Heating room temperature prediction method, medium, equipment and heating room temperature prediction system

Abstract

The invention relates to the technical field of heating temperature prediction, and relates to a heating room temperature prediction method, a medium, equipment and a heating room temperature prediction system, and the method comprises the steps: obtaining target information corresponding to a target house; inputting the target information into a predetermined heating room temperature prediction model to obtain a predicted room temperature corresponding to the target house; wherein the heating room temperature prediction model is a model obtained by training according to first historical heating data, and the first historical heating data comprises house information and actual room temperature of a plurality of houses, historical water supply temperature and historical water return temperature of a heating station for supplying heat to the houses, and historical meteorological information when the actual room temperature is collected; the actual room temperature is acquired by temperature sensors deployed in the houses, and the plurality of houses are part of houses in the building. According to the method, the prediction precision is higher, the hardware cost and the installation cost are remarkably reduced, the power consumption of the heating system is reduced, and the external temperature change can be responded in real time.

Description

Technical Field

[0001] The present application relates to the technical field of heating temperature prediction, and more specifically, to a heating room temperature prediction method, a storage medium, an electronic device and a heating room temperature prediction system. Background Art

[0002] With the advancement of urbanization and the improvement of people's quality of life, winter heating has become a basic need for residents in cold areas. However, traditional heating systems have problems such as high energy consumption and inability to respond to external temperature changes in real time, making it difficult to meet modern energy conservation and emission reduction and user comfort requirements.

[0003] Although existing technologies have made some progress in winter heating and indoor temperature prediction, there are still many shortcomings in large-scale residential applications. First, the high cost of installing IoT sensors in each household and the complex maintenance requirements make it difficult to promote them on a large scale. Second, the prediction accuracy of existing prediction models cannot meet the needs of precise heating. Therefore, a more economical and efficient solution is urgently needed to achieve precise heating and energy consumption optimization in large-scale residential areas. Summary of the invention

[0004] The embodiments of the present application provide a heating room temperature prediction method, a storage medium, an electronic device, and a heating room temperature prediction system. The method of the embodiments of the present application has higher prediction accuracy, significantly reduces hardware cost and installation cost, reduces power consumption of the heating system, and can respond to external temperature changes in real time.

[0005] In a first aspect, this embodiment provides a method for predicting heating room temperature, including:

[0006] Acquire target information corresponding to the target house; wherein the target information includes house information of the target house, target water supply temperature and target return water temperature of a heating station that provides heat to the target house, and target meteorological information;

[0007] Inputting the target information into a predetermined heating room temperature prediction model to obtain a predicted room temperature corresponding to the target house output by the heating room temperature prediction model;

[0008] Among them, the heating room temperature prediction model is a model trained according to the first historical heating data, the first historical heating data includes house information and actual room temperatures of multiple houses, historical water supply temperature and historical return water temperature of the heating station that provides heating for the houses, and historical meteorological information when the actual room temperature is collected. The actual room temperature is collected by temperature sensors deployed in the houses, and the multiple houses are part of the houses in the building.

[0009] Optionally, the house information includes at least any one of the following: the floor where the house is located, the house type, the area of ​​the house, and the building to which the house belongs.

[0010] Optionally, the method further comprises:

[0011] According to the predicted room temperature, the actual water supply temperature of the heating station is adjusted.

[0012] Optionally, the predetermined heating room temperature prediction model is generated in the following manner:

[0013] Acquire the first historical heating data set;

[0014] Obtain multiple parameter value combinations of hyperparameters, wherein the hyperparameters include at least one of the following: a learning rate, a maximum depth of a tree, a number of iterations, a sample subset ratio, a feature subset ratio, an L1 regularization parameter, and an L2 regularization parameter;

[0015] Based on each of the parameter value combinations, respectively, a training set in the first historical heating data set is used to perform model training, and a first evaluation result of the first heating room temperature prediction model obtained through training is determined;

[0016] The predetermined heating room temperature prediction model is determined according to the first evaluation result corresponding to each parameter value combination.

[0017] Optionally, the heating room temperature prediction model is a model established based on an XGBoost algorithm or a support vector regression algorithm.

[0018] Optionally, the obtaining of a first historical heating data set includes:

[0019] Obtaining a second historical heating data set;

[0020] Preprocessing the second historical heating data set to obtain a first historical heating data set;

[0021] Wherein, the preprocessing includes at least one of the following:

[0022] Detecting whether an attribute value of a field in each data record in the second historical heating data set is missing, and if so, deleting the data record;

[0023] Detecting whether there are duplications between data records in the historical heating data set, and deleting the duplicate data records if there are duplications;

[0024] Performing feature standardization on the attributes of all fields in the historical heating data set;

[0025] Anomaly detection is performed on each data record in the historical heating data set, and when an anomaly is detected in a data record, the data record is deleted.

[0026] Optionally, the performing model training using a training set in the first historical heating data set to determine a first evaluation result of the first heating room temperature prediction model obtained through training includes:

[0027] Using a K-fold cross validation method, using the training set in the first historical heating data set, performing K model trainings, and determining a second evaluation result of the second heating room temperature prediction model obtained by each training;

[0028] The first evaluation result of the first heating room temperature prediction model is determined according to the second evaluation result obtained in each training.

[0029] Optionally, determining the predetermined heating room temperature prediction model according to the first evaluation result corresponding to each parameter value combination includes:

[0030] Determining a target heating room temperature prediction model according to the first evaluation result corresponding to each of the parameter value combinations;

[0031] Using a test set in the first historical heating data set, evaluating the target heating room temperature prediction model to obtain a third evaluation result;

[0032] When the third evaluation result meets the requirements, the target heating room temperature prediction model is used as the predetermined heating room temperature prediction model.

[0033] In a second aspect, this embodiment provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method as described in any one of the first aspects of the present application.

[0034] In a third aspect, this embodiment provides an electronic device, comprising a memory and a processor, wherein the memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute a method as described in any one of the first aspects of the present application.

[0035] In a fourth aspect, this embodiment provides a heating room temperature prediction system, comprising the electronic device as described in the third aspect.

[0036] The embodiment of the present application trains a heating room temperature prediction model based on the house information and actual room temperatures of multiple houses in the historical heating data, the historical water supply temperature and historical return water temperature of the heating station that provides heating for the houses, and the historical meteorological information when the actual room temperature is collected. The prediction accuracy of the heating room temperature prediction model obtained by training with multi-dimensional heating data is higher. In addition, the actual room temperature in the historical heating data in this embodiment is collected by temperature sensors deployed in some houses in the building, so there is no need to deploy temperature sensors in all houses, and the hardware cost and installation cost are significantly reduced. In addition, through the heating room temperature prediction method of this embodiment, the predicted room temperature can be predicted in advance based on the target information, and the actual water supply temperature of the heating station can be dynamically adjusted based on the predicted room temperature, thereby reducing the power consumption of the heating system and responding to external temperature changes in real time.

[0037] Other features and advantages of the present application will become apparent from the following detailed description of exemplary embodiments of the present application with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the application and, together with the description, serve to explain the principles of the application.

[0039] Figure 1 A schematic flow chart of a heating room temperature prediction method provided in an embodiment of the present application is shown.

[0040] Figure 2 A schematic flow chart of a heating room temperature prediction method provided in another embodiment of the present application is shown.

[0041] Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the present application is shown. DETAILED DESCRIPTION

[0042] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present application.

[0043] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present application, its application, or uses.

[0044] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0045] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0046] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0047] With the advancement of urbanization and the improvement of people's quality of life, winter heating has become a basic need for residents in cold areas. However, traditional heating systems have problems such as high energy consumption and inability to respond to external temperature changes in real time, making it difficult to meet modern energy conservation and emission reduction and user comfort requirements.

[0048] Specifically, traditional heating systems usually use fixed parameter settings, such as setting fixed water supply and outlet temperatures at heating stations, which cannot be adjusted dynamically according to actual needs, resulting in energy waste. Moreover, this static method is also difficult to cope with complex and changing environmental conditions, and cannot respond to external temperature changes in real time. Indoor temperature fluctuates greatly, affecting the quality of life of residents.

[0049] Although existing technologies have made some progress in winter heating and indoor temperature prediction, there are still many shortcomings in large-scale residential applications. First, the high cost of installing IoT sensors in each household and the complex maintenance requirements make it difficult to promote them on a large scale. Second, the prediction accuracy of existing prediction models cannot meet the needs of precise heating. Therefore, a more economical and efficient solution is urgently needed to achieve precise heating and energy consumption optimization in large-scale residential areas.

[0050] At present, machine learning, as an advanced data analysis and prediction technology, is gradually being introduced into the heating system to achieve more efficient energy management and a comfortable indoor environment. Based on this, the embodiment of the present application uses a machine learning method to train a prediction model by collecting the actual room temperature of some houses in the building, thereby reducing hardware costs and installation costs. In addition, the embodiment of the present application also aims to obtain a heating room temperature prediction model with higher prediction accuracy through multi-dimensional heating data training.

[0051] The heating room temperature prediction method of the embodiment of the present application is as follows Figure 1 As shown, it may include steps S110 to S120.

[0052] Step S110, obtaining target information corresponding to the target house; wherein the target information includes house information of the target house, target water supply temperature and target return water temperature of the heating station that provides heating for the target house, and target meteorological information.

[0053] In this embodiment, the target house may be at least one house in a building. The building may be a building within the area of ​​any community or any heating station.

[0054] Generally, the indoor temperature of a house varies depending on its spatial location. The spatial location of a house, for example, is the house located in a side unit or a middle unit, or is the house located on a high floor or a low floor. In addition, the size of the house and the distance from the building to which the house belongs to the heating station will also affect the room temperature of the house after heating. Therefore, the house information of the target house may include at least any one of the following: the floor where the house is located, the house type, the area of ​​the house, and the building to which the house belongs. The house type may include side units and middle units. As an example, the house information may be Room 101, Unit 1, Building 22.

[0055] The target water supply temperature of the heating station is the target temperature flowing out of the heating station. The target water supply temperature may be the actual water supply temperature of the current heating station, or the water supply temperature of the heating station to be adjusted. When the target water supply temperature is the actual water supply temperature of the current heating station, the room temperature of the target house can be predicted by the prediction method of this embodiment, thereby realizing the heating monitoring function. By predicting the room temperature of the target house, the overall room temperature can be further predicted. The overall room temperature is the average temperature of a specific area, such as the average room temperature of houses in a community, or the average room temperature of houses in a building. When the overall room temperature is lower than the set temperature, an actual water supply temperature higher than the current heating station can be configured as the target water supply temperature, and the predicted room temperature can be predicted by continuing the prediction method of this embodiment, thereby further predicting the overall room temperature. When the overall room temperature is higher than the set temperature, an actual water supply temperature lower than the current heating station can be configured as the target water supply temperature, and the predicted room temperature can be predicted by continuing the prediction method of this embodiment, thereby further predicting the overall room temperature.

[0056] The target return water temperature of the heating station is the target temperature of the water flowing into the heating station. The target return water temperature may be the actual return water temperature of the current heating station or the return water temperature of the heating station to be adjusted. In this embodiment, the relationship between the supply water temperature and the return water temperature of the heating station may be predetermined, so as to determine the target return water temperature through the relationship and the target supply water temperature.

[0057] The target meteorological information is a weather phenomenon, and may include at least one of the following, for example: weather temperature, wind speed, and precipitation.

[0058] In this embodiment, optionally, the target information corresponding to the target house can be obtained by receiving information input by the user on the interface. Optionally, the target water supply temperature and target return water temperature of the heating station in the target information are input by the user on the client, the weather information can be obtained by calling the client weather software, and the house information can be obtained by reading the house information pre-configured by the system.

[0059] Step S120, inputting the target information into a predetermined heating room temperature prediction model, and obtaining a predicted room temperature corresponding to the target house output by the heating room temperature prediction model.

[0060] In this embodiment, the heating room temperature prediction model is a model trained based on the first historical heating data. The first historical heating data may include house information and actual room temperature of multiple houses, historical water supply temperature and historical return water temperature of the heating station that provides heating for the houses, and historical meteorological information when the actual room temperature is collected. The actual room temperature is collected by a temperature sensor deployed in the house. The multiple houses are part of the houses in a building.

[0061] The house information in the first historical heating data, the historical water supply temperature and historical return water temperature of the heating station that heats the house, and the historical meteorological information when the actual room temperature is collected are characteristic variables, which are used as the input of the model. The actual room temperature of the house is the target variable of the model. The predicted room temperature of the house is used as the output of the model.

[0062] In this embodiment, several houses at key locations in the building can be selected in advance, and temperature sensors (such as Internet of Things temperature sensors) can be deployed in the houses at the key locations, so as to collect the actual room temperature of the houses through these temperature sensors. The heating room temperature prediction model obtained by training the room temperature of houses at key locations can be used to predict the room temperature of other houses in the building where no temperature sensors are deployed, so that the accurate prediction of the room temperature of each house in the building is achieved at a lower cost.

[0063] The key positions are determined based on the degree of influence of the room temperature of the house on the overall room temperature prediction result. Usually, the room temperature of the house at the key position is close to the overall room temperature, or the average value of the room temperature of the houses at all key positions is close to the overall room temperature. In this embodiment, the key positions for deploying temperature sensors can be determined by a clustering algorithm. The clustering algorithm can be, for example, a K-Means clustering algorithm. By adopting an efficient data collection and distribution strategy, the extensiveness and accuracy of the collected data can be ensured.

[0064] In this embodiment, the predicted temperature may be sent to the client for display so that the user can more intuitively view the predicted room temperature of the target house in the building.

[0065] The embodiment of the present application trains a heating room temperature prediction model based on the house information and actual room temperatures of multiple houses in the historical heating data, the historical water supply temperature and historical return water temperature of the heating station that provides heating for the houses, and the historical meteorological information when the actual room temperature is collected. The prediction accuracy of the heating room temperature prediction model obtained by training with multi-dimensional heating data is higher. In addition, the actual room temperature in the historical heating data in this embodiment is collected by temperature sensors deployed in some houses in the building, so there is no need to deploy temperature sensors in all houses, and the hardware cost and installation cost are significantly reduced. In addition, through the heating room temperature prediction method of this embodiment, the predicted room temperature can be predicted in advance according to the target information, and the actual water supply temperature of the heating station can be dynamically adjusted according to the predicted room temperature, thereby reducing the power consumption of the heating system and responding to external temperature changes in real time.

[0066] like Figure 2 As shown, the heating room temperature prediction method of the embodiment of the present application may also include step S130: adjusting the actual water supply temperature of the heating station according to the predicted room temperature of the target house.

[0067] In this embodiment, the predicted room temperature of the target house can be predicted based on the actual water supply temperature, water outlet temperature and current meteorological information of the heating station, and the overall room temperature can be determined. When the overall room temperature is lower than the set temperature, a higher actual water supply temperature than the current heating station can be configured as the target water supply temperature, and the predicted room temperature can be predicted by the prediction method of this embodiment, thereby further predicting the overall room temperature. When the further predicted overall room temperature reaches the set temperature, the actual water supply temperature of the heating station is adjusted to the target water supply temperature. In this way, the temperature changes of the outside world can be responded to in a timely manner, and the indoor temperature can be maintained stable, thereby improving the comfort of the residents.

[0068] When the overall room temperature is higher than the set temperature, a lower actual water supply temperature than the current heating station can be configured as the target water supply temperature, and the predicted room temperature is predicted by the prediction method of this embodiment, thereby further predicting the overall room temperature. When the predicted overall room temperature drops to the set temperature, the actual water supply temperature of the heating station is adjusted to the target water supply temperature. In this way, the power consumption of the heating system can be reduced and energy waste can be reduced.

[0069] The heating room temperature prediction model of the embodiment of the present application can be generated through steps S210 to S220.

[0070] Step S210, obtaining a first historical heating data set.

[0071] Step S220, using the first historical heating data set to perform model training to obtain a predetermined heating room temperature prediction model.

[0072] The first historical heating data set is a set of first historical heating data. The first historical heating data set includes multiple data records, each of which includes multiple fields such as house information of a target house, target water supply temperature of a heating station providing heating for the target house, target return water temperature, and target meteorological information.

[0073] In this embodiment, the first historical heating data set can be divided into a training set and a test set according to a set ratio. The training set is used to train the model to obtain the target heating room temperature prediction model with the best performance. The test set is used to evaluate the target heating room temperature prediction model. When the evaluation result meets the requirements, the target heating room temperature prediction model is used as the predetermined heating room temperature prediction model.

[0074] The rationality of the division of the training set and the test set can ensure the reliability of the model evaluation results, and the specific value of the set ratio can be determined according to actual needs. As an example, the set ratio can be 8:2.

[0075] Taking into account the impact of the complexity of massive data processing and analysis on the real-time and accuracy of the system, the embodiment of the present application can establish a heating room temperature prediction model based on the XGBoost algorithm or the support vector regression algorithm. The predetermined heating room temperature prediction model is a model trained based on the XGBoost algorithm or the support vector regression algorithm. The XGBoost prediction model is an efficient gradient boosting tree model suitable for regression tasks, capable of processing large-scale data and having strong prediction capabilities. The objective function of the model is the square error between the predicted room temperature and the actual room temperature.

[0076] In order to ensure data quality and consistency, the embodiment of the present application needs to pre-process the first historical heating data set in advance. That is, step S210 in the embodiment of the present application may include steps S211 to S212.

[0077] Step S211, obtaining a second historical heating data set.

[0078] Step S212, preprocessing the second historical heating data set to obtain the first historical heating data set.

[0079] In this embodiment, the second historical heating data set is a set of original historical heating data that has not been preprocessed, and the first historical heating data set is a set of preprocessed historical heating data.

[0080] Preprocessing the second historical heating data set may include at least one of the following: detecting whether there is a missing attribute value of a field in each data record in the second historical heating data set, and deleting the data record if there is a missing attribute value; detecting whether there is duplication between data records in the historical heating data set, and deleting the duplicate data record if there is duplication; performing feature standardization on the attributes of all fields in the historical heating data set; performing anomaly detection on each data record in the historical heating data set, and deleting the data record when an anomaly is detected in the data record; and segmenting the fields in each data record in the second historical heating data set into feature variables and target variables.

[0081] In this embodiment, feature standardization can improve the convergence speed and prediction effect of the model, and avoid the influence of some features on the model due to too large or too small value range. As an example, StandardScaler can be used to standardize the feature values ​​of the features so that the features have the same dimension.

[0082] The presence of outliers may have a negative impact on model training, so it is necessary to remove these data before modeling. As an example, IsolationForest can be used to detect outliers, identify and remove abnormal data, and ensure data quality.

[0083] House information, historical water supply temperature of the heating station, historical return water temperature of the heating station, and meteorological information can be used as feature variables, and the actual indoor temperature of the house corresponding to these feature variables can be used as the target variable. The relationship between the feature variables and the target variables should be clarified to prepare for subsequent modeling.

[0084] In some embodiments, before acquiring the second historical heating data set, the file format storing the second historical heating data set may be checked to determine whether it is a file in a set format. If the file format storing the second historical heating data set is not a file in a set format, it is converted into a file in a set format.

[0085] The embodiment of the present application preprocesses the historical heating room temperature data to ensure the quality and consistency of the data in the training set.

[0086] Considering that the existing prediction model has weak generalization ability in different cells and is difficult to adapt to diverse environments and needs, step S220 of the embodiment of the present application may include steps S221 to S223.

[0087] Step S221, obtaining multiple parameter value combinations of hyperparameters.

[0088] In this embodiment, the hyperparameter may include at least one of the following: a learning rate, a maximum depth of a tree, a number of iterations, a sample subset ratio, a feature subset ratio, an L1 regularization parameter, and an L2 regularization parameter.

[0089] Among them, the learning rate is the step size scaling factor of the model at each iteration, which is used to prevent overfitting. The value of the learning rate affects the amount of weight adjustment of the model at each iteration. A smaller learning rate can make the model converge more stably, but it means that more iterations are required. A larger learning rate can make the model converge quickly, but it may also skip the optimal solution or make the model unstable. The value of the learning rate can be set according to actual needs. As an example, the learning rate can take a value between 0.01 and 0.2.

[0090] The maximum depth of the tree limits the number of nodes in the tree. The larger the maximum depth of the tree, the more complex the tree. To prevent overfitting, the maximum depth of the tree cannot be too large. As an example, the maximum depth of the tree can take values ​​between 3 and 10.

[0091] The number of iterations is the number of times the algorithm is repeated during model training, that is, the total number of trees to be trained. Too many trees may lead to overfitting, so the number of iterations cannot be too large. As an example, the number of iterations can be between 500 and 1000.

[0092] The sample subset ratio is the ratio of samples used to train each tree to the total training set samples. In order to prevent overfitting or underfitting, an appropriate sample subset ratio needs to be selected. As an example, the sample subset ratio can be a value between 0.5 and 1.

[0093] The feature subset ratio is the ratio of the feature subsamples used in each tree training to the entire feature set. In order to prevent overfitting, it is necessary to select an appropriate feature subset ratio. As an example, the feature subset ratio can take a value between 0.5 and 1.

[0094] By increasing the L1 regularization parameter and the L2 regularization parameter, it can help reduce overfitting.

[0095] In this embodiment, optionally, a parameter value range can be set for each hyperparameter, and a step length for traversing the parameter value can be set. From the parameter value range of each hyperparameter, a parameter value of the hyperparameter is obtained according to the traversal step length, and then the obtained parameter values ​​of each hyperparameter are combined to obtain a parameter value combination. The traversal is continued until the end, and multiple parameter value combinations can be obtained. It can be understood that each parameter value combination includes the parameter values ​​of all hyperparameters.

[0096] Optionally, multiple specific parameter values ​​can be set for each hyperparameter. A parameter value is randomly selected from the parameter values ​​of each hyperparameter, and then the obtained parameter values ​​of each hyperparameter are combined to obtain a parameter value combination. After permutations and combinations, multiple parameter value combinations can be obtained.

[0097] As an example, the learning rate can have values ​​of 0.01, 0.05, 0.1, 0.2. The maximum depth of the tree can have values ​​of 3, 5, 7, 9. The number of iterations can have values ​​of 50, 100, 200, 300. The sample subset ratio can have values ​​of 0.7, 0.8, 0.9, 1. The feature subset ratio can have values ​​of 0.7, 0.8, 0.9, 1. The L1 regularization parameter can have values ​​of 0, 0.1, 0.5, 1. The L2 regularization parameter can have values ​​of 1, 1.5, 2, 3. The above 7 hyperparameters, each with 4 parameter values, can give 4 7 parameter value combinations.

[0098] In this embodiment, the parameter value of each hyperparameter can be input into GridSearchCV, and GridSearchCV is used to obtain multiple parameter value combinations.

[0099] Step S222: Based on each parameter value combination, model training is performed using a training set in the first historical heating data set to determine a first evaluation result of the trained first heating room temperature prediction model.

[0100] In this embodiment, the specific performance metric of the first heating room temperature prediction model can be set according to actual needs, which is not limited here. As an example, the performance metric can be the accuracy of the model output result. The accuracy of the model output result represents the difference between the actual room temperature and the predicted room temperature output by the model. As another example, the performance metric can also be the negative mean square error between the model prediction result and the actual result. According to the value of the performance metric obtained by training, the first evaluation result can be obtained.

[0101] Step S223, determining a predetermined heating room temperature prediction model according to the first evaluation result of the first heating room temperature prediction model corresponding to each parameter value combination.

[0102] For the above 4 7 Each parameter value combination in the parameter value combinations is trained using the training set in the first historical heating data set, and a first evaluation result can be obtained. 7 The combination of parameter values ​​can result in 4 7 The first evaluation result. 7 The first evaluation result can determine the optimal parameter value combination, thereby determining the predetermined heating room temperature prediction model.

[0103] With GridSearchCV, you can systematically try various parameter combinations to find the optimal parameter settings.

[0104] The embodiment of the present application comprehensively considers the influence of multiple hyperparameters on the model performance, and the obtained heating room temperature prediction model is more stable. In addition, model training is performed based on different parameter value combinations to determine the optimal parameter value combination of the hyperparameters, thereby determining the target heating room temperature prediction model with the best performance. By tuning the hyperparameters, the performance of the target heating room temperature prediction model can be improved.

[0105] Furthermore, in order to ensure the stability and generalization ability of the model under different data partitions, in step S222 of the embodiment of the present application, the training set in the first historical heating data set is used to perform model training, and the first evaluation result of the first heating room temperature prediction model obtained by training is determined, which may include steps S310 to S320.

[0106] Step S310, adopting the K-fold cross validation method, using the training set in the first historical heating data set, performing K times of model training, and determining the second evaluation result of the second heating room temperature prediction model obtained by each training.

[0107] In this embodiment, specifically, in step S310, the training set in the first historical heating data set can be divided into K subsets. The K subsets can be obtained by random and uniform division. From the K subsets, one of the subsets is selected in turn as the test subset, and the remaining K-1 subsets are used as the training subsets. The model is trained once using the training subset to obtain the second heating room temperature prediction model, and the performance of the second heating room temperature prediction model is tested using the test subset to obtain a second evaluation result of the second heating room temperature prediction model. This process is repeated K times to ensure that each subset can be used as a test subset. After K times of model training, K second evaluation results can be obtained.

[0108] In this embodiment, K is a positive integer greater than or equal to 2. As an example, K may be equal to 5. A 5-fold cross validation method is adopted, and the training set in the first historical heating data set is used to perform model training 5 times, and the second evaluation result of the model obtained after each training is determined.

[0109] The second evaluation result of the second heating room temperature prediction model can also be reflected by a specific performance metric. As an example, it can be the accuracy of the model output result and / or the negative mean square error between the model prediction result and the actual result.

[0110] Step S320, determining the first evaluation result of the first heating room temperature prediction model according to the second evaluation result of the second heating room temperature prediction model obtained by each training.

[0111] In this embodiment, in step S320, the second evaluation results of the second heating room temperature prediction model obtained from each training can be averaged, and the average of the second evaluation results can be used as the first evaluation result. Taking 5-fold cross validation as an example, the average of 5 second evaluation results obtained from 5 tests can be calculated, and the average can be used as the first evaluation result of the first heating room temperature prediction model.

[0112] In step S320, a mean square error may also be calculated for the second evaluation result of the trained second heating room temperature prediction model, and the mean square error may be used as the first evaluation result of the first heating room temperature prediction model.

[0113] Of course, in step S320, other calculations may be performed on the second evaluation result of the trained second heating room temperature prediction model, and the calculation result may be used as the first evaluation result of the first heating room temperature prediction model. This embodiment does not limit the specific implementation of step S320, and it may be performed according to actual needs.

[0114] In this embodiment, for each parameter value combination in step S222, K-fold cross validation is performed to obtain the second evaluation results of K second heating room temperature prediction models, and the first evaluation result of the first heating room temperature prediction model is determined based on the second evaluation results of the K second heating room temperature prediction models. In step S223, the first evaluation results of the first heating room temperature prediction model corresponding to all parameter value combinations are combined to determine the target parameter value combination, thereby determining the target heating room temperature prediction model.

[0115] With 4 7 For example, 7 For each parameter value combination, the K-fold cross-validation method is used to train the model and obtain K second evaluation results. Based on the K second evaluation results, 1 first evaluation result is obtained. 7 The combination of parameter values ​​can result in 4 7 The first evaluation results. 7 Among the first evaluation results, the target parameter value combination with the best first evaluation result can be determined, and the model corresponding to the target parameter value combination is used as the target heating room temperature prediction model.

[0116] The embodiment of the present application uses hyperparameter tuning and / or K-fold cross validation to make the trained heating room temperature prediction model have better adaptability and generalization ability in different communities and environments.

[0117] In order to further improve the prediction accuracy and reliability of the model, step S223 in the embodiment of the present application may include steps S330 to S350.

[0118] Step S330: determining a target heating room temperature prediction model according to the first evaluation result corresponding to each parameter value combination.

[0119] In this embodiment, optionally, a target parameter value combination corresponding to the best first evaluation result can be determined from the first evaluation results corresponding to all parameter value combinations. Optionally, a target parameter value combination corresponding to the first evaluation result when the first evaluation result meets the set requirements can be determined from the first evaluation results corresponding to all parameter value combinations. After obtaining the target parameter value combination, a heating room temperature prediction model corresponding to the target parameter value combination is determined as the target heating room temperature prediction model.

[0120] Step S340: Use the test set in the first historical heating data set to evaluate the target heating room temperature prediction model to obtain a third evaluation result.

[0121] Step S350: When the third evaluation result meets the requirements, the target heating room temperature prediction model is used as the predetermined heating room temperature prediction model.

[0122] In this embodiment, the target heating room temperature prediction model is evaluated using the test set in the first historical heating data set to obtain a third evaluation result. If the third evaluation result does not meet the requirements, continue to perform step S221 to obtain a parameter value combination of the hyperparameter. If the third evaluation result meets the requirements, determine the predetermined heating room temperature prediction model.

[0123] The third evaluation result can be set to a specific evaluation index according to actual needs, which is not specifically limited here. As an example, the third evaluation index can be the mean square error of the target heating room temperature prediction model, or the R 2 The score can also be used to calculate the mean square error and R of the target heating room temperature prediction model. 2 In this example, the mean square error and / or R 2 When the score meets the requirements, the target heating room temperature prediction model is used as the predetermined heating room temperature prediction model.

[0124] In this embodiment, the mean square error of the model is used to measure the difference between the predicted room temperature output by the model and the actual room temperature, R 2 The score is used to measure the model's ability to explain changes in the target variable. The third evaluation result can reflect the prediction accuracy and reliability of the model. The smaller the mean square error value, the smaller the error between the predicted room temperature and the actual room temperature. 2 The closer the score value is to 1, the more variation the model can explain, which means that the model fits the data better and the predicted room temperature is very close to the actual room temperature.

[0125] In some embodiments, the heating room temperature prediction model and the normalizer obtained after training and evaluation can be saved for subsequent use. Saving the model can facilitate subsequent prediction tasks without retraining. Saving the normalizer can ensure that the feature processing of new data is consistent with the training data. As an example, the Joblib library can be used to save the model and the normalizer.

[0126] The embodiment of the present application uses the XGBoost algorithm to predict indoor temperature, making full use of its advantages in processing high-dimensional data and nonlinear relationships, integrating multi-dimensional data, such as floor height, heating station temperature, weather conditions and apartment information, to achieve multivariate analysis. The system improves heating efficiency and resident comfort by dynamically adjusting heating parameters. By adopting efficient data collection and distribution strategies, the breadth and accuracy of collected data can be ensured. Through algorithm optimization and performance improvement, accurate prediction and energy conservation and environmental protection can be achieved, significantly improving the intelligent management level and user experience of the heating system.

[0127] An embodiment of the present application further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described in any one of the above method embodiments is implemented.

[0128] The present application also provides an electronic device. Figure 3 As shown, the electronic device 100 includes a memory 110 and a processor 120, wherein the memory 110 is used to store computer instructions, and the processor 120 is used to call the computer instructions from the memory 110 to execute the method as described in any one of the above embodiments.

[0129] An embodiment of the present application further provides a computer program product, including a computer program, which implements the method described in any one of the above embodiments when executed by a processor.

[0130] The embodiment of the present application further provides a heating room temperature prediction system, comprising the electronic device 100 as described in the above embodiment.

[0131] Each embodiment in this application is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and equipment embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0132] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0133] The present application may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present application.

[0134] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0135] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0136] The computer program instructions for performing the operation of the present application can be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as SmallTalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions can be executed completely on a user's computer, partially on a user's computer, executed as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet). In certain embodiments, by using the state information of a computer-readable program instruction to customize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA) or a programmable logic array (PLA), the electronic circuit can execute a computer-readable program instruction, thereby realizing various aspects of the present application.

[0137] Various aspects of the present application are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0138] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0139] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0140] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a part of a module, program segment or instruction, and the part of the module, program segment or instruction contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that it is equivalent to implement it by hardware, implement it by software, and implement it by combining software and hardware.

[0141] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein. The scope of the present application is defined by the appended claims.

Claims

1. A method for predicting heating room temperature, characterized in that: include: Acquire target information corresponding to the target house; wherein the target information includes house information of the target house, target water supply temperature and target return water temperature of a heating station that provides heat to the target house, and target meteorological information; Inputting the target information into a predetermined heating room temperature prediction model to obtain a predicted room temperature corresponding to the target house output by the heating room temperature prediction model; Among them, the heating room temperature prediction model is a model trained according to the first historical heating data, the first historical heating data includes house information and actual room temperatures of multiple houses, historical water supply temperature and historical return water temperature of the heating station that provides heating for the houses, and historical meteorological information when the actual room temperature is collected. The actual room temperature is collected by temperature sensors deployed in the houses, and the multiple houses are part of the houses in the building.

2. The method according to claim 1, characterized in that The house information includes at least any one of the following: the floor where the house is located, the house type, the area of ​​the house, and the building to which the house belongs.

3. The method according to claim 1, characterized in that The method further comprises: According to the predicted room temperature, the actual water supply temperature of the heating station is adjusted.

4. The method according to claim 1, characterized in that: The heating room temperature prediction model is a model established based on the XGBoost algorithm or the support vector regression algorithm.

5. The method according to claim 1, characterized in that The predetermined heating room temperature prediction model is generated in the following manner: Acquire the first historical heating data set; Obtain multiple parameter value combinations of hyperparameters, wherein the hyperparameters include at least one of the following: a learning rate, a maximum depth of a tree, a number of iterations, a sample subset ratio, a feature subset ratio, an L1 regularization parameter, and an L2 regularization parameter; Based on each of the parameter value combinations, respectively, a training set in the first historical heating data set is used to perform model training, and a first evaluation result of the first heating room temperature prediction model obtained through training is determined; The predetermined heating room temperature prediction model is determined according to the first evaluation result corresponding to each parameter value combination.

6. The method according to claim 5, characterized in that The obtaining of the first historical heating data set comprises: Obtaining a second historical heating data set; Preprocessing the second historical heating data set to obtain a first historical heating data set; Wherein, the preprocessing includes at least one of the following: Detecting whether an attribute value of a field in each data record in the second historical heating data set is missing, and if so, deleting the data record; Detecting whether there are duplications between data records in the historical heating data set, and deleting the duplicate data records if there are duplications; Performing feature standardization on the attributes of all fields in the historical heating data set; Anomaly detection is performed on each data record in the historical heating data set, and when an anomaly is detected in a data record, the data record is deleted.

7. The method according to claim 5, characterized in that The using the training set in the first historical heating data set to perform model training and determine a first evaluation result of the first heating room temperature prediction model obtained through training, comprises: Using a K-fold cross validation method, using the training set in the first historical heating data set, performing K model trainings, and determining a second evaluation result of the second heating room temperature prediction model obtained by each training; The first evaluation result of the first heating room temperature prediction model is determined according to the second evaluation result obtained in each training.

8. The method according to claim 5, characterized in that Determining the predetermined heating room temperature prediction model according to the first evaluation result corresponding to each parameter value combination includes: Determining a target heating room temperature prediction model according to the first evaluation result corresponding to each of the parameter value combinations; Using a test set in the first historical heating data set, evaluating the target heating room temperature prediction model to obtain a third evaluation result; When the third evaluation result meets the requirements, the target heating room temperature prediction model is used as the predetermined heating room temperature prediction model.

9. A storage medium, characterized in that: A computer program is stored thereon, which implements the method according to any one of claims 1 to 8 when executed by a processor.

10. An electronic device, characterized in that: including memory and processor, The memory is used to store computer instructions, and the processor is used to call the computer instructions from the memory to execute the method according to any one of claims 1 to 8.

11. A heating room temperature prediction system, characterized in that: Comprising the electronic device as claimed in claim 10.

Images