A PID control method for smart buildings

Abstract

The present invention discloses a smart building PID control method, which relates to the technical field of building intelligent control, and comprises: S101, according to n groups of outdoor environment data and indoor environment data collected in a time period [T-K, T], obtaining n groups of indoor and outdoor environment difference data and constructing an indoor and outdoor environment difference distribution map; S102, inputting the indoor and outdoor environment difference distribution map into a pre-constructed first machine learning model to predict the indoor optimum PID control temperature at the moment [T+K, T+2K]; S103, inputting the indoor ventilation coefficient collected in the time period [T-K, T] and the generated optimum PID control temperature into a pre-constructed second machine learning model to predict the indoor optimum control humidity in the time period [T+K, T+2K]; S104, correcting the indoor temperature and the indoor humidity according to the optimum PID control temperature and the indoor control humidity. The present invention combines and optimizes the machine learning model with the PID control algorithm to predict the optimum PID control temperature and the indoor control humidity in the future time period, thereby improving the accuracy of indoor environment control.

Description

Technical Field

[0001] The present invention relates to the technical field of building intelligent control, and in particular to a smart building PID control method and system. Background Art

[0002] With the acceleration of urbanization and the continuous upgrading of building functions, modern buildings have an increasingly urgent need for intelligent management and energy-saving control. Smart building systems integrate various sensors, controllers and communication technologies to achieve comprehensive monitoring and intelligent regulation of the building environment, thereby improving energy efficiency and the comfort of living and office environments.

[0003] At present, smart building control systems mostly use traditional control methods, such as proportional-integral-differential (PID) control methods. PID controllers are widely used in industrial control and building environment regulation due to their simplicity and robustness. Traditional PID control methods achieve precise control of environmental parameters such as temperature, humidity, and air quality by adjusting proportional, integral, and differential parameters. However, research and practical application of the above methods and existing technologies have found that the above methods and existing technologies have at least the following defects:

[0004] (1) Indoor PID control cannot respond to environmental changes in a timely manner, which leads to reduced comfort of the building's indoor environment and a low level of intelligence in the building's indoor temperature and humidity control;

[0005] (2) The existing PID control method can only monitor indoor and outdoor temperature and humidity data in real time, and adjust and control the current indoor temperature and humidity based on the current feedback data. However, when adjusting the indoor temperature and humidity, the indoor residents have already felt the abnormal temperature and humidity. The dynamic response capability of the control system is low, and there is a lag in the regulation. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a smart building PID control method.

[0007] A smart building PID control method, comprising:

[0008] S101, according to n groups of outdoor environment data and indoor environment data collected in the time period [TK, T], obtain n groups of indoor and outdoor environment difference data and construct an indoor and outdoor environment difference distribution map;

[0009] S102, inputting the indoor and outdoor environment difference distribution map into a pre-built first machine learning model to predict the optimal PID control temperature of the room at time [T+K, T+2K];

[0010] S103, inputting the indoor ventilation coefficient collected in the time period [TK, T] and the generated optimal PID control temperature into the pre-built second machine learning model to predict the optimal indoor control humidity in the time period [T+K, T+2K];

[0011] S104, correcting the indoor temperature and the indoor humidity according to the optimum PID controlled temperature and the indoor regulated humidity.

[0012] Furthermore, the outdoor environmental data includes outdoor carbon dioxide concentration, outdoor temperature and outdoor light intensity, and the indoor environmental data includes indoor carbon dioxide concentration, indoor temperature and indoor light intensity.

[0013] Furthermore, the indoor and outdoor environment difference data include carbon dioxide concentration difference data, temperature difference data and light intensity difference data, and the acquisition logic of the carbon dioxide concentration difference data is: ;△ is the carbon dioxide concentration difference data, is the outdoor carbon dioxide concentration, is the indoor carbon dioxide concentration; the acquisition logic of the temperature difference data is:

[0014] ; is the temperature difference data, is the outdoor temperature, is the indoor temperature; the logic for obtaining the light intensity difference data is: ; is the light intensity difference data, is the outdoor light intensity, is the indoor light intensity; taking time as the x-axis and the environmental difference data as the y-axis, the n groups of indoor and outdoor environmental difference data collected in the time period [TK, T] are respectively constructed into two-dimensional graphs as indoor and outdoor environmental difference distribution graphs, wherein the indoor and outdoor environmental difference distribution graphs include indoor and outdoor carbon dioxide concentration difference distribution graphs, indoor and outdoor temperature difference distribution graphs and indoor and outdoor light intensity difference distribution graphs.

[0015] Furthermore, the generation logic of the first machine learning model is:

[0016] Acquire historical indoor and outdoor environment difference training data, and divide the historical indoor and outdoor environment difference training data into an indoor and outdoor environment difference training set and an indoor and outdoor environment difference test set; the historical indoor and outdoor environment difference training data includes an indoor and outdoor environment difference distribution map and its corresponding predicted optimal PID control temperature;

[0017] Constructing a first regression network, taking the indoor and outdoor environment difference distribution map in the indoor and outdoor environment difference training set as the first regression network input data, and taking the optimal PID control temperature in the indoor and outdoor environment difference training set as the first regression network output data, training the first regression network, and obtaining an initial first machine learning network;

[0018] The initial first machine learning network is model verified using the indoor and outdoor environment difference test set, and the initial first machine learning network with a value less than or equal to a preset first test error threshold is output as the pre-constructed first machine learning model.

[0019] Furthermore, the logic for obtaining the optimal PID control temperature is:

[0020] The m groups of temperature acquisition values ​​Tc(m) acquired in real time according to the time series are compared with the preset temperature target value Tb to generate temperature error data , the temperature error data The acquisition logic is: = (Tc(m) – Tb) β; β is the weight value; according to the error data obtained Modify the PID algorithm to obtain the optimal PID control temperature , ,in, , and are the matrices of proportional gain, integral gain and differential gain respectively.

[0021] Furthermore, the indoor ventilation coefficient is calculated by The concentration value is calculated, and the logic for obtaining the indoor ventilation coefficient is: Ach= ; Where Ach is the indoor ventilation coefficient, is a weight value greater than 0, is the indoor area, K is the time difference within the time period [TK,T], Collected for TK Moments Concentration value, Collected at time T Concentration value.

[0022] Furthermore, the generation logic of the second machine learning model is:

[0023] Acquire historical humidity control training data, and divide the historical humidity control training data into a humidity control training set and a humidity control test set; the historical humidity control training data includes an indoor ventilation coefficient, a PID controlled temperature and its corresponding predicted indoor controlled humidity;

[0024] Constructing a second regression network, taking the indoor ventilation coefficient and PID control temperature in the humidity control training set as input data of the first regression network, and taking the indoor control humidity in the humidity control training set as output data of the second regression network, training the second regression network, and obtaining an initial second machine learning network;

[0025] The initial second machine learning network is model verified using the indoor and outdoor environment difference test set, and the initial second machine learning network with a value less than or equal to a preset second test error threshold is output as the pre-constructed second machine learning model.

[0026] Furthermore, the acquisition logic of the indoor humidity control is:

[0027] P1: Set the initial humidity of the room to G, which is greater than zero;

[0028] P2: Adjust the initial humidity, set J = G + a, and obtain the indoor humidity control J, where J and a are greater than zero;

[0029] P3: Under the indoor controlled humidity J, obtain the dryness value of the surface layer of the human skin in the room. If the dryness value is within the preset dryness range, the adjusted humidity J is used as the optimal controlled humidity; if the dryness value is not within the preset dryness range, set a=a+k, and return to step P2, where k is an integer greater than zero;

[0030] P4: Repeat the above steps P2 to P3 until the optimal indoor humidity is obtained, and then end the cycle.

[0031] Furthermore, a smart building PID control system is implemented based on any one of the smart building PID control methods, comprising:

[0032] The difference analysis module obtains n sets of indoor and outdoor environment difference data and constructs a distribution map of indoor and outdoor environment differences based on n sets of outdoor environment data and indoor environment data collected in the time period [TK, T];

[0033] The first prediction module inputs the indoor and outdoor environment difference distribution map into the pre-built first machine learning model to predict the optimal PID control temperature in the room at the time [T+K, T+2K];

[0034] The second prediction module inputs the indoor ventilation coefficient collected in the time period [TK, T] and the generated optimal PID control temperature into the pre-built second machine learning model to predict the optimal indoor control humidity in the time period [T+K, T+2K].

[0035] The temperature and humidity control module corrects the indoor temperature and humidity according to the optimal PID control temperature and indoor control humidity.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] (1) By combining and optimizing the machine learning model with the PID control algorithm, the optimal PID control temperature and indoor humidity control in the future time period can be predicted, the accuracy of indoor environmental control can be improved, and environmental changes can be responded to in a timely manner, further ensuring the comfort of the building's indoor environment and improving the intelligence of indoor temperature and humidity control in the building;

[0038] (2) It can monitor indoor and outdoor environmental data in real time and make advance adjustments to the future indoor environment based on the current feedback data, thereby ensuring that the indoor environment is always in the best state, further improving the dynamic response capability of the control system, and avoiding the lag in indoor temperature and humidity control. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0040] Figure 1 A flowchart of a smart building PID control method provided in Example 1 of the present invention;

[0041] Figure 2 A module diagram of a smart building PID control system provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Example 1

[0044] See also Figure 1 As shown, this embodiment discloses a smart building PID control method, the method comprising:

[0045] S101, according to n groups of outdoor environment data and indoor environment data collected in the time period [TK, T], obtain n groups of indoor and outdoor environment difference data and construct an indoor and outdoor environment difference distribution map;

[0046] Specifically, the outdoor environment data includes outdoor carbon dioxide concentration, outdoor temperature and outdoor light intensity, and the indoor environment data includes indoor carbon dioxide concentration, indoor temperature and indoor light intensity;

[0047] It should be noted that:

[0048] Outdoor / indoor carbon dioxide concentration can be collected by CO2 sensors, which are installed outdoors / indoors to monitor the CO2 concentration in the air in real time. For example, non-dispersive infrared (NDIR) sensors are often used to measure CO2 concentration;

[0049] The outdoor / indoor temperature can be measured using temperature sensors, such as thermistors and RTDs (platinum resistance thermometers), which can be installed outdoors / indoors to monitor outdoor / indoor temperature changes in real time;

[0050] Outdoor / indoor light intensity: Light sensors can be used to collect light. Light sensors such as photodiodes or photoresistors can be used to measure light intensity. Light sensors can be installed outside / indoors of buildings to monitor sunlight intensity and changes.

[0051] Specifically, the indoor and outdoor environment difference data include carbon dioxide concentration difference data, temperature difference data and light intensity difference data. The acquisition logic of the carbon dioxide concentration difference data is:

[0052] ;

[0053] Among them, △ is the carbon dioxide concentration difference data, is the outdoor carbon dioxide concentration, is the indoor carbon dioxide concentration;

[0054] The acquisition logic of the temperature difference data is:

[0055] ;

[0056] in, is the temperature difference data, is the outdoor temperature, is the indoor temperature;

[0057] The logic for obtaining the light intensity difference data is:

[0058] ;

[0059] in, is the light intensity difference data, is the outdoor light intensity, is the indoor light intensity;

[0060] Taking time as the x-axis and environmental difference data as the y-axis, construct two-dimensional graphs as indoor and outdoor environmental difference distribution graphs for n groups of indoor and outdoor environmental difference data collected in the time period [TK, T], respectively. The indoor and outdoor environmental difference distribution graphs include indoor and outdoor carbon dioxide concentration difference distribution graphs, indoor and outdoor temperature difference distribution graphs, and indoor and outdoor light intensity difference distribution graphs;

[0061] It should be noted that the distribution map of indoor and outdoor environment differences can be drawn using Matplotlib;

[0062] S102, inputting the indoor and outdoor environment difference distribution map into a pre-built first machine learning model to predict the optimal PID control temperature of the room at time [T+K, T+2K];

[0063] Specifically, the generation logic of the first machine learning model is:

[0064] Acquire historical indoor and outdoor environment difference training data, and divide the historical indoor and outdoor environment difference training data into an indoor and outdoor environment difference training set and an indoor and outdoor environment difference test set; the historical indoor and outdoor environment difference training data includes an indoor and outdoor environment difference distribution map and its corresponding predicted optimal PID control temperature;

[0065] Specifically, the logic for obtaining the optimal PID control temperature is:

[0066] The m groups of temperature acquisition values ​​Tc(m) acquired in real time according to the time series are compared with the preset temperature target value Tb to generate temperature error data , the temperature error data The acquisition logic is: = (Tc(m) – Tb) β; β is the weight value; according to the error data obtained Modify the PID algorithm to obtain the optimal PID control temperature , ,in, , and are the matrices of proportional gain, integral gain and differential gain respectively;

[0067] It should be noted that the temperature acquisition value Tc is acquired by a temperature sensor;

[0068] Constructing a first regression network, taking the indoor and outdoor environment difference distribution map in the indoor and outdoor environment difference training set as the first regression network input data, and taking the optimal PID control temperature in the indoor and outdoor environment difference training set as the first regression network output data, training the first regression network, and obtaining an initial first machine learning network;

[0069] The initial first machine learning network is model verified using the indoor and outdoor environment difference test set, and the initial first machine learning network with a value less than or equal to a preset first test error threshold is output as the pre-constructed first machine learning model.

[0070] S103, inputting the indoor ventilation coefficient collected in the time period [TK, T] and the generated optimal PID control temperature into the pre-built second machine learning model to predict the optimal indoor control humidity in the time period [T+K, T+2K];

[0071] Specifically, the indoor ventilation coefficient is calculated by The concentration value is calculated, and the logic for obtaining the indoor ventilation coefficient is: Ach= ; Where Ach is the indoor ventilation coefficient, is a weight value greater than 0, is the indoor area, K is the time difference within the time period [TK,T], Collected for TK Moments Concentration value, Collected at time T Concentration value;

[0072] Specifically, the generation logic of the second machine learning model is:

[0073] Acquire historical humidity control training data, and divide the historical humidity control training data into a humidity control training set and a humidity control test set; the historical humidity control training data includes an indoor ventilation coefficient, a PID controlled temperature and its corresponding predicted indoor controlled humidity;

[0074] Specifically, the acquisition logic of the indoor humidity control is:

[0075] P1: Set the initial humidity of the room to G, which is greater than zero;

[0076] P2: Adjust the initial humidity, set J = G + a, and obtain the indoor controlled humidity J, where J and a are greater than zero;

[0077] P3: Under the indoor controlled humidity J, obtain the dryness value of the surface layer of the human skin in the room. If the dryness value is within the preset dryness range, the adjusted humidity J is used as the optimal controlled humidity; if the dryness value is not within the preset dryness range, set a=a+k, and return to step P2, where k is an integer greater than zero;

[0078] It should be noted that: the dryness value of the human skin surface is based on the monitoring of environmental humidity and direct or indirect measurement of the degree of skin dryness, including but not limited to the skin resistance method and the skin moisture content method; the dryness value is that the human body surface is within the preset dryness range, that is, the control humidity J is the appropriate control humidity. If not, the control humidity J has severe humidification or does not meet the humidification standard.

[0079] P4: Repeat the above steps P2 to P3 until the optimal indoor humidity is obtained, and then end the cycle;

[0080] Constructing a second regression network, taking the indoor ventilation coefficient and PID control temperature in the humidity control training set as input data of the first regression network, and taking the indoor control humidity in the humidity control training set as output data of the second regression network, training the second regression network, and obtaining an initial second machine learning network;

[0081] The initial second machine learning network is model verified using the indoor and outdoor environment difference test set, and the initial second machine learning network with a value less than or equal to a preset second test error threshold is output as the pre-constructed second machine learning model.

[0082] It should be noted that the second test error threshold and the first test error threshold are both set by experimenters, which is existing technology and will not be described in detail.

[0083] S104, correcting the indoor temperature and the indoor humidity according to the optimum PID controlled temperature and the indoor regulated humidity.

[0084] It should be noted that the indoor heating equipment, exhaust fan or air conditioner is regulated according to the optimal PID control temperature to achieve the optimal indoor temperature, and the indoor humidifier is controlled according to the indoor humidity control until the indoor humidity reaches the optimal indoor humidity.

[0085] Example 2

[0086] See also Figure 2 As shown, based on the same inventive concept, this embodiment discloses a smart building PID control system, including:

[0087] The difference analysis module S201 obtains n sets of indoor and outdoor environment difference data and constructs an indoor and outdoor environment difference distribution map according to n sets of outdoor environment data and indoor environment data collected in the time period [TK, T];

[0088] Specifically, the outdoor environment data includes outdoor carbon dioxide concentration, outdoor temperature and outdoor light intensity, and the indoor environment data includes indoor carbon dioxide concentration, indoor temperature and indoor light intensity;

[0089] Outdoor / indoor carbon dioxide concentration can be collected by CO2 sensors, which are installed outdoors / indoors to monitor the CO2 concentration in the air in real time. For example, non-dispersive infrared (NDIR) sensors are often used to measure CO2 concentration;

[0090] The outdoor / indoor temperature can be measured using temperature sensors, such as thermistors and RTDs (platinum resistance thermometers), which can be installed outdoors / indoors to monitor outdoor / indoor temperature changes in real time;

[0091] Outdoor / indoor light intensity: Light sensors can be used to collect light. Light sensors such as photodiodes or photoresistors can be used to measure light intensity. Light sensors can be installed outside / indoors of buildings to monitor sunlight intensity and changes.

[0092] Specifically, the indoor and outdoor environment difference data include carbon dioxide concentration difference data, temperature difference data and light intensity difference data. The acquisition logic of the carbon dioxide concentration difference data is:

[0093] ;

[0094] Among them, △ is the carbon dioxide concentration difference data, is the outdoor carbon dioxide concentration, is the indoor carbon dioxide concentration;

[0095] The acquisition logic of the temperature difference data is:

[0096] ;

[0097] in, is the temperature difference data, is the outdoor temperature, is the indoor temperature;

[0098] The logic for obtaining the light intensity difference data is:

[0099] ;

[0100] in, is the light intensity difference data, is the outdoor light intensity, is the indoor light intensity;

[0101] Taking time as the x-axis and environmental difference data as the y-axis, construct two-dimensional graphs as indoor and outdoor environmental difference distribution graphs for n groups of indoor and outdoor environmental difference data collected in the time period [TK, T], respectively. The indoor and outdoor environmental difference distribution graphs include indoor and outdoor carbon dioxide concentration difference distribution graphs, indoor and outdoor temperature difference distribution graphs, and indoor and outdoor light intensity difference distribution graphs;

[0102] The first prediction module S202 inputs the indoor and outdoor environment difference distribution map into a pre-built first machine learning model to predict the optimal PID control temperature of the room at the time [T+K, T+2K];

[0103] Specifically, the generation logic of the first machine learning model is:

[0104] Acquire historical indoor and outdoor environment difference training data, and divide the historical indoor and outdoor environment difference training data into an indoor and outdoor environment difference training set and an indoor and outdoor environment difference test set; the historical indoor and outdoor environment difference training data includes an indoor and outdoor environment difference distribution map and its corresponding predicted optimal PID control temperature;

[0105] Specifically, the logic for obtaining the optimal PID control temperature is:

[0106] The m groups of temperature acquisition values ​​Tc(m) acquired in real time according to the time series are compared with the preset temperature target value Tb to generate temperature error data , the temperature error data The acquisition logic is: = (Tc(m) – Tb) β; β is the weight value; according to the error data obtained Modify the PID algorithm to obtain the optimal PID control temperature , ,in, , and are the matrices of proportional gain, integral gain and differential gain respectively;

[0107] Constructing a first regression network, taking the indoor and outdoor environment difference distribution map in the indoor and outdoor environment difference training set as the first regression network input data, and taking the optimal PID control temperature in the indoor and outdoor environment difference training set as the first regression network output data, training the first regression network, and obtaining an initial first machine learning network;

[0108] The initial first machine learning network is model verified using the indoor and outdoor environment difference test set, and the initial first machine learning network with a value less than or equal to a preset first test error threshold is output as the pre-constructed first machine learning model.

[0109] The second prediction module S203 inputs the indoor ventilation coefficient collected in the time period [TK, T] and the generated optimal PID control temperature into the pre-built second machine learning model to predict the optimal indoor control humidity in the time period [T+K, T+2K];

[0110] Specifically, the indoor ventilation coefficient is calculated by The concentration value is calculated, and the logic for obtaining the indoor ventilation coefficient is: Ach= ; Where Ach is the indoor ventilation coefficient, is a weight value greater than 0, is the indoor area, K is the time difference within the time period [TK,T], Collected for TK Moments Concentration value, Collected at time T Concentration value;

[0111] Specifically, the generation logic of the second machine learning model is:

[0112] Acquire historical humidity control training data, and divide the historical humidity control training data into a humidity control training set and a humidity control test set; the historical humidity control training data includes an indoor ventilation coefficient, a PID controlled temperature and its corresponding predicted indoor controlled humidity;

[0113] Specifically, the acquisition logic of the indoor humidity control is:

[0114] P1: Set the initial humidity of the room to G, which is greater than zero;

[0115] P2: Adjust the initial humidity, set J = G + a, and obtain the indoor controlled humidity J, where J and a are greater than zero;

[0116] P3: Under the indoor controlled humidity J, obtain the dryness value of the surface layer of the human skin in the room. If the dryness value is within the preset dryness range, the adjusted humidity J is used as the optimal controlled humidity; if the dryness value is not within the preset dryness range, set a=a+k, and return to step P2, where k is an integer greater than zero;

[0117] P4: Repeat the above steps P2 to P3 until the optimal indoor humidity is obtained, and then end the cycle;

[0118] Constructing a second regression network, taking the indoor ventilation coefficient and PID control temperature in the humidity control training set as input data of the first regression network, and taking the indoor control humidity in the humidity control training set as output data of the second regression network, training the second regression network, and obtaining an initial second machine learning network;

[0119] The initial second machine learning network is model verified using the indoor and outdoor environment difference test set, and the initial second machine learning network with a value less than or equal to a preset second test error threshold is output as the pre-constructed second machine learning model.

[0120] The temperature and humidity control module S204 corrects the indoor temperature and indoor humidity according to the optimum PID control temperature and indoor control humidity.

[0121] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired network or a wireless network. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD) or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0122] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0123] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0124] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a smart building PID control method. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0125] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0126] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0127] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0128] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A smart building PID control method, characterized in that: The method comprises: S101, according to n groups of outdoor environment data and indoor environment data collected in the time period [TK, T], obtain n groups of indoor and outdoor environment difference data and construct an indoor and outdoor environment difference distribution map; The indoor environmental data include indoor carbon dioxide concentration, indoor temperature and indoor light intensity; S102, inputting the indoor and outdoor environment difference distribution map into a pre-built first machine learning model to predict the optimal PID control temperature of the room at the time [T+K, T+2K]; The logic for obtaining the optimal PID control temperature is: The m groups of temperature acquisition values ​​Tc(m) acquired in real time according to the time series are compared with the preset temperature target value Tb to generate temperature error data , the temperature error data The acquisition logic is: = (Tc(m) – Tb) β; β is the weight value; according to the error data obtained Modify the PID algorithm to obtain the optimal PID control temperature , ,in, , and are the matrices of proportional gain, integral gain and differential gain respectively; S103, inputting the indoor ventilation coefficient collected in the time period [TK, T] and the generated optimal PID control temperature into the pre-built second machine learning model to predict the optimal indoor control humidity in the time period [T+K, T+2K]; The indoor ventilation coefficient is calculated by The concentration value is calculated, and the logic for obtaining the indoor ventilation coefficient is: Ach= ; Where Ach is the indoor ventilation coefficient, is a weight value greater than 0, is the indoor area, K is the time difference within the time period [TK,T], Collected for TK Moments Concentration value, Collected at time T Concentration value; S104, correcting the indoor temperature and the indoor humidity according to the optimum PID controlled temperature and the indoor regulated humidity.

2. A smart building PID control method according to claim 1, characterized in that: The outdoor environment data includes outdoor carbon dioxide concentration, outdoor temperature and outdoor light intensity.

3. A smart building PID control method according to claim 2, characterized in that: The indoor and outdoor environment difference data include carbon dioxide concentration difference data, temperature difference data and light intensity difference data. The acquisition logic of the carbon dioxide concentration difference data is: ;△ is the carbon dioxide concentration difference data, is the outdoor carbon dioxide concentration, is the indoor carbon dioxide concentration; the acquisition logic of the temperature difference data is: ; is the temperature difference data, is the outdoor temperature, is the indoor temperature; taking time as the x-axis and the environmental difference data as the y-axis, the n groups of indoor and outdoor environmental difference data collected in the time period [TK, T] are respectively constructed into two-dimensional graphs as indoor and outdoor environmental difference distribution graphs, wherein the indoor and outdoor environmental difference distribution graphs include indoor and outdoor carbon dioxide concentration difference distribution graphs, indoor and outdoor temperature difference distribution graphs and indoor and outdoor light intensity difference distribution graphs.

4. A smart building PID control method according to claim 3, characterized in that: The generation logic of the first machine learning model is: Acquire historical indoor and outdoor environment difference training data, and divide the historical indoor and outdoor environment difference training data into an indoor and outdoor environment difference training set and an indoor and outdoor environment difference test set; the historical indoor and outdoor environment difference training data includes an indoor and outdoor environment difference distribution map and its corresponding predicted optimal PID control temperature; Constructing a first regression network, taking the indoor and outdoor environment difference distribution map in the indoor and outdoor environment difference training set as the first regression network input data, and taking the optimal PID control temperature in the indoor and outdoor environment difference training set as the first regression network output data, training the first regression network, and obtaining an initial first machine learning network; The initial first machine learning network is model verified using the indoor and outdoor environment difference test set, and the initial first machine learning network with a value less than or equal to a preset first test error threshold is output as the pre-constructed first machine learning model.

5. A smart building PID control method according to claim 4, characterized in that: The generation logic of the second machine learning model is: Acquire historical humidity control training data, and divide the historical humidity control training data into a humidity control training set and a humidity control test set; the historical humidity control training data includes an indoor ventilation coefficient, a PID controlled temperature and its corresponding predicted indoor controlled humidity; Constructing a second regression network, taking the indoor ventilation coefficient and PID control temperature in the humidity control training set as input data of the first regression network, and taking the indoor control humidity in the humidity control training set as output data of the second regression network, training the second regression network, and obtaining an initial second machine learning network; The initial second machine learning network is model verified using the indoor and outdoor environment difference test set, and the initial second machine learning network with a value less than or equal to a preset second test error threshold is output as the pre-constructed second machine learning model.

6. A smart building PID control method according to claim 5, characterized in that: The acquisition logic of the indoor humidity control is: P1: Set the initial humidity of the room to G, which is greater than zero; P2: Adjust the initial humidity, set J = G + a, and obtain the indoor controlled humidity J, where J and a are greater than zero; P3: Under the indoor controlled humidity J, obtain the dryness value of the surface layer of the human skin in the room. If the dryness value is within the preset dryness range, the adjusted humidity J is used as the optimal controlled humidity; if the dryness value is not within the preset dryness range, set a=a+k, and return to step P2, where k is an integer greater than zero; P4: Repeat the above steps P2 to P3 until the optimal indoor humidity is obtained, and then end the cycle.

7. A smart building PID control system, implemented based on the smart building PID control method according to claim 1, characterized in that: include: The difference analysis module obtains n sets of indoor and outdoor environment difference data and constructs a distribution map of indoor and outdoor environment differences based on n sets of outdoor environment data and indoor environment data collected in the time period [TK, T]; The first prediction module inputs the indoor and outdoor environment difference distribution map into the pre-built first machine learning model to predict the optimal PID control temperature of the room at the time [T+K, T+2K]; The second prediction module inputs the indoor ventilation coefficient collected in the time period [TK, T] and the generated optimal PID control temperature into the pre-built second machine learning model to predict the optimal indoor control humidity in the time period [T+K, T+2K]. The temperature and humidity control module corrects the indoor temperature and humidity according to the optimal PID control temperature and indoor control humidity.

Images