Fresh air temperature and humidity adjusting method and system

Abstract

The invention provides a fresh air temperature and humidity adjusting method and system, and relates to the technical field of temperature and humidity adjustment, and the method comprises the steps of data acquisition, first modeling, first model training, prediction, adjustment and the like. The system comprises a data acquisition module, a first modeling module, a first model training module, a prediction module and an adjustment module, the temperature and humidity parameters of the fresh air system are further adjusted through environmental data, and intelligent management and optimal control over the fresh air system are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of temperature and humidity regulation, and in particular to a method and system for regulating the temperature and humidity of fresh air. Background Art

[0002] With the development of Internet of Things technology and economy, most people have a higher pursuit of living environment quality. The quality of indoor environment also directly affects the health of residents. Therefore, the monitoring and regulation of indoor environment of houses has become an important part of smart home. In order to improve the quality of indoor environment of houses, it is necessary to manage the monitoring and regulation of indoor environment of houses. At present, the indoor environment monitoring of houses mainly monitors the corresponding air environment of houses, and then uses the fresh air system to regulate the indoor air environment to achieve the purpose of improving the indoor environment.

[0003] The Chinese invention patent with the application publication date of February 13, 2024 and the publication number of CN117555376A provides a method and system for energy-saving regulation of residential building environment based on big data. The patent obtains the corresponding human body physical data information and imports it into the constructed neural network model, outputs the suitable temperature and humidity data of each person, obtains the corresponding person's critical illness value and the calculated suitable temperature and humidity data of the corresponding person, imports it into the temperature and humidity value calculation strategy to calculate the final temperature and humidity, and the temperature and humidity controller in the building adjusts the residential building environment to the final temperature and humidity in real time, analyzes the physical characteristics of the human body in the residential building, and constructs a deep learning neural network model to accurately calculate and export the suitable temperature and humidity in the residential environment. However, indoor temperature and humidity data are also affected by environmental factors, such as light brightness, season, weather and other environmental factors, so this patent still has certain limitations on the indoor temperature and humidity regulation of the house. Summary of the invention

[0004] In order to improve the accuracy of temperature and humidity regulation of a fresh air system, the present application provides a fresh air temperature and humidity regulation method and system.

[0005] In the first aspect, the present application provides a method for adjusting the temperature and humidity of fresh air, which adopts the following technical solution: A method for adjusting fresh air temperature and humidity, comprising the following steps: Data collection: including real-time data collection and historical data collection; Collect real-time data: collect real-time indoor temperature and humidity data, real-time environmental data, and real-time fresh air temperature and humidity data; Collect historical data: collect historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data; First modeling: Building a BI-LSTM model based on machine learning algorithms; First model training: Use historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data to train the BI-LSTM model; Prediction: Input the real-time indoor temperature and humidity data and the real-time environmental data into the trained BI-LSTM model, and output the predicted fresh air temperature and humidity data, which is recorded as the first data; Adjustment: Adjust the real-time fresh air temperature and humidity data according to the first data.

[0006] By adopting the above technical solution, the present application takes environmental factors into consideration when predicting the fresh air temperature and humidity data, and improves the adjustment accuracy. In the prediction process, there are complex time-dependent relationships between environmental factors, and the BI-LSTM model can better capture these relationships, thereby improving the accuracy and reliability of the prediction. Through the training of historical data, the BI-LSTM model can learn the complex mapping relationship between environmental data and fresh air temperature and humidity. This training process enables the BI-LSTM model to predict the future fresh air temperature and humidity based on the current state, providing a scientific basis for subsequent adjustments. The real-time indoor temperature and humidity data and the real-time environmental data are input into the trained BI-LSTM model, and the BI-LSTM model can output the predicted fresh air temperature and humidity data. According to the fresh air temperature and humidity data predicted by the BI-LSTM model, the real-time fresh air temperature and humidity data are automatically adjusted to achieve a more comfortable effect.

[0007] Optionally, the step of data collection further includes: Collect physical data: collect current personnel's physical data, historical personnel's physical data, and historical personnel's suitable fresh air temperature and humidity data.

[0008] By adopting the above technical solution, this application not only considers the impact of environmental data on fresh air temperature and humidity data, but also pays attention to the physical data of personnel, and further provides a personalized adjustment method.

[0009] Optionally, after performing the step of training the first model and before performing the step of predicting, the method further includes: Transfer learning: including second modeling and second model training; Second modeling: establishing a prediction model, the prediction model includes an attention layer, a base model, a migration model and a fully connected layer, the base model is a trained BI-LSTM model, and the migration model is a feedforward neural network model; Second model training: freeze the base model, and use historical personnel physical data and historical personnel suitable fresh air temperature and humidity data to train the migration model.

[0010] By adopting the above technical solution, the present application constructs a prediction model, which introduces an attention layer to give the prediction model the ability to pay attention to important parts of the input data, which helps the prediction model to understand complex data relationships more accurately, especially when processing physical data and fresh air temperature and humidity data. The migration model (feedforward neural network model) is a new learning component that can be optimized for specific tasks (predicting suitable fresh air temperature and humidity based on personnel physical data). By freezing the base model (trained BI-LSTM model), the knowledge learned by the base model will not be changed when training the migration model, thereby ensuring the stability and accuracy of the prediction model when processing basic environmental data (such as indoor temperature and humidity, environmental data). The migration model is trained using historical personnel physical data and historical personnel suitable fresh air temperature and humidity data, so that it can learn the specific relationship between personnel physical data and fresh air temperature and humidity suitable for the human body. By introducing migration learning, historical data can be used more effectively, including physical data that may not have been fully utilized before, thereby improving the use value of the data and the prediction accuracy of the system.

[0011] Optionally, in the prediction step, real-time indoor temperature and humidity data, real-time environmental data, and current personnel physical data are input into the trained prediction model to output new predicted fresh air temperature and humidity data.

[0012] By adopting the above technical solution, based on the processing and analysis of the input data, the prediction model will output a new predicted fresh air temperature and humidity data. This fresh air temperature and humidity data is the optimal or most suitable fresh air temperature and humidity data predicted by the prediction model based on the current environmental conditions and personnel physical data.

[0013] Optionally, the new predicted fresh air temperature and humidity data calculation model is as follows: ; ; in, The new predicted fresh air temperature data; is the temperature data predicted by the base model; n is the number of people in the room; is the body temperature data of the i-th person; is the indoor temperature data; The new predicted fresh air humidity data; is the humidity data predicted by the base model; M is the energy metabolism rate of the human body; is the convection mass exchange coefficient; P is the pressure of indoor water vapor formation, and the calculation model is as follows: .

[0014] By adopting the above-mentioned technical solution, this application calculates the fresh air temperature and humidity based on the output data of the base model according to the current environment and personnel physical data, so that the prediction model can more accurately predict the fresh air temperature and humidity required by each human body, thereby improving the overall environmental comfort.

[0015] Optionally, after executing the first model training step and before executing the prediction step, the method further includes: Set loss function: Set the loss function of the BI-LSTM model; Setting an evaluation matrix: Setting an evaluation matrix, wherein each element in the evaluation matrix represents the probability that a sample belonging to the i-th fresh air temperature and humidity data is misclassified as the j-th fresh air temperature and humidity data, recorded as the first probability, wherein the sample includes: historical indoor temperature and humidity data, historical environmental data, and the diagonal elements of the evaluation matrix are 0; Adjust the loss function: Adjust the weights in the loss function according to the evaluation matrix.

[0016] By adopting the above technical solution, this application sets an evaluation matrix to quantify the accuracy of the BI-LSTM model. Each element in the evaluation matrix represents the probability that the BI-LSTM model should classify the i-th fresh air temperature and humidity data as the j-th fresh air temperature and humidity data, and the diagonal element is the probability that the i-th fresh air temperature and humidity data is mistakenly classified as the i-th fresh air temperature and humidity data. This is not a case of misclassification, so the diagonal element is set to 0. Adjusting the weights in the loss function according to the evaluation matrix can make the BI-LSTM model pay more attention to those samples or intervals that have a greater impact on the prediction performance during the training process. Optimizing the performance of the model on the evaluation matrix by adjusting the loss function helps to improve the generalization ability of the model on different data sets.

[0017] Optionally, after executing the step of setting the evaluation matrix and before executing the step of adjusting the loss function, the method further includes: Sum: Calculate the sum of the elements in each row of the evaluation matrix and record it as the second data; Normalization: Divide each element by the second data of the corresponding row to obtain a new evaluation matrix.

[0018] By adopting the above technical solution, the present application divides each element in the evaluation matrix by the "second data" of the row in which it is located (i.e., the sum of the row), standardizes the elements of each row to the range of 0 to 1, or more generally, converts it into a ratio relative to the total sum in the row, and then converts the evaluation matrix into a conditional probability matrix.

[0019] Optionally, after the step of adjusting the loss function and before the step of predicting, the method further includes: The third modeling: Building a Bayesian network model based on historical indoor temperature and humidity data, historical environmental data, historical fresh air temperature and humidity data, and the first probability; Reasoning: The Bayesian network model is used for reasoning to predict the fresh air temperature and humidity data given the real-time indoor temperature and humidity data and the real-time environmental data, which is recorded as the third data.

[0020] By adopting the above technical solution, the Bayesian network model can learn and simulate the causal relationship between indoor temperature and humidity, environmental data and fresh air temperature and humidity. By using the Bayesian network model for reasoning, it can quickly and accurately predict fresh air temperature and humidity data (recorded as the third data) based on real-time indoor temperature and humidity data and real-time environmental data. Since the Bayesian network model can capture the complex relationship and causal relationship between variables, its prediction results are often more accurate than simple statistical models or machine learning models, which helps to improve overall performance and reliability.

[0021] Optionally, after the step of performing the prediction and before the step of performing the adjustment, the method further includes: Calculate data: assign weights to the first data and the third data, and calculate final fresh air temperature and humidity data, and use the final fresh air temperature and humidity data as new predicted fresh air temperature and humidity data.

[0022] By adopting the above technical solution, assigning weights to the first data and the third data actually identifies and distinguishes the relative importance of these two sets of data in the final prediction. By combining the first data and the third data and considering their weights, the final fresh air temperature and humidity data calculated is a result of combining multiple information sources and prediction methods. This method can usually provide more accurate and reliable predictions because it reduces the deviations and uncertainties that may be caused by a single data source or prediction method.

[0023] In the second aspect, the present application provides a fresh air temperature and humidity adjustment system, which adopts the following technical solution: A fresh air temperature and humidity adjustment system, comprising: Data collection module, including real-time data collection unit and historical data collection unit; Real-time data collection unit, used to collect real-time indoor temperature and humidity data, real-time environmental data, and real-time fresh air temperature and humidity data; The historical data collection unit is used to collect historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data; The first modeling module is used to build a BI-LSTM model; The first model training module is connected to the historical data collection unit and the first modeling module for training the BI-LSTM model using historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data; The prediction module is connected to the real-time data collection unit and the first model training module, and is used to input the real-time indoor temperature and humidity data and the real-time environmental data into the trained BI-LSTM model, and output the predicted fresh air temperature and humidity data; The adjustment module is connected to the real-time data collection unit and the prediction module for adjusting the real-time fresh air temperature and humidity data according to the predicted fresh air temperature and humidity data.

[0024] By adopting the above technical solutions, this system integrates several modules including data collection, model training, prediction and adjustment. Through comprehensive collection of real-time and historical data, the bidirectional long short-term memory network (BI-LSTM) model is used for training to learn the law of environmental changes, thereby realizing accurate prediction of fresh air temperature and humidity. The prediction results drive the adjustment module to respond quickly and intelligently adjust the real-time fresh air temperature and humidity, thereby effectively maintaining the comfort and stability of the indoor environment, while improving energy efficiency and operational economy.

[0025] In summary, the present application includes at least one of the following beneficial technical effects: 1. This application takes environmental factors into consideration when predicting fresh air temperature and humidity data, and improves the adjustment accuracy. In the prediction process, there are complex time-dependent relationships between environmental factors, and the BI-LSTM model can better capture these relationships, thereby improving the accuracy and reliability of the prediction. Through the training of historical data, the BI-LSTM model can learn the complex mapping relationship between environmental data and fresh air temperature and humidity. This training process enables the BI-LSTM model to predict future fresh air temperature and humidity based on the current environmental state, providing a scientific basis for subsequent adjustments. The real-time indoor temperature and humidity data and the real-time environmental data are input into the trained BI-LSTM model, and the BI-LSTM model can output the predicted fresh air temperature and humidity data. According to the fresh air temperature and humidity data predicted by the BI-LSTM model, the real-time fresh air temperature and humidity data can be automatically adjusted to achieve a more comfortable effect.

[0026] 2. This application constructs a prediction model that introduces an attention layer to give the prediction model the ability to focus on important parts of the input data, which helps the prediction model to understand complex data relationships more accurately. In particular, when processing physical data and fresh air temperature and humidity data, the migration model (feedforward neural network model) is a new learning component that can be optimized for specific tasks (predicting appropriate fresh air temperature and humidity based on personnel physical data). By freezing the base model (trained BI-LSTM model), the knowledge that has been learned will not be changed when training the migration model, thereby ensuring the stability and accuracy of the prediction model when processing basic environmental data (such as indoor temperature and humidity, environmental data). The migration model is trained using historical personnel physical data and historical personnel suitable fresh air temperature and humidity data, so that it can learn the specific relationship between personnel physical data and fresh air temperature and humidity suitable for the human body. By introducing migration learning, the system can more effectively utilize historical data, including physical data that may not have been fully utilized before, thereby improving the use value of the data and the prediction accuracy of the system.

[0027] 3. This application sets an evaluation matrix to quantify the accuracy of the BI-LSTM model. Each element in the evaluation matrix represents the probability that the BI-LSTM model should classify the i-th fresh air temperature and humidity data as the j-th fresh air temperature and humidity data, while the diagonal elements are the probability that the i-th fresh air temperature and humidity data is mistakenly classified as the i-th fresh air temperature and humidity data. This is not a case of misclassification, so the diagonal elements are set to 0. Adjusting the weights in the loss function according to the evaluation matrix can make the BI-LSTM model pay more attention to samples or intervals that have a greater impact on performance during training. Optimizing the performance of the model on the evaluation matrix by adjusting the loss function helps improve the generalization ability of the model on different data sets. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a flow chart of Example 1 of the present application; Figure 2 is a flow chart of Example 2 of the present application; Figure 3 This is a flow chart of Example 3 of the present application. DETAILED DESCRIPTION

[0029] The following combination Figures 1 to 3 This application is described in further detail.

[0030] Embodiment 1: This embodiment discloses a method for adjusting the temperature and humidity of fresh air, referring to Figure 1The adjustment method includes: S1 data collection, S2 first modeling, S3 first model training, S4 prediction and S5 adjustment. First, data collection is performed, then a BI-LSTM model is established, then the BI-LSTM model is trained, and then the trained BI-LSTM model is used to predict fresh air temperature and humidity data based on real-time indoor temperature and humidity data and real-time environmental data, and then the real-time fresh air temperature and humidity data is adjusted. This embodiment includes the following steps: S1 data collection, including S11 real-time data collection and S12 historical data collection.

[0031] S11 collects real-time data, including real-time indoor temperature and humidity data, real-time environmental data, and real-time fresh air temperature and humidity data.

[0032] Real-time indoor temperature and humidity data: The temperature and humidity sensors installed indoors collect the current indoor temperature and humidity information in real time.

[0033] Real-time environmental data: including light intensity, CO 2 Concentration, PM2.5 concentration, etc. These data are important factors affecting indoor comfort and air quality, and are monitored in real time through corresponding sensors.

[0034] Real-time fresh air temperature and humidity data: When the fresh air system introduces external air, its temperature and humidity are also key factors affecting the indoor environment. These real-time data are obtained through temperature and humidity sensors installed at the entrance of the fresh air system.

[0035] S12 collects historical data, including historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data.

[0036] Historical indoor temperature and humidity data: Collect indoor temperature and humidity data over a period of time. These data are used to analyze the long-term change trends and seasonal characteristics of the indoor environment.

[0037] Historical environmental data: Corresponding to the real-time environmental data, it also collects past light intensity, CO 2 Historical data of environmental parameters such as concentration and PM2.5 concentration.

[0038] Historical fresh air temperature and humidity data: records the temperature and humidity data of the fresh air system over a period of time to help understand changes in fresh air quality and its impact on the indoor environment.

[0039] S2 first builds a BI-LSTM model based on machine learning algorithm.

[0040] S3 first model training uses the historical data (historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data) collected by S12 as the training data set to train the BI-LSTM model. During the training process, the historical indoor temperature and humidity data and the historical environmental data collected at the same time are used as input data, and the historical fresh air temperature and humidity data are used as the true label of the input data to train the BI-LSTM model. During the training process, the BI-LSTM model will learn the rules and patterns in these historical data so that it can make predictions on the fresh air temperature and humidity data based on new input data in the future, that is, predict the labels of the real-time indoor temperature and humidity data and the real-time environmental data.

[0041] The historical fresh air temperature and humidity data mentioned in this step are fresh air temperature and humidity data in a stable state. The fresh air temperature and humidity data collected during the adjustment process (i.e., fresh air temperature and humidity data in a changing process) will be filtered out and will not be used for training the BI-LSTM model.

[0042] S4 predicts that real-time indoor temperature and humidity data and real-time environmental data are input into the trained BI-LSTM model. The trained BI-LSTM model will output the predicted fresh air temperature and humidity data, i.e. the first data, based on these real-time data and the historical laws and patterns it has learned.

[0043] S5 adjustment, adjusts the real-time fresh air temperature and humidity data according to the first data, that is: if the prediction result shows that the first data is not equal to the real-time fresh air temperature and humidity data, the settings of the fresh air system are automatically adjusted, and then the temperature or humidity of the fresh air is adjusted to make the indoor environment achieve the best comfort and air quality.

[0044] This embodiment collects real-time indoor temperature and humidity, real-time environmental data, and real-time fresh air temperature and humidity data, and uses historical data (historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data) to train a BI-LSTM model, which can predict fresh air temperature and humidity based on real-time data. Then, the fresh air system is automatically adjusted according to the results predicted by the BI-LSTM model to achieve the best comfort level for the indoor environment, thus realizing intelligent environmental control and management.

[0045] Example 2: Reference Figure 2 The difference between this embodiment and embodiment 1 is that, in S1 data collection, it also includes: S13 collects physical data, including current personnel's physical data, historical personnel's physical data, and historical personnel's suitable fresh air temperature and humidity data.

[0046] This step aims to collect the physical data of the current personnel (such as age, gender, height, weight, etc.), the physical data of historical personnel (i.e. the physical data of other personnel), and the suitable fresh air temperature and humidity data of historical personnel (i.e. the suitable fresh air temperature and humidity data of other personnel). The suitable fresh air temperature and humidity data of historical personnel are based on statistical analysis. For example, the fresh air temperature and humidity data applicable to the person at different time periods stored in the database reflect the most suitable fresh air temperature and humidity data for people with different physiques under different conditions.

[0047] When the amount of personnel physical fitness data is insufficient, a generative adversarial network is also used to generate similar data to enrich the quantity and types of personnel physical fitness data.

[0048] After executing S3 first model training and before executing S4 prediction, it also includes: S31 transfer learning, including S311 second modeling and S312 second model training.

[0049] S311 is the second modeling, which is to establish a prediction model. The prediction model includes an attention layer, a base model, a migration model and a fully connected layer. The base model is a trained BI-LSTM model, and the migration model is a feedforward neural network model.

[0050] The input of the attention layer is used to receive input data. The output of the attention layer is connected to the input of the base model and the input of the migration model respectively. The input of the fully connected layer is connected to the output of the base model and the output of the migration model respectively. The fully connected layer outputs the final prediction result. The attention layer is used to highlight important information (by assigning different weights to different input data). The base model provides basic time series analysis capabilities and provides a prediction value of fresh air temperature and humidity data based on time patterns. The migration model is used to capture the impact of physical data on the demand for fresh air temperature and humidity, and provides a prediction value of fresh air temperature and humidity data based on physical data. The fully connected layer is responsible for integrating all this information to output the final prediction result.

[0051] S312 Second model training, freeze the base model, that is, the parameters of the base model remain unchanged during the training process to avoid destroying the rules of time series data (various historical data) that it has learned. Then use the historical personnel physical data and the historical personnel suitable fresh air temperature and humidity data to train the migration model so that it can learn how the physical data affects the suitability of fresh air temperature and humidity. In the process of training the migration model, the historical personnel physical data is used as the input data of the migration model, and the historical personnel suitable fresh air temperature and humidity data is used as the true label of the historical personnel physical data.

[0052] In the S4 prediction, the real-time indoor temperature and humidity data, real-time environmental data, and current personnel physical data are input into the trained prediction model. The prediction model will comprehensively consider these input information, and output new predicted fresh air temperature and humidity data through the synergy of the attention layer, base model, migration model, and fully connected layer. The calculation model of the new predicted fresh air temperature and humidity data is as follows: ; ; in, The new predicted fresh air temperature data; is the temperature data predicted by the base model; n is the number of people in the room; The temperature data of the i-th person is obtained through wearable devices, such as smart bracelets, thermometers, etc., or contactlessly obtained through infrared thermometers; Indoor temperature data is measured by placing temperature sensors indoors; The new predicted fresh air humidity data; Humidity data predicted by the base model.

[0053] M is the energy metabolism rate of the human body. When the indoor personnel are in an active state, its calculation model is as follows: ; is oxygen consumption, obtained by respirometry; is the oxygen thermal value, and the specific value is obtained by consulting the oxygen thermal value table corresponding to NPRQ. NPRQ is the non-protein respiratory quotient, and its calculation formula is as follows: ; in, is the amount of carbon dioxide produced, obtained by respirometry.

[0054] When the indoor personnel are in an inactive state, that is, in a resting state, the calculation model of the human energy metabolism rate M is as follows: ; Among them, TZ is weight, unit is kg; SG is height, unit is cm; NL is age, unit is year.

[0055] is the mass exchange coefficient, in units of (W·m) / kJ. The calculation model is as follows: ; in, is the fluid density, in kg / m 3 , the value range is [1.1,1.3]; is the specific heat capacity at constant pressure, in kJ / (kg·K), with a value of 1.005; h is the heat transfer coefficient of the convective heat transfer surface, in W / (m 2 ·K), the value range is [5,50].

[0056] In other embodiments, the convective mass exchange coefficient It can also be calculated using the Butz formula or estimated by using the Sherwood criterion.

[0057] P is the pressure of water vapor formed in the room, and the calculation model is as follows: .

[0058] This embodiment adds the collection of current and historical personnel physical data, as well as the record of historical personnel suitable fresh air temperature and humidity. After the BI-LSTM model is trained, the prediction ability of the prediction model is further improved through transfer learning technology (including establishing a prediction model combining the attention layer, the frozen BI-LSTM base model, the migration model and the fully connected layer, and using physical and historical data for training). In the S4 prediction, the real-time environment, indoor temperature and humidity, and current personnel physical data are input into the prediction model to predict more personalized and accurate fresh air temperature and humidity data.

[0059] Example 3: Reference Figure 3 The difference between this embodiment and embodiment 1 is that after executing S3 first model training and before executing S4 prediction, it also includes: S61 sets a loss function, and sets the loss function of the BI-LSTM model. The loss function is the key to optimizing the BI-LSTM model, and measures the difference between the predicted value and the actual value, and trains the BI-LSTM model by minimizing the difference.

[0060] S62 sets an evaluation matrix, sets an evaluation matrix, each element in the evaluation matrix represents the probability that a sample belonging to the i-th fresh air temperature and humidity data is misclassified as the j-th fresh air temperature and humidity data, recorded as the first probability, and the sample includes: historical indoor temperature and humidity data, historical environmental data, and the diagonal elements of the evaluation matrix are 0.

[0061] Each element in the evaluation matrix represents the probability that a sample belonging to the i-th fresh air temperature and humidity data is misclassified as the j-th fresh air temperature and humidity data, that is, the first probability represents: the probability that the prediction result of the BI-LSTM model is a certain wrong result.

[0062] S63 is to calculate the sum of the elements in each row of the evaluation matrix and record it as the second data.

[0063] S64 normalizes, dividing each element by the second data of the corresponding row to achieve normalization and obtain a new evaluation matrix.

[0064] S65 adjusts the loss function and adjusts the weights in the loss function according to the evaluation matrix.

[0065] The calculation model of the loss function is set as follows: ; Among them, L is the first loss function, m is the number of input data of BI-LSTM model, is the true label of the i-th input data (i.e., historical fresh air temperature and humidity data), is the predicted label of the BI-LSTM model for the i-th input data (i.e. the predicted fresh air temperature and humidity data).

[0066] The calculation model of the loss function after adjusting the weights is as follows: ; m is the number of input data of BI-LSTM model; when i=j, ; is the value of the element in the evaluation matrix that classifies the predicted label of the i-th input data as the predicted label of the j-th input data; For the BI-LSTM model The predicted label (i.e., the wrong label) of .

[0067] S66: third modeling, establishing a Bayesian network model based on historical indoor temperature and humidity data, historical environmental data, historical fresh air temperature and humidity data, and the first probability. The nodes of the Bayesian network model are: historical indoor temperature and humidity data, historical environmental data, historical fresh air temperature and humidity data, and the value of the directed edge of the Bayesian network model is equal to one minus the first probability.

[0068] S67 reasoning uses the Bayesian network model for reasoning to predict the fresh air temperature and humidity data when given the real-time indoor temperature and humidity data and the real-time environmental data, which is recorded as the third data.

[0069] Then, S4 prediction is performed, and after performing S4 prediction, this embodiment further includes: S7 calculates data: by assigning weights to the prediction results of the BI-LSTM model (first data) and the inference results of the Bayesian network model (third data), and calculating the weighted average, the final real-time fresh air temperature and humidity data is obtained, and the final fresh air temperature and humidity data is used as the new predicted fresh air temperature and humidity data, and S5 adjustment is executed.

[0070] The final real-time fresh air temperature and humidity data calculation model is as follows: ; ; in, For the final real-time fresh air temperature and humidity data, is the weight of the first data, is the value of the first data; is the weight of the third data, is the value of the third data.

[0071] After the BI-LSTM model is trained, this embodiment sets and adjusts the loss function, constructs an evaluation matrix and analyzes prediction errors, introduces a Bayesian network model for independent reasoning, and finally integrates the prediction results of the two models to adjust and optimize the prediction process of fresh air temperature and humidity in a more refined manner, aiming to improve the accuracy and reliability of the prediction.

[0072] Embodiment 4: This embodiment discloses a fresh air temperature and humidity adjustment system, the system comprising: The data acquisition module includes a real-time data acquisition unit and a historical data acquisition unit.

[0073] The real-time data acquisition unit is used to capture real-time data, mainly including real-time indoor temperature and humidity data, real-time environmental data, and real-time fresh air temperature and humidity data. Real-time indoor temperature and humidity data reflects the current indoor temperature and humidity data, while real-time environmental data (such as outdoor temperature, humidity, air pressure, etc.) provides external factors that may affect the indoor environment. These data are acquired in real time through hardware devices such as sensors, ensuring the timeliness and accuracy of the data.

[0074] The historical data collection unit is used to collect data from the past period of time, including historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data. Historical data are crucial for model training because they provide rich samples to help the model learn and identify complex relationships between data. With long-term accumulation of historical data, the model can more accurately predict future states.

[0075] The first modeling module is used to establish a BI-LSTM model. The BI-LSTM model is a special recurrent neural network that can handle long-term dependency problems in sequence data and consider the context information of the sequence at the same time. In this embodiment, the BI-LSTM model is used to learn the complex relationship between indoor temperature and humidity, environmental data, and fresh air temperature and humidity.

[0076] The first model training module is connected to the historical data collection unit and the first modeling module, and is used to train the BI-LSTM model using the historical data provided by the historical data collection unit. Through continuous iteration and optimization, the BI-LSTM model can gradually learn the inherent laws and patterns between various data, thereby improving the accuracy of prediction. After the training is completed, the BI-LSTM model will have the ability to predict the temperature and humidity of fresh air based on real-time data.

[0077] The prediction module is communicated with the real-time data collection unit and the first model training module, and is used to input the real-time indoor temperature and humidity data and the real-time environmental data into the trained BI-LSTM model. The BI-LSTM model calculates the predicted value of the fresh air temperature and humidity based on the input data, and outputs the value to the subsequent modules.

[0078] The adjustment module is connected to the real-time data collection unit and the prediction module for adjusting the real-time fresh air temperature and humidity data according to the predicted fresh air temperature and humidity data.

[0079] The data acquisition module of this embodiment is responsible for collecting real-time and historical data, including indoor temperature and humidity, environment, and fresh air temperature and humidity. The first modeling module establishes a BI-LSTM model, and the first model training module trains this model using historical data. The prediction module inputs real-time data into the trained model and outputs the predicted value of fresh air temperature and humidity. The adjustment module automatically adjusts the real-time fresh air temperature and humidity data according to the prediction results to realize intelligent environmental control.

[0080] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for adjusting fresh air temperature and humidity, characterized in that: include: Data collection: including real-time data collection and historical data collection; Collect real-time data: collect real-time indoor temperature and humidity data, real-time environmental data, and real-time fresh air temperature and humidity data; Collect historical data: collect historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data; First modeling: Building a BI-LSTM model based on machine learning algorithms; First model training: Use historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data to train the BI-LSTM model; Prediction: Input the real-time indoor temperature and humidity data and the real-time environmental data into the trained BI-LSTM model, and output the predicted fresh air temperature and humidity data, which is recorded as the first data; Adjustment: Adjust the real-time fresh air temperature and humidity data according to the first data.

2. The fresh air temperature and humidity adjustment method according to claim 1, characterized in that: The data collection steps also include: Collect physical data: collect current personnel's physical data, historical personnel's physical data, and historical personnel's suitable fresh air temperature and humidity data.

3. The fresh air temperature and humidity adjustment method according to claim 2, characterized in that: After performing the first model training step and before performing the prediction step, the method further includes: Transfer learning: including second modeling and second model training; Second modeling: establishing a prediction model, the prediction model includes an attention layer, a base model, a migration model and a fully connected layer, the base model is a trained BI-LSTM model, and the migration model is a feedforward neural network model; Second model training: freeze the base model, and use historical personnel physical data and historical personnel suitable fresh air temperature and humidity data to train the migration model.

4. The fresh air temperature and humidity adjustment method according to claim 3, characterized in that: In the prediction step, the real-time indoor temperature and humidity data, real-time environmental data, and current personnel physical data are input into the trained prediction model, and new predicted fresh air temperature and humidity data are output.

5. The fresh air temperature and humidity adjustment method according to claim 4, characterized in that: The new predicted fresh air temperature and humidity data calculation model is as follows: ； ； in, The new predicted fresh air temperature data; is the temperature data predicted by the base model; n is the number of people in the room; is the body temperature data of the i-th person; is the indoor temperature data; The new predicted fresh air humidity data; is the humidity data predicted by the base model; M is the energy metabolism rate of the human body; is the convection mass exchange coefficient; P is the pressure of indoor water vapor formation, and the calculation model is as follows: 。 6. The fresh air temperature and humidity adjustment method according to any one of claims 1 to 5, characterized in that: After executing the first model training step and before executing the prediction step, the method further includes: Set loss function: Set the loss function of the BI-LSTM model; Setting an evaluation matrix: Setting an evaluation matrix, wherein each element in the evaluation matrix represents the probability that a sample belonging to the i-th fresh air temperature and humidity data is misclassified as the j-th fresh air temperature and humidity data, recorded as the first probability, wherein the sample includes: historical indoor temperature and humidity data, historical environmental data, and the diagonal elements of the evaluation matrix are 0; Adjust the loss function: Adjust the weights in the loss function according to the evaluation matrix.

7. The fresh air temperature and humidity adjustment method according to claim 6, characterized in that: After executing the step of setting the evaluation matrix and before executing the step of adjusting the loss function, the method further includes: Sum: Calculate the sum of the elements in each row of the evaluation matrix and record it as the second data; Normalization: Divide each element by the second data of the corresponding row to obtain a new evaluation matrix.

8. The fresh air temperature and humidity adjustment method according to claim 7, characterized in that: After the step of adjusting the loss function and before the step of predicting, it also includes: The third modeling: Building a Bayesian network model based on historical indoor temperature and humidity data, historical environmental data, historical fresh air temperature and humidity data, and the first probability; Reasoning: The Bayesian network model is used for reasoning to predict the fresh air temperature and humidity data given the real-time indoor temperature and humidity data and the real-time environmental data, which is recorded as the third data.

9. The fresh air temperature and humidity adjustment method according to claim 8, characterized in that: After the step of performing prediction and before the step of performing adjustment, the method further includes: Calculate data: assign weights to the first data and the third data, and calculate final fresh air temperature and humidity data, and use the final fresh air temperature and humidity data as new predicted fresh air temperature and humidity data.

10. A fresh air temperature and humidity adjustment system, the system being applicable to the method according to any one of claims 1 to 9, characterized in that: include: Data collection module, including real-time data collection unit and historical data collection unit; Real-time data collection unit, used to collect real-time indoor temperature and humidity data, real-time environmental data, and real-time fresh air temperature and humidity data; The historical data collection unit is used to collect historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data; The first modeling module is used to build a BI-LSTM model; The first model training module is connected to the historical data collection unit and the first modeling module for training the BI-LSTM model using historical indoor temperature and humidity data, historical environmental data, and historical fresh air temperature and humidity data; The prediction module is connected to the real-time data collection unit and the first model training module, and is used to input the real-time indoor temperature and humidity data and the real-time environmental data into the trained BI-LSTM model, and output the predicted fresh air temperature and humidity data; The adjustment module is connected to the real-time data collection unit and the prediction module for adjusting the real-time fresh air temperature and humidity data according to the predicted fresh air temperature and humidity data.

Images