Building environment automatic control method and device based on distributed architecture

Abstract

The present application discloses a method and device for building environment self-control based on distributed architecture, the method comprising: receiving real-time environmental parameters fed back by various sensors pre-set in the building to be self-controlled; fusing the real-time environmental parameters with the historical environmental parameters to obtain the environmental parameters to be analyzed; inputting the environmental parameters to be analyzed into a pre-trained building environment self-control model, and outputting the target operating parameters of different environmental equipment in the preset time period corresponding to the current moment; the pre-trained building environment self-control model is deployed in the target subsystem, the target subsystem is selected from multiple subsystems of the distributed architecture, and the pre-trained building environment self-control model is a mathematical model for predicting the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment; based on the target operating parameters of different environmental equipment, adjusting the current operating parameters of different environmental equipment. Therefore, the accuracy of building environment control can be improved by adopting the embodiments of the present application.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method and device for automatic control of a building environment based on a distributed architecture. Background Art

[0002] In modern building automation systems, environmental control technology is widely used in commercial buildings, residential buildings, public facilities, etc. By automatically adjusting the operating parameters of environmental equipment that affects indoor temperature, humidity and lighting in response to changes in indoor and outdoor environments, precise control of the building environment can be achieved through automation, thereby improving the comfort of residents.

[0003] In the related technologies, existing building environment control systems usually rely on traditional automation technologies. These systems receive historical environmental parameters over a period of time through a central control unit and adjust the operating parameters of environmental equipment according to preset algorithms. These systems can adjust the operating parameters of environmental equipment that affects the environmental parameters in the building according to historical environmental parameters to adapt to environmental changes.

[0004] However, the central control unit is a centralized control system. When the central control unit fails, automatic control cannot be achieved. At the same time, when the weather environment changes too quickly, such as extreme weather events, the existing system cannot adjust the operating parameters of the environmental equipment in time to adapt to the rapid changes in the weather environment, thereby reducing the accuracy of building environmental control. Summary of the invention

[0005] The embodiments of the present application provide a method and device for building environment automatic control based on a distributed architecture. In order to have a basic understanding of some aspects of the disclosed embodiments, a simple summary is given below. This summary is not a general review, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its only purpose is to present some concepts in a simple form as a preface to the detailed description that follows.

[0006] In a first aspect, an embodiment of the present application provides a building environment automatic control method based on a distributed architecture, the method comprising:

[0007] Receive real-time environmental parameters fed back by various sensors pre-set in the building to be controlled;

[0008] The real-time environmental parameters are integrated with the historical environmental parameters to obtain the environmental parameters to be analyzed; wherein the historical environmental parameters are the environmental parameters of the target time period in a preset number of historical years, and the middle moment of the target time period is the same as the current moment;

[0009] Input the environmental parameters to be analyzed into the pre-trained building environment automatic control model, and output the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment; wherein the pre-trained building environment automatic control model is deployed in the target subsystem, the target subsystem is selected from multiple subsystems of the distributed architecture, and the pre-trained building environment automatic control model is a mathematical model for predicting the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment;

[0010] Based on the target operating parameters of different environmental devices, adjust the current operating parameters of different environmental devices.

[0011] Optionally, a pre-trained building environment automatic control model is generated by following the steps below, including:

[0012] Obtaining the first environmental parameter in each historical preset period in each historical year in the building to be automatically controlled;

[0013] Determine multiple period observation points from each historical preset period in each historical year;

[0014] Using the first environmental parameter of each periodic observation point as the real-time environmental parameter of each periodic observation point;

[0015] Using the first environmental parameters of multiple historical preset periods before each periodic observation point as the historical environmental parameters of each periodic observation point;

[0016] Construct model training samples based on the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point;

[0017] A pre-trained building environment automatic control model is generated based on the model training samples.

[0018] Optionally, a model training sample is constructed based on the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point, including:

[0019] Determine the optimal operating parameters of different environmental equipment at each periodic observation point based on the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point;

[0020] Using the optimal operating parameters of different environmental equipment at each periodic observation point, the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point are marked to obtain data samples of each periodic observation point;

[0021] The data samples of each period observation point are classified to obtain model training samples.

[0022] Optionally, according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point, the optimal operating parameters of different environmental devices at each periodic observation point are determined, including:

[0023] Construct a feature vector according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point;

[0024] Create constraints for different environmental devices;

[0025] Define objective functions for measuring the effects of equipment operating parameters in different environments;

[0026] Initialize the operating parameters of different environment equipment and obtain the operating parameters of different environment equipment;

[0027] According to the eigenvector, the gradient of the objective function at the operating parameter is calculated;

[0028] Update the operating parameters of equipment in different environments according to the gradient;

[0029] When the updated operating parameters of the different environmental devices do not satisfy the constraints of the different environmental devices, continue to perform the step of calculating the gradient of the objective function at the updated operating parameters according to the characteristic vector until the updated operating parameters of the different environmental devices satisfy the constraints of the different environmental devices; wherein,

[0030] The update formula of the operating parameters is:

[0031]

[0032] in, is the updated parameter, To update the previous parameters, is the learning rate, which is used to control the magnitude of each update step. is the gradient of the objective function at the operating parameter.

[0033] Optionally, the gradient of the objective function at the operating parameter is calculated based on the eigenvector, including:

[0034] Calculate energy consumption and user comfort based on the characteristic vector and operating parameters;

[0035] Calculate the target value of the objective function based on energy consumption and user comfort;

[0036] The gradient descent algorithm is used to calculate the partial derivative of the target value of the objective function with respect to the eigenvector, and the local rate of change of the objective function on the eigenvector is obtained;

[0037] The local rate of change is taken as the gradient of the objective function at the operating parameter; where,

[0038] The calculation formula of the objective value of the objective function is:

[0039]

[0040] in, is the target value of the objective function, are the operating parameters, is the eigenvector, and is a weighting factor used to balance the importance of energy consumption and user comfort; is the energy consumption, which depends on the operating parameters and the eigenvector ; is user comfort, which depends on operating parameters and the eigenvector .

[0041] Optionally, calculate energy consumption and user comfort based on the feature vector and operating parameters, including:

[0042] Calculate the power consumption of equipment in different environments according to the operating parameters of equipment in different environments;

[0043] The power consumption of the equipment in different environments is multiplied by the preset duration to obtain the energy consumption of the equipment in different environments;

[0044] Based on the feature vector, determine the weight factors of energy consumption of different environment equipment;

[0045] The energy consumption of different environmental devices is summed up using weight factors to obtain the energy consumption;

[0046] Calculate the comfort score based on the feature vector and operating parameters;

[0047] The comfort score is used as the user comfort; where,

[0048] The comfort score calculation formula is:

[0049]

[0050] in, Rate the comfort level. , , are the comfort score functions of temperature, humidity and light obtained according to the eigenvector and operating parameters, , , is the actual temperature, humidity and light intensity, , as well as is the weight coefficient of temperature, humidity and light intensity, are the operating parameters, is the eigenvector.

[0051] Optionally, the model training samples include data samples of multiple periodic observation points, and the data samples of each periodic observation point include label data and sample data, the label data are optimal operating parameters of different environmental devices at each periodic observation point, and the sample data are real-time environmental parameters of each periodic observation point and historical environmental parameters of each periodic observation point;

[0052] Generate a pre-trained building environment automatic control model based on the model training samples, including:

[0053] Use neural networks to create a building environment self-control model;

[0054] Input the data samples of each period observation point into the building environment automatic control model, and output the model cross entropy loss value;

[0055] When the cross entropy loss value of the model reaches the minimum, a pre-trained building environment automatic control model is generated;

[0056] Among them, the loss function of the building environment automatic control model is:

[0057]

[0058] in, is the model cross entropy loss value, is the number of data samples at multiple periodic observation points, For the Real-time environmental parameters of the observation points in the first cycle, The predicted value corresponding to the historical environmental parameters of the observation point in the period, For the The optimal operating parameters of equipment in different environments at each observation point in a cycle.

[0059] Optionally, before receiving the real-time environmental parameters fed back by various sensors pre-set in the building to be controlled, the method further includes:

[0060] Obtain system operation information of each subsystem from multiple subsystems of the distributed architecture;

[0061] Calculate the operational health of each subsystem based on the system operation information of each subsystem;

[0062] The subsystem with the highest running health is selected as the target subsystem;

[0063] Deploy the pre-trained building environment autonomous control model to the target subsystem.

[0064] Optionally, system operation information for each subsystem includes CPU and memory usage as well as error logs and failure rates;

[0065] Based on the system operation information of each subsystem, the operation health of each subsystem is calculated, including:

[0066] The CPU and memory usage as well as the error log and failure rate are summed to get the operational health of each subsystem.

[0067] In a second aspect, an embodiment of the present application provides a building environment automatic control device based on a distributed architecture, the device comprising:

[0068] A parameter receiving module is used to receive real-time environmental parameters fed back by various sensors pre-set in the building to be controlled;

[0069] A parameter fusion module is used to fuse the real-time environmental parameters with the historical environmental parameters to obtain the environmental parameters to be analyzed; wherein the historical environmental parameters are environmental parameters of the target time period in a preset number of historical years, and the middle moment of the target time period is the same as the current moment;

[0070] The parameter processing module is used to input the environmental parameters to be analyzed into the pre-trained building environment automatic control model, and output the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment; wherein the pre-trained building environment automatic control model is deployed in the target subsystem, the target subsystem is selected from multiple subsystems of the distributed architecture, and the pre-trained building environment automatic control model is a mathematical model for predicting the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment;

[0071] The environment equipment adjustment module is used to adjust the current operating parameters of different environment equipment based on the target operating parameters of different environment equipment.

[0072] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0073] In the embodiments of the present application, on the one hand, the present application adopts a distributed architecture. When a subsystem fails, other subsystems can continue to operate. By electing the subsystem with the highest health as the target subsystem, the stability and reliability of the overall system are guaranteed. On the other hand, the present application pre-trains the building environment self-control model. The pre-trained building environment self-control model is a mathematical model that predicts the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment. The self-control model deployed on the target subsystem can use real-time data and historical data to predict environmental changes and automatically adjust equipment operating parameters to achieve optimal energy use and indoor comfort, thereby improving the adaptability of the entire building and the comfort of residents when facing rapid changes in weather conditions.

[0074] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0076] Figure 1 It is a method flow diagram of a building environment automatic control method based on a distributed architecture provided in an embodiment of the present application;

[0077] Figure 2 is a schematic diagram of a data table before and after adjustment of operating parameters of an air-conditioning device and a lighting device provided in an embodiment of the present application;

[0078] Figure 3 It is a flow chart of a building environment automatic control model training method provided by the present application;

[0079] Figure 4 It is a schematic diagram of a data table of first environmental parameters in each historical preset period in each historical year provided by the present application;

[0080] Figure 5 It is a schematic diagram of a data table for periodic observation point settings provided by the present application;

[0081] Figure 6 It is a structural schematic diagram of a building environment automatic control device based on a distributed architecture provided by the present application;

[0082] Figure 7 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0083] The following description and the accompanying drawings sufficiently illustrate specific embodiments of the present application to enable those skilled in the art to practice them.

[0084] It should be clear that the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0085] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Instead, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0086] In the description of the present application, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances. In addition, in the description of the present application, unless otherwise specified, "multiple" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0087] At present, existing building environment control systems usually rely on traditional automation technology. These systems receive historical environmental parameters over a period of time through a central control unit and adjust the operating parameters of environmental equipment according to preset algorithms. These systems can adjust the operating parameters of environmental equipment that affects environmental parameters in the building based on historical environmental parameters to adapt to environmental changes.

[0088] The inventors realized that the central control unit is a centralized control, and when the central control unit fails, automatic control cannot be achieved. At the same time, when the weather environment changes too quickly, such as extreme weather events, the existing system cannot adjust the operating parameters of the environmental equipment in time to adapt to the rapid changes in the weather environment, thereby reducing the accuracy of building environmental control.

[0089] In order to solve the above problems, the present application provides a building environment automatic control method and device based on a distributed architecture to solve the problems existing in the above-mentioned related technical problems. In the embodiment of the present application, on the one hand, the present application adopts a distributed architecture. When a subsystem fails, other subsystems can continue to operate. By electing the subsystem with the highest health as the target subsystem, the stability and reliability of the overall system are guaranteed. On the other hand, the present application pre-trains the building environment automatic control model. The pre-trained building environment automatic control model is a mathematical model for predicting the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment, so that the automatic control model deployed on the target subsystem can use real-time data and historical data to predict environmental changes, and automatically adjust the equipment operating parameters to achieve optimal energy use and indoor comfort, thereby improving the adaptability of the entire building and the comfort of the residents when facing rapid changes in the weather environment. The following uses an exemplary embodiment for detailed description.

[0090] The following will be combined with the attached Figure 1 -Attached Figure 5 , a building environment automatic control method based on a distributed architecture provided by an embodiment of the present application is introduced in detail. The method can be implemented by a computer program and can be run on a building environment automatic control device based on a distributed architecture based on a von Neumann system. The computer program can be integrated into an application or run as an independent tool application.

[0091] See also Figure 1 , which is a flow chart of a building environment automatic control method based on a distributed architecture according to an embodiment of the present application. Figure 1 As shown, the method of the embodiment of the present application may include the following steps:

[0092] S101, receiving real-time environmental parameters fed back by various sensors pre-installed in the building to be controlled;

[0093] Among them, buildings to be controlled refer to buildings that are about to be equipped with an automated control system, which will automatically adjust and control the environment inside the building. Various sensors refer to devices installed inside the building to monitor environmental conditions, such as temperature sensors, humidity sensors, light sensors, and air quality sensors. Real-time environmental parameters refer to data monitored and fed back by sensors in real time, which reflect the current environmental conditions inside the building, such as the current temperature, humidity, light intensity, and air quality index.

[0094] In some embodiments of the present application, a variety of sensors are installed at different locations in a building. For example, temperature and humidity sensors are installed on each floor, light sensors are installed near windows, and air quality sensors are installed near ventilation systems. These sensors are equipped with data acquisition modules that can convert the monitored environmental parameters into electrical signals. A communication network is constructed so that the sensors can send the collected data to a local controller. The sensors monitor the environmental parameters in real time and send the data streams to the local controller through the communication network. The local controller receives the data streams from the sensors and obtains the real-time environmental parameters fed back by various sensors.

[0095] For example, in a large commercial office building, in order to achieve automatic control of the indoor environment, the management decided to install an automated control system. 10 temperature sensors, 10 humidity sensors, 5 light sensors, and 5 air quality sensors were installed on each floor. Each sensor is connected to a small microcontroller, which is responsible for the initial processing of the data and sending the data through a wireless network. The central control room of the office building has a local controller that continuously receives data from all sensors, thereby obtaining real-time environmental parameters fed back by various sensors.

[0096] In an embodiment of the present application, before receiving the real-time environmental parameters fed back by various sensors pre-set in the building to be automatically controlled, it is also necessary to deploy a model in the local controller, specifically including: obtaining the system operation information of each subsystem from multiple subsystems of the distributed architecture of the local controller; calculating the operating health of each subsystem based on the system operation information of each subsystem; selecting the subsystem with the highest operating health as the target subsystem; and deploying the pre-trained building environment automatic control model in the target subsystem.

[0097] The system operation information includes various data during the operation of the subsystem, such as performance indicators, error logs, resource utilization, etc. Operation health is a comprehensive indicator calculated based on this operation information, which is used to evaluate the performance status of the subsystem. The target subsystem is the subsystem selected as the best one among all subsystems based on health evaluation, and is used to deploy the self-control model.

[0098] This application collects operating information from multiple subsystems of the building automation system and calculates the operating health of each subsystem to ensure that the selected subsystem is in the best operating state. This method can not only improve the reliability and efficiency of the entire building environment automatic control system, but also ensure that the automatic control model operates in the optimal environment, thereby achieving more precise environmental control and higher energy efficiency. In addition, this method provides system resilience, ensuring that even if a subsystem fails, other subsystems can still operate normally, ensuring the comfort and continuity of management of the building environment.

[0099] The system operation information of each subsystem includes CPU and memory usage as well as error logs and failure rates.

[0100] Specifically, according to the system operation information of each subsystem, the specific process of calculating the operation health of each subsystem includes: summing up the CPU and memory usage as well as the error log and the failure rate to obtain the operation health of each subsystem.

[0101] S102, fusing the real-time environmental parameters with the historical environmental parameters to obtain the environmental parameters to be analyzed; wherein the historical environmental parameters are environmental parameters of a target time period in a preset number of historical years, and the middle moment of the target time period is the same as the current moment;

[0102] Among them, historical environmental parameters refer to environmental data collected during the same period of the past few years. Environmental parameters to be analyzed refer to the data set that combines real-time parameters with historical parameters for further analysis. Target time period refers to the past time interval corresponding to the current moment.

[0103] In some embodiments of the present application, a past time interval is determined according to the current moment, and the length of this time interval can be set as needed, such as the past week, month or year. The environmental parameter data corresponding to the target time period in the past few years are extracted from the database. The real-time environmental parameters are combined with the historical environmental parameters in a preset manner to obtain the environmental parameters to be analyzed. The fusion method can be to use a specific identifier for connection.

[0104] S103, inputting the environmental parameters to be analyzed into the pre-trained building environment automatic control model, and outputting the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment; wherein the pre-trained building environment automatic control model is deployed in the target subsystem, the target subsystem is selected from multiple subsystems of the distributed architecture, and the pre-trained building environment automatic control model is a mathematical model for predicting the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment;

[0105] The environmental parameters to be analyzed refer to the environmental data that have been prepared and used as model input, which will be used to predict and optimize the control strategy. The future preset period refers to a period of time starting from the current moment, such as the next hour, day or week. Different environmental equipment includes various equipment in the building, such as air conditioning, lighting, heating systems, etc. Their operating parameters need to be optimized to improve energy efficiency and comfort.

[0106] In some embodiments of the present application, after the pre-trained building environment automatic control model has been deployed in the target subsystem, the environmental parameters to be analyzed are input into the automatic control model. These parameters may include the current temperature, humidity, expected external temperature changes, etc. The automatic control model can calculate the target operating parameters of different environmental equipment in the future preset time period based on the input environmental parameters and its pre-trained algorithm. The model outputs the predicted target operating parameters, which are the set values ​​to which the equipment should be adjusted to achieve optimal energy efficiency and environmental comfort.

[0107] S104, adjusting current operating parameters of the different environment devices based on the target operating parameters of the different environment devices.

[0108] In some embodiments of the present application, the automatic control system receives target operating parameters for different environmental equipment from the building environment automatic control model. The system queries the current operating status and parameters of each environmental equipment, compares the target operating parameters with the current operating parameters, calculates the values ​​that need to be adjusted, and generates adjustment instructions based on the calculated differences. The instructions include the equipment parameters and adjustment amounts that need to be changed. The adjustment instructions are sent to the corresponding environmental equipment through the communication network of the control system. After receiving the adjustment instructions, each environmental equipment executes the parameter adjustment operation. After the equipment adjustment is completed, the new operating status and parameters are transmitted back to the automatic control system.

[0109] For example, the automatic control model outputs the target operating parameters, for example, the target temperature of the air conditioner is set to 22°C and the target brightness of the lighting system is 70%. The automatic control system calculates that the air conditioner temperature needs to be lowered by 2°C and the lighting system brightness needs to be lowered by 20%. The automatic control system generates adjustment instructions, which are "adjust the air conditioner temperature to 22°C" and "adjust the lighting brightness to 70%". The automatic control system sends adjustment instructions to the air conditioner and lighting systems through the building automation network. After receiving the instructions, the air conditioner and lighting systems adjust their respective operating parameters. After the adjustment is completed, the air conditioner and lighting systems feed back the new operating status to the automatic control system.

[0110] Schematic diagram of the data table before and after the adjustment of the operating parameters of the air conditioning equipment and lighting equipment. Figure 2 shown.

[0111] In the embodiments of the present application, on the one hand, the present application adopts a distributed architecture. When a subsystem fails, other subsystems can continue to operate. By electing the subsystem with the highest health as the target subsystem, the stability and reliability of the overall system are guaranteed. On the other hand, the present application pre-trains the building environment self-control model. The pre-trained building environment self-control model is a mathematical model that predicts the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment. The self-control model deployed on the target subsystem can use real-time data and historical data to predict environmental changes and automatically adjust equipment operating parameters to achieve optimal energy use and indoor comfort, thereby improving the adaptability of the entire building and the comfort of residents when facing rapid changes in weather conditions.

[0112] See also Figure 3 , which is a flow chart of a method for training a building environment automatic control model according to an embodiment of the present application. Figure 3 As shown, the method of the embodiment of the present application may include the following steps:

[0113] S201, obtaining the first environmental parameter in each historical preset period in each historical year in the building to be automatically controlled;

[0114] Among them, each historical year refers to each year in the past, such as 2019, 2020, 2021, etc., which is used to collect and analyze historical data. The historical preset period refers to a pre-defined time period in the historical year, which is used for data sampling and analysis. These periods can be divided by hours, days, weeks or months. The first environmental parameter refers to the environmental parameter value recorded at the beginning of each historical preset period. For example, if the period is every hour, the first environmental parameter may be the temperature reading at the beginning of the hour. These parameters are generally used to represent the initial environmental state of the entire cycle. Environmental parameters include but are not limited to temperature, humidity, light intensity, air quality index, etc., which describe the environmental conditions inside the building.

[0115] For example Figure 4 As shown, Figure 4 This application provides a data representation of the first environmental parameters in each historical preset period in each historical year. The data table can provide historical environmental parameters for the building automatic control system, help the system learn environmental change trends, predict future environmental changes, and automatically adjust the equipment operating parameters in the building to maintain a comfortable and energy-saving indoor environment.

[0116] S202, determining a plurality of period observation points from each historical preset period in each historical year;

[0117] Among them, the periodic observation point is a specific time point determined in the historical preset cycle, which is used to collect and analyze environmental data.

[0118] In some embodiments of the present application, 2018, 2019 and 2020 are first selected as historical years, and then the first day of each month is determined as the historical preset cycle, and finally on the first day of each month, several key time points are selected as periodic observation points, such as 00:00, 06:00, 12:00 and 18:00.

[0119] For example Figure 5 As shown, Figure 5 This is a data representation intent for setting a periodic observation point provided by the present application. Periodic observation points can be set within preset periods in different years.

[0120] S203, using the first environmental parameter of each periodic observation point as the real-time environmental parameter of each periodic observation point;

[0121] S204, taking the first environmental parameters of a plurality of historical preset periods before each periodic observation point as the historical environmental parameters of each periodic observation point;

[0122] S205, constructing a model training sample according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point;

[0123] In some embodiments of the present application, the specific process of constructing a model training sample based on the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point includes: determining the optimal operating parameters of different environmental equipment at each periodic observation point based on the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point; using the optimal operating parameters of different environmental equipment at each periodic observation point to annotate the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point to obtain data samples for each periodic observation point; classifying the data samples of each periodic observation point to obtain model training samples.

[0124] Specifically, the specific process of determining the optimal operating parameters of different environmental devices at each periodic observation point according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point includes: constructing a feature vector according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point; creating constraints for different environmental devices; defining an objective function for measuring the effects of operating parameters of different environmental devices; initializing the operating parameters of different environmental devices to obtain the operating parameters of different environmental devices; calculating the gradient of the objective function at the operating parameters according to the feature vector; updating the operating parameters of different environmental devices according to the gradient; when the updated operating parameters of different environmental devices do not meet the constraints of different environmental devices, continuing to execute the step of calculating the gradient of the objective function at the updated operating parameters according to the feature vector until the updated operating parameters of different environmental devices meet the constraints of different environmental devices; wherein the updating formula of the operating parameters is:

[0125]

[0126] in, is the updated parameter, To update the previous parameters, is the learning rate, which is used to control the magnitude of each update step. is the gradient of the objective function at the operating parameter.

[0127] Specifically, according to the eigenvector, the specific process of calculating the gradient of the objective function at the operating parameters includes: calculating the energy consumption and user comfort according to the eigenvector and the operating parameters; calculating the target value of the objective function based on the energy consumption and the user comfort; using the gradient descent algorithm to calculate the partial derivative of the target value of the objective function with respect to the eigenvector to obtain the local change rate of the objective function on the eigenvector; using the local change rate as the gradient of the objective function at the operating parameters; wherein the calculation formula of the target value of the objective function is:

[0128]

[0129] in, is the target value of the objective function, are the operating parameters, is the eigenvector, and is a weighting factor used to balance the importance of energy consumption and user comfort; is the energy consumption, which depends on the operating parameters and the eigenvector ; is user comfort, which depends on operating parameters and the eigenvector .

[0130] Specifically, the specific process of calculating energy consumption and user comfort according to the characteristic vector and the operating parameters includes: calculating the power consumption of different environmental equipment according to the operating parameters of different environmental equipment; multiplying the power consumption of different environmental equipment by the preset time to obtain the energy consumption of different environmental equipment; determining the weight factor of the energy consumption of different environmental equipment based on the characteristic vector; using the weight factor to sum the energy consumption of different environmental equipment to obtain the energy consumption; calculating the comfort score according to the characteristic vector and the operating parameters; using the comfort score as the user comfort; wherein the comfort score calculation formula is:

[0131]

[0132] in, Rate the comfort level. , , are the comfort score functions of temperature, humidity and light obtained according to the eigenvector and operating parameters, , , is the actual temperature, humidity and light intensity, , as well as is the weight coefficient of temperature, humidity and light intensity, are the operating parameters, is the eigenvector.

[0133] S206: Generate a pre-trained building environment automatic control model based on the model training samples.

[0134] Among them, the model training samples include data samples from multiple periodic observation points. The data samples of each periodic observation point include label data and sample data. The label data is the optimal operating parameters of different environmental equipment at each periodic observation point. The sample data is the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point.

[0135] In some embodiments of the present application, the specific process of generating a pre-trained building environment automatic control model according to the model training sample includes: using a neural network to create a building environment automatic control model; inputting the data sample of each period observation point into the building environment automatic control model, and outputting the model cross entropy loss value; when the model cross entropy loss value reaches the minimum, generating a pre-trained building environment automatic control model; wherein the loss function of the building environment automatic control model is:

[0136]

[0137] in, is the model cross entropy loss value, is the number of data samples at multiple periodic observation points, For the Real-time environmental parameters of the observation points in the first cycle, The predicted value corresponding to the historical environmental parameters of the observation point in the period, For the The optimal operating parameters of equipment in different environments at each observation point in a cycle.

[0138] In an embodiment of the present application, by obtaining the first environmental parameters of a preset cycle of each historical year in the building to be controlled, and determining multiple periodic observation points, and then defining the initial environmental parameters of these observation points as real-time environmental parameters, and defining the environmental parameters of the same time point of the previous cycle as historical environmental parameters, a model training sample that integrates historical and real-time data can be constructed. This training sample can provide rich information for the building automatic control model, allowing the model to learn the patterns and trends of environmental changes. The pre-trained building environment automatic control model trained with these data can more accurately predict future environmental changes and automatically adjust the building system to maintain optimal comfort and energy efficiency, thereby improving the intelligence level and response efficiency of building environmental control.

[0139] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.

[0140] See also Figure 6 , which shows a schematic diagram of the structure of a building environment automatic control device based on a distributed architecture provided by an exemplary embodiment of the present application. The building environment automatic control device based on a distributed architecture can be implemented as all or part of an electronic device through software, hardware or a combination of both. The device 1 includes a parameter receiving module 10, a parameter fusion module 20, a parameter processing module 30, and an environmental equipment adjustment module 40.

[0141] The parameter receiving module 10 is used to receive real-time environmental parameters fed back by various sensors pre-set in the building to be controlled;

[0142] The parameter fusion module 20 is used to fuse the real-time environmental parameters with the historical environmental parameters to obtain the environmental parameters to be analyzed; wherein the historical environmental parameters are environmental parameters of a target time period in a preset number of historical years, and the middle moment of the target time period is the same as the current moment;

[0143] The parameter processing module 30 is used to input the environmental parameters to be analyzed into the pre-trained building environment automatic control model, and output the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment; wherein the pre-trained building environment automatic control model is deployed in the target subsystem, and the target subsystem is selected from multiple subsystems of the distributed architecture, and the pre-trained building environment automatic control model is a mathematical model for predicting the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment;

[0144] The environmental equipment adjustment module 40 is used to adjust the current operating parameters of different environmental equipment based on the target operating parameters of different environmental equipment.

[0145] It should be noted that the building environment automatic control device based on a distributed architecture provided in the above embodiment only uses the division of the above functional modules as an example when executing the building environment automatic control method based on a distributed architecture. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the building environment automatic control device based on a distributed architecture provided in the above embodiment and the building environment automatic control method based on a distributed architecture belong to the same concept. The implementation process is detailed in the method embodiment and will not be repeated here.

[0146] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0147] In the embodiments of the present application, on the one hand, the present application adopts a distributed architecture. When a subsystem fails, other subsystems can continue to operate. By electing the subsystem with the highest health as the target subsystem, the stability and reliability of the overall system are guaranteed. On the other hand, the present application pre-trains the building environment self-control model. The pre-trained building environment self-control model is a mathematical model that predicts the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment. The self-control model deployed on the target subsystem can use real-time data and historical data to predict environmental changes and automatically adjust equipment operating parameters to achieve optimal energy use and indoor comfort, thereby improving the adaptability of the entire building and the comfort of residents when facing rapid changes in weather conditions.

[0148] The present application also provides a computer-readable medium having program instructions stored thereon, which, when executed by a processor, implements the building environment automatic control method based on a distributed architecture provided by the above-mentioned various method embodiments.

[0149] The present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute the building environment automatic control method based on a distributed architecture of each of the above method embodiments.

[0150] See also Figure 7 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 7 As shown, the electronic device 1000 may include: at least one processor 1001 , at least one network interface 1004 , a user interface 1003 , a memory 1005 , and at least one communication bus 1002 .

[0151] The communication bus 1002 is used to realize the connection and communication between these components.

[0152] The user interface 1003 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 1003 may also include a standard wired interface and a wireless interface.

[0153] The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0154] Among them, the processor 1001 may include one or more processing cores. The processor 1001 uses various interfaces and lines to connect various parts of the entire electronic device 1000, and executes various functions and processes data of the electronic device 1000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1005, and calling data stored in the memory 1005. Optionally, the processor 1001 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 1001 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 1001, and it can be implemented separately through a chip.

[0155] Among them, the memory 1005 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 1005 includes a non-transitory computer-readable storage medium. The memory 1005 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1005 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1005 may also be optionally at least one storage system located away from the aforementioned processor 1001. As Figure 7 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a building environment automatic control application based on a distributed architecture.

[0156] exist Figure 7 In the electronic device 1000 shown, the user interface 1003 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 1001 can be used to call the building environment automatic control application based on the distributed architecture stored in the memory 1005, and specifically perform the following operations:

[0157] Receive real-time environmental parameters fed back by various sensors pre-set in the building to be controlled;

[0158] The real-time environmental parameters are integrated with the historical environmental parameters to obtain the environmental parameters to be analyzed; wherein the historical environmental parameters are the environmental parameters of the target time period in a preset number of historical years, and the middle moment of the target time period is the same as the current moment;

[0159] Input the environmental parameters to be analyzed into the pre-trained building environment automatic control model, and output the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment; wherein the pre-trained building environment automatic control model is deployed in the target subsystem, the target subsystem is selected from multiple subsystems of the distributed architecture, and the pre-trained building environment automatic control model is a mathematical model for predicting the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment;

[0160] Based on the target operating parameters of different environmental devices, adjust the current operating parameters of different environmental devices.

[0161] In one embodiment, when the processor 1001 generates a pre-trained building environment automatic control model, the processor 1001 specifically performs the following operations:

[0162] Obtaining the first environmental parameter in each historical preset period in each historical year in the building to be automatically controlled;

[0163] Determine multiple period observation points from each historical preset period in each historical year;

[0164] Using the first environmental parameter of each periodic observation point as the real-time environmental parameter of each periodic observation point;

[0165] Using the first environmental parameters of multiple historical preset periods before each periodic observation point as the historical environmental parameters of each periodic observation point;

[0166] Construct model training samples based on the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point;

[0167] A pre-trained building environment automatic control model is generated based on the model training samples.

[0168] In one embodiment, when the processor 1001 constructs the model training sample according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point, the processor 1001 specifically performs the following operations:

[0169] Determine the optimal operating parameters of different environmental equipment at each periodic observation point based on the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point;

[0170] Using the optimal operating parameters of different environmental equipment at each periodic observation point, the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point are marked to obtain data samples of each periodic observation point;

[0171] The data samples of each period observation point are classified to obtain model training samples.

[0172] In one embodiment, when the processor 1001 determines the optimal operating parameters of different environmental devices at each periodic observation point according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point, the processor 1001 specifically performs the following operations:

[0173] Construct a feature vector according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point;

[0174] Create constraints for different environmental devices;

[0175] Define objective functions for measuring the effects of equipment operating parameters in different environments;

[0176] Initialize the operating parameters of different environment equipment and obtain the operating parameters of different environment equipment;

[0177] According to the eigenvector, the gradient of the objective function at the operating parameter is calculated;

[0178] Update the operating parameters of equipment in different environments according to the gradient;

[0179] When the updated operating parameters of the different environmental devices do not satisfy the constraints of the different environmental devices, continue to perform the step of calculating the gradient of the objective function at the updated operating parameters according to the characteristic vector until the updated operating parameters of the different environmental devices satisfy the constraints of the different environmental devices; wherein,

[0180] The update formula of the operating parameters is:

[0181]

[0182] in, is the updated parameter, To update the previous parameters, is the learning rate, which is used to control the magnitude of each update step. is the gradient of the objective function at the operating parameter.

[0183] In one embodiment, when the processor 1001 calculates the gradient of the objective function at the operating parameter according to the feature vector, the processor 1001 specifically performs the following operations:

[0184] Calculate energy consumption and user comfort based on the characteristic vector and operating parameters;

[0185] Calculate the target value of the objective function based on energy consumption and user comfort;

[0186] The gradient descent algorithm is used to calculate the partial derivative of the target value of the objective function with respect to the eigenvector, and the local rate of change of the objective function on the eigenvector is obtained;

[0187] The local rate of change is taken as the gradient of the objective function at the operating parameter; where,

[0188] The calculation formula of the objective value of the objective function is:

[0189]

[0190] in, is the target value of the objective function, are the operating parameters, is the eigenvector, and is a weighting factor used to balance the importance of energy consumption and user comfort; is the energy consumption, which depends on the operating parameters and the eigenvector ; is user comfort, which depends on operating parameters and the eigenvector .

[0191] In one embodiment, when the processor 1001 calculates energy consumption and user comfort according to the feature vector and the operating parameters, the processor 1001 specifically performs the following operations:

[0192] Calculate the power consumption of equipment in different environments according to the operating parameters of equipment in different environments;

[0193] The power consumption of the equipment in different environments is multiplied by the preset duration to obtain the energy consumption of the equipment in different environments;

[0194] Based on the feature vector, determine the weight factors of energy consumption of different environment equipment;

[0195] The energy consumption of different environmental devices is summed up using weight factors to obtain the energy consumption;

[0196] Calculate the comfort score based on the feature vector and operating parameters;

[0197] The comfort score is used as the user comfort; where,

[0198] The comfort score calculation formula is:

[0199]

[0200] in, Rate the comfort level. , , are the comfort score functions of temperature, humidity and light obtained according to the eigenvector and operating parameters, , , is the actual temperature, humidity and light intensity, , as well as is the weight coefficient of temperature, humidity and light intensity, are the operating parameters, is the eigenvector.

[0201] In one embodiment, when the processor 1001 generates a pre-trained building environment automatic control model according to the model training sample, the processor 1001 specifically performs the following operations:

[0202] Use neural networks to create a building environment self-control model;

[0203] Input the data samples of each period observation point into the building environment automatic control model, and output the model cross entropy loss value;

[0204] When the cross entropy loss value of the model reaches the minimum, a pre-trained building environment automatic control model is generated;

[0205] Among them, the loss function of the building environment automatic control model is:

[0206]

[0207] in, is the model cross entropy loss value, is the number of data samples at multiple periodic observation points, For the Real-time environmental parameters of the observation points in the first cycle, The predicted value corresponding to the historical environmental parameters of the observation point in the period, For the The optimal operating parameters of equipment in different environments at each observation point in a cycle.

[0208] In one embodiment, before receiving the real-time environmental parameters fed back by various sensors pre-set in the building to be controlled, the processor 1001 further performs the following operations:

[0209] Obtain system operation information of each subsystem from multiple subsystems of the distributed architecture;

[0210] Calculate the operational health of each subsystem based on the system operation information of each subsystem;

[0211] The subsystem with the highest running health is selected as the target subsystem;

[0212] Deploy the pre-trained building environment autonomous control model to the target subsystem.

[0213] In one embodiment, when the processor 1001 calculates the operation health of each subsystem according to the system operation information of each subsystem, the processor 1001 specifically performs the following operations:

[0214] The CPU and memory usage as well as the error log and failure rate are summed to get the operational health of each subsystem.

[0215] In the embodiments of the present application, on the one hand, the present application adopts a distributed architecture. When a subsystem fails, other subsystems can continue to operate. By electing the subsystem with the highest health as the target subsystem, the stability and reliability of the overall system are guaranteed. On the other hand, the present application pre-trains the building environment self-control model. The pre-trained building environment self-control model is a mathematical model that predicts the target operating parameters of different environmental equipment in the future preset time period corresponding to the current moment. The self-control model deployed on the target subsystem can use real-time data and historical data to predict environmental changes and automatically adjust equipment operating parameters to achieve optimal energy use and indoor comfort, thereby improving the adaptability of the entire building and the comfort of residents when facing rapid changes in weather conditions.

[0216] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program of building environment automatic control based on a distributed architecture can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium of the program of building environment automatic control based on a distributed architecture can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0217] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. A building environment automatic control method based on distributed architecture, characterized in that: The method comprises: Receive real-time environmental parameters fed back by various sensors pre-set in the building to be controlled; The real-time environmental parameters are integrated with the historical environmental parameters to obtain the environmental parameters to be analyzed; wherein the historical environmental parameters are environmental parameters of a target time period in a preset number of historical years, and the middle moment of the target time period is the same as the current moment; Input the environmental parameters to be analyzed into a pre-trained building environment automatic control model, and output the target operating parameters of different environmental equipment in a future preset time period corresponding to the current moment; wherein the pre-trained building environment automatic control model is deployed in a target subsystem, and the target subsystem is selected from multiple subsystems of a distributed architecture, and the pre-trained building environment automatic control model is a mathematical model for predicting the target operating parameters of different environmental equipment in a future preset time period corresponding to the current moment; Based on the target operating parameters of the different environment devices, the current operating parameters of the different environment devices are adjusted; wherein, Follow these steps to generate a pre-trained building environment autonomous control model, including: Obtaining the first environmental parameter in each historical preset period in each historical year in the building to be automatically controlled; Determining a plurality of period observation points from each of the historical preset periods in each of the historical years; Using the first environmental parameter of each periodic observation point as the real-time environmental parameter of each periodic observation point; Using the first environmental parameters of multiple historical preset periods before each periodic observation point as the historical environmental parameters of each periodic observation point; Constructing a model training sample according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point; According to the model training samples, a pre-trained building environment automatic control model is generated; wherein, The constructing of model training samples according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point includes: Determine optimal operating parameters of different environmental equipment at each periodic observation point according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point; Using the optimal operating parameters of different environmental equipment at each periodic observation point, marking the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point, and obtaining data samples of each periodic observation point; The data samples of each period observation point are classified to obtain model training samples; wherein, The determining the optimal operating parameters of different environmental devices at each periodic observation point according to the real-time environmental parameters at each periodic observation point and the historical environmental parameters at each periodic observation point includes: Constructing a feature vector according to the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point; Creating constraints for the different environmental devices; defining an objective function for measuring the effects of the operating parameters of the different environmental equipment; Initializing the operating parameters of the different environment devices to obtain the operating parameters of the different environment devices; Calculating the gradient of the objective function at the operating parameter according to the characteristic vector; According to the gradient, updating the operating parameters of the different environment equipment; When the updated operating parameters of the different environmental devices do not satisfy the constraints of the different environmental devices, continue to execute the step of calculating the gradient of the objective function at the updated operating parameters according to the feature vector until the updated operating parameters of the different environmental devices satisfy the constraints of the different environmental devices.

2. The method according to claim 1, characterized in that in, The updating formula of the operating parameters is: in, is the updated parameter, To update the previous parameters, is the learning rate, which is used to control the magnitude of each update step. is the gradient of the objective function at the operating parameter.

3. The method according to claim 1, characterized in that The step of calculating the gradient of the objective function at the operating parameter according to the characteristic vector includes: Calculating energy consumption and user comfort according to the characteristic vector and the operating parameters; Calculating a target value of the target function based on the energy consumption and user comfort; Using a gradient descent algorithm, calculate the partial derivative of the target value of the objective function with respect to the eigenvector to obtain a local rate of change of the objective function on the eigenvector; The local change rate is used as the gradient of the objective function at the operating parameter; wherein, The calculation formula of the target value of the objective function is: in, is the target value of the objective function, are the operating parameters, is the eigenvector, and is a weighting factor used to balance the importance of energy consumption and user comfort; is the energy consumption, which depends on the operating parameters and the eigenvector ; is user comfort, which depends on operating parameters and the eigenvector .

4. The method according to claim 3, characterized in that The calculating of energy consumption and user comfort according to the characteristic vector and the operating parameters includes: Calculating the power consumption of the different environment devices according to the operating parameters of the different environment devices; The power consumption of the equipment in different environments is multiplied by the preset time to obtain the energy consumption of the equipment in different environments; Based on the feature vector, determining the weight factor of the energy consumption of the different environment devices; Using the weight factor, summing the energy consumption of the different environment devices to obtain energy consumption; Calculating a comfort score according to the characteristic vector and the operating parameters; The comfort score is used as the user comfort; wherein, The comfort score calculation formula is: in, Rate the comfort level. , , are respectively the comfort score functions of temperature, humidity and light obtained according to the characteristic vector and the operating parameters, , , is the actual temperature, humidity and light intensity, , as well as is the weight coefficient of temperature, humidity and light intensity, are the operating parameters, is the eigenvector.

5. The method according to claim 1, characterized in that The model training samples include data samples of multiple periodic observation points, and the data samples of each periodic observation point include label data and sample data. The label data are the optimal operating parameters of different environmental equipment at each periodic observation point, and the sample data are the real-time environmental parameters of each periodic observation point and the historical environmental parameters of each periodic observation point. The step of generating a pre-trained building environment automatic control model according to the model training sample comprises: Use neural networks to create a building environment self-control model; Inputting the data samples of each periodic observation point into the building environment automatic control model, and outputting the model cross entropy loss value; When the cross entropy loss value of the model reaches a minimum, generating a pre-trained building environment automatic control model; Among them, the loss function of the building environment automatic control model is: in, is the model cross entropy loss value, is the number of data samples at multiple periodic observation points, For the Real-time environmental parameters of the observation points in the first cycle, The predicted value corresponding to the historical environmental parameters of the observation point in the period, For the The optimal operating parameters of equipment in different environments at each observation point in a cycle.

6. The method according to claim 1, characterized in that Before receiving the real-time environmental parameters fed back by various sensors pre-set in the building to be controlled, the method further includes: Obtaining system operation information of each subsystem from multiple subsystems of the distributed architecture; Calculating the operation health of each subsystem according to the system operation information of each subsystem; Selecting the subsystem with the highest running health as the target subsystem; The pre-trained building environment automatic control model is deployed in the target subsystem.

7. The method according to claim 6, characterized in that The system operation information of each subsystem includes CPU and memory usage as well as error logs and failure rates; The calculating the operation health of each subsystem according to the system operation information of each subsystem includes: The CPU and memory usage rates as well as the error log and failure rates are summed to obtain the operational health of each subsystem.

8. A building environment automatic control device based on a distributed architecture implemented using the method described in any one of claims 1 to 7, characterized in that: The device comprises: A parameter receiving module is used to receive real-time environmental parameters fed back by various sensors pre-set in the building to be controlled; A parameter fusion module, used to fuse the real-time environmental parameters with historical environmental parameters to obtain environmental parameters to be analyzed; wherein the historical environmental parameters are environmental parameters of a target time period in a preset number of historical years, and the middle moment of the target time period is the same as the current moment; A parameter processing module, used for inputting the environmental parameters to be analyzed into a pre-trained building environment automatic control model, and outputting target operating parameters of different environmental equipment in a future preset period corresponding to the current moment; wherein the pre-trained building environment automatic control model is deployed in a target subsystem, and the target subsystem is selected from multiple subsystems of a distributed architecture, and the pre-trained building environment automatic control model is a mathematical model for predicting target operating parameters of different environmental equipment in a future preset period corresponding to the current moment; The environmental equipment adjustment module is used to adjust the current operating parameters of the different environmental equipment based on the target operating parameters of the different environmental equipment.

Images