A photovoltaic rolling curtain system for residential buildings in cold regions and its dynamic intelligent control method

Abstract

The present invention proposes a photovoltaic rolling curtain system for residential buildings in cold regions and its dynamic intelligent control method. The method includes: collecting building geometric information and data on the external environmental light and weather conditions, and performing optical and thermal environment simulation and analysis through the RHINO-LBT software platform; using indoor infrared cameras and illuminance meters to collect real-time data and establishing a measured and simulated data set; training through Convolutional Neural Networks (CNN) to construct a prediction model for the comfort of the indoor optical and thermal environment; quickly and real-time evaluating the indoor optical and thermal comfort according to the prediction model, and dynamically and intelligently controlling the photovoltaic rolling curtain system based on the evaluation results. This method not only optimizes the utilization of natural light and indoor temperature control, but also converts solar energy into electrical energy, improves the building's energy self-sufficiency ability, and reduces greenhouse gas emissions. The invention has theoretical value and practical significance. By intelligently adjusting the rolling curtain, it significantly improves the comfort of the indoor optical and thermal environment, provides a healthier and more comfortable living environment for residents in cold regions, and at the same time supports the realization of the carbon neutrality goal.

Description

Technical Field

[0001] The present invention belongs to the fields of building energy conservation and building technology, and particularly relates to a photovoltaic rolling curtain system for residential buildings in severe cold regions and a dynamic intelligent regulation method thereof. The method is an exploration of a dynamic regulation method for building sunshade components based on a convolutional neural network. Background Art

[0002] In recent years, building science has made remarkable progress in the context of carbon neutrality and carbon peaking. In particular, it has shown good implementability in promoting green buildings and energy conservation and emission reduction strategies, and has brought good economic and social benefits. This progress is reflected in the extensive use of sustainable materials, the promotion of high-energy-efficiency design concepts, and the application of intelligent energy management systems in new buildings, especially in the fields of public and office buildings. However, in the face of a large number of existing buildings, especially residential buildings, the further development of related building technologies involving renovation still faces major challenges. These challenges include the high technical difficulty of renovating old buildings, the relatively high initial investment cost, the lack of unified standards and policy support, and the extensiveness of renovation and operation and maintenance plans. These problems jointly restrict the progress speed and efficiency of building science in achieving the carbon neutrality and carbon peaking goals.

[0003] The evaluation of indoor light environment and indoor thermal comfort, as the most common evaluation methods for building indoor environment, has also developed. At the same time, there have been many studies on the multi-factor coupling evaluation of indoor environment in recent years. The evaluation of indoor light-thermal coupling environment involves comprehensive considerations of multiple aspects such as the thermal performance of buildings, natural lighting, indoor temperature distribution, and human comfort. Through the evaluation of these factors, building design can be guided, such as the size and position of windows, the selection of thermal insulation materials for walls and roofs, and the optimization of indoor layout, so that the building can effectively utilize renewable resources for heating and lighting, while maintaining a good indoor temperature and humidity environment, and improving the comfort and health level of residents.

[0004] In addition, the evaluation of indoor light-thermal coupling environment can also provide a scientific basis for the formulation of energy conservation and emission reduction policies and promote the development of sustainable building technologies. With the dual challenges of energy and environment brought about by global climate change, the evaluation of light-thermal coupling environment in cold-region residential buildings not only helps to reduce building energy consumption, but also promotes the formation of a green and low-carbon lifestyle, contributing to the response to climate change.

[0005] In residential buildings in cold regions, studying the coupling of indoor light and heat environment is of great significance for achieving efficient energy utilization and improving living comfort. Buildings in cold regions face severe energy challenges, including high insulation requirements, high heating energy consumption, and extensive operation and maintenance after renovation. Therefore, how to effectively utilize natural light and heat energy and reduce energy consumption through architectural design and technical means is the key to improving building energy efficiency. At the same time, traditional indoor light and heat environment simulation and evaluation of buildings takes a long time, requires high computing performance of the simulation platform, and cannot provide real-time feedback and dynamic adjustment.

[0006] Therefore, the research on the system and method described in the present invention is not only of theoretical value, but also of practical significance. Relying on this method, rapid feedback can be given according to the indoor environment, so that photovoltaic blinds can increase the photoelectric conversion efficiency when there is sufficient sunlight, and provide self-produced and self-used clean energy for buildings. When the light is insufficient or excessive, the angle is adjusted to control the light and heat entering the room, thereby effectively controlling the indoor temperature and reducing energy consumption. The application of this method helps to improve the energy self-sufficiency of buildings, reduce greenhouse gas emissions, and support the achievement of carbon neutrality goals. Summary of the invention

[0007] The purpose of the present invention is to solve the problems in the prior art, and a photovoltaic roller shutter system for residential buildings in severe cold regions and a dynamic intelligent control method thereof are proposed. The method inputs building geometry information and reads sky environment module data, performs light and heat environment simulation and analysis based on the RHINO-LBT software platform, and collects real-time indoor infrared images and illuminance values based on indoor infrared cameras and illuminance meters, constructs measured data images and simulation data sets, and trains through convolutional neural networks (CNN) in machine learning to obtain a prediction model for indoor light and heat environment comfort based on infrared images and illuminance data, and evaluates the real-time light and heat comfort based on the model, thereby constructing a photovoltaic roller shutter system for residential buildings in severe cold regions based on the implementation of light and heat comfort feedback evaluation and a dynamic intelligent control method thereof.

[0008] The present invention is realized by the following technical scheme. The present invention proposes a photovoltaic roller shutter system for residential buildings in severe cold regions. The system includes a building physical information input module, a sky environment module, an indoor visual information input module, a non-visual information input module, a prediction model establishment module for indoor light and heat environment comfort based on infrared images and illumination data, and a photovoltaic roller shutter dynamic control module;

[0009] The building physical information input module: divides the grid for the indoor floor plan and the window opening sizes of the building facades, assigns a unique RFID number to each room according to its functional use in the indoor space of the building, assigns a unique WID number to the windows of each room according to the window size information, and simultaneously loads the longitude and latitude coordinates of the location where the building is located and the three-dimensional environmental coordinates of the building to determine the building orientation and the skylight environmental characteristics of the location where it is located;

[0010] The sky environment module: is used to collect external environmental light and weather condition data for the RHINO-LBT software platform to conduct light and heat environment simulation and analysis;

[0011] The indoor visual information input module: numbers the infrared information collection points IR-x and the indoor illuminance collection points Lux-x, uses an indoor infrared camera as the information input terminal to record the indoor infrared images in real time, standardizes the infrared images to make them suitable for the input requirements of the CNN prediction model, and collects the illuminance values through an illuminometer to provide a data basis for the CNN model training;

[0012] The non-visual information input module: according to the requirements of the indoor light environment and heat environment under different room functions and different intensities of human activities, detects the comfortable illuminance values and indoor thermal comfort ranges through questionnaires and human physiological information, and assigns a number TSID to each possible indoor environment setting and human thermal sensation model based on different scenarios and personal preferences, and normalizes the measured data;

[0013] The prediction model establishment module: trains through the convolutional neural network CNN in machine learning to obtain a prediction model for the indoor light and heat environment comfort based on infrared images and illuminance data;

[0014] The photovoltaic roller shutter dynamic regulation module: relies on this prediction model to dynamically regulate the photovoltaic roller shutter system to optimize the indoor light and heat environment.

[0015] Further, the adjustment rules of the building photovoltaic roller shutter system are specifically as follows: number different controllers respectively, and each controller of the photovoltaic roller shutter is assigned a unique CID number; set the goal of indoor comfort, with the indoor thermal comfort and light environment as the optimal range and energy collection as the reference index; dynamically adjust the priority of comfort and energy-saving factors under different circumstances.

[0016] Further, the corresponding relationships among the building room number RFID, window number WID, human thermal sensation model number TSID, infrared information collection point IR-x, indoor illuminance collection point Lux-x, and controller number CID in the system are as follows:

[0017] According to the input building space data, the building room numbers and window numbers with different functions respectively correspond to different human thermal sensation model numbers TSID, and are in one-to-one correspondence with the infrared information collection points IR-x, indoor illuminance collection points Lux-x and controller numbers CID.

[0018] The present invention also proposes a dynamic intelligent control method based on the photovoltaic rolling curtain system for residential buildings in cold regions, and the method includes the following steps:

[0019] Step 1: Collect and input the geometric information of the building, and obtain the external environmental light and weather condition data through the sky detection module;

[0020] Step 2: Install infrared cameras and illuminometers indoors to collect indoor infrared images and illuminance values in real time for monitoring the indoor light and heat environment status;

[0021] Step 3: Based on the RHINO-LBT software platform and the measured data, conduct light and heat environment simulation and analysis, and train a convolutional neural network CNN through machine learning methods to establish a prediction model for the indoor light and heat environment comfort level;

[0022] Step 4: According to the results of the prediction model, dynamically and intelligently adjust the state of the photovoltaic rolling curtain system to optimize the indoor light and heat environment and achieve the goals of energy conservation and comfort improvement.

[0023] Further, the specific content of Step 1 is: Use 3D modeling software combined with building information modeling tools to perform geometric modeling on the target building. At the same time, adopt external environmental monitoring equipment to collect external environmental data in real time, including solar radiation intensity, cloud cover and other meteorological conditions related to the light and heat environment.

[0024] Further, the specific content of Step 2 is: Arrange multiple infrared cameras and illuminometers in key areas of the building. These devices can provide real-time indoor environment images and light data, and these data are sent to the central processing unit for subsequent analysis.

[0025] Further, the specific architecture of the prediction model for the indoor light and heat environment comfort level is:

[0026] Input layer: Design two input channels, one for receiving infrared images and the other for receiving illuminance data;

[0027] Convolution layer: The convolution layer is used for feature extraction. Given an input feature map I and a convolution kernel F, the calculation formula for each element of the output feature map O of the convolution operation is:

[0028]

[0029] Among them, I represents the input feature map, K is the convolutional kernel, b is the bias term, a and b are respectively half of the convolutional kernel size, and x and y represent the position coordinates of the output feature map;

[0030] The convolutional layer captures the indoor average temperature by learning the features of the infrared image, and simultaneously processes the data related to illuminance to obtain the feature information of the indoor lighting conditions;

[0031] Pooling layer: The pooling layer is used to reduce the feature dimension and the number of parameters, and improve the generalization ability of the model. The pooling operations include max pooling and average pooling;

[0032] Max pooling selects the maximum value within the pooling window as the output of the window:

[0033] O xy = max (h，w)∈W I (x+h)(y+w)

[0034] Average pooling calculates the average value of all values within the pooling window:

[0035]

[0036] Among them, H and W are respectively the height and width of the pooling window;

[0037] Fully connected layer: The fully connected layer is used to map the features extracted by the convolutional and pooling layers to the final output. The output y of the fully connected layer can be calculated by the following formula:

[0038] y = W x + b

[0039] Among them, W is the weight matrix, x is the input vector, and b is the bias vector;

[0040] The fully connected layer can summarize the features extracted from the infrared image and illuminance data, and perform classification or regression analysis on the comfort level of the indoor light and heat environment;

[0041] Softmax function: In the output layer of multi-classification problems, the Softmax function converts the linear output of the fully connected layer into a probability distribution:

[0042]

[0043] Among them, Zi is the input value of the i-th output unit, and the denominator is the sum of the exponents of the input values of all output units, ensuring that the sum of the output values is 1, representing probability;

[0044] This enables the model to output the probability of each comfort level, helping decision-makers understand the comfort level of the indoor environment and adjust the photovoltaic roller shutter system accordingly.

[0045] Further, step 4 is specifically as follows: automatically adjust the opening degree and angle of the photovoltaic rolling shutter according to the output of the prediction model established in step 3.

[0046] The present invention provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the dynamic intelligent control method of the photovoltaic rolling shutter system for residential buildings in cold regions are implemented.

[0047] The present invention also provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the steps of the dynamic intelligent control method of the photovoltaic rolling shutter system for residential buildings in cold regions are implemented.

[0048] Advantages of the present invention:

[0049] The present invention provides a photovoltaic rolling shutter system for residential buildings and its dynamic intelligent control method based on indoor thermal images and illuminance detection. The present invention uses an indoor infrared camera and an illuminance meter as detection means for the indoor physical environment, constructs a rapid prediction model for indoor thermal comfort and light environment through a Convolutional Neural Networks (CNN) algorithm, significantly improves the sensitivity and response speed of the photovoltaic blinds to the indoor environment of the building, avoids the complex data processing process in the previous simulation prediction process, and can significantly improve the comfort of the indoor light and heat environment by applying the photovoltaic rolling shutter system and its dynamic intelligent control method in residential buildings. By intelligently adjusting the rolling shutter to optimize the utilization of natural light and temperature control, while converting solar energy into electrical energy, it reduces building energy consumption and dependence on traditional energy. This not only enhances the sustainability and economic benefits of the building, but also provides a healthier and more comfortable living environment for residents. Description of the Drawings

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to the provided drawings.

[0051] Figure 1 It is a schematic diagram of the prediction model for the comfort of the indoor light and heat environment based on infrared images and illuminance data of the present invention.

[0052] Figure 2 It is a schematic diagram of the prediction model architecture of the present invention.

[0053] Figure 3 It is a schematic diagram of arranging multiple infrared cameras and illuminance meters in key areas of the building.

[0054] Figure 4 It is a schematic diagram of the state for dynamically and intelligently adjusting a photovoltaic rolling curtain system. Specific implementation manners

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0056] In combination with Figures 1 - 4 , the present invention provides a photovoltaic rolling curtain system for residential buildings in cold regions. The system includes a building physical information input module, a sky detection module, an indoor visual information input module, a non-visual information input module, a prediction model establishment module for indoor light and heat environment comfort based on infrared images and illuminance data, and a photovoltaic rolling curtain dynamic regulation module;

[0057] The building physical information input module: divides the grid for the indoor plane and the window opening size of the building facade of the building, assigns a unique RFID number to each room according to its function based on the usage function of the indoor space of the building, assigns a unique WID number to the windows of each room according to the size information of the windows, and simultaneously loads the longitude and latitude coordinates of the location where the building is located and the three-dimensional environmental coordinates of the building to determine the orientation of the building and the skylight environment characteristics of the location where it is located;

[0058] The sky detection module: is used to collect external environmental light and weather condition data for the RHINO-LBT software platform to perform light and heat environment simulation and analysis;

[0059] The indoor visual information input module: numbers the infrared information collection points IR-x and the indoor illuminance collection points Lux-x, uses an indoor infrared camera as the information input end to record the indoor infrared image in real time, and performs standardization processing on the infrared image, including size adjustment, normalization, etc., to make it suitable for the input requirements of the CNN prediction model, and collects the illuminance value through an illuminometer to provide a data basis for CNN training;

[0060] The non-visual information input module: according to the requirements of the indoor light environment and heat environment under different room functions and different intensities of human activities, detects the comfortable illuminance value and the indoor thermal comfort range through questionnaires and human physiological information, and assigns a number TSID to each possible indoor environment setting and human thermal sensation model based on different scenarios and personal preferences, and normalizes the measured data;

[0061] The prediction model establishment module: It is trained through the convolutional neural network (CNN) in machine learning to obtain a prediction model for the indoor light and heat environment comfort based on infrared images and illuminance data;

[0062] The photovoltaic roller shutter dynamic regulation module: Relying on this prediction model, it dynamically regulates the photovoltaic roller shutter system to optimize the indoor light and heat environment.

[0063] The adjustment rules of the building photovoltaic roller shutter system are specifically as follows: Different controllers are numbered respectively, and each controller of the photovoltaic roller shutter is assigned a unique number CID. The main controllers are: lifting motors and steering motors; Set the goal of indoor comfort, with the optimal range of indoor thermal comfort and light environment, and energy collection as a reference index; In different situations, the priorities of comfort and energy-saving factors are dynamically adjusted. For example, in extremely high temperatures, priority is given to ensuring indoor temperature comfort; When the light is sufficient and the indoor thermal comfort is good, priority is given to the collection and utilization of energy.

[0064] The corresponding relationships among the building room number RFID, window number WID, human thermal sensation model number TSID, infrared information collection point IR-x, indoor illuminance collection point Lux-x and controller number CID in the system are as follows:

[0065] According to the input building space data, the building room numbers and window numbers with different functions respectively correspond to different human thermal sensation model numbers TSID, and are in one-to-one correspondence with the infrared information collection point IR-x, indoor illuminance collection point Lux-x and controller number CID.

[0066] Establish the corresponding relationships among the lifting height, louver angle of the photovoltaic roller shutter louver system and the indoor light and heat environment. According to the detection of indoor infrared images and illuminance, use the prediction model trained by the convolutional neural network (Convolutional Neural Networks, CNN) to quickly evaluate the real-time light and heat comfort in the room and adjust the state of the photovoltaic roller shutter louver system. Combining with user habits, it can be controlled according to the time dimension.

[0067] The present invention also proposes an operation method for the residential building photovoltaic roller shutter system with real-time optimization of the indoor light and heat environment, and the method is specifically as follows:

[0068] S1. Install the building photovoltaic roller shutter louver, indoor infrared camera and illuminometer. The general installation position of the illuminometer is convenient for measuring the working plane, and input user parameters.

[0069] S2. Based on the meteorological data of the location, the indoor thermal environment conditions and solar radiation heat conditions under passive conditions in the building, determine the influence range of solar radiation heat on the indoor environment.

[0070] S3. Build a rapid prediction model for indoor light and heat environment.

[0071] S4. Run the system, take real-time indoor thermal images, and measure the indoor illuminance in real time with an illuminometer. Input the data of the above two into the rapid prediction model to determine the adjustment range of the photovoltaic louver system.

[0072] S5. Drive the motor through the built-in battery and real-time photovoltaic power generation to complete the adjustment of the photovoltaic louvers.

[0073] In step S1, when installing the building photovoltaic louvers, determine its size according to the size of the lighting window and select the power of the drive motor.

[0074] In step S1, the indoor illuminance detector should be installed on the normal working plane and away from situations with extreme illuminance such as window sills.

[0075] In step S1, the input of user parameters is mainly the practical functions of the house, the user's behavior preferences, and the indoor clothing preferences, so as to build a human thermal sensation model for the user.

[0076] In step S2, different evaluation data are used for the evaluation of indoor thermal environment and light environment. For the thermal environment, PMV-PPD indoor thermal comfort is often used as the main evaluation. The method of the present invention mainly calculates PMV based on objective data, and its calculation formula is as follows:

[0077] PMV=(0.303e -0.036M +0.028)×L

[0078] Where L is the heat exchange amount between the heat generated by the human body and the environment, which can be calculated by the following formula:

[0079] L = M - W - R - C - E sk -E re -C res

[0080] ·M: Metabolic rate, unit is W / m 2

[0081] ·W: External workload, which can be ignored in general indoor environments

[0082] ·R: Radiant heat loss

[0083] ·C: Convective heat loss

[0084] ·E sk : Heat loss through skin evaporation

[0085] ·E re : Heat loss through respiration evaporation

[0086] ·C res : Heat loss through respiratory convection

[0087] In step S3, the dataset should be prepared in advance, and the indoor thermal images and illuminance data in the dataset should correspond to the human body thermal sensation evaluation data.

[0088] The present invention also proposes a dynamic intelligent regulation method based on the photovoltaic rolling curtain system for residential buildings in cold regions, and the method includes the following steps:

[0089] Step 1, collect and input the geometric information of the building, and obtain the external environmental light and weather condition data through the sky detection module;

[0090] Step 2, install infrared cameras and illuminometers indoors to collect indoor infrared images and illuminance values in real time for monitoring the indoor light and heat environment status;

[0091] Step 3, based on the RHINO-LBT software platform and the measured data, conduct light and heat environment simulation and analysis, and train the convolutional neural network CNN through machine learning methods to establish a prediction model for the indoor light and heat environment comfort based on indoor thermal images;

[0092] Step 4, according to the results of the prediction model, dynamically and intelligently adjust the state of the photovoltaic rolling curtain system to optimize the indoor light and heat environment and achieve the goals of energy conservation and comfort improvement.

[0093] The specific content of step 1 is: use 3D modeling software combined with building information modeling tools to perform geometric modeling on the target building. At the same time, adopt external environmental monitoring equipment to collect external environmental data in real time, including solar radiation intensity, cloud cover, and other meteorological conditions related to the light and heat environment.

[0094] The specific content of step 2 is: arrange multiple infrared cameras and illuminometers in key areas of the building. These devices can provide real-time indoor environmental images and light data, and these data are sent to the central processing unit for subsequent analysis.

[0095] The convolutional neural network (CNN) is a deep learning model specifically designed to process data with a grid structure, eliminating the need for a large amount of preprocessing of complex images. It automatically detects and learns spatial hierarchical features by applying convolutional operations in multiple layers of the network. CNN is constructed through a series of convolutional layers, activation layers, pooling layers, and fully connected layers, and can extract information layer by layer from simple features to complex patterns. It is widely used in image and video recognition, image classification, object detection, and many other visual tasks. The specific architecture of the prediction model for the indoor light and heat environment comfort based on indoor thermal images and real-time illuminance data is as follows:

[0096] Input layer: Design two input channels, one for receiving infrared images and the other for receiving illuminance data;

[0097] Convolution Layer: The convolution layer is used for feature extraction. Given an input feature map I and a convolution kernel (or filter) F, the formula for calculating each element of the output feature map O of the convolution operation is as follows:

[0098]

[0099] where I represents the input feature map, K is the convolution kernel (or filter), b is the bias term, a and b are respectively half of the convolution kernel size (assuming the convolution kernel size is odd), and x and y represent the position coordinates of the output feature map;

[0100] The convolution layer captures the average indoor temperature by learning the features of the infrared image, and at the same time processes the data related to illuminance to obtain the feature information of the indoor lighting conditions;

[0101] Pooling Layer: The pooling layer is used to reduce the feature dimension and the number of parameters, and improve the generalization ability of the model. The pooling operations include max pooling and average pooling;

[0102] Max Pooling selects the maximum value within the pooling window as the output of the window:

[0103] O xy =max (h，w)∈W I (x+h)(y+w)

[0104] Average Pooling calculates the average value of all values within the pooling window:

[0105]

[0106] where H and W are respectively the height and width of the pooling window;

[0107] This helps to reduce the computational amount while maintaining the key parts of the features, which is particularly important for the global understanding of indoor thermal comfort and lighting.

[0108] Fully Connected Layer: The fully connected layer is used to map the features extracted by the convolution and pooling layers to the final output (such as classification labels). The output y of the fully connected layer can be calculated by the following formula:

[0109] y=W x +b

[0110] Among them, W is the weight matrix, x is the input vector, and b is the bias vector;

[0111] The fully connected layer can summarize the features extracted from the infrared image and illuminance data, and classify or perform regression analysis on the comfort level of the indoor light and heat environment;

[0112] Softmax function: In the output layer of multi-classification problems, the Softmax function converts the linear output of the fully connected layer into a probability distribution:

[0113]

[0114] Among them, Zi is the input value of the i-th output unit, and the denominator is the sum of the exponents of the input values of all output units, ensuring that the sum of the output values is 1, representing probability;

[0115] This enables the model to output the probability of each comfort level, helping decision-makers understand the comfort level of the indoor environment and adjust the photovoltaic roller blind system accordingly.

[0116] Specifically, step 4 is: automatically adjusting the opening degree and angle of the photovoltaic roller blind according to the output of the prediction model established in step 3.

[0117] The present invention proposes a photovoltaic roller blind system for residential buildings in cold regions and its dynamic intelligent control method, aiming to improve the comfort of the indoor light and heat environment of buildings while reducing energy consumption. In response to the challenges faced by buildings in cold regions, such as high demand for thermal insulation and large heating energy consumption, the present invention adopts a deep learning model based on convolutional neural network (CNN) to quantitatively and rapidly evaluate and feedback regulate the indoor light and heat environment of residential buildings through indoor infrared images and illuminance data. This method includes: collecting building geometric information and external environmental light and weather condition data, and performing light and heat environment simulation and analysis through the RHINO-LBT software platform; using indoor infrared cameras and illuminance meters to collect real-time data and establish measured and simulated data sets; training through convolutional neural network to construct a prediction model for the comfort level of the indoor light and heat environment; dynamically and intelligently regulating the photovoltaic roller blind system according to the results of the prediction model. This system not only optimizes the utilization of natural light and indoor temperature control, but also converts solar energy into electrical energy, improves the building's energy self-sufficiency ability, and reduces greenhouse gas emissions. The invention has theoretical value and practical significance. By intelligently adjusting the roller blind, it significantly improves the comfort of the indoor light and heat environment, provides a healthier and more comfortable living environment for residents in cold regions, and supports the achievement of the carbon neutrality goal at the same time.

[0118] Embodiment

[0119] The present invention provides a dynamic intelligent control method based on the above-mentioned photovoltaic roller blind system for residential buildings in cold regions, and this method includes the following steps:

[0120] Step 1: Collect and input the geometric information of the building, and obtain the data of external environmental illumination and weather conditions through the sky detection module;

[0121] Specifically, Step 1 is as follows: Use 3D modeling software (such as RHINO) combined with building information modeling tools (such as Ladybug Tools) to perform geometric modeling on the target building. At the same time, adopt external environmental monitoring equipment, such as a full sky scanner, to collect external environmental data in real time, including solar radiation intensity, cloud cover, and other meteorological conditions related to the light and heat environment.

[0122] Step 2: Install infrared cameras and illuminometers indoors to collect indoor infrared images and illuminance values in real time for monitoring the indoor light and heat environment status;

[0123] Specifically, Step 2 is as follows: Arrange multiple infrared cameras and illuminometers in key areas of the building, such as near windows and places where indoor activities are frequent, as Figure 3 shown. These devices can provide real-time indoor environmental images and lighting data, which are sent to the central processing unit for subsequent analysis.

[0124] Step 3: Based on the RHINO-LBT software platform and the measured data, conduct simulation and analysis of the light and heat environment, and train a convolutional neural network (CNN) through machine learning methods to establish a prediction model for indoor light and heat environment comfort;

[0125] Specifically, Step 3 is as follows: Use the RHINO-LBT software platform to conduct light and heat environment simulation, combine the measured data collected in Step 2, and integrate the simulation data and measured data through data fusion technology. Then, use a convolutional neural network (CNN) to analyze and learn these data to train a model that can accurately predict the indoor light and heat environment comfort.

[0126] Step 4: According to the results of the prediction model, dynamically and intelligently adjust the state (including the opening degree and angle) of the photovoltaic roller shutter system to optimize the indoor light and heat environment and achieve the goals of energy conservation and comfort improvement.

[0127] Specifically, Step 4 is as follows: Automatically adjust the opening degree and angle of the photovoltaic roller shutter according to the output of the prediction model established in Step 3. This process takes into account the temperature difference between indoors and outdoors, the current light intensity, and the predicted indoor comfort to ensure the best energy efficiency and living comfort.

[0128] Through the above technical solutions, the present invention provides a new type of photovoltaic roller shutter system and its dynamic intelligent control method for residential buildings in cold regions, effectively integrating the energy utilization of buildings and the regulation of living comfort, demonstrating innovative applications in the field of modern building technology.

[0129] In this embodiment, the proposed photovoltaic rolling shutter system for residential buildings in cold regions and its dynamic intelligent control method demonstrate an innovative solution that integrates software and hardware technologies to improve residential comfort and energy efficiency. The implementation of this system can be achieved through software, hardware, firmware, or any combination thereof. In particular, when implemented using software, the method can be transformed, in whole or in part, into the form of a computer program product.

[0130] The present invention proposes an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the dynamic intelligent control method based on the photovoltaic rolling shutter system for residential buildings in cold regions described above.

[0131] The present invention also proposes a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, they implement the steps of the dynamic intelligent control method based on the photovoltaic rolling shutter system for residential buildings in cold regions described above.

[0132] This computer program product contains a series of computer-executable instructions. When these instructions are executed on a computer or other programmable device, they cause the device to operate according to the processes or functions described in the embodiments of the present invention. These operations include receiving building geometry information, reading data from the sky detection module, performing optical and thermal environment simulation and analysis through the RHINO-LBT software platform, and predicting the indoor environmental comfort using a convolutional neural network and adjusting the photovoltaic rolling shutter system accordingly.

[0133] These computer instructions can be stored on a computer-readable storage medium or transmitted from one computer-readable storage medium to another by wire or wireless means. These media may include, but are not limited to, magnetic media (such as hard disks, floppy disks, magnetic tapes), optical media (such as CDs, DVDs), semiconductor media (such as solid-state drives), and they can be part of data storage devices such as servers and data centers.

[0134] During implementation, the method steps described in the present invention can be implemented through hardware logic circuits integrated in the processor or software instructions. These steps can be executed by a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices. The software module may be located in mature storage media such as random access memory (RAM), flash memory, read-only memory (ROM), etc., and when the processor reads and executes the instructions in these memories, it completes the steps of the above method in combination with its hardware.

[0135] Through this design, the present invention not only improves the energy use efficiency of residential buildings in cold regions and the comfort of occupants, but also demonstrates the forefront application of the combination of modern building technology with artificial intelligence and Internet of Things technology, providing a new solution for the fields of green buildings and smart homes.

[0136] The memory in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read only memory (ROM), a programmable ROM (PROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DRRAM). It should be noted that the memory of the method described in the present invention is intended to include but not limited to these and any other suitable types of memory.

[0137] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can 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. When the computer instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can 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 can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a high-density digital video disc (DVD)), or a semiconductor medium (such as a solid state disc (SSD)), etc.

[0138] In the implementation process, the steps of the above method can be completed by the integrated logic circuit of the hardware in the processor or the instructions in the form of software. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware processor, or executed and completed by a combination of the hardware and software modules in the processor. The software module can be located in a mature storage medium in the art such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, register, etc. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method. To avoid repetition, it will not be described in detail here.

[0139] It should be noted that the processor in the embodiments of the present application can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiments can be completed by the integrated logic circuit in the hardware of the processor or instructions in software form. The above-mentioned processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. It can implement or execute each method, step, and logic block diagram disclosed in the embodiments of the present application. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present application can be directly embodied as being executed and completed by the hardware decoding processor, or can be executed and completed by a combination of the hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. This storage medium is located in the memory, and the processor reads the information in the memory and combines its hardware to complete the steps of the above method.

[0140] The above has introduced in detail a photovoltaic rolling curtain system for residential buildings in cold regions and its dynamic intelligent control method proposed by the present invention. Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A photovoltaic rolling curtain system for residential buildings in cold regions, characterized in that, The system includes a building physical information input module, a sky environment module, an indoor visual information input module, a non-visual information input module, a prediction model establishment module for indoor light and heat environment comfort based on infrared images and illuminance data, and a photovoltaic roller shutter dynamic regulation module; The building physical information input module: divides the grid of the indoor plane and the window opening size of the building elevation, assigns a unique RFID number to each room according to its function based on the usage function of the indoor space of the building, assigns a unique WID number to the windows of each room according to the window size information, and loads the longitude and latitude coordinates of the location where the building is located and the three-dimensional environment coordinates of the building to determine the orientation of the building and the skylight environment characteristics of the location; The sky environment module: is used to collect external environmental light and weather condition data for the RHINO-LBT software platform to perform light and heat environment simulation and analysis; The indoor visual information input module: numbers the infrared information collection points IR-x and the indoor illuminance collection points Lux-x, uses the indoor infrared camera as the information input end to record the indoor infrared images in real time, standardizes the infrared images to make them suitable for the input requirements of the CNN prediction model, and collects the illuminance values through the illuminance meter to provide a data basis for the CNN model training; The non-visual information input module: according to the requirements of the indoor light environment and heat environment under different room functions and different human activity intensities, detects the comfortable illuminance values and indoor thermal comfort ranges through questionnaires and human physiological information, and assigns a TSID number to each possible indoor environment setting and human thermal sensation model based on different scenarios and personal preferences, and normalizes the measured data; The prediction model establishment module: trains through the convolutional neural network CNN in machine learning to obtain a prediction model for indoor light and heat environment comfort based on infrared images and illuminance data; The photovoltaic roller shutter dynamic regulation module: relies on this prediction model to dynamically regulate the photovoltaic roller shutter system to optimize the indoor light and heat environment; The adjustment rules of the building photovoltaic roller shutter system are specifically as follows: number different controllers respectively, and each controller of the photovoltaic roller shutter is assigned a unique CID number; set the goal of indoor comfort, with the indoor thermal comfort and light environment as the optimal range and energy collection as the reference index; dynamically adjust the priority of comfort and energy-saving factors under different conditions.

2. The system according to claim 1, wherein The corresponding relationships among the building room number RFID, window number WID, human thermal sensation model number TSID, infrared information collection point IR-x, indoor illuminance collection point Lux-x and controller number CID in the system are as follows: According to the input building space data, the building room numbers and window numbers with different functions respectively correspond to different human thermal sensation model numbers TSID, and are in one-to-one correspondence with the infrared information collection point IR-x, indoor illuminance collection point Lux-x and controller number CID.

3. A dynamic intelligent control method for a photovoltaic rolling curtain system of residential buildings in cold regions according to claim 1, characterized in that, The method includes the following steps: Step 1, collect and input the geometric information of the building, and obtain the external environmental light and weather condition data through the sky environment module; Step 2: Install infrared cameras and illuminometers indoors in the building to collect indoor infrared images and illuminance values in real time for monitoring the indoor light and heat environment status; Step 3: Based on the RHINO-LBT software platform and the measured data, conduct light and heat environment simulation and analysis, and train a convolutional neural network CNN through machine learning methods to establish a prediction model for indoor light and heat environment comfort; Step 4: According to the results of the prediction model, dynamically and intelligently adjust the state of the photovoltaic roller shutter system to optimize the indoor light and heat environment and achieve the goals of energy conservation and comfort improvement.

4. The method according to claim 3, characterized in that, The specific content of Step 1 is as follows: Use 3D modeling software combined with building information modeling tools to perform geometric modeling on the target building. At the same time, adopt external environment monitoring equipment to collect external environment data in real time, including solar radiation intensity, cloud cover, and other meteorological conditions related to the light and heat environment.

5. The method according to claim 3, characterized in that, The specific content of Step 2 is as follows: Arrange multiple infrared cameras and illuminometers in key areas of the building. These devices can provide real-time indoor environment images and lighting data, which are sent to the central processing unit for subsequent analysis.

6. The method according to claim 3, wherein The specific architecture of the prediction model for indoor light and heat environment comfort is as follows: Input layer: Design two input channels, one for receiving infrared images and the other for receiving illuminance data; Convolutional layer: The convolutional layer is used for feature extraction. Given an input feature map I and a convolutional kernel F, the calculation formula for each element of the output feature map O of the convolution operation is: where I represents the input feature map, K is the convolutional kernel, b is the bias term, a and b are respectively half of the convolutional kernel size, and x and y represent the position coordinates of the output feature map; The convolutional layer captures the indoor average temperature by learning the features of the infrared image, and at the same time processes the data related to illuminance to obtain the feature information of the indoor lighting conditions; Pooling layer: The pooling layer is used to reduce the feature dimension and the number of parameters and improve the generalization ability of the model. The pooling operations include max pooling and average pooling; Max pooling selects the maximum value within the pooling window as the output of the window: O xy = max (h,w)∈W I (x+h)(y+w) Average pooling calculates the average value of all values within the pooling window: where H and W are respectively the height and width of the pooling window; Fully connected layer: The fully connected layer is used to map the features extracted by the convolutional and pooling layers to the final output. The output y of the fully connected layer can be calculated by the following formula: y = Wx + b where W is the weight matrix, x is the input vector, and b is the bias vector; The fully connected layer can summarize the features extracted from the infrared image and illuminance data and perform classification or regression analysis on the indoor light and heat environment comfort; Softmax function: In the output layer of multi-classification problems, the Softmax function converts the linear output of the fully connected layer into a probability distribution: where Zi is the input value of the i-th output unit, and the denominator is the sum of the exponents of the input values of all output units, ensuring that the sum of the output values is 1, representing probability; This enables the model to output the probability of each comfort level, helping decision-makers understand the comfort level of the indoor environment and adjust the photovoltaic roller shutter system accordingly.

7. The method according to claim 3, characterized in that, The specific content of Step 4 is as follows: Automatically adjust the opening degree and angle of the photovoltaic roller shutter according to the output of the prediction model established in Step 3.

8. An electronic device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 3-7 are implemented.

9. A computer-readable storage medium for storing computer instructions, characterized in that, When the computer instructions are executed by the processor, the steps of the method according to any one of claims 3-7 are implemented.

Images