CN-122015239-A - HVAC intelligent regulation and control method and system based on personnel dynamic prediction

CN122015239ACN 122015239 ACN122015239 ACN 122015239ACN-122015239-A

Abstract

The application discloses an HVAC intelligent regulation and control method and system based on personnel dynamic prediction, and relates to the field of intelligent buildings, wherein the method comprises the steps of deploying a plurality of monitoring nodes in a building regulation and control area through a sensing layer, collecting personnel quantity and environmental parameter data in the area in real time, and transmitting the data to a comprehensive regulation and control module according to a standardized format; the method comprises the steps of predicting the number of people in a short-term dimension and a historical dimension, outputting a weighted and fused comprehensive predicted number of people, calculating a target temperature set value according to the number of people in real time and the comprehensive predicted number of people through a decision layer, combining a thermal comfort temperature and an energy-saving temperature, generating a regional HVAC (heating and ventilation) equipment regulation and control instruction based on a PID (proportion integration differentiation) control algorithm, adjusting the running parameters of the HVAC equipment in a corresponding region according to the regulation and control instruction through an execution layer, synchronously calculating the total load of a building air conditioner to adjust the power of an air conditioner host, analyzing historical running data to optimize the regulation and control parameter, detecting abnormal data and generating a fault diagnosis report, and realizing iterative optimization.

Inventors

CUI HEYANG
CUI XIAOFEI

Assignees

青岛和泰精密工业有限公司

Dates

Publication Date: 20260512
Application Date: 20260325
Priority Date: 20250327

Claims (10)

1. An intelligent HVAC regulation and control method based on personnel dynamic prediction is characterized by comprising the following steps: Step S1, deploying a plurality of monitoring nodes in a building regulation area through a sensing layer, collecting the number of personnel in the area and environmental parameter data in real time, and transmitting the data to a comprehensive regulation module according to a standardized format; Step S2, based on the acquired data of the sensing layer, respectively predicting the number of people in a short-term dimension and a history dimension through a prediction layer, wherein the short-term dimension predicts the number of people in a second preset time period in the future for the data in a first preset time period in the past by adopting an LSTM model, the history dimension predicts the number of people in the same period through time clustering and a linear regression model, and dynamically distributes weights according to prediction errors, and outputs the comprehensive predicted number after weighted fusion; step S3, calculating a target temperature set value by combining the thermal comfort temperature and the energy-saving temperature according to the number of real-time personnel and the comprehensive predicted number of people through a decision layer, and generating a regional HVAC (heating ventilation and air conditioning) equipment regulation and control instruction based on a PID (proportion integration differentiation) control algorithm; S4, adjusting the operation parameters of the HVAC equipment in the corresponding area according to the regulation and control instruction of the HVAC equipment in the area through an execution layer, and synchronously calculating the total load of the building air conditioner so as to adjust the power of an air conditioner host; And S5, optimizing the regulation and control parameters by analyzing the historical operation data through the evaluation layer, detecting abnormal data and generating a fault diagnosis report to realize iterative optimization.
2. The intelligent HVAC control method based on personnel dynamic prediction according to claim 1 is characterized in that in the step S1, each node comprises a visual sensor, an infrared sensor, a CO 2 sensor and a temperature and humidity sensor, the deployment of the monitoring nodes is particularly carried out by uniformly arranging the monitoring nodes on the ceilings of the control area in a preset area unit to ensure full coverage, carrying out hierarchical spatial coding on the building area, synchronizing clocks of the sensors through an NTP protocol, and setting time stamp errors.
3. The HVAC intelligent regulation method based on personnel dynamic prediction according to claim 1, wherein in the step S2, the short-term dimension prediction process specifically includes: Performing linear interpolation of missing values and abnormal value elimination processing on the sensor data; the personnel number is standardized by adopting a Z-score normalization method, an LSTM model is input for training, and the model structure comprises an input layer, a 32-neuron LSTM layer and a fully-connected output layer; And taking the minimized MAE as a target training model, and outputting the predicted population in the future second time period after the inverse normalization.
4. The HVAC intelligent regulation method based on personnel dynamic prediction according to claim 1, wherein in the step S2, the history dimension prediction process specifically includes: Clustering the timestamps into monday to sunday and holidays, and filtering special weather periods by associating historical meteorological data; and encoding the historical data of the same category by minutes, inputting a linear regression model, and outputting the predicted number of people in a second time period in the future.
5. The intelligent HVAC control method based on dynamic prediction of personnel according to claim 1, wherein in step S2, the weight distribution formula is: Where a is the short-term prediction error, b is the historical prediction error, Is minimum and avoids denominator of 0.
6. The intelligent HVAC control method based on dynamic prediction of personnel according to claim 1, wherein in step S3, the calculation formula of the target temperature set value is: Wherein, the Target temperature set for the zone based on the energy saving and comfort targets; The standard use temperature of the functional area is the primary thermal comfort level range; the energy-saving temperature is the lower limit of the temperature range of the secondary thermal comfort level; k is the proportion of the trigger area adjusted to the comfortable temperature; Capacity or number of regular users is designed for the area.
7. The intelligent HVAC control method based on dynamic prediction of personnel according to claim 1, wherein in step S5, the abnormality detection process specifically includes: setting a threshold range of personnel number, environment parameters and equipment operation parameters; and analyzing abnormal data characteristics through a correlation rule algorithm, positioning a sensor fault or network interruption source, and automatically generating a diagnosis report.
8. The method of intelligent regulation and control of HVAC based on dynamic prediction of personnel according to claim 1, further comprising automatically switching back to the intelligent regulation and control mode when the zone enters the manual regulation and control mode for a third preset period of time, and triggering energy saving temperature setting if the zone unmanned state continues for a super-threshold time.
9. An HVAC intelligent regulation system based on personnel dynamic prediction, the system comprising: The sensing module is composed of a plurality of monitoring nodes, and each node is integrated with a visual sensor, an infrared sensor, a CO 2 sensor and a temperature and humidity sensor and is used for collecting the number of people and environmental parameter data in real time; The prediction module is deployed at the edge computing node and comprises a short-term prediction unit and a history prediction unit, and outputs the comprehensive predicted number of people after weighted fusion through an LSTM model and a linear regression model respectively; The decision module calculates a target temperature set value based on the real-time data and the prediction result, and generates a regulation and control instruction through a PID algorithm; The execution module is used for adjusting the running parameters of the regional HVAC equipment according to the instruction and dynamically adjusting the total power of the air conditioner; And the evaluation module is used for optimizing the regulation and control parameters by utilizing machine learning, detecting abnormal data and generating a fault report.
10. The intelligent HVAC control system based on personnel dynamic prediction according to claim 9, wherein the data transmission rule of the sensing module is that the real-time uploading is triggered when the number of personnel changes, the temperature and humidity data are uploaded according to fixed frequency, and the transmission format is that space coding, time stamping, number of people, temperature and humidity, and the data are compressed and transmitted after noise reduction through Kalman filtering.

Description

HVAC intelligent regulation and control method and system based on personnel dynamic prediction Technical Field The application relates to the technical field of intelligent buildings, in particular to an HVAC intelligent regulation and control method and system based on personnel dynamic prediction. Background Numerous studies have shown that most HVAC (Heating, ventilation and Air Conditioning, heating ventilation and air conditioning) systems mostly rely on a single temperature and humidity sensor, or simple feedback control regulation, and the problem of regional personnel count is not considered, resulting in excessive refrigeration and Heating of equipment, for example, different regional personnel distributions in different time periods of a market, but the prior art generally regulates and controls the areas uniformly, resulting in overheating or supercooling of parts of the areas, such as no one's office in lunch time periods in office areas, and the system still operates in a conventional mode, and according to study statistics, such a stiff operation mode results in 20% -40% energy waste. When the number of persons in a room suddenly changes, HVAC equipment cannot react quickly due to thermodynamic inertia, for example, a conference room suddenly holds a conference, and the equipment needs to run for a period of time to adjust the temperature to be within a comfortable range, resulting in a decrease in human comfort. Meanwhile, the traditional system performs reactive regulation and control according to the current data, and the lack of prejudgment on future demands causes energy waste due to response delay, for example, a few minutes before the conference is finished, equipment still operates at relatively high power, and the predictability of power reduction cannot be achieved. The existing regulation and control method is regulated and controlled according to fixed manual setting parameters, lacks a dynamic optimization mechanism, is difficult to adapt to environmental changes of long-term operation, such as different traffic patterns of staff on holidays and workdays of a market, and causes energy waste caused by excessive regulation or insufficient comfort caused by insufficient regulation due to the rigidity of an HVAC regulation and control method. In addition, the traditional people number prediction method is mostly a single prediction model, and the error between the predicted value and the actual value is large. Disclosure of Invention The application provides an HVAC intelligent regulation and control method and system based on personnel dynamic prediction, which are used for solving the technical problem that the existing HVAC equipment regulation and control method is easy to cause energy waste. In one aspect, the application provides an intelligent HVAC control method based on personnel dynamic prediction, comprising the following steps: Step S1, deploying a plurality of monitoring nodes in a building regulation area through a sensing layer, collecting the number of personnel in the area and environmental parameter data in real time, and transmitting the data to a comprehensive regulation module according to a standardized format; Step S2, based on the acquired data of the sensing layer, respectively predicting the number of people in a short-term dimension and a history dimension through a prediction layer, wherein the short-term dimension predicts the number of people in a second preset time period in the future for the data in a first preset time period in the past by adopting an LSTM model, the history dimension predicts the number of people in the same period through time clustering and a linear regression model, and dynamically distributes weights according to prediction errors, and outputs the comprehensive predicted number after weighted fusion; step S3, calculating a target temperature set value by combining the thermal comfort temperature and the energy-saving temperature according to the number of real-time personnel and the comprehensive predicted number of people through a decision layer, and generating a regional HVAC (heating ventilation and air conditioning) equipment regulation and control instruction based on a PID (proportion integration differentiation) control algorithm; S4, adjusting the operation parameters of the HVAC equipment in the corresponding area according to the regulation and control instruction of the HVAC equipment in the area through an execution layer, and synchronously calculating the total load of the building air conditioner so as to adjust the power of an air conditioner host; And S5, optimizing the regulation and control parameters by analyzing the historical operation data through the evaluation layer, detecting abnormal data and generating a fault diagnosis report to realize iterative optimization. In one implementation mode of the application, in the step S1, each node comprises a visual sensor, an infrared sensor, a CO 2 sensor and a te