Network control method, device and system based on Internet of Things smart heating technology

Abstract

The present invention relates to a networking control method, device and system based on the Internet of Things smart heating technology, which is applicable to the field of smart heating technology. The present invention inputs user heating related parameters into a prediction model trained under theoretical conditions to output a prediction value, and obtains an environmental bias parameter that can sum up all complex parameters affecting the user's indoor heating temperature by comparing the difference between the prediction value and the corresponding actual value. Based on the obtained environmental bias parameter, the user's subsequent heating demand temperature value is accurately compensated and corrected, so that when the primary heating side is heating, the heating temperature deviation caused by the user's own environmental parameters can be accurately reserved, thereby making the heating more accurate and reasonable, successfully ensuring the user's heating demand without wasting heating resources, and improving the efficiency of smart heating.

Description

Technical Field

[0001] The present invention is applicable to the field of smart heating technology, and in particular to a networking control method, device and system based on the Internet of Things smart heating technology. Background Art

[0002] With the development of urban heating, the complexity of the physical equipment network of heating is also increasing. It is difficult for traditional heating systems to meet the needs of refined operations and personalized services, and they must inevitably move towards intelligence. As a result, a smart heating model based on the intelligence of the heating system has been developed.

[0003] Smart heating is based on the deep integration of digital, networked, intelligent information technology and advanced heating technology, takes user needs as its target, and features low carbon, comfort, and efficiency as its main characteristics. It is a modern heating model with technical characteristics such as self-perception, self-analysis, self-diagnosis, self-decision-making, and self-learning.

[0004] However, the heating network includes primary and secondary heating networks. The primary network refers to the networking of the heat source and the trunk pipeline, that is, the primary side of the heating, and the secondary network refers to the networking between the trunk pipeline and the user, that is, the secondary side of the heating. Different users have different heat demands and the time for adjusting the heat demand is not fixed. At the same time, the indoor temperature maintenance conditions of different users are different, making it difficult to effectively and accurately determine the corresponding primary side heating supply. In the current smart heating solution, whether it is a heating control method that determines the primary side heating supply based on the maximum user heat demand in order to ensure that all users on the secondary side can get enough heating, or a heating control method that uses the average user heat demand for heating in order to avoid waste of heat energy, both will lead to waste of thermal energy resources or insufficient heating for some users, resulting in the current smart heating solution having a technical problem of low heating efficiency. Summary of the invention

[0005] In view of this, the embodiments of the present invention provide a networking control method, device and system based on the Internet of Things smart heating technology to solve the problem of low efficiency of smart heating.

[0006] In a first aspect, a network control method based on the Internet of Things smart heating technology is provided, the method comprising:

[0007] The historical demand temperature information, historical ambient temperature information, historical indoor temperature information, preset temperature rise time and bias parameter initialized to zero of any user are input into the prediction model trained under theoretical conditions, and the bias parameter is iterated according to the error between the predicted value output by the prediction model and the corresponding actual value in the historical data, and the iterated bias parameter is recorded as the environmental bias parameter corresponding to the user;

[0008] All users are clustered based on the environmental bias parameters of each user, and the mean value of the environmental bias parameters of all users in any cluster set is used as the cluster center value of the cluster set;

[0009] Constructing a first weight proportional to the number of users in the cluster set and a second weight proportional to the cluster center value of the cluster set;

[0010] Determine the average demand temperature of all users in any cluster set within a preset future time period, multiply the average demand temperature with the first weight and the second weight to obtain a sub-demand value of the cluster set, add the sub-demand values ​​corresponding to all cluster sets to obtain a target demand value, and adjust the heating parameters on the primary side of the heating according to the target demand value feedback.

[0011] In a second aspect, a network control device based on the Internet of Things smart heating technology is provided, which is applied to any of the network control methods based on the Internet of Things smart heating technology, including:

[0012] A user indoor heat loss characterization module is used to input any user's historical demand temperature information, historical ambient temperature information, historical indoor temperature information, preset temperature rise time, and a bias parameter initialized to zero into a prediction model trained under theoretical conditions, and iterate the bias parameter according to the error between the predicted value output by the prediction model and the corresponding actual value in the historical data, and record the iterated bias parameter as the environmental bias parameter corresponding to the user;

[0013] A user classification module is used to cluster all users based on the environmental bias parameters of each user, and take the mean value of the environmental bias parameters of all users in any cluster set as the cluster center value of the cluster set;

[0014] A modified weight acquisition module, used to construct a first weight proportional to the number of users in the cluster set, and a second weight proportional to the cluster center value of the cluster set;

[0015] The primary side heat supply regulation module is used to determine the average demand temperature of all users in any cluster set within a preset future time period, multiply the average demand temperature with the first weight and the second weight to obtain the sub-demand value of the cluster set, add the sub-demand values ​​corresponding to all cluster sets to obtain the target demand value, and adjust the heating parameters of the primary side of the heating according to the target demand value feedback.

[0016] In the third aspect, the present invention provides a networked control system based on the Internet of Things smart heating technology, including a computer device and a computer-readable storage medium, for controlling the networked control device based on the Internet of Things smart heating technology, characterized in that the computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, any one of the networked control methods based on the Internet of Things smart heating technology is implemented, and when the computer program is executed by the processor, any one of the networked control methods based on the Internet of Things smart heating technology is implemented.

[0017] Compared with the prior art, the present invention has the following beneficial effects:

[0018] The present invention utilizes the adaptability of the neural network, and obtains the environmental bias parameters based on the training iteration of the neural network that can sum up the complex parameters that are originally unquantifiable or difficult to quantify and affect the indoor heating temperature of each user, and then classifies the users according to the bias parameters, and compensates and corrects the user demand temperature values ​​under different categories based on the size of the bias parameters, so that when determining the heating parameters of the primary heating network, the corresponding heating compensation amount can be accurately reserved for the heating temperature deviation caused by the user's own environmental parameters, so that the heating is more accurate and reasonable, and the heating demand of the user is successfully guaranteed without causing waste of heating resources. The conventional method avoids the situation where the average user demand is used for heating, resulting in insufficient heating for a large number of users, and the conventional method avoids the situation where the maximum user demand is used for heating, resulting in waste of heating resources, thereby improving the efficiency of smart heating. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0020] Figure 1 This is a schematic diagram of an application environment of a network control method based on the Internet of Things smart heating technology provided in the first embodiment of the present invention;

[0021] Figure 2 It is a flow chart of a networking control method based on the Internet of Things smart heating technology provided by the first embodiment of the present invention;

[0022] Figure 3 is a schematic diagram of the structure of a prediction model provided by Embodiment 2 of the present invention;

[0023] Figure 4is a flow chart of a method for determining an environmental bias parameter corresponding to a user provided in Embodiment 2 of the present invention;

[0024] Figure 5 It is a structural schematic diagram of a network control device based on the Internet of Things smart heating technology provided by the third embodiment of the present invention;

[0025] Figure 6 It is a structural diagram of a computer device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0026] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0027] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0028] It should also be understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0029] As used in the present specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]", depending on the context.

[0030] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0031] References to "one embodiment" or "some embodiments" etc. described in the present specification mean that one or more embodiments of the present invention include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0032] The embodiments of the present invention can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0033] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0034] It should be understood that the order of execution of the steps in the following embodiments does not imply a precedence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0035] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0036] The networking control method based on the Internet of Things smart heating technology provided in the first embodiment of the present invention can be applied in the following aspects: Figure 1In the application environment, the server communicates with the sensing device and the client through the network respectively. The server can be implemented with an independent server or a server cluster composed of multiple servers. The above-mentioned smart heating method based on the Internet of Things can be applied to the server in Figure 1. The computer device corresponding to the server is connected to the corresponding database, rule base, etc. to obtain the corresponding data in the database, and at the same time obtains the data collected by the sensing device and the request data sent by the client through the network. Among them, the sensing device is a device installed in a heating environment for collecting relevant temperature data in the heating environment. For example, the sensing device can be a temperature sensor and an intelligent heating valve, wherein the temperature sensor may include an indoor temperature sensor and an outdoor temperature sensor. The sensing device is connected to the server through the network, etc., so that the server can obtain the corresponding data and store and process the data. The client, the server, the sensing device, etc. form the entire smart heating system. The server is used to collect data and perform data analysis and processing, thereby realizing smart heating.

[0037] See also Figure 2 , is a flow chart of a network control method based on the Internet of Things smart heating technology provided in the first embodiment of the present invention. The smart heating method based on the Internet of Things may include the following steps:

[0038] Step S101, input any user's historical demand temperature information, historical ambient temperature information, historical indoor temperature information, preset temperature rise time and bias parameter initialized to zero into a prediction model trained under theoretical conditions, and iterate the bias parameter according to the error between the predicted value output by the prediction model and the corresponding actual value in the historical data, and record the iterated bias parameter as the environmental bias parameter corresponding to the user.

[0039] One of the main reasons for the low heating efficiency in the current smart heating solution is that on the secondary side of the heating, that is, the user side, the indoor temperature maintenance conditions of each user are different. For example, different users have different indoor area sizes, different house types, and whether the heating is turned on for neighbors on the same floor and above or below. As a result, different users on the user side have different abilities to maintain indoor temperatures. In other words, the degree of indoor heat loss of different users on the user side during the heating process is different, so it is impossible to make a sufficiently refined determination of the heating amount, resulting in a waste of heating resources or a low degree of satisfaction of user heating needs, which ultimately leads to low heating efficiency.

[0040] To this end, this embodiment first constructs a number of training samples under theoretical conditions. The so-called training samples under theoretical conditions refer to training samples constructed under the condition that the heat loss of different users in the heating process is different and the heat exchange parameters or heat exchange conditions between each user and the environment are assumed to be the same. The prediction model trained by such training samples can output the theoretical supply temperature value after inputting clear indoor temperature value, ambient temperature value, temperature rise time and demand temperature value, and can output the theoretical room temperature value after inputting clear indoor temperature value, supply temperature value, ambient temperature value and temperature rise time.

[0041] Therefore, after inputting any user's historical demand temperature information, historical ambient temperature information, historical indoor temperature information and preset temperature rise time into the prediction model, the corresponding theoretical temperature supply value can be obtained, or the predicted temperature supply value under theoretical conditions can be obtained; then, based on the predicted temperature supply value obtained under theoretical conditions, and continuing to combine the historical indoor temperature information, historical ambient temperature information and preset temperature rise time input into the prediction model, the theoretical indoor temperature value after the preset temperature rise time compared to the historical indoor temperature information input into the prediction model, or the predicted indoor temperature value, can be obtained.

[0042] That is, after this embodiment inputs any user's historical demand temperature information, historical ambient temperature information, historical indoor temperature information, and preset temperature rise duration into the prediction model trained under theoretical conditions, the predicted indoor temperature value finally output by the prediction model is correspondingly the theoretical indoor temperature value under the theoretical conditions. The predicted indoor temperature value under the theoretical conditions is easily understood to be a prediction result of an indoor temperature value that does not take into account the difference in the user's actual indoor temperature maintenance conditions during the heating process, or the user's actual indoor heat loss degree. Then, by comparing the predicted indoor temperature value with the historical indoor temperature value corresponding to the time point in the historical indoor temperature information input into the prediction model and taking the difference, the actual indoor heat loss degree of the user can be characterized.

[0043] Therefore, by constructing an iterative loss function about the difference between the predicted indoor temperature value and the historical indoor temperature value corresponding to the time point, the bias parameter of the input prediction model initialized to zero can be iteratively adjusted. When the iterative loss function converges, the obtained bias parameter is recorded as the environmental bias parameter corresponding to the user. At this time, the environmental bias parameter can accurately represent the actual degree of indoor heat loss of the corresponding user. The obtained environmental bias parameter can accurately represent various quantifiable and / or non-quantifiable factors that affect the user's indoor heating temperature, such as the user's indoor environment layout, the heating conditions of other users going up and down the stairs, etc.

[0044] It should be noted that the historical demand temperature information refers to the demand temperature values ​​issued by the user at different historical time points. Usually, the interval between adjacent historical time points is a fixed value, which can be 1 hour, 2 hours, etc.; the historical ambient temperature information refers to the outdoor temperature value corresponding to the user at each historical time point. The outdoor temperature value can be obtained through the temperature sensor deployed outdoors in the Internet of Things, or determined based on historical meteorological data; the historical indoor temperature information refers to the indoor temperature value corresponding to the user at each historical time point. The indoor temperature value can be obtained through the temperature sensor deployed indoors in the Internet of Things.

[0045] Through this step of the present embodiment, the environmental bias parameters corresponding to each user can be determined, so as to accurately quantify the degree of indoor heat loss of each user during the heating process, and provide a basis for the subsequent classification of all users regarding the indoor temperature maintenance conditions during the heating process.

[0046] Step S102: clustering all users based on the environmental bias parameters of each user, and taking the mean value of the environmental bias parameters of all users in any cluster set as the cluster center value of the cluster set.

[0047] The size of the user's environmental bias parameter represents the specific degree of indoor heat loss of the user during the heating process. Therefore, in order to achieve refined heating and thus improve heating efficiency, it is necessary to provide different additional heating amounts to users with different degrees of indoor heat loss to ensure that they can reach the required temperature or the expected temperature without providing too much heating amount to cause heating waste.

[0048] To this end, it is necessary to cluster all users based on the environmental bias parameters of each user to distinguish the degree of indoor heat loss. Optionally, the absolute value of the difference between the environmental bias parameters of any two users is used as the distance between the two users, and then DBSCAN clustering without specifying the number of clusters is performed on all users based on the constructed distance to obtain several cluster sets, so as to determine the cluster set to which each user belongs.

[0049] In order to characterize the overall indoor heat loss degree of all users in the cluster set, the mean value of the environmental bias parameters of all users in the cluster set is taken as the cluster center value of the cluster set.

[0050] Step S103: constructing a first weight that is proportional to the number of users in the cluster set and a second weight that is proportional to the cluster center value of the cluster set.

[0051] It is easy to understand that the larger the cluster center value of the cluster set is, the greater the degree of indoor heat loss of the users in the cluster set during the heating process. Therefore, in order to ensure that the indoor temperature of the users can reach their expected temperature during the heating process, the amount of heat provided to the users in the cluster set must be larger than their original required temperature value. Therefore, in order to accurately quantify the degree of deviation from the original required temperature value, a second weight proportional to the cluster center value of the cluster set is constructed to correct the required temperature values ​​issued by subsequent users.

[0052] And, it is also easy to understand that the more users there are in a certain cluster set, the more heating the cluster set as a whole requires, so a first weight proportional to the number of users in the cluster set is constructed.

[0053] Optionally, in one embodiment, the method for obtaining the first weight and the second weight is:

[0054] Taking the ratio of the number of users in the current cluster set to the number of all users as the first weight of the current cluster set;

[0055] The second weight is:

[0056]

[0057] in, represents the second weight of the cluster set, represents the gain factor, represents the adjustment factor, represents the cluster center value of the cluster set, Indicates the natural constant As the base number, is an exponential function of an exponent.

[0058] Gain factor The value range is [0,0.1], and its function is to adjust the value of the second weight and adjust the coefficient is a positive number, which is used to set the cluster center value of the cluster set Before calculation, try to map to the range of independent variables near 0 where the change rate of the corresponding function value is large or the change trend is more obvious, so as to improve the overall adjustment sensitivity of the second weight. The value range of the second weight is (1, c+1). It can be understood that due to the heat loss in the user's room during the heating process, the demand mean corresponding to the cluster set needs to be increased to meet the actual temperature demand value of each user in the cluster set. The greater the degree of heat loss corresponding to the cluster set, that is, the greater the cluster center value, the greater the multiplier of the actual temperature demand value.

[0059] Step S104, determine the average demand temperature of all users in any cluster set within a preset future time period, multiply the average demand temperature with the first weight and the second weight to obtain a sub-demand value of the cluster set, add the sub-demand values ​​corresponding to all cluster sets to obtain a target demand value, and adjust the heating parameters on the primary side of the heating according to the target demand value feedback.

[0060] The overall purpose of the steps before this step is essentially to determine and quantify the degree of indoor temperature loss of different user groups during the heating process, and then obtain the compensation weights for compensating for the different degrees of indoor temperature loss between different user groups, so as to make targeted compensation adjustments for the required temperature values ​​of subsequent users, thereby ensuring heating for users without wasting heating resources.

[0061] Among them, the demand temperature value of subsequent users is generally sent by the user to the server through the client at the current moment, but the time corresponding to the demand temperature value sent by the user is generally not too far away from the current moment, generally in units of hours or up to one day. Considering that the heating system is relatively large, the parameter control interval on the primary heating side is generally much longer than the interval between the demand temperature value sent by the user and the current moment, so this embodiment preferably predicts the demand temperature value of each user within the preset future time length based on the historical demand temperature value of each user, so as to determine the average demand temperature of all users in the cluster set within the preset future time length, wherein the preset future time length corresponds to the heating strategy control interval length of the primary heating side, or corresponds to the heating parameter control interval length of the primary heating side.

[0062] Of course, it is easy to understand that determining the average temperature demanded by the user within the preset future duration in a predictive manner is only a preferred method, and its purpose is to match the determined duration corresponding to the average temperature demanded by the user, i.e., the preset future duration, with the parameter control interval on the primary heating side, so as to match the parameter control frequency on the primary heating side, and facilitate the implementation of heating on the primary heating side. When the preset future duration is set to the interval duration between the time corresponding to the user sending the required temperature value and the current time, then at this time, there is no need to predict the required temperature value and the required temperature value sent by the user can be directly obtained.

[0063] When determining the subsequent average temperature demand of the user in a predictive manner, optionally, the method for determining the average temperature demand of all users in any cluster set within a preset future time period is:

[0064] Performing time series prediction on the required temperature value of each user in the cluster set based on the historical required temperature value of the user, and determining the predicted required temperature value of the user in the preset future time period;

[0065] Calculate the mean of the predicted demand temperature values ​​of the current user in the cluster set at the preset future time, then average the mean of the demand temperature values ​​of each user in the cluster set at the preset future time, and determine the mean demand temperature value of all users in the current cluster set within the preset future time.

[0066] The above-mentioned time series prediction model can adopt existing RNN, LSTM and other models, and the predicted demand temperature value obtained by prediction can be a continuous value at a preset future time length, or a discrete point value distributed at a fixed sampling interval at a preset future time length. However, no matter which form, after obtaining the predicted demand temperature value of the user at the preset future time length, the average of the predicted demand temperature value of the user at the preset future time length can be obtained.

[0067] After determining the mean of the predicted required temperature values ​​of each user in the cluster set within the preset future time, the means corresponding to all users in the cluster set are averaged again to determine the mean required temperature values ​​of all users in the current cluster set within the preset future time.

[0068] The above-mentioned average demand temperature of all users in the current cluster set within the preset future time represents the average demand temperature of all users in the cluster set. The number of users in different cluster sets is different, and the degree of overall indoor temperature loss of users corresponding to different cluster sets is also different. Therefore, it is necessary to multiply the average demand temperature of all users in the current cluster set within the preset future time by the first weight and the second weight corresponding to the cluster set to make targeted corrections to the average demand temperature of all users in the current cluster set within the preset future time to obtain the sub-demand value of the cluster set.

[0069] This sub-demand value takes into account the number of users in the cluster set and the compensation for the indoor temperature loss of the users in the cluster set. After summing up the sub-demand values ​​corresponding to all cluster sets to obtain the target demand value, the heating on the primary side of the heating supply is adjusted according to the target demand value to ensure that the required temperature requirements of each user are met without wasting heating resources.

[0070] Among them, according to the target demand value, the heating parameters of the primary side of the heating supply can be adjusted by feedback through the existing primary side control model, which will not be described here. In actual use, if the user's real-time required temperature is received, the valve adjustment method corresponding to the user can be determined according to the real-time valve parameters, required temperature and target demand value, and the room temperature of the user can be adjusted intelligently in real time.

[0071] In summary, this embodiment can comprehensively characterize the user's indoor heating loss during heating only through specially obtained bias parameters without actually obtaining various complex parameters that affect the user's indoor heating temperature, such as the user's room location orientation, the user's apartment distribution, the user's surrounding heating environment and other parameters, and then classify the users according to the bias parameters, and compensate and correct the user demand temperature values ​​under different categories based on the size of the bias parameters, so that when determining the heating parameters of the primary heating network, it is more accurate and reasonable, avoiding the situation in which a large number of users are insufficiently heated due to the use of the average user demand for heating in the conventional method, and also avoiding the situation in which the maximum user demand for heating is used for heating in the conventional method, resulting in a waste of heating resources, thereby improving the efficiency of smart heating.

[0072] See also Figure 3 , is a structural diagram of a prediction model provided in Embodiment 2 of the present invention, wherein the prediction model includes a first prediction model and a second prediction model.

[0073] Among them, the first prediction model is used to predict the temperature supply value under theoretical conditions. Based on this purpose, the method for obtaining the first prediction model in this embodiment is:

[0074] The historical data is sampled at a set sampling interval, and the historical demand temperature value of any user at the i-th sampling moment, the historical ambient temperature value at the i-1-th sampling moment, the historical indoor temperature value at the i-1-th sampling moment, and the sampling interval duration in the historical data are used as training samples input into the first prediction model, and the theoretical temperature values ​​corresponding to the training samples input into the first prediction model are used as labels of the training samples input into the first prediction model. A first loss function about the difference between the output temperature value output by the first prediction model and the theoretical temperature value is constructed, and the first prediction model is trained according to the first loss function to obtain the pre-trained first prediction model.

[0075] The sampling interval can be set to match the interval length of the user's required temperature value, such as 1 hour, 2 hours, etc. After sampling the historical data at the set sampling interval, a large amount of historical sampling data information for each user can be obtained from the historical data, such as historical required temperature values, historical ambient temperature values, historical indoor temperature values, historical actual temperature values, etc.

[0076] The so-called theoretical conditions in which the first prediction model predicts the heating value under theoretical conditions refer to the theoretical situation or ideal situation in which the actual differences in users due to actual apartment types, room orientations, and the existence of other heating users around are not taken into account, and the heat exchange parameters between the user's indoor and outdoor environments are assumed to be fixed. In this theoretical situation, the temperature rise time between the previous and next moments is clearly defined as the sampling interval time, and then the indoor temperature value and ambient temperature value at the current moment (corresponding to the i-1 sampling moment) and the required temperature value at the next moment (corresponding to the i-th sampling moment) are added. The temperature rise time, that is, the theoretical heating value in the sampling interval time, can be calculated according to the theoretical conditions.

[0077] Since the first loss function is a function of the difference between the theoretical temperature supply value and the output temperature supply value output by the first prediction model, the first prediction model trained by the first loss function is a model for predicting the temperature supply value under theoretical conditions.

[0078] The second prediction model is used to predict the indoor temperature value under theoretical conditions. Based on this purpose, the method for obtaining the second prediction model in this embodiment is:

[0079] The historical indoor temperature value at the i-th sampling moment of any user in the historical data, the historical actual temperature value at the i-th sampling moment, the historical ambient temperature value at the i-th sampling moment and the sampling interval duration are used as training samples input into the second prediction model, and the theoretical indoor temperature value corresponding to the training sample input into the second prediction model is used as the label of the training sample input into the second prediction model, and a second loss function about the difference between the output indoor temperature value output by the first prediction model and the theoretical indoor temperature value is constructed, and the second prediction model is trained according to the second loss function to obtain the pre-trained second prediction model.

[0080] Similar to the training process of the first prediction model, the second prediction model acquires historical data using the same set sampling interval as the first prediction model in the process of acquiring historical data, and the so-called theoretical conditions in which the second prediction model predicts the indoor temperature value under theoretical conditions are the same as the so-called theoretical conditions in which the first prediction model predicts the heating value under theoretical conditions.

[0081] Analogous to the training of the first prediction model, in this theoretical case, after the temperature rise time between the previous and next moments is clearly defined as the sampling interval time, the indoor temperature value, ambient temperature value and actual heating value at the current moment (corresponding to the i-th sampling moment) can be added to obtain the theoretical indoor temperature value at the next moment (corresponding to the i+1-th sampling moment not specified in the above-mentioned second prediction model acquisition method) after the temperature rise time, that is, the sampling interval time.

[0082] Since the second loss function is a function of the difference between the theoretical indoor temperature value and the output indoor temperature value output by the second prediction model, the second prediction model trained by the second loss function is a model for predicting the indoor temperature value under theoretical conditions.

[0083] The reason why the prediction ability of the first and second prediction models under such theoretical conditions is specially constructed in this embodiment is to continue to input historical data into the first and second prediction models in the future, and then train the bias parameters input into the first and second prediction models by comparing the differences between the prediction outputs of the first and second prediction models and the corresponding historical actual data, so as to obtain environmental bias parameters that can characterize the difference between theoretical conditions and actual conditions after training. Each user can obtain an environmental bias parameter, which is the sum of various complex parameters that affect the user's indoor heating temperature.

[0084] For details, see Figure 4 , is a flow chart of a method for determining an environmental bias parameter corresponding to a user provided in Embodiment 2 of the present invention. The method for determining an environmental bias parameter corresponding to a user is implemented based on the first prediction model and the second prediction model pre-constructed as described above. Figure 2 As shown, the method for determining the environmental bias parameter may include the following steps:

[0085] Step S201, taking the sampling interval duration as the preset temperature rise duration, inputting the historical demand temperature value at the j-th sampling moment, the historical ambient temperature value at the j-1-th sampling moment, the historical indoor temperature value at the j-1-th sampling moment, the sampling interval duration and the bias parameter initialized to zero of any user in the historical data into the pre-trained first prediction model to obtain the predicted temperature supply value corresponding to the j-th sampling moment.

[0086] After taking the above-mentioned sampling interval as the preset temperature rise time, based on the historical ambient temperature value and the historical indoor temperature value of the user selected in the historical data at the j-1th sampling moment, and the historical demand temperature value at the jth sampling moment, the predicted temperature supply value at the jth sampling moment after the preset temperature rise time can be obtained by the pre-trained first prediction model under the condition that the bias parameter is initialized to 0.

[0087] Step S202, the predicted heating value corresponding to the j-th sampling moment, the historical indoor temperature value of any user at the j-th sampling moment in the historical data, the historical ambient temperature value at the j-th sampling moment, and the sampling interval duration are input into the pre-trained second prediction model to obtain the predicted indoor temperature value corresponding to the j+1-th sampling moment, and an iterative loss function is constructed about the difference between the predicted indoor temperature value corresponding to the j+1-th sampling moment and the historical indoor temperature value of any user at the j+1-th sampling moment. The bias parameter is iterated according to the iterative loss function until the iterative loss function converges, and the converged bias parameter is used as the environmental bias parameter corresponding to the user.

[0088] Continue to input the predicted heating value at the j-th sampling moment obtained in step S201 into the pre-trained second prediction model, and at the same time input the historical indoor temperature value and historical ambient temperature value of the same user at the j-th sampling moment in the historical data. Under the condition that the bias parameter is initialized to 0, the pre-trained second prediction model can obtain the predicted indoor temperature value at the j+1-th sampling moment after the preset temperature rise time.

[0089] As mentioned above, since the predicted indoor temperature value output by the first and second prediction models trained under theoretical conditions is a prediction of the indoor temperature value under ideal conditions, it is obviously not accurate. Then, the predicted indoor temperature value of the user at the j+1th sampling moment is subtracted from the historical indoor temperature value of the user at the j+1th sampling moment, and an iterative loss function about the obtained difference is constructed. The bias parameter input to the prediction model is iteratively trained with the iterative loss function. When the iterative loss function converges, it indicates that the predicted indoor temperature value output by the prediction model after the bias parameter correction is considered to be consistent with the actual historical indoor temperature value. Accordingly, the bias parameter obtained at this time can be considered to accurately realize the sum representation of various complex parameters that affect the user's indoor heating temperature, and the bias parameter obtained at this time is recorded as the environmental bias parameter corresponding to the user.

[0090] This embodiment utilizes the adaptability of the neural network and obtains, based on the training iteration of the neural network, an environmental bias parameter that can sum up the complex parameters that originally cannot be quantified or are difficult to quantify and affect the indoor heating temperature of each user, thereby providing an accurate basis for subsequent compensation and adjustment of the heating demands proposed by the users, thereby providing a basis for ultimately achieving an improvement in the heating efficiency.

[0091] Corresponding to the method of the above embodiment, Figure 5The structural block diagram of a smart heating device based on the Internet of Things provided by the third embodiment of the present invention is shown. The heating device is applied to a computer device, and the computer device is connected to a target database through a preset application program interface. When the target database is driven to execute a corresponding task, a corresponding task log is generated, and the task log can be collected through the API. For the convenience of explanation, only the part related to the embodiment of the present invention is shown.

[0092] See also Figure 5 The network control device based on the Internet of Things smart heating technology includes:

[0093] The user indoor heat loss characterization module 31 is used to input the historical demand temperature information, historical ambient temperature information, historical indoor temperature information, preset temperature rise time and the bias parameter initialized to zero of any user into the prediction model trained under theoretical conditions, and iterate the bias parameter according to the error between the predicted value output by the prediction model and the corresponding actual value in the historical data, and record the iterated bias parameter as the environmental bias parameter corresponding to the user;

[0094] A user classification module 32, used to cluster all users based on the environmental bias parameters of each user, and use the mean value of the environmental bias parameters of all users in any cluster set as the cluster center value of the cluster set;

[0095] A modified weight acquisition module 33, used to construct a first weight proportional to the number of users in the cluster set, and a second weight proportional to the cluster center value of the cluster set;

[0096] The primary side heat supply regulation module 34 is used to determine the average demand temperature of all users in any cluster set within a preset future time period, multiply the average demand temperature with the first weight and the second weight to obtain the sub-demand value of the cluster set, add the sub-demand values ​​corresponding to all cluster sets to obtain the target demand value, and adjust the heating parameters of the primary side of the heating according to the target demand value feedback.

[0097] Optionally, the above-mentioned network control device based on the Internet of Things smart heating technology also includes:

[0098] Predictive model building blocks, including:

[0099] A first prediction model acquisition unit is used to sample historical data at a set sampling interval, use the historical demand temperature value at the i-th sampling moment of any user in the historical data, the historical ambient temperature value at the i-1-th sampling moment, the historical indoor temperature value at the i-1-th sampling moment, and the sampling interval duration as training samples input to the first prediction model, use the theoretical temperature supply value corresponding to the training sample input to the first prediction model as the label of the training sample input to the first prediction model, construct a first loss function about the difference between the output temperature supply value output by the first prediction model and the theoretical temperature supply value, and train the first prediction model according to the first loss function to obtain the pre-trained first prediction model;

[0100] The second prediction model acquisition unit is used to use the historical indoor temperature value at the i-th sampling moment of any user in the historical data, the historical actual temperature value at the i-th sampling moment, the historical ambient temperature value at the i-th sampling moment and the sampling interval as the training samples input to the second prediction model, and use the theoretical indoor temperature value corresponding to the training sample input to the second prediction model as the label of the training sample input to the second prediction model, construct a second loss function about the difference between the output indoor temperature value output by the first prediction model and the theoretical indoor temperature value, and train the second prediction model according to the second loss function to obtain the pre-trained second prediction model.

[0101] Optionally, the user indoor heat loss characterization module 31 includes:

[0102] A temperature supply value prediction unit is used to use the sampling interval duration as the preset temperature rise duration, input the historical demand temperature value at the j-th sampling moment of any user in the historical data, the historical ambient temperature value at the j-1-th sampling moment, the historical indoor temperature value at the j-1-th sampling moment, the sampling interval duration and the bias parameter initialized to zero into the pre-trained first prediction model, and obtain the predicted temperature supply value corresponding to the j-th sampling moment;

[0103] An environmental bias parameter determination unit is used to input the predicted heating value corresponding to the j-th sampling moment, the historical indoor temperature value of any user at the j-th sampling moment in the historical data, the historical environmental temperature value at the j-th sampling moment, and the sampling interval duration into the pre-trained second prediction model to obtain the predicted indoor temperature value corresponding to the j+1-th sampling moment, construct an iterative loss function about the difference between the predicted indoor temperature value corresponding to the j+1-th sampling moment and the historical indoor temperature value of any user at the j+1-th sampling moment, iterate the bias parameter according to the iterative loss function until the iterative loss function converges, and use the converged bias parameter as the environmental bias parameter corresponding to the user.

[0104] Optionally, the prediction model building module further includes:

[0105] The parameter limiting unit is used to limit the heat exchange parameters between each user and the environment to be the same during the calculation process of the theoretical heating value and the theoretical indoor temperature value.

[0106] Optionally, the modified weight acquisition module 33 includes:

[0107] A first weight calculation unit, configured to use a ratio of the number of users in the current cluster set to the number of all users as a first weight of the current cluster set;

[0108] A second weight calculation unit is used to limit the second weight to:

[0109]

[0110] in, represents the second weight of the cluster set, represents the gain factor, represents the adjustment factor, represents the cluster center value of the cluster set, Indicates the natural constant As the base number, is an exponential function of an exponent.

[0111] Optionally, the user classification module 32 includes:

[0112] The user clustering unit is used to use the absolute value of the difference between the environmental bias parameters of any two users as the distance between the any two users, perform DBSAN clustering on all users based on the distance, and determine the cluster set to which each user belongs.

[0113] Optionally, the primary side heat supply regulating module 34 includes:

[0114] A user predicted demand determination unit, configured to perform time series prediction on the required temperature value of each user in the cluster set based on the historical required temperature value of the user, and determine the predicted required temperature value of the user in the preset future time period;

[0115] The cluster set predicted demand determination unit is used to calculate the mean of the predicted demand temperature values ​​of the current users in the cluster set within the preset future time, and then average the mean of the demand temperature values ​​of each user in the cluster set within the preset future time, to determine the mean demand temperature value of all users in the current cluster set within the preset future time.

[0116] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0117] Figure 6 This is a schematic diagram of the structure of the computer equipment in the networked control system based on the Internet of Things smart heating technology provided in the fourth embodiment of the present invention. Figure 6 As shown, the computer device of this embodiment includes: at least one processor ( Figure 6 Only one is shown in the figure), a memory, and a computer program stored in the memory and executable on at least one processor, wherein when the processor executes the computer program, the steps in any of the above-mentioned health prediction method embodiments are implemented.

[0118] The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that Figure 6 This is merely an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components than shown in the figure, or a combination of certain components, or different components.

[0119] The processor may be a CPU, or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0120] The memory includes a readable storage medium, an internal memory, etc., wherein the internal memory may be the memory of a computer device, and the internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The readable storage medium may be a hard disk of a computer device, and in other embodiments, it may also be an external storage device of the computer device, for example, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device. Further, the memory may also include both an internal storage unit of the computer device and an external storage device. The memory is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of a computer program, etc. The memory may also be used to temporarily store data that has been output or is to be output.

[0121] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the above-mentioned method embodiment, which will not be repeated here. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiment can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device capable of carrying computer program code, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0122] The present invention implements all or part of the processes in the above-mentioned embodiment method, and may also be completed through a computer program product. When the computer program product runs on a computer device, the computer device implements the steps in the above-mentioned method embodiment when executing.

[0123] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0124] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0125] In the embodiments provided by the present invention, it should be understood that the disclosed devices / computer equipment and methods can be implemented in other ways. For example, the device / computer equipment embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0126] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0127] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A networking control method based on the Internet of Things smart heating technology, characterized in that: The method comprises: The historical demand temperature information, historical ambient temperature information, historical indoor temperature information, preset temperature rise time and bias parameter initialized to zero of any user are input into the prediction model trained under theoretical conditions, and the bias parameter is iterated according to the error between the predicted value output by the prediction model and the corresponding actual value in the historical data, and the iterated bias parameter is recorded as the environmental bias parameter corresponding to the user; All users are clustered based on the environmental bias parameters of each user, and the mean value of the environmental bias parameters of all users in any cluster set is used as the cluster center value of the cluster set; Constructing a first weight proportional to the number of users in the cluster set and a second weight proportional to the cluster center value of the cluster set; Determine the average demand temperature of all users in any cluster set within a preset future time period, multiply the average demand temperature by the first weight and the second weight to obtain a sub-demand value of the cluster set, add the sub-demand values ​​corresponding to all cluster sets to obtain a target demand value, and adjust the heating parameters of the primary side of the heating according to the target demand value feedback; The prediction model includes a pre-trained first prediction model and a pre-trained second prediction model; The method for obtaining the pre-trained first prediction model is: Sampling historical data at a set sampling interval, using the historical demand temperature value at the i-th sampling moment of any user in the historical data, the historical ambient temperature value at the i-1-th sampling moment, the historical indoor temperature value at the i-1-th sampling moment, and the sampling interval duration as training samples input to the first prediction model, using the theoretical temperature supply value corresponding to the training sample input to the first prediction model as the label of the training sample input to the first prediction model, constructing a first loss function about the difference between the output temperature supply value output by the first prediction model and the theoretical temperature supply value, and training the first prediction model according to the first loss function to obtain the pre-trained first prediction model; The method for obtaining the pre-trained second prediction model is: Taking the historical indoor temperature value at the i-th sampling moment of any user in the historical data, the historical actual temperature value at the i-th sampling moment, the historical ambient temperature value at the i-th sampling moment, and the sampling interval duration as the training samples input into the second prediction model, taking the theoretical indoor temperature value corresponding to the training sample input into the second prediction model as the label of the training sample input into the second prediction model, constructing a second loss function about the difference between the output indoor temperature value output by the first prediction model and the theoretical indoor temperature value, and training the second prediction model according to the second loss function to obtain the pre-trained second prediction model; The method comprises inputting the historical required temperature information, the historical ambient temperature information, the historical indoor temperature information, the preset temperature rise time and the bias parameter initialized to zero of any user into the prediction model trained under theoretical conditions, iterating the bias parameter according to the error between the predicted value output by the prediction model and the corresponding actual value in the historical data, and recording the iterated bias parameter as the environmental bias parameter corresponding to the user, including: The sampling interval duration is used as the preset temperature rise duration, and the historical demand temperature value at the j-th sampling moment, the historical ambient temperature value at the j-1-th sampling moment, the historical indoor temperature value at the j-1-th sampling moment, the sampling interval duration and the bias parameter initialized to zero of any user in the historical data are input into the pre-trained first prediction model to obtain the predicted temperature supply value corresponding to the j-th sampling moment; The predicted heating value corresponding to the j-th sampling moment, the historical indoor temperature value of any user at the j-th sampling moment in the historical data, the historical ambient temperature value at the j-th sampling moment, and the sampling interval duration are input into the pre-trained second prediction model to obtain the predicted indoor temperature value corresponding to the j+1-th sampling moment, and an iterative loss function is constructed about the difference between the predicted indoor temperature value corresponding to the j+1-th sampling moment and the historical indoor temperature value of any user at the j+1-th sampling moment. The bias parameter is iterated according to the iterative loss function until the iterative loss function converges, and the converged bias parameter is used as the environmental bias parameter corresponding to the user.

2. The networking control method based on the Internet of Things smart heating technology according to claim 1 is characterized in that: In the calculation process of the theoretical heating value and the theoretical indoor temperature value, it is preset that the heat exchange parameters between each user and the environment are the same.

3. The networking control method based on the Internet of Things smart heating technology according to claim 1 is characterized in that: Taking the ratio of the number of users in the current cluster set to the number of all users as the first weight of the current cluster set; The second weight is: in, represents the second weight of the cluster set, represents the gain factor, represents the adjustment factor, represents the cluster center value of the cluster set, Indicates the natural constant As the base number, is an exponential function of an exponent.

4. The networking control method based on the Internet of Things smart heating technology according to claim 1 is characterized in that: The method for clustering all users based on the environmental bias parameters of each user is: The absolute value of the difference between the environmental bias parameters of any two users is used as the distance between the any two users, and DBSAN clustering is performed on all users based on the distance to determine the cluster set to which each user belongs.

5. The networking control method based on the Internet of Things smart heating technology according to claim 1 is characterized in that: The method for determining the average required temperature of all users in any cluster set within a preset future time period is: Performing time series prediction on the required temperature value of each user in the cluster set based on the historical required temperature value of the user, and determining the predicted required temperature value of the user in the preset future time period; Calculate the mean of the predicted demand temperature values ​​of the current user in the cluster set at the preset future time, then average the mean of the demand temperature values ​​of each user in the cluster set at the preset future time, and determine the mean demand temperature value of all users in the current cluster set within the preset future time.

6. A network control device based on the Internet of Things smart heating technology, applied to the network control method based on the Internet of Things smart heating technology according to any one of claims 1 to 5, characterized in that: include: A user indoor heat loss characterization module is used to input any user's historical demand temperature information, historical ambient temperature information, historical indoor temperature information, preset temperature rise time, and a bias parameter initialized to zero into a prediction model trained under theoretical conditions, and iterate the bias parameter according to the error between the predicted value output by the prediction model and the corresponding actual value in the historical data, and record the iterated bias parameter as the environmental bias parameter corresponding to the user; A user classification module is used to cluster all users based on the environmental bias parameters of each user, and take the mean value of the environmental bias parameters of all users in any cluster set as the cluster center value of the cluster set; A modified weight acquisition module, used to construct a first weight proportional to the number of users in the cluster set, and a second weight proportional to the cluster center value of the cluster set; The primary side heat supply regulation module is used to determine the average demand temperature of all users in any cluster set within a preset future time period, multiply the average demand temperature with the first weight and the second weight to obtain the sub-demand value of the cluster set, add the sub-demand values ​​corresponding to all cluster sets to obtain the target demand value, and adjust the heating parameters of the primary side of the heating according to the target demand value feedback.

7. A networked control system based on the Internet of Things smart heating technology, comprising a computer device and a computer-readable storage medium, for controlling the networked control device based on the Internet of Things smart heating technology according to claim 6, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the network control method based on the smart heating technology of the Internet of Things as described in any one of claims 1 to 5 is implemented. When the computer program is executed by the processor, the network control method based on the smart heating technology of the Internet of Things as described in any one of claims 1 to 5 is implemented.

Images