Environment intelligent control system based on far infrared heating technology

By acquiring and processing real-time data, combined with multilayer perceptrons and PID control algorithms, the problems of inaccurate prediction and slow response in traditional environmental control systems have been solved, achieving precise control of environmental parameters and efficient use of energy, thus improving user comfort.

CN118963448AActive Publication Date: 2024-11-15JIANGSU SENHAI ELECTRIC CO LTD

Patent Information

Application Number
CN202411047461.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-11-15
Estimated Expiration
2044-08-01

AI Technical Summary

Technical Problem

Traditional environmental control systems cannot accurately predict future changes in ambient temperature and lack real-time feedback and adjustment mechanisms, resulting in low energy efficiency and poor user comfort.

Method used

The system acquires environmental parameters in real time through a data acquisition module, performs preprocessing and predictive analysis using a data processing module, formulates heating strategies using a multilayer perceptron algorithm, adjusts the heater output in real time using a PID control algorithm, and integrates a feedback regulation module for real-time temperature monitoring and adjustment.

Benefits of technology

It achieves precise control of environmental parameters and efficient use of energy, improving system response speed and user comfort.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963448A_ABST
    Figure CN118963448A_ABST
Patent Text Reader

Abstract

The invention discloses an environment intelligent control system based on a far infrared heating technology, and relates to the technical field of intelligent control, and the system comprises the following components: a data collection module which is used for collecting indoor and outdoor environment parameters in real time, is connected with an external data source through various sensors and interfaces, guarantees the accuracy and real-time performance of data, and transmits the data to a server; the data processing module integrates the linear regression model, the system can accurately predict the future environment temperature based on historical data and environment parameters collected in real time, the parameter space in the linear regression model can be automatically explored by using the grid search optimization technology, the optimal parameter combination can be found, and therefore the prediction precision is further improved, and the prediction efficiency is improved. According to the method, the model with the optimal performance can be selected by comparing MSE values under different parameter settings so as to ensure the accuracy and reliability of a prediction result, and the system can utilize an optimization algorithm to formulate an optimal heating strategy based on the accurate prediction result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of intelligent control technology, in particular to an environmental intelligent control system based on far-infrared heating technology. Background Art

[0002] In the current business environment, environmental intelligent control systems using far-infrared heating technology play a vital role. Traditional environmental control systems often rely on simple mathematical models or empirical formulas when predicting future ambient temperatures, which results in limited accuracy of the prediction results. Since it is impossible to accurately predict the changing trend of future ambient temperatures, it is difficult for the system to formulate the optimal heating or cooling strategy, which affects energy efficiency and environmental comfort.

[0003] In the environmental control process, existing systems usually lack a real-time feedback adjustment mechanism, which means that the system cannot perceive changes in ambient temperature in a timely manner and make corresponding adjustments. When the actual temperature deviates from the set range, the system often takes a long time to respond, resulting in large fluctuations in ambient temperature, affecting the user's comfort experience.

[0004] In addition, many existing environmental control systems also have deficiencies in real-time response capabilities. When the ambient temperature changes, the system may not be able to quickly perceive and adjust the heating or cooling strategy, resulting in the environmental status not being corrected in time, which in turn affects the user experience. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and provide an environmental intelligent control system based on far-infrared heating technology. The system obtains indoor and outdoor environmental parameters in real time through a data acquisition module, and uses a data processing module for preprocessing and predictive analysis. The decision control module formulates a heating strategy based on the prediction results and real-time environmental parameters through a multi-layer perceptron (MLP) algorithm, and adjusts the heater output in real time through a PID control algorithm. By integrating advanced prediction and optimization technologies, precise control of environmental parameters and efficient use of energy are achieved, significantly improving the environmental comfort and system response speed.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: an environmental intelligent control system based on far-infrared heating technology, the system comprising the following components:

[0007] Data acquisition module: used to collect indoor and outdoor environmental parameters in real time, including temperature, humidity, light, weather, and pedestrian flow, and connect with external data sources through various sensors and interfaces to ensure the accuracy and real-time nature of the data;

[0008] Data processing module: Use preprocessing techniques including removing duplicate data, filling missing values, and processing outliers to ensure data accuracy and completeness, convert preprocessed data into a standard scale, extract data features using a construction tree, predict future ambient temperature trends and predicted values ​​using a linear regression model, use grid search to continuously optimize the model, and use mean square error (MSE) to evaluate model performance;

[0009] Decision-making control module: According to the ambient temperature prediction value output by the data processing module and the indoor and outdoor environmental parameters, the appropriate heating strategy is formulated through the multi-layer perceptron (MLP) algorithm, and the output power or operating time of the heater is calculated in combination with the current system status, and the output of the far-infrared heater is dynamically adjusted;

[0010] Feedback adjustment module: It can monitor the changes in ambient temperature in real time and compare the actual temperature with the set temperature. When the actual temperature deviates from the set range, the output of the decision control module is adjusted through the feedback mechanism to ensure that the system can respond quickly and correct the temperature deviation.

[0011] Furthermore, the environmental intelligent control system based on far-infrared heating technology is characterized in that the data acquisition module sets the start time, frequency and duration of data acquisition according to the system's measurement range, accuracy and sampling frequency requirements, selects appropriate sensors and interfaces to connect to external data sources, collects environmental parameters such as temperature, humidity and light in real time through sensors, uses interfaces to obtain weather and crowd flow data from external data sources (such as meteorological stations, crowd counting systems), and stores the collected data in a database.

[0012] Furthermore, the environmental intelligent control system based on far-infrared heating technology is characterized in that the data processing module extracts key features that can reflect the law of environmental temperature changes from the pre-processed data through a decision tree model, divides the feature data into different subsets, such as a training set, a validation set and a test set, and uses a linear regression model to predict the future environmental temperature change trend and predicted value, optimizes the regression model through grid search using the training set, selects the mean square error (MSE) to evaluate the performance of the model through the validation set, uses the trained model to predict new and unseen data, and obtains the future environmental temperature change trend and predicted value, wherein the mathematical formula of the linear regression model is: y=β0+β1x1+β2x2+…+β p x p Where: y is the dependent variable such as ambient temperature, x1, x2, ..., x p are independent variables such as temperature, humidity and traffic volume, β0,β1,…,β pare regression coefficients (or parameters), which indicate the degree of influence of the independent variables on the dependent variable. β0 is the intercept term, which indicates the predicted value of the dependent variable when all independent variables are 0.

[0013] Furthermore, the environmental intelligent control system based on far-infrared heating technology is characterized in that the decision tree model extracts key features that can reflect the law of ambient temperature changes from the preprocessed data, and starts to build a decision tree based on the selected features. The optimal splitting feature is selected by calculating the information gain of each feature. By comparing the information gains of different features, the algorithm can select the feature with the strongest correlation with the target variable (ambient temperature) for splitting. For each subset after splitting, the algorithm will recursively select the next optimal feature for splitting until the stopping condition is met (all samples belong to the same category), and finally construct a decision tree. The features corresponding to the splitting nodes in the decision tree are key features, which can significantly affect the value of the target variable (ambient temperature). The data can be divided into a training set, a validation set and a test set through the decision tree, and the mathematical formula of the information gain is as follows: Where: D is the data set, A is a feature, Values(A) is all possible values ​​of feature A, D v is the sample subset in the dataset whose feature A takes the value v. Entropy(D) is the entropy of dataset D (which measures the impurity of the dataset). v ) is a subset D v The entropy of .

[0014] Furthermore, the environmental intelligent control system based on far-infrared heating technology is characterized in that the grid search optimizes the linear regression model in the data processing module, and the specific steps of the grid search are:

[0015] (1) Define parameter ranges for weight coefficients and regularization parameters in the linear regression model according to ambient temperature prediction requirements;

[0016] (2) Divide the parameter space into a series of discrete grid points, each of which represents a set of possible parameter values;

[0017] (3) For each set of parameter values ​​in the parameter space, a regression analysis model is trained and the model performance is evaluated using the mean square error (MSE) on the validation set;

[0018] (4) Select the parameters corresponding to the model with the best performance on the validation set as the optimal parameters;

[0019] (5) Use the test set and the optimal parameter combination to verify the performance of the linear regression model, make predictions on new, unseen data, and obtain future trends and predictions of ambient temperature.

[0020] Furthermore, the mean square error (MSE) is calculated by calculating the residual between the predicted temperature and the actual temperature of each sample in the validation set, squaring these residuals to eliminate the influence of negative values ​​and ensure that the error is always positive in value. The squared residuals of all samples are added and divided by the number of samples N to obtain the mean square error (MSE). By comparing the MSE values ​​under different models or different parameter settings, the model with the best performance can be selected. The mathematical formula of the mean square error (MSE) is as follows: Where N is the number of samples in the validation set, y pred,i is the model's predicted temperature for the i-th sample, y true,i is the true temperature of the i-th sample.

[0021] Furthermore, the decision control module receives the predicted value of the ambient temperature and indoor and outdoor environmental parameters (such as outdoor temperature, humidity, wind speed, and light intensity) from the data processing module, sets the target temperature range and comfort index, and uses the multi-layer perceptron (MLP) algorithm to predict the heating strategy. According to the preset evaluation criteria (such as energy consumption, comfort, and temperature fluctuation), the optimal heating strategy is found, and the output power or operating time of the heater is calculated through the PID control algorithm and the current system status (such as error size and change trend), and the control instruction is sent to the far-infrared heater to execute the heating strategy.

[0022] Furthermore, the decision control module uses a multi-layer perceptron (MLP) algorithm to predict the optimal heating strategy based on historical data and the set target temperature range and comfort index. The steps of the multi-layer perceptron (MLP) algorithm prediction are:

[0023] (1) Determine the architecture of the MLP, with an input layer having (n) neurons, a hidden layer having (m) neurons, and an output layer having (1) neuron for heating power or heating time;

[0024] (2) For neurons in the hidden layer, the Sigmoid function is selected as the activation function. The mathematical formula of the Sigmoid function is:

[0025] (3) Set the mean square error (MSE) as the loss function to calculate the difference between the predicted result and the actual result. The mathematical formula of MSE is: Where (N) is the number of samples, y i is a real value (such as heating power or heating time in historical data), is the model prediction value;

[0026] (4) Adjust model parameters through the back-propagation algorithm to minimize the loss function;

[0027] (5) Evaluate the MLP model using the same loss function (MSE) as in the training process to calculate the difference between the predicted value and the true value;

[0028] (6) If the model performance does not meet the requirements, you can adjust the model architecture (such as increasing or decreasing the number of neurons in the hidden layer), change the activation function, adjust the loss function or optimize the parameters of the algorithm, and then retrain the model;

[0029] (7) The real-time predicted ambient temperature and indoor and outdoor environmental parameters are input into the trained MLP model to output the optimal heating strategy.

[0030] Furthermore, the decision control module uses the PID control algorithm to combine the heating strategy with the current system state and convert it into a specific control instruction. According to the physical limitations of the far-infrared heater (such as the maximum and minimum values ​​of the heating power), the heating power is dynamically adjusted by setting the proportional (P), integral (I) and differential (D) coefficients, and the deviation between the current predicted temperature and the target temperature range is calculated. The temperature deviation formula is: e(t) = T target -T pred (t), where: T target is the predicted value of ambient temperature, T pred (t) is the target temperature. According to the PID control algorithm, the output u(k) of the controller is calculated, and the output u(k) is converted into a specific control signal and sent to the far-infrared heater. The mathematical formula of the PID control algorithm is: Where: (u(k)) is the output of the controller at the (k)th sampling moment, (e(k)) is the error at the (k)th sampling moment, (K_p) is the proportional coefficient, (K i ) is the integral coefficient, (K_d) is the differential coefficient, and (T) is the sampling period.

[0031] Furthermore, the environmental intelligent control system based on far-infrared heating technology according to claim 1 is characterized in that the feedback adjustment module obtains the value of the ambient temperature in real time through the temperature sensor, compares the actual temperature monitored in real time with the set temperature, and calculates the temperature error, and the temperature deviation formula is: e(t)=T target -T pred (t), where: T target is the predicted value of ambient temperature, T pred(t) is the target temperature. The absolute value of the deviation is used to determine whether the actual temperature deviates from the set range. If the error exceeds a predetermined threshold, feedback adjustment is triggered. According to the calculated error value, the output of the decision control module is adjusted through the feedback mechanism to adjust the power and operating time of the heating or cooling equipment to correct the temperature deviation. After the adjustment, the ambient temperature is continuously monitored and the above steps are repeated until the actual temperature returns to the set range or the error is reduced to an acceptable range.

[0032] Compared with the existing technology, the environmental intelligent control system based on far-infrared heating technology has the following beneficial effects:

[0033] 1. The data processing module of the present invention integrates a linear regression model. The system can accurately predict the future ambient temperature based on historical data and environmental parameters collected in real time. By using grid search optimization technology, it can automatically explore the parameter space in the linear regression model and find the optimal parameter combination, thereby further improving the prediction accuracy. By comparing the MSE values ​​under different parameter settings, the model with the best performance can be selected to ensure the accuracy and reliability of the prediction results. Based on the accurate prediction results, the system can use optimization algorithms to formulate the optimal heating strategy.

[0034] 2. The feedback regulation module of the present invention monitors the changes of ambient temperature in real time through temperature sensors and other equipment to ensure the system's real-time perception of the environmental status. The system compares the actual temperature monitored in real time with the set temperature and calculates the temperature deviation. When the actual temperature deviates from the set range, the system quickly calculates the adjustment amount of the heater through the PID (proportional-integral-differential) control algorithm. The system uses a fast response mechanism to ensure that the heating or cooling strategy can be quickly adjusted when the ambient temperature changes to maintain a stable and comfortable environment.

[0035] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 This is a flow chart of an environmental intelligent control system based on far-infrared heating technology. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] In the first aspect, an environmental intelligent control system based on far infrared heating technology comprises the following components:

[0040] Data acquisition module: used to collect indoor and outdoor environmental parameters in real time, including temperature, humidity, light, weather, and pedestrian flow, and connect with external data sources through various sensors and interfaces to ensure the accuracy and real-time nature of the data;

[0041] Data processing module: Use preprocessing techniques including removing duplicate data, filling missing values, and processing outliers to ensure data accuracy and completeness, convert preprocessed data into a standard scale, extract data features using a construction tree, predict future ambient temperature trends and predicted values ​​using a linear regression model, use grid search to continuously optimize the model, and use mean square error (MSE) to evaluate model performance;

[0042] Decision-making control module: According to the ambient temperature prediction value output by the data processing module and the indoor and outdoor environmental parameters, the appropriate heating strategy is formulated through the multi-layer perceptron (MLP) algorithm, and the output power or operating time of the heater is calculated in combination with the current system status, and the output of the far-infrared heater is dynamically adjusted;

[0043] Feedback adjustment module: It can monitor the changes in ambient temperature in real time and compare the actual temperature with the set temperature. When the actual temperature deviates from the set range, the output of the decision control module is adjusted through the feedback mechanism to ensure that the system can respond quickly and correct the temperature deviation.

[0044] Embodiment 1:

[0045] The present embodiment provides an environmental intelligent control system based on far-infrared heating technology. The system obtains indoor and outdoor environmental parameters in real time through a data acquisition module, and uses a data processing module to extract data features. The future environmental temperature change trend and predicted value are predicted through a linear regression model, and the regression model is optimized by grid search. The mean square error (MSE) is selected as an indicator for evaluating model performance. A heating strategy is formulated through a multi-layer perceptron (MLP) algorithm, and a PID control algorithm is used to adjust the heater output in real time. The feedback adjustment module monitors the ambient temperature in real time and compares it with the set value, and responds quickly to temperature deviations through a feedback mechanism to make rapid adjustments.

[0046] In the specific implementation, the data acquisition module sets the start time, frequency and duration of data collection according to the system's measurement range, accuracy and sampling frequency requirements. It collects environmental parameters such as temperature, humidity and light in real time through sensors, and uses interfaces to obtain weather and pedestrian flow data from external data sources (such as weather stations and crowd counting systems). The collected data will be stored in the database for subsequent processing.

[0047] The data processing module first preprocesses the collected data, including removing duplicate data, filling missing values, processing outliers and other steps to ensure the accuracy and completeness of the data. Then, the data is standardized to facilitate subsequent use of the data. The decision tree model is used to extract key features that can reflect the law of ambient temperature changes from the preprocessed data. Based on the selected features, the decision tree is constructed. The optimal splitting feature of the data is selected by calculating the information gain of the feature. By comparing the information gain of different features, the algorithm can select the feature with the strongest correlation with the ambient temperature for splitting. For each subset after splitting, the algorithm will select the next optimal feature for splitting in turn until all samples belong to the same category, construct a decision tree, and finally divide the data into training set, validation set and test set. The mathematical formula of information gain is: The future ambient temperature change trend and predicted value are predicted by feature selection linear regression model. The mathematical formula of linear regression is: y = β0 + β1x1 + β2x2 + ... + β p x p , use the data in the training set to calculate the linear regression analysis model, use grid search to optimize the regression model, and define the parameter range for the weight coefficient and regularization parameter in the linear regression model according to the ambient temperature requirements. The parameter space is divided into a series of discrete grid points, each grid point represents a set of possible parameter values. For each set of parameters, a linear regression analysis model is trained, and the model performance is evaluated using the mean square error (MSE) on the validation set. The mean square error (MSE) is calculated by calculating the residual between the predicted temperature and the actual temperature of each sample in the validation set, squaring these residuals, adding the squared residuals of all samples and dividing them by the number of samples N to obtain the mean square error (MSE). By comparing the MSE values ​​under different models or different parameter settings, the model with the best performance can be selected. The mathematical formula of the mean square error (MSE) is as follows: Use the trained linear regression model to predict new, unseen data to obtain future ambient temperature change trends and predicted values.

[0048] The decision control module receives the ambient temperature prediction value and indoor and outdoor environmental parameters from the data processing module, sets the target temperature range and comfort index, evaluates the heating strategy using the multi-layer perceptron (MLP) algorithm, and determines the MLP architecture, with an input layer with (n) neurons (including ambient temperature prediction value and indoor and outdoor environmental parameters), a hidden layer with (m) neurons for learning the complex relationship between features, an output layer with (1) neuron for heating power or heating time, and a Sigmoid function is selected. As an activation function, the input content is transformed nonlinearly, and the mean square error MSE is set as the loss function to predict the gap between the result and the actual result. The mathematical formula of MSE is: The model parameters are adjusted through the back propagation algorithm, and the loss function (MSE) is used to calculate the difference between the predicted value and the true value to evaluate the MLP model. If the model performance does not meet the requirements, the number of hidden layer neurons is increased or decreased, and the parameters of the loss function or optimization algorithm are adjusted. The real-time ambient temperature prediction value and indoor and outdoor environmental parameters are input into the trained MLP model, and the optimal heating strategy is output. The heating power is dynamically adjusted by setting the proportional (P), integral (I) and differential (D) coefficients through the PID control algorithm and the current system status (such as error size, change trend), and the deviation between the current predicted temperature and the target temperature range is calculated. The temperature deviation formula is: e(t) = T target -T pred (t), according to the PID control algorithm The output u(k) of the controller is calculated, and the output u(k) is converted into a specific control signal and sent to the far-infrared heater.

[0049] The feedback regulation module obtains the ambient temperature value in real time through the temperature sensor, compares the actual temperature monitored in real time with the set temperature, and calculates the temperature error. If the error exceeds the predetermined threshold, the feedback regulation mechanism is triggered. According to the calculated error value, the output of the decision control module is adjusted through the feedback mechanism, and the power and operating time of the heating or cooling equipment are adjusted to correct the temperature deviation. After the adjustment, the ambient temperature continues to be monitored and the above steps are repeated until the actual temperature returns to the set range or the error is reduced to an acceptable range.

[0050] Embodiment 2:

[0051] In the environmental intelligent control system based on far-infrared heating technology in this embodiment, especially the data processing module predicts the future ambient temperature. First, the system will preprocess the original data, including data cleaning, missing value filling, and outlier processing. The decision tree model is used to extract key features that can reflect the law of ambient temperature changes from the preprocessed data. Based on the selected features, a decision tree is constructed, and the optimal splitting feature is selected by calculating the information gain of each feature. The next optimal feature is recursively selected for splitting until the stopping condition is met. The features corresponding to the splitting nodes in the decision tree are key features. These features significantly affect the value of the ambient temperature. The data is divided into a training set, a validation set, and a test set through the decision tree model.

[0052] Use the training set to build a linear regression model, the model form is: y = β0 + β1x1 + β2x2 + ... + β p x p Where: y is the dependent variable such as ambient temperature, x1, x2, ..., x p are independent variables such as temperature, humidity, β0,β1,…,β p is the regression coefficient, and the regression coefficients β0, β1, …, β are estimated by the least squares method (OLS). p , and then the linear regression analysis model is obtained.

[0053] Subsequently, the model is optimized through grid search. First, the parameter range is defined for the weight coefficient and regularization parameter in the linear regression model. The parameter space is divided into a series of discrete grid points. Each grid point represents a set of possible parameter values. For each set of parameter values ​​in the parameter space, a linear regression model is trained, and the model performance is evaluated using the mean square error (MSE) on the validation set. The parameters corresponding to the model with the best performance on the validation set are selected as the optimal parameters. The model is retrained using the optimal parameters, and new, unseen data are predicted to obtain the future trend of ambient temperature changes and predicted values.

[0054] In addition, the mean square error (MSE) is used as the evaluation indicator The performance of the model on the validation set is calculated by calculating the residual between the predicted temperature and the actual temperature of each sample in the validation set, squaring these residuals, adding the squared residuals of all samples and dividing by the number of samples N to obtain the mean square error (MSE). By comparing the MSE values ​​under different models or different parameter settings, the model with the best performance can be selected.

[0055] Embodiment three:

[0056] Based on the first and second embodiments, the specific process steps of the system are as follows:

[0057] (1) Collect indoor and outdoor environmental parameters in real time through sensors and external data sources;

[0058] (2) Clean the collected data, remove outliers and fill in missing values ​​to ensure the accuracy and completeness of the data;

[0059] (3) Use linear regression models and other machine learning techniques to predict future ambient temperature changes, and optimize the prediction model through methods such as grid search;

[0060] (4) Based on the prediction results and current environmental parameters, the system formulates the optimal heating strategy through the multi-layer perceptron (MLP) algorithm and dynamically adjusts the output of the heater through the PID control algorithm;

[0061] (5) The system monitors the ambient temperature in real time and compares it with the set value. When the actual temperature deviates from the set range, the heating strategy is adjusted through the feedback mechanism to ensure that the ambient temperature remains within the set range.

[0062] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An environmental intelligent control system based on far infrared heating technology, It is characterized in that The system consists of the following components: Data acquisition module: used to collect indoor and outdoor environmental parameters in real time, including temperature, humidity, light, weather, and pedestrian flow, and connect with external data sources through various sensors and interfaces to ensure the accuracy and real-time nature of the data; Data processing module: Use preprocessing techniques including removing duplicate data, filling missing values, and processing outliers to ensure data accuracy and completeness, convert preprocessed data into a standard scale, extract data features using a construction tree, predict future ambient temperature trends and predicted values ​​using a linear regression model, use grid search to continuously optimize the model, and use mean square error (MSE) to evaluate model performance; Decision-making control module: According to the ambient temperature prediction value output by the data processing module and the indoor and outdoor environmental parameters, the appropriate heating strategy is formulated through the multi-layer perceptron (MLP) algorithm, and the output power or operating time of the heater is calculated in combination with the current system status, and the output of the far-infrared heater is dynamically adjusted; Feedback adjustment module: It can monitor the changes in ambient temperature in real time and compare the actual temperature with the set temperature. When the actual temperature deviates from the set range, the output of the decision control module is adjusted through the feedback mechanism to ensure that the system can respond quickly and correct the temperature deviation.

2. According to claim 1, the environmental intelligent control system based on far infrared heating technology, It is characterized in that The data acquisition module sets the start time, frequency and duration of data acquisition according to the system's measurement range, accuracy and sampling frequency requirements, selects appropriate sensors and interfaces to connect to external data sources, collects environmental parameters such as temperature, humidity and light in real time through sensors, uses interfaces to obtain weather and pedestrian flow data from external data sources (such as meteorological stations and crowd counting systems), and stores the collected data in a database.

3. The environmental intelligent control system based on far infrared heating technology according to claim 1, It is characterized in that The data processing module extracts key features that can reflect the law of ambient temperature change from the preprocessed data through a decision tree model, divides the feature data into different subsets, such as a training set, a validation set, and a test set, and uses a linear regression model to predict the future ambient temperature change trend and predicted value. The training set is used to optimize the regression model through grid search, and the mean square error (MSE) is selected to evaluate the performance of the model through the validation set. The trained model is used to predict new and unseen data to obtain the future ambient temperature change trend and predicted value. The mathematical formula of the linear regression model is: y = β 0 +β 1 x 1 +β 2 x 2 +…+β p x p Where: y is the dependent variable such as ambient temperature, x 1 ,x 2 ,…,x p are independent variables such as temperature, humidity and traffic volume, β 0 ,β 1 ,…,β p are regression coefficients (or parameters), which indicate the degree of influence of the independent variable on the dependent variable, β 0 is the intercept term, which represents the predicted value of the dependent variable when all independent variables are 0.

4. The environmental intelligent control system based on far infrared heating technology according to claim 3, It is characterized in that The decision tree model extracts key features that can reflect the law of ambient temperature changes from the preprocessed data. Based on the selected features, the decision tree is constructed. The optimal split feature is selected by calculating the information gain of each feature. By comparing the information gain of different features, the algorithm can select the feature with the strongest correlation with the target variable (ambient temperature) for splitting. For each subset after splitting, the algorithm will recursively select the next optimal feature for splitting until the stop condition is met (all samples belong to the same category). Finally, a decision tree is constructed. The features corresponding to the split nodes in the decision tree are key features, which can significantly affect the value of the target variable (ambient temperature). The data can be divided into a training set, a validation set, and a test set through the decision tree. The mathematical formula of information gain is as follows: Where: D is the data set, A is a feature, Values(A) is all possible values ​​of feature A, D v is the sample subset in the dataset whose feature A takes the value v. Entropy(D) is the entropy of dataset D (which measures the impurity of the dataset). v ) is a subset D v The entropy of .

5. The environmental intelligent control system based on far infrared heating technology according to claim 3, It is characterized in that The grid search optimizes the linear regression model in the data processing module, and the specific steps of the grid search are: (1) Define parameter ranges for weight coefficients and regularization parameters in the linear regression model according to ambient temperature prediction requirements; (2) Divide the parameter space into a series of discrete grid points, each of which represents a set of possible parameter values; (3) For each set of parameter values ​​in the parameter space, a regression analysis model is trained and the model performance is evaluated using the mean square error (MSE) on the validation set; (4) Select the parameters corresponding to the model with the best performance on the validation set as the optimal parameters; (5) Use the test set and the optimal parameter combination to verify the performance of the linear regression model, make predictions on new, unseen data, and obtain future trends and predictions of ambient temperature.

6. The environmental intelligent control system based on far infrared heating technology according to claim 3, It is characterized in that The mean square error (MSE) is calculated by calculating the residual between the predicted temperature and the actual temperature of each sample in the validation set, squaring these residuals to eliminate the influence of negative values ​​and ensure that the error is always positive in value. The squared residuals of all samples are added and divided by the number of samples N to obtain the mean square error (MSE). By comparing the MSE values ​​under different models or different parameter settings, the model with the best performance can be selected. The mathematical formula of the mean square error (MSE) is as follows: Where N is the number of samples in the validation set, y pred,i is the model's predicted temperature for the i-th sample, y true,i is the true temperature of the i-th sample.

7. The environmental intelligent control system based on far infrared heating technology according to claim 1, It is characterized in that The decision control module receives the predicted value of the ambient temperature and indoor and outdoor environmental parameters (such as outdoor temperature, humidity, wind speed, and light intensity) from the data processing module, sets the target temperature range and comfort index, and uses the multi-layer perceptron (MLP) algorithm to predict the heating strategy. According to the preset evaluation criteria (such as energy consumption, comfort, and temperature fluctuation), the optimal heating strategy is found, and the output power or operating time of the heater is calculated through the PID control algorithm and the current system status (such as error size and change trend), and the control instruction is sent to the far-infrared heater to execute the heating strategy.

8. The environmental intelligent control system based on far infrared heating technology according to claim 1, It is characterized in that The decision control module uses a multi-layer perceptron (MLP) algorithm to predict the optimal heating strategy based on historical data and the set target temperature range and comfort index. The steps of the multi-layer perceptron (MLP) algorithm prediction are: (1) Determine the architecture of the MLP, with an input layer having (n) neurons (including ambient temperature prediction values ​​and indoor and outdoor environmental parameters), a hidden layer having (m) neurons for learning complex relationships between features, and an output layer having (1) neuron for heating power or heating time; (2) For neurons in the hidden layer, the Sigmoid function is selected as the activation function to perform nonlinear transformation on the input content. The mathematical formula of the Sigmoid function is: (3) Set the mean square error (MSE) as the loss function to calculate the difference between the predicted result and the actual result. The mathematical formula of MSE is: Where (N) is the number of samples, y i is a real value (such as heating power or heating time in historical data), is the model prediction value; (4) According to the error calculated by the loss function, the model parameters are adjusted through the back propagation algorithm to minimize the loss function; (5) Evaluate the MLP model using the same loss function (MSE) as in the training process to calculate the difference between the predicted value and the true value; (6) If the model performance does not meet the requirements, you can adjust the model architecture (such as increasing or decreasing the number of neurons in the hidden layer), change the activation function, adjust the loss function or optimize the parameters of the algorithm, and then retrain the model; (7) The real-time predicted ambient temperature and indoor and outdoor environmental parameters are input into the trained MLP model to output the optimal heating strategy.

9. According to claim 7, the decision control module uses the PID control algorithm to combine the heating strategy with the current system state and convert it into a specific control instruction. According to the physical limitations of the far-infrared heater (such as the maximum and minimum values ​​of the heating power), the heating power is dynamically adjusted by setting the proportional (P), integral (I) and differential (D) coefficients, and the deviation between the current predicted temperature and the target temperature range is calculated. The temperature deviation formula is: e(t) = T target -T pred (t), in: T target is the predicted value of ambient temperature, T pred (t) is the target temperature. According to the PID control algorithm, the output u(k) of the controller is calculated, and the output u(k) is converted into a specific control signal and sent to the far-infrared heater. The mathematical formula of the PID control algorithm is: Where: (u(k)) is the output of the controller at the (k)th sampling moment, (e(k)) is the error at the (k)th sampling moment, (K_p) is the proportional coefficient, (K i ) is the integral coefficient, (K_d) is the differential coefficient, and (T) is the sampling period.

10. The environmental intelligent control system based on far infrared heating technology according to claim 1, It is characterized in that The feedback adjustment module obtains the value of the ambient temperature in real time through the temperature sensor, compares the actual temperature monitored in real time with the set temperature, and calculates the temperature error. The temperature deviation formula is: e(t) = T target -T pred (t), where: T target is the predicted value of ambient temperature, T pred (t) is the target temperature. The absolute value of the deviation is used to determine whether the actual temperature deviates from the set range. If the error exceeds a predetermined threshold, feedback adjustment is triggered. According to the calculated error value, the output of the decision control module is adjusted through the feedback mechanism to adjust the power and operating time of the heating or cooling equipment to correct the temperature deviation. After the adjustment, the ambient temperature is continuously monitored and the above steps are repeated until the actual temperature returns to the set range or the error is reduced to an acceptable range.

Citation Information

Patent Citations

  • Decision tree-based method for extracting key characteristic variables of finish rolling temperature control process

    CN102049420A

  • Decision-making-tree-algorithm-based analysis and evaluation method for operation risk of power equipment

    CN105894177A

  • A text processing method and system

    CN106407406A

  • Online learning based air conditioning temperature non-inductive control method

    CN111486554A

  • Heat exchange adjusting system of multi-split air conditioner

    CN117146369A

Cited By

  • Temperature control method and system for gas-liquid separation device of distributed flow calorimeter

    CN119178076A

  • Temperature control method and system for gas-liquid separation device of distributed flow calorimeter

    CN119178076B

  • Dynamic heat dissipation adjusting method and system for wind direction predictor of wind energy storage cabinet

    CN119277738A

  • Ultrathin ceramic tile batching optimization method and system based on real-time data acquisition

    CN119314602A

  • Domain control optimization method and system based on cockpit internal environment recognition

    CN119336107A