy saving rate ηeff based on a comparative analysis of energy regulation consumption density nEd and energy consumption density Ed, ensuring the effectiveness of the optimization strategy. The energy efficiency assessment and reporting module summarizes and analyzes energy consumption before and after optimization, generating visual reports to provide data support and assist management in decision-making.

[0141] By optimizing building energy dispatch strategies through intelligent prediction and adaptive control, energy utilization efficiency can be improved by 15% to 30%, effectively reducing building energy consumption. The use of cloud computing and AI algorithms automates and intelligentizes the energy management process, reducing reliance on manual control and improving management efficiency.

[0142] Example 2

[0143] This embodiment is an explanation based on Embodiment 1. Please refer to it. Figure 1 and Figure 4 Specifically: the environmental perception module includes a data acquisition unit and a data preprocessing unit;

[0144] The data acquisition unit collects environmental data of the smart building through radiation sensors, temperature sensors, smart meters, and wind speed sensors. Specifically, it collects atmospheric radiation intensity Ra through radiation sensors, temperature gradient ∇T through temperature sensors, energy consumption density Ed through smart meters, and wind speed Wou through wind speed sensors, and fits them into the original data set Q.

[0145] The temperature gradient ∇T is obtained using the following formula:

[0146] ;

[0147] In the formula, Tin represents the indoor temperature, Tou represents the outdoor temperature, η represents the radiation absorptivity of the building's exterior walls, and Cp represents the specific heat capacity. Indicates air density;

[0148] Energy consumption density Ed is obtained using the following formula;

[0149] ;

[0150] In the formula, Pe represents the power consumption of lighting equipment, Pha represents the energy consumption of air conditioning, heating and ventilation, Pw represents the power consumption of elevator system, and A represents the total building area.

[0151] The data preprocessing unit cleans and standardizes the acquired raw data set Q to obtain the environmental dataset QH;

[0152] Cleaning includes outlier detection and missing value imputation. Outlier detection is performed by processing the original data set Q using an outlier detection algorithm to remove outliers from Q. Missing value imputation is performed by filling in missing data in the original data set Q using temporal interpolation.

[0153] Standardization is achieved by using Min-Max normalization to unify the dimensions of the data in the original dataset Q to the same dimension.

[0154] The energy forecasting and dispatching module uses time series forecasting models and regression models to predict the power load forecast Ppre and the population density forecast Dus for different periods in the future, based on the environmental dataset QH.

[0155] First, the acquired environmental dataset QH is fitted with historical power load data and historical population density data to form a training set, which is then used to train the time series prediction model and the regression model.

[0156] The power load forecast value Ppre is obtained by using a time series forecasting model to predict the power load.

[0157] The electricity load forecast value Ppre is obtained using the following formula:

[0158] ;

[0159] In the formula, Ppre(t) represents the predicted power load at time t, QHi(td) represents the value of the i-th data in the environmental dataset QH at time (td), αi represents the time series prediction model coefficient of the i-th data, C represents the error compensation term, and n represents the total number of data.

[0160] The predicted population density, Dus, is obtained by using a regression model to predict population density.

[0161] The predicted population density value Dus is obtained using the following formula:

[0162] ;

[0163] In the formula, Dus(t) represents the predicted population density at time t, m represents the total number of time points, and Dus(td) represents the population density at time (td). This represents the regression weight value at the j-th time point. Let B represent the regression model coefficient for the i-th data point, and let B represent the error term.

[0164] In this embodiment, an environmental sensing module is employed to collect real-time environmental data of the building through radiation sensors, temperature sensors, smart meters, and wind speed sensors, improving the real-time performance and accuracy of the data. By using a temperature gradient calculation formula, combined with the building's exterior wall radiation absorptivity, air specific heat capacity, and air density, the indoor-outdoor temperature difference is accurately estimated, optimizing the cooling / heating regulation strategy of the air conditioning, heating, and ventilation system. Furthermore, by using an energy consumption density calculation formula, the energy consumption per unit area of ​​the building is calculated by integrating the energy consumption of lighting, elevators, and the air conditioning, heating, and ventilation systems, providing data support for subsequent energy consumption optimization.

[0165] Outlier detection algorithms automatically remove noise and abnormal data from sensor data, preventing incorrect system adjustments to equipment operation due to sensor malfunctions. Time-series interpolation fills in missing data, improving data integrity, making time-series prediction model training more stable, and enhancing prediction accuracy. Min-Max normalization is used to standardize the data, avoiding the impact of data dimensionality differences on the prediction model and improving algorithm computational stability.

[0166] By employing time-series forecasting models based on environmental datasets (QH) and historical load data, power load is predicted in advance, allowing for optimized energy dispatch and avoiding unnecessary equipment operation. Furthermore, by using regression models combined with historical population density data, future population movement trends are accurately predicted, enabling air conditioning, heating, ventilation, elevators, lighting, and other equipment to adapt to usage demands in advance and optimize operational efficiency.

[0167] By predicting electricity load (Ppre), energy supply can be optimized in advance to avoid maintaining high power operation during off-peak hours, reducing energy waste and improving the overall energy efficiency of buildings. By predicting occupancy density (Dus), air conditioning, heating, ventilation, elevators, and lighting systems can be optimized. Through precise forecasting and intelligent control, building energy dispatch strategies can be optimized, improving energy utilization efficiency by 15%–30% and effectively reducing building energy consumption. This reduces human intervention, improves management efficiency, and automatically optimizes dispatch strategies, minimizing human error and lowering operation and maintenance costs. Finally, it enhances building environmental friendliness, reduces carbon emissions, and meets the requirements of green building and sustainable development.

[0168] Example 3

[0169] This embodiment is an explanation based on Embodiment 2. Please refer to it. Figure 1 and Figure 5 Specifically: the user behavior analysis module includes a peak period identification unit and a device usage unit;

[0170] The peak period identification unit calculates the comprehensive load index Iode inside the building based on the obtained power load forecast value Ppre and personnel density forecast value Dus.

[0171] The composite load index Iode is obtained using the following formula:

[0172] ;

[0173] In the formula, These represent the preset weight values ​​for the predicted power load Ppre and the predicted population density Dus, respectively.

[0174] By using a time series clustering algorithm, the load index change rate ΔIod between adjacent time points is obtained, and the peak index IPE is calculated and compared with the preset peak threshold TIP to identify peak times.

[0175] The load index change rate ΔIod is obtained by the difference between the comprehensive load index Iode(t) at time t and the comprehensive load index Iode(t-1) at time t-1.

[0176] The Peak Index (IPE) is obtained using the following formula:

[0177] ;

[0178] In the formula, max(ΔIod) represents the peak value of the load index change rate ΔIod;

[0179] Peak hours are obtained through the following methods:

[0180] When the peak index IPE is greater than the peak threshold TIP, it indicates that the peak time has arrived.

[0181] The equipment usage unit obtains user behavior patterns based on the acquired peak index IPE and the predicted personnel density Dus, including peak hours, elevator usage frequency, and lighting usage Lon.

[0182] Peak periods are obtained by combining peak time and dwell time Ts;

[0183] The residence time Ts is obtained using the following formula:

[0184] ;

[0185] In the formula, Dus(t, (ai, Tavg)) represents the population density at the ai-th sampling point at time t, which is corrected based on the average dwell time Tavg, where Tavg represents the average dwell time, and Na represents the total number of sampling points.

[0186] The peak time period is obtained by comparing the obtained residence time Ts with the time threshold HTs.

[0187] When the dwell time Ts > the time threshold HTs, it indicates that the period is at its peak.

[0188] By fitting peak time periods with peak hours, peak periods can be obtained;

[0189] When the dwell time Ts > the time threshold HTs and the peak index IPE > the peak threshold TIP, it indicates that the period is in a peak period.

[0190] Elevator usage frequency is obtained by the ratio of elevator usage count DNE to the fixed cycle TG; elevator usage count DNE is obtained by the following formula:

[0191] ;

[0192] In the formula, HfL(a) represents the floor height of the a-th floor, N represents the total number of floors, and Cee represents the elevator capacity;

[0193] Lighting usage Lon is obtained by comparing the current lighting power with the maximum lighting power in the building. The correlation between lighting usage Lon and the predicted population density Dus (Corr) is used to determine the correlation between lighting and population distribution.

[0194] Correlation (Corr) is obtained using the following formula:

[0195] ;

[0196] In the formula, pDus represents the average value of the predicted population density, and pLon represents the average proportion of light usage;

[0197] The correlation between lighting and personnel distribution was obtained through the following methods:

[0198] When 0 < Correlation (Corr) < 0.3, it indicates that there is no correlation between the matching of lighting and personnel distribution;

[0199] When 0.3 ≤ Correlation < 1, it indicates that the matching between lighting and personnel distribution is correlated.

[0200] In this embodiment, a comprehensive load index (Iode) calculation formula is used, combined with power load forecasting (Ppre) and population density forecasting (Dus), to construct a building-wide load assessment model, improving the accuracy of peak period identification. A time-series clustering algorithm is employed to calculate the load index change rate (ΔIod) and peak index (IPE), ensuring more accurate peak period identification and avoiding misjudgments caused by fixed peak time settings in traditional systems. By comparing the peak index (IPE) with the peak threshold (TIP), the specific time point when the building is in a peak period is dynamically determined, making equipment scheduling more flexible.

[0201] A peak-hour identification mechanism is adopted, combining the peak index IPE and dwell time Ts to dual-determine the true peak period, avoiding misjudgments caused by a single indicator. This method can accurately identify the occupancy status of people in different building functional areas such as conference rooms, office areas, and shopping malls, optimizing regional energy scheduling and improving energy-saving effects. An elevator usage frequency calculation method is used, based on the number of elevator uses DNE within a fixed period TG and the floor height HfL(a) of the a-th floor, to accurately assess the elevator operating load. Elevator scheduling is adjusted in advance during peak hours to reduce waiting time and improve transportation efficiency. An energy-saving mode is adopted during off-peak hours to reduce idle operation and lower energy consumption.

[0202] The Lon algorithm calculates lighting usage and dynamically adjusts lighting brightness based on the ratio of current lighting power to the building's maximum lighting power, thereby improving energy efficiency.

[0203] A matching analysis of lighting and population distribution is employed, using a correlation coefficient (CorrCorrCorr) to calculate the degree of matching between lighting usage and population density: 0 < Corr < 0.3: Lighting and population distribution are mismatched, suggesting optimization of the lighting scheduling strategy to reduce ineffective lighting. 0.3 ≤ Corr < 1: Lighting and population distribution are matched, indicating that the current lighting control strategy is relatively reasonable. Compared to traditional time-based dimming or fixed sensor dimming modes, this optimization strategy can more accurately adapt to the actual scene, reduce unnecessary power consumption, and improve user comfort.

[0204] By optimizing equipment operation modes, building energy management efficiency can be improved: reducing energy waste by 15%–30%, lowering electricity costs, and increasing economic benefits. It also extends equipment lifespan, reduces maintenance costs, and minimizes wear and tear caused by excessive equipment operation. Furthermore, it reduces carbon emissions, meeting the requirements of green building and sustainable development.

[0205] Example 4

[0206] This embodiment is an explanation based on Embodiment 3. Please refer to it. Figure 1 Specifically: The adaptive energy consumption control module adjusts the operating mode of the equipment based on the acquired user behavior patterns and energy consumption density Ed, including air conditioning, heating and ventilation adjustment, intelligent lighting adjustment and elevator operation adjustment, and acquires the adjusted air conditioning, heating and ventilation energy consumption nPha, the adjusted lighting energy consumption nPe and the adjusted elevator operation energy consumption nPele.

[0207] Air conditioning, heating, and ventilation regulation includes temperature adjustment strategies and adaptive fan speed control.

[0208] Temperature adjustment strategies are obtained through the following methods:

[0209] When it is during peak hours and the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, the ventilation volume and carbon dioxide concentration are adjusted, and the fan speed is reduced after the peak hours.

[0210] Adaptive wind speed adjustment is obtained through the following methods:

[0211] When the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, and the temperature gradient ∇T is greater than the preset temperature threshold H∇T, then the wind speed Wou is adjusted.

[0212] The adjusted energy consumption (nPha) for air conditioning, heating, and ventilation can be obtained using the following formula.

[0213] ;

[0214] In the formula, C1, C2, and C3 represent the load adjustment coefficients of the air conditioning, heating, and ventilation system;

[0215] Intelligent lighting adjustment is obtained through the following methods:

[0216] If the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, and the lighting usage Lon is less than the preset lighting usage threshold TLon, then increase the lighting brightness.

[0217] When the predicted population density Dus(t) at time t is less than the preset density threshold TDu, the lighting brightness is reduced.

[0218] The adjusted light energy consumption nPe is obtained using the following formula:

[0219] ;

[0220] In the formula, D1 and D2 represent the lighting adjustment coefficients;

[0221] Elevator operation adjustments are obtained through the following methods:

[0222] When the elevator usage frequency exceeds the preset frequency threshold during peak hours, a priority target scheduling method is adopted to control the elevator to run empty and adjust the passenger carrying efficiency.

[0223] When it is not during peak hours, the elevator is adjusted in energy-saving mode;

[0224] Elevator operating energy consumption nPele is obtained using the following formula:

[0225] ;

[0226] In the formula, HfL(a) represents the floor height of the a-th floor, N represents the total number of floors, Cee represents the elevator capacity, and Efl represents the energy consumption of a single floor elevator operation.

[0227] The energy regulation consumption density nEd is calculated by fitting the adjusted air conditioning heating and ventilation energy consumption nPha, the adjusted lighting energy consumption nPe, and the adjusted elevator operation energy consumption nPele.

[0228] Energy regulation consumption density nEd is obtained using the following formula:

[0229] .

[0230] The feedback and self-optimization module includes an execution performance evaluation unit and a model adaptive optimization unit;

[0231] The performance evaluation unit calculates the energy saving rate ηeff based on the obtained energy regulation consumption density nEd and energy consumption density Ed, and judges the effectiveness of the equipment operation mode through the energy saving rate ηeff;

[0232] The energy saving rate ηeff is obtained using the following formula:

[0233] ;

[0234] The validity of the device's operating mode is obtained through the following methods:

[0235] When the energy saving rate ηeff > 20%, it indicates that it is in the first effective level;

[0236] When 20% ≥ energy saving rate ηeff ≥ 5%, it indicates that it is in the second effective level;

[0237] When 5% > energy saving rate ηeff > 0, it indicates that it is in the third effective level.

[0238] Specific examples:

[0239] The energy consumption density obtained is Ed = 200;

[0240] Energy regulation consumption density nEd = 150;

[0241] Calculate and obtain the energy saving rate ηeff;

[0242] ;

[0243] The energy saving rate ηeff = 25%, which is in the first effective level;

[0244] Table 1: Building Energy Efficiency Analysis Table

[0245] In this embodiment, intelligent ventilation is implemented based on the predicted population density value Dus. During peak hours, if the population density exceeds the density threshold TDu, the system automatically adjusts the ventilation volume and CO2 concentration to ensure air quality. After the peak period ends, the fan speed is reduced to minimize unnecessary energy consumption.

[0246] The system features adaptive wind speed adjustment. When the predicted personnel density (Dus) exceeds the density threshold (TDu) and the temperature gradient (∇T) exceeds the temperature threshold (H∇T), the system dynamically adjusts the wind speed (Wou) to ensure a balance between comfort and energy efficiency. By dynamically optimizing temperature and wind speed adjustment strategies, it reduces cooling / heating energy consumption by 10%–20%, improving building energy efficiency. If the predicted personnel density (Dus) exceeds the density threshold (TDu) and the lighting usage (Lon) is less than the preset lighting usage threshold (TLon), the system automatically increases lighting brightness to ensure a comfortable working environment. If the predicted personnel density (Dus) exceeds the density threshold (TDu), the system automatically reduces lighting brightness to decrease lighting energy consumption.

[0247] During peak hours: If elevator usage frequency exceeds a set threshold, a priority target scheduling algorithm is used to reduce idle operation and improve carrying efficiency. During off-peak hours: An energy-saving mode is used to reduce frequent starts and stops and lower elevator standby energy consumption.

[0248] If the energy saving rate is too low, the system automatically adjusts the control strategies for air conditioning, lighting, and elevators, and optimizes the time series forecasting model to improve the accuracy of energy consumption management. Through adaptive optimization strategies, the system makes energy use within the building more precise: reducing total building energy consumption by 15% to 30%, lowering electricity costs, and improving operational economics.

[0249] Example 5

[0250] This embodiment is an explanation based on Embodiment 4. Please refer to it. Figure 1 Specifically: the model adaptive optimization unit adjusts the parameters of the time series prediction model and the regression model according to the level of the energy saving rate ηeff;

[0251] When the energy saving rate ηeff is at the third effective level, the gradient descent algorithm is used to adjust the parameters of the time series prediction model and the regression model, including adjusting the time parameter α of the time series prediction model and the regression parameter β of the regression model; obtaining the new time parameter nα and the new regression parameter nβ, and returning the new time parameter nα and the new regression parameter nβ to the time series prediction model and the regression model to recalculate the predicted value of electricity load Ppre and the predicted value of population density Dus;

[0252] The new time parameter nα is obtained using the following formula:

[0253] ;

[0254] In the formula, θ represents the learning rate, and ∂ represents the partial derivative;

[0255] The new regression parameter nβ is obtained using the following formula:

[0256] .

[0257] The energy efficiency assessment and reporting module summarizes the parameters of the adjusted equipment operation mode, including energy regulation consumption density nEd, new time parameter nα, and new regression parameter nβ, with the energy saving rate ηeff, and generates a visual report.

[0258] Visualized reports include line charts and pie charts;

[0259] The line graph shows energy consumption over different time periods;

[0260] The pie chart illustrates the contribution of adjusting equipment operating modes to energy conservation.

[0261] In this embodiment, the energy saving rate ηeff is used as the optimization evaluation standard to automatically determine whether the current equipment operating mode has achieved the energy saving target: if the energy saving rate ηeff is at the third effective level, it indicates that the prediction model has failed to accurately match the actual energy consumption trend and the parameters need to be optimized. If the energy saving rate ηeff is at the second or first level, it indicates that the existing prediction model is usable and does not require significant adjustments, only continuous monitoring.

[0262] The gradient descent algorithm is used to optimize the parameters α of the time series prediction model and β of the regression model, thereby improving the model's predictive ability. Through parameter optimization, the system becomes more adaptable to changes in the building operating environment, and the self-learning ability of the prediction model is enhanced.

[0263] Based on the optimized new parameters nα and nβ, the following are recalculated: the electricity load forecast value Ppre, ensuring that energy supply matches actual demand and avoiding oversupply or undersupply; and the personnel density forecast value Dus, improving the accuracy of peak and off-peak period identification and enabling more precise equipment operation strategies. The energy efficiency assessment and reporting module aggregates the optimized energy regulation consumption density nEd, the new time parameter nα, and the new regression parameter nβ, generating a visual report to help managers make more accurate optimization decisions.

[0264] Line chart: Displays energy consumption trends over different time periods, visually illustrating the impact of optimization strategies on energy consumption.

[0265] Advantages: It can analyze energy consumption changes during peak and off-peak hours and optimize equipment start-up and shutdown strategies.

[0266] By monitoring long-term data, we can identify cyclical changes in energy consumption and provide data support for future optimization.

[0267] Pie chart: Shows the contribution of equipment operation mode adjustment to energy saving, and quantifies the energy saving effect of optimization strategies for each piece of equipment.

[0268] It helps managers intuitively understand the energy consumption ratio of equipment such as air conditioners, lighting, and elevators, and identify key areas for optimization.

[0269] Identify high-energy-consuming equipment to provide data support for further optimizing equipment management.

[0270] Example 6

[0271] A cloud computing-based integrated control method for smart buildings; please refer to [reference needed]. Figure 2 and Figure 3 Specifically, it includes the following steps:

[0272] Step 1: The environmental perception module collects environmental data of the smart building in real time through sensors, including atmospheric radiation intensity Ra, temperature gradient ∇T, energy consumption density Ed, and wind speed Wou, fits them into the original data set Q, and performs preprocessing to obtain the environmental dataset QH.

[0273] Step 2: The energy forecasting and scheduling module uses the acquired environmental dataset QH to predict energy demand for different time periods in the future through time series forecasting models and regression models, and obtains the predicted power load value Ppre and the predicted population density value Dus.

[0274] Step 3: The user behavior analysis module analyzes peak periods of pedestrian traffic and equipment usage habits inside the building based on the predicted power load value Ppre and the predicted personnel density value Dus, and obtains user behavior patterns.

[0275] Step 4: The adaptive energy consumption control module adjusts the operating mode of the equipment based on the user's behavior pattern and the energy consumption density Ed, and obtains the energy regulation consumption density nEd.

[0276] Step 5: The feedback and self-optimization module obtains the energy saving rate ηeff by comparing the energy regulation consumption density nEd with the energy consumption density Ed.

[0277] Step Six: The Energy Efficiency Assessment and Reporting module summarizes the parameters and energy saving rate ηeff of the adjusted equipment operating mode and generates a visual report.

[0278] In this embodiment, a data preprocessing method is employed to detect outliers, impute missing values, and standardize the raw data, effectively removing sensor errors, improving data quality, and making subsequent prediction models more accurate. This improvement enables the system to make energy consumption optimization decisions based on high-quality data, avoiding control failures due to data anomalies or incompleteness, and improving the accuracy of energy efficiency management. The energy forecasting and scheduling module uses time series prediction models and regression models, based on the environmental dataset QH, to predict the electricity load forecast Ppre and the population density forecast Dus for different time periods in the future.

[0279] By accurately predicting peak electricity consumption periods and peak pedestrian traffic, the system can optimize energy dispatch in advance, reducing oversupply or undersupply and improving energy efficiency. Compared to the passive adjustment model of traditional building management, this method achieves predictive control, enabling the system to optimize energy allocation before peak periods arrive, improving response speed and reducing energy waste.

[0280] The user behavior analysis module uses power load forecasting and personnel density forecasting to analyze peak pedestrian flow and equipment usage habits inside the building, accurately identifying peak times, equipment usage frequency, and lighting usage.

[0281] The adaptive energy consumption control module dynamically adjusts the equipment operation mode based on user behavior patterns and energy consumption density Ed: it automatically adjusts the air supply rate and temperature setpoint according to factors such as population density, room temperature changes, and external ambient temperature to reduce unnecessary cooling / heating energy consumption.

[0282] Adjust lighting brightness according to people's activities, so that lights turn on when people are there and turn off when people leave, thereby improving lighting efficiency.

[0283] Based on peak hour forecasts and elevator usage frequency, optimize elevator scheduling to reduce empty runs and improve transport efficiency.

[0284] The feedback and self-optimization module compares the energy regulation consumption density nEd with the energy consumption density Ed to calculate the energy saving rate ηeff and evaluate the effectiveness of the optimization strategy.

[0285] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A cloud computing-based intelligent building integrated control system, characterized in that: It includes an environmental perception module, an energy prediction and scheduling module, a user behavior analysis module, an adaptive energy consumption control module, a feedback and self-optimization module, and an energy efficiency assessment and reporting module; The environmental perception module collects environmental data of the smart building in real time through sensors, including atmospheric radiation intensity Ra, temperature gradient ∇T, energy consumption density Ed, and wind speed Wou, fits them into the original data set Q, and performs preprocessing to obtain the environmental dataset QH. The energy forecasting and scheduling module uses the acquired environmental dataset QH to predict energy demand for different time periods in the future through time series forecasting models and regression models, and obtains the predicted power load value Ppre and the predicted population density value Dus. The user behavior analysis module analyzes peak periods of pedestrian traffic and equipment usage habits inside the building based on the predicted power load value Ppre and the predicted personnel density value Dus, and obtains user behavior patterns. The adaptive energy consumption control module adjusts the operating mode of the equipment based on user behavior patterns and energy consumption density Ed, and obtains the energy regulation consumption density nEd. The feedback and self-optimization module obtains the energy saving rate ηeff by comparing the energy regulation consumption density nEd with the energy consumption density Ed. The energy efficiency assessment and reporting module summarizes the parameters and energy saving rate ηeff of the adjusted equipment operating mode and generates a visual report.

2. A cloud-based intelligent building integrated control system according to claim 1, characterized in that: The environmental perception module includes a data acquisition unit and a data preprocessing unit; The data acquisition unit collects environmental data of the smart building through radiation sensors, temperature sensors, smart meters, and wind speed sensors. Specifically, it collects atmospheric radiation intensity Ra through radiation sensors, temperature gradient ∇T through temperature sensors, energy consumption density Ed through smart meters, and wind speed Wou through wind speed sensors, and fits them into the original data set Q. The temperature gradient ∇T is obtained using the following formula: ； In the formula, Tin represents the indoor temperature, Tou represents the outdoor temperature, η represents the radiation absorptivity of the building's exterior walls, and Cp represents the specific heat capacity. Indicates air density; Energy consumption density Ed is obtained using the following formula; ； In the formula, Pe represents the power consumption of lighting equipment, Pha represents the energy consumption of air conditioning, heating and ventilation, Pw represents the power consumption of elevator system, and A represents the total building area. The data preprocessing unit cleans and standardizes the acquired raw data set Q to obtain the environmental dataset QH; Cleaning includes outlier detection and missing value imputation. Outlier detection is performed by processing the original data set Q using an outlier detection algorithm to remove outliers from Q. Missing value imputation is performed by filling in missing data in the original data set Q using temporal interpolation. Standardization is achieved by using Min-Max normalization to unify the dimensions of the data in the original dataset Q to the same dimension.

3. A cloud-based intelligent building integrated control system according to claim 2, characterized in that: The energy forecasting and dispatching module uses time series forecasting models and regression models to predict the power load forecast Ppre and the population density forecast Dus for different periods in the future, based on the environmental dataset QH. First, the acquired environmental dataset QH is fitted with historical power load data and historical population density data to form a training set, which is then used to train the time series prediction model and the regression model. The power load forecast value Ppre is obtained by using a time series forecasting model to predict the power load. The electricity load forecast value Ppre is obtained using the following formula: ； In the formula, Ppre(t) represents the predicted power load at time t, QHi(td) represents the value of the i-th data in the environmental dataset QH at time (td), αi represents the time series prediction model coefficient of the i-th data, C represents the error compensation term, and n represents the total number of data. The predicted population density, Dus, is obtained by using a regression model to predict population density. The predicted population density value Dus is obtained using the following formula: ； In the formula, Dus(t) represents the predicted population density at time t, m represents the total number of time points, and Dus(td) represents the population density at time (td). This represents the regression weight value at the j-th time point. Let B represent the regression model coefficient for the i-th data point, and let B represent the error term.

4. A cloud-based intelligent building integrated control system according to claim 3, characterized in that: The user behavior analysis module includes a peak period identification unit and a device usage unit; The peak period identification unit calculates the comprehensive load index Iode inside the building based on the obtained power load forecast value Ppre and personnel density forecast value Dus. The composite load index Iode is obtained using the following formula: ； In the formula, These represent the preset weight values ​​for the predicted power load Ppre and the predicted population density Dus, respectively. By using a time series clustering algorithm, the load index change rate ΔIod between adjacent time points is obtained, and the peak index IPE is calculated and compared with the preset peak threshold TIP to identify peak times. The load index change rate ΔIod is obtained by the difference between the comprehensive load index Iode(t) at time t and the comprehensive load index Iode(t-1) at time t-1. The Peak Index (IPE) is obtained using the following formula: ； In the formula, max(ΔIod) represents the peak value of the load index change rate ΔIod; Peak hours are obtained through the following methods: When the peak index IPE is greater than the peak threshold TIP, it indicates that the peak time has arrived.

5. A cloud-based intelligent building integrated control system according to claim 4, characterized in that: The equipment usage unit obtains user behavior patterns based on the acquired peak index IPE and the predicted personnel density Dus, including peak hours, elevator usage frequency, and lighting usage Lon. Peak periods are obtained by combining peak time and dwell time Ts; The residence time Ts is obtained using the following formula: ； In the formula, Dus(t, (ai, Tavg)) represents the population density at the ai-th sampling point at time t, which is corrected based on the average dwell time Tavg, where Tavg represents the average dwell time, and Na represents the total number of sampling points. The peak time period is obtained by comparing the obtained residence time Ts with the time threshold HTs. When the dwell time Ts > the time threshold HTs, it indicates that the period is at its peak. By fitting peak time periods with peak hours, peak periods can be obtained; When the dwell time Ts > the time threshold HTs and the peak index IPE > the peak threshold TIP, it indicates that the period is in a peak period. Elevator usage frequency is obtained by the ratio of elevator usage count DNE to fixed cycle TG; The elevator usage count DNE is obtained using the following formula: ； In the formula, HfL(a) represents the floor height of the a-th floor, N represents the total number of floors, and Cee represents the elevator capacity; Lighting usage Lon is obtained by comparing the current lighting power with the maximum lighting power in the building. The correlation between lighting usage Lon and the predicted population density Dus (Corr) is used to determine the correlation between lighting and population distribution. Correlation (Corr) is obtained using the following formula: ； In the formula, pDus represents the average value of the predicted population density, and pLon represents the average proportion of light usage; The correlation between lighting and personnel distribution was obtained through the following methods: When 0 < Correlation (Corr) < 0.3, it indicates that there is no correlation between the matching of lighting and personnel distribution; When 0.3 ≤ Correlation < 1, it indicates that the matching between lighting and personnel distribution is correlated.

6. A cloud-based intelligent building integrated control system according to claim 5, characterized in that: The adaptive energy consumption control module adjusts the operating mode of the equipment based on the acquired user behavior patterns and energy consumption density Ed, including air conditioning, heating and ventilation adjustment, intelligent lighting adjustment and elevator operation adjustment, and acquires the adjusted air conditioning, heating and ventilation energy consumption nPha, the adjusted lighting energy consumption nPe and the adjusted elevator operation energy consumption nPele. Air conditioning, heating, and ventilation regulation includes temperature adjustment strategies and adaptive fan speed control. Temperature adjustment strategies are obtained through the following methods: When it is during peak hours and the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, the ventilation volume and carbon dioxide concentration are adjusted, and the fan speed is reduced after the peak hours. Wind speed adaptive adjustment is obtained through the following methods: When the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, and the temperature gradient ∇T is greater than the preset temperature threshold H∇T, then the wind speed Wou is adjusted. The adjusted energy consumption (nPha) for air conditioning, heating, and ventilation can be obtained using the following formula. ； In the formula, C1, C2, and C3 represent the load adjustment coefficients of the air conditioning, heating, and ventilation system; Intelligent lighting adjustment is obtained through the following methods: If the predicted population density Dus(t) at time t is greater than the preset density threshold TDu, and the lighting usage Lon is less than the preset lighting usage threshold TLon, then increase the lighting brightness. When the predicted population density Dus(t) at time t is less than the preset density threshold TDu, the lighting brightness is reduced. The adjusted light energy consumption nPe is obtained using the following formula: ； In the formula, D1 and D2 represent the lighting adjustment coefficients; Elevator operation adjustments are obtained through the following methods: When the elevator usage frequency exceeds the preset frequency threshold during peak hours, a priority target scheduling method is adopted to control the elevator to run empty and adjust the passenger carrying efficiency. When it is not during peak hours, the elevator is adjusted in energy-saving mode; Elevator operating energy consumption nPele is obtained using the following formula: ； In the formula, HfL(a) represents the floor height of the a-th floor, N represents the total number of floors, Cee represents the elevator capacity, and Efl represents the energy consumption of a single floor elevator operation. The energy regulation consumption density nEd is calculated by fitting the adjusted air conditioning heating and ventilation energy consumption nPha, the adjusted lighting energy consumption nPe, and the adjusted elevator operation energy consumption nPele. Energy regulation consumption density nEd is obtained using the following formula: 。 7. A cloud-based intelligent building integrated control system according to claim 6, characterized in that: The feedback and self-optimization module includes an execution performance evaluation unit and a model adaptive optimization unit; The performance evaluation unit calculates the energy saving rate ηeff based on the obtained energy regulation consumption density nEd and energy consumption density Ed, and judges the effectiveness of the equipment operation mode through the energy saving rate ηeff; The energy saving rate ηeff is obtained using the following formula: ； The validity of the device's operating mode is obtained through the following methods: When the energy saving rate ηeff > 20%, it indicates that it is in the first effective level; When 20% ≥ energy saving rate ηeff ≥ 5%, it indicates that it is in the second effective level; When 5% > energy saving rate ηeff > 0, it indicates that it is in the third effective level.

8. A cloud-based intelligent building integrated control system according to claim 7, characterized in that: The model adaptive optimization unit adjusts the parameters of the time series prediction model and the regression model according to the level of the energy saving rate ηeff; When the energy saving rate ηeff is at the third effective level, the gradient descent algorithm is used to adjust the parameters of the time series prediction model and the regression model, including adjusting the time parameter α of the time series prediction model and the regression parameter β of the regression model; obtaining the new time parameter nα and the new regression parameter nβ, and returning the new time parameter nα and the new regression parameter nβ to the time series prediction model and the regression model to recalculate the predicted value of electricity load Ppre and the predicted value of population density Dus; The new time parameter nα is obtained using the following formula: ； In the formula, θ represents the learning rate, and ∂ represents the partial derivative; The new regression parameter nβ is obtained using the following formula: 。 9. A cloud-based intelligent building integrated control system according to claim 1, characterized in that: The energy efficiency assessment and reporting module summarizes the parameters of the adjusted equipment operation mode, including energy regulation consumption density nEd, new time parameter nα, and new regression parameter nβ, with the energy saving rate ηeff, and generates a visual report. Visualized reports include line charts and pie charts; The line graph shows energy consumption over different time periods; The pie chart illustrates the contribution of adjusting equipment operating modes to energy conservation.

10. A cloud computing-based intelligent building integrated control method, applied to a cloud computing-based intelligent building integrated control system according to any one of claims 1 to 9, characterized in that: Includes the following steps: Step 1: The environmental perception module collects environmental data of the smart building in real time through sensors, including atmospheric radiation intensity Ra, temperature gradient ∇T, energy consumption density Ed, and wind speed Wou, fits them into the original data set Q, and performs preprocessing to obtain the environmental dataset QH. Step 2: The energy forecasting and scheduling module uses the acquired environmental dataset QH to predict energy demand for different time periods in the future through time series forecasting models and regression models, and obtains the predicted power load value Ppre and the predicted population density value Dus. Step 3: The user behavior analysis module analyzes peak periods of pedestrian traffic and equipment usage habits inside the building based on the predicted power load value Ppre and the predicted personnel density value Dus, and obtains user behavior patterns. Step 4: The adaptive energy consumption control module adjusts the operating mode of the equipment based on the user's behavior pattern and the energy consumption density Ed, and obtains the energy regulation consumption density nEd. Step 5: The feedback and self-optimization module obtains the energy saving rate ηeff by comparing the energy regulation consumption density nEd with the energy consumption density Ed. Step Six: The Energy Efficiency Assessment and Reporting module summarizes the parameters and energy saving rate ηeff of the adjusted equipment operating mode and generates a visual report.

Images