CN111797980A - Self-adaptive learning method for personalized floor heating use habits - Google Patents
Self-adaptive learning method for personalized floor heating use habits Download PDFInfo
- Publication number
- CN111797980A CN111797980A CN202010695337.4A CN202010695337A CN111797980A CN 111797980 A CN111797980 A CN 111797980A CN 202010695337 A CN202010695337 A CN 202010695337A CN 111797980 A CN111797980 A CN 111797980A
- Authority
- CN
- China
- Prior art keywords
- data
- floor heating
- user
- personalized
- operation parameters
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010438 heat treatment Methods 0.000 title claims abstract description 119
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013528 artificial neural network Methods 0.000 claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 31
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 230000007613 environmental effect Effects 0.000 claims description 40
- 238000013136 deep learning model Methods 0.000 claims description 25
- 230000015654 memory Effects 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 8
- 230000007787 long-term memory Effects 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000005265 energy consumption Methods 0.000 abstract description 4
- 125000004122 cyclic group Chemical group 0.000 abstract 1
- 230000002708 enhancing effect Effects 0.000 abstract 1
- 230000006403 short-term memory Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Abstract
The invention provides a self-adaptive learning method for personalized floor heating use habits, which comprises the steps of obtaining historical user use data in a preset time period, preprocessing the historical user use data to obtain user use data containing environment information, inputting the user use data containing the environment information into a preset neural network, training and learning by using the neural network, predicting the user use data containing the environment information to obtain the predicted personalized user use habits, setting floor heating operation parameters in the next working time, adjusting and producing new floor heating operation parameters by a user according to the floor heating operation parameters, repeating the steps for cyclic iteration habits, improving the accuracy of the personalized user use prediction, improving the intelligence, reducing the energy consumption and enhancing the user use experience.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a self-adaptive learning method for personalized floor heating use habits of users.
Background
In the current society, along with people's daily life water improves, ground warms up and has become one of indispensable life electrical apparatus, and the work of ground heating receives external environmental condition to and the different influence of every user custom, make people constantly adjust ground according to the impression of self and warm up the setting, intelligent degree is low, thereby has reduced user and has used the impression, and the corresponding energy consumption that warms up also increases thereupon.
Based on the current situation, a high-intelligence low-energy-consumption self-adaptive learning method which can be used according to the personalized floor heating use habit of the user and is combined with the external environment is needed to be provided so as to enhance the use experience of the user.
Disclosure of Invention
In order to solve the technical problem, an embodiment of the present invention provides a method for adaptively learning a personalized floor heating use habit, including:
acquiring recorded data obtained by measuring a preset measuring object; step two, acquiring historical user use data of a preset time period from the recorded data; preprocessing the historical user use data to obtain user use data containing environment information; inputting the user use data containing the environmental information into a preset neural network, and using the neural network to train and learn the user use data containing the environmental information to generate a personalized deep learning model; step five, predicting the user use data containing the environmental information by using the personalized deep learning model to obtain the predicted personalized user use habit; setting floor heating operation parameters in the next working time according to the use habits of the personalized users; step seven, in the using process of the next working time, a user adjusts and produces new floor heating operation parameters for the floor heating operation parameters, and the new floor heating operation parameters are used as new user historical using data to be added into the user historical using data of the preset time period again; and repeating the first step to the seventh step, preprocessing the historical use data of the user again to obtain new use data of the user containing the environmental information, inputting the new use data of the user into the neural network again for learning, and updating the personalized deep learning model until the user does not adjust the floor heating operation parameters in the seventh step any more so as to improve the accuracy of predicting the use habits of the personalized user.
In some embodiments of this embodiment, the step one of acquiring recorded data obtained by measuring a predetermined measurement object includes:
the method comprises the steps of continuously detecting a measuring object so as to obtain the setting state data of the floor heating and the environment parameter data of the environment where the measuring object is located, wherein the preset measuring object comprises indoor temperature, outdoor temperature, measuring time, floor heating setting temperature, floor heating working mode and floor heating switch condition.
In some embodiments of this embodiment, in step two, obtaining the historical usage data of the user for the predetermined period of time from the recorded data includes:
the method comprises the steps of obtaining floor heating setting state data of a preset time period and environment parameter data of the environment where the floor heating is located, wherein the starting time of the preset time period is not later than the last N hours in the recorded data, the floor heating setting state data comprise the set temperature, the working mode and the switching condition, and the environment parameter data comprise the indoor temperature, the outdoor temperature and the recorded time.
In some embodiments of this embodiment, in step three, the method for preprocessing the historical user usage data to obtain user usage data including environment information includes:
and performing set state data correction on the historical user use data so as to obtain user use data containing environment information, wherein the set state data correction comprises the step of expanding the floor heating set state data at the moment A forward by a time length B-A according to the change of the environment parameter data after the moment A starts and the moment B finishes, wherein the moment B is the moment when the indoor temperature data reaches the stability or the moment when the floor heating set state data changes again.
In some embodiments of this embodiment, in step four, the predetermined neural network is a simple long-term and short-term memory neural network, and the method for generating the personalized deep learning model through training and learning includes:
collecting historical user use data of the preset time period to form an original data set, wherein parameters in use of the original user comprise indoor temperature, outdoor temperature, time record, set temperature, working mode and switching condition;
executing set state data correction on the collected original data set to obtain the usage record data set containing the environmental information, and executing data serialization on the usage record data set containing the environmental information to generate a training data set; the set state data correction comprises the step of expanding the floor heating set state data at the moment A forward by a time length B-A according to the change of the environment parameter data after the moment A starts and the moment B finishes, wherein the moment B is the moment when the indoor temperature data is stable or the floor heating set state data is changed again; the data serialization comprises the step of converting the data in the usage record data set containing the environment information to obtain a training data set, wherein the training data set contains sample number, time step and attribute;
training the training data set by using a simple long-time memory neural network so as to generate the personalized deep learning model.
In some implementations of this embodiment, the method of generating a personalized machine learning model further includes:
before model training, presetting the number of hidden layers, the number of nodes in the hidden layers, a hidden layer activation function and a learning rate of the simple long-time and short-time memory neural network, and randomly initializing parameters of the simple long-time and short-time memory neural network, wherein the hidden layer activation function is a tanh function;
in the model training process, forward propagation is executed, the propagation loss is calculated by using a Mean Absolute Error (MAE) function, the learning rate is adjusted by using an Adam optimization algorithm, and the weight and the bias of the model are adjusted by using an error back propagation algorithm.
In some embodiments of this embodiment, in step five, the method for predicting the user usage data including the environmental information by using the personalized deep learning model to obtain the predicted personalized user usage habit includes:
collecting the user use data containing the environmental information in a preset time length to form a test data set, carrying out data serialization on the user use data containing the environmental information, and inputting the data into the personalized deep learning model for prediction to obtain the use habits of the personalized user; the data serialization includes transforming data in a test data set, the test data set including a sample number, a time step, and an attribute; the personalized user use habits comprise the set temperature, the working mode, the switch condition and the set time of the floor heating.
In some embodiments of this embodiment, step six, the method for setting floor heating operation parameters in the next working time based on the usage habits of the personalized users includes:
and according to the personalized user habits, when the next working time reaches the set time, the floor heating operation parameters are modified to be the same as the set temperature, the working mode and the switch condition, wherein the floor heating operation parameters comprise the set temperature, the working mode and the switch condition.
In some embodiments of this embodiment, in the seventh step, in the next working time, the user adjusts and produces new floor heating operation parameters for the floor heating operation parameters, and the new floor heating operation parameters are added to the recorded data as new recorded data, including:
if the floor heating operation parameters are found to be not in line with expectations at the moment in the using process of a user, the floor heating operation parameters can be manually adjusted to generate new floor heating operation parameters, wherein the new floor heating operation parameters comprise set temperature, working modes and switch conditions; and generating new floor heating setting state data according to the new floor heating operation parameters, acquiring the environmental parameter data of the environment where the floor heating is located at the moment, and combining the environmental parameter data with the new floor heating setting state data to generate new recorded data.
In some embodiments of the present embodiment, the steps one to seven are repeated until the user does not adjust the floor heating operation parameter any more in the step seven.
The embodiment of the invention adopts at least one technical scheme which can achieve the following beneficial effects:
acquiring recorded data obtained by measuring a preset measuring object; step two, acquiring historical user use data of a preset time period from the recorded data; preprocessing the historical user use data to obtain user use data containing environment information; inputting the user use data containing the environmental information into a preset neural network, and using the neural network to train and learn the user use data containing the environmental information to generate a personalized deep learning model; step five, predicting the user use data containing the environmental information by using the personalized deep learning model to obtain the predicted personalized user use habit; setting floor heating operation parameters in the next working time according to the use habits of the personalized users; step seven, in the using process of the next working time, a user adjusts and produces new floor heating operation parameters for the floor heating operation parameters, and the new floor heating operation parameters are used as new user historical using data to be added into the user historical using data of the preset time period again; repeating the first step to the seventh step, preprocessing the historical user use data again to obtain new user use data containing environment information, inputting the new user use data into the neural network again for learning, and updating the personalized deep learning model until the user does not adjust the floor heating operation parameters in the seventh step; the using habits of the personalized users are accurately predicted and adaptively adjusted, the parameters are scientific and accurate and are more intelligent, the energy consumption is reduced, and the using experience of the users is enhanced.
Drawings
Fig. 1 is a schematic flow chart of an adaptive learning method for personalized floor heating use habits according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a simple long-term and short-term memory neural network according to an embodiment of the present invention.
Detailed Description
In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present specification.
Fig. 1 is a schematic diagram of an adaptive learning method for an individualized floor heating use habit provided by an embodiment of the present invention, and the method may specifically include the following steps:
acquiring recorded data obtained by measuring a preset measuring object; step two, acquiring historical user use data of a preset time period from the recorded data; preprocessing the historical user use data to obtain user use data containing environment information; inputting the user use data containing the environmental information into a preset neural network, and using the neural network to train and learn the user use data containing the environmental information to generate a personalized deep learning model; step five, predicting the user use data containing the environmental information by using the personalized deep learning model to obtain the predicted personalized user use habit; setting floor heating operation parameters in the next working time according to the use habits of the personalized users; step seven, in the using process of the next working time, a user adjusts and produces new floor heating operation parameters for the floor heating operation parameters, and the new floor heating operation parameters are used as new user historical using data to be added into the user historical using data of the preset time period again; and repeating the first step to the seventh step, preprocessing the historical use data of the user again to obtain new use data of the user containing the environmental information, inputting the new use data of the user into the neural network again for learning, and updating the personalized deep learning model until the user does not adjust the floor heating operation parameters in the seventh step any more so as to improve the accuracy of predicting the use habits of the personalized user.
In some embodiments of this embodiment, the step one of acquiring recorded data obtained by measuring a predetermined measurement object includes:
the measurement object is continuously detected, so that the setting state data of the floor heating and the environmental parameter data of the environment where the floor heating is located are obtained.
Further, in some embodiments, the predetermined measurement objects include an indoor temperature, an outdoor temperature, a measurement time, a floor heating setting temperature, a floor heating operation mode, and a floor heating switch condition.
In some embodiments of this embodiment, in step two, obtaining the historical usage data of the user for the predetermined period of time from the recorded data includes:
the method comprises the steps of obtaining floor heating setting state data of a preset time period and environment parameter data of the environment where the floor heating is located, wherein the starting time of the preset time period is not later than the last N hours in the recorded data, the floor heating setting state data comprise the set temperature, the working mode and the switching condition, and the environment parameter data comprise the indoor temperature, the outdoor temperature and the recorded time.
Further, in some embodiments, setting the execution frequency of the example, for example, to execute once every 24 hours, requires N ≧ 24. In practical application, a user can acquire weather data of the location of the user recorded on the internet through networking so as to acquire environment parameter data, can input the environment parameter data through a terminal (such as a mobile phone, a computer and the like), and can also realize acquisition of the environment parameter data by integrating a temperature sensor on a floor heating device.
In some embodiments of this embodiment, in step three, the method for preprocessing the historical user usage data to obtain user usage data including environment information includes:
and performing set state data correction on the historical user use data so as to obtain user use data containing environment information, wherein the set state data correction comprises the step of expanding the floor heating set state data at the moment A forward by a time length B-A according to the change of the environment parameter data after the moment A starts and the moment B finishes, wherein the moment B is the moment when the indoor temperature data reaches the stability or the moment when the floor heating set state data changes again.
Further, in some embodiments, due to the change of environmental parameter data, the limitation of floor heating temperature setting, the temperature control algorithm of the floor heating itself, and the like, the indoor temperature cannot keep a stable value, but fluctuates up and down within a certain temperature interval. In this case, if no change in floor heating setting state data occurs in the process, it can be considered that the indoor temperature has reached a stable temperature.
In some embodiments of this embodiment, in step four, the predetermined neural network is a simple long-term and short-term memory neural network, and the method for generating the personalized deep learning model through training and learning includes:
collecting historical user use data of the preset time period to form an original data set, wherein parameters in use of the original user comprise indoor temperature, outdoor temperature, time record, set temperature, working mode and switching condition;
executing set state data correction on the collected original data set to obtain the usage record data set containing the environmental information, and executing data serialization on the usage record data set containing the environmental information to generate a training data set; the set state data correction comprises the step of expanding the floor heating set state data at the moment A forward by a time length B-A according to the change of the environment parameter data after the moment A starts and the moment B finishes, wherein the moment B is the moment when the indoor temperature data is stable or the floor heating set state data is changed again; the data serialization comprises the step of converting the data in the usage record data set containing the environment information to obtain a training data set, wherein the training data set contains sample number, time step and attribute;
training the training data set by using a simple long-time memory neural network so as to generate the personalized deep learning model.
In some implementations of this embodiment, the method of generating a personalized machine learning model further includes:
before model training, presetting the number of hidden layers, the number of nodes in the hidden layers, a hidden layer activation function and a learning rate of the simple long-time and short-time memory neural network, and randomly initializing parameters of the simple long-time and short-time memory neural network, wherein the hidden layer activation function is a tanh function;
in the model training process, forward propagation is executed, the propagation loss is calculated by using a Mean Absolute Error (MAE) function, the learning rate is adjusted by using an Adam optimization algorithm, and the weight and the bias of the model are adjusted by using an error back propagation algorithm.
Further, in the fourth step, the training data set is trained by using the simple long-short term memory neural network, so as to generate the simple long-short term memory neural network model. The following describes the simple long-term and short-term memory neural network with reference to fig. 2, which is a schematic structural diagram of the simple long-term and short-term memory neural network provided in an embodiment of the present invention, and the diagram may include the following contents:
the simple long-short time memory neural network consists of an input layer (InputLayer), a long-short time memory Layer (LSTM) and three full connection layers (Dense), which are named LSTM _13_ input, LSTM _13, Dense _37, Dense _38 and Dense _39 respectively. The first value of the input dimension (input) and the output dimension (output) of each layer of neural network in the figure represents the number of samples selected by the layer network in one training, and the value shown as None in the figure represents that the samples selected by the layer network in the current training are not set when the training is not started, so that the scale of the neural network can be more easily modified to adapt to different operating environments and hardware configurations.
Further, in some embodiments, the predicted object is next day user usage habits and the training data set time step is 10 minutes. According to calculation, 144 pieces of training data are needed for predicting the use habit of the user with the length of 24 hours, so that the sample input dimension of the input layer is (144, 1). Since the input layer does not perform any arithmetic operation, the output dimension of the input layer is also (144, 1), and the input dimension of the LSTM layer that receives the output data of the input layer is also (144, 1). And setting the output dimension of the LSTM layer to be 50 according to the preset number of LSTM layer hiding layers. And setting the output dimensions of the three fully-connected layers as 128, 64 and 1 respectively according to the fact that the dimensions of the fully-connected layers are reduced layer by layer to extract features and predict the use habits of users to belong to a regression problem. Thereby obtaining a simple long-term and short-term memory neural network structure as shown in fig. 2.
In some embodiments of this embodiment, in step five, the method for predicting the user usage data including the environmental information by using the personalized deep learning model to obtain the predicted personalized user usage habit includes:
collecting the user use data containing the environmental information in a preset time length to form a test data set, carrying out data serialization on the user use data containing the environmental information, and inputting the data into the personalized deep learning model for prediction to obtain the use habits of the personalized user; the data serialization includes transforming data in a test data set, the test data set including a sample number, a time step, and an attribute; the personalized user use habits comprise the set temperature, the working mode, the switch condition and the set time of the floor heating.
Further, in some embodiments, the execution frequency of the embodiment is set, for example, once every 24 hours, and the time length is required to be not shorter than 24 hours.
In some embodiments of this embodiment, step six, the method for setting floor heating operation parameters in the next working time based on the usage habits of the personalized users includes:
and according to the personalized user habits, when the next working time reaches the set time, the floor heating operation parameters are modified to be the same as the set temperature, the working mode and the switch condition, wherein the floor heating operation parameters comprise the set temperature, the working mode and the switch condition.
In some embodiments of this embodiment, in the seventh step, in the next working time, the user adjusts and produces new floor heating operation parameters for the floor heating operation parameters, and the new floor heating operation parameters are added to the recorded data as new recorded data, including:
if the floor heating operation parameters are found to be not in line with expectations at the moment in the using process of a user, the floor heating operation parameters can be manually adjusted to generate new floor heating operation parameters, wherein the new floor heating operation parameters comprise set temperature, working modes and switch conditions; and generating new floor heating setting state data according to the new floor heating operation parameters, acquiring the environmental parameter data of the environment where the floor heating is located at the moment, and combining the environmental parameter data with the new floor heating setting state data to generate new recorded data.
In some embodiments of the present embodiment, the steps one to seven are repeated until the user does not adjust the floor heating operation parameter any more in the step seven.
The described embodiments are only some embodiments of the invention, not all embodiments. Embodiments based on the principles and applications of this invention, or all other embodiments obtained without inventive faculty, are intended to be included within the scope of protection of the invention.
Claims (10)
1. A self-adaptive learning method for personalized floor heating use habits comprises the following steps:
acquiring recorded data obtained by measuring a preset measuring object; step two, acquiring historical user use data of a preset time period from the recorded data; preprocessing the historical user use data to obtain user use data containing environment information; inputting the user use data containing the environmental information into a preset neural network, and using the neural network to train and learn the user use data containing the environmental information to generate a personalized deep learning model; step five, predicting the user use data containing the environmental information by using the personalized deep learning model to obtain the predicted personalized user use habit; setting floor heating operation parameters in the next working time according to the use habits of the personalized users; step seven, in the using process of the next working time, a user adjusts and produces new floor heating operation parameters for the floor heating operation parameters, and the new floor heating operation parameters are used as new user historical using data to be added into the user historical using data of the preset time period again; and repeating the first step to the seventh step, preprocessing the historical use data of the user again to obtain new use data of the user containing the environmental information, inputting the new use data of the user into the neural network again for learning, and updating the personalized deep learning model until the user does not adjust the floor heating operation parameters in the seventh step any more so as to improve the accuracy of predicting the use habits of the personalized user.
2. The method according to claim 1, wherein the step one of acquiring the record data measured by the predetermined measurement object comprises:
the method comprises the steps of continuously detecting a measuring object so as to obtain the setting state data of the floor heating and the environment parameter data of the environment where the measuring object is located, wherein the preset measuring object comprises indoor temperature, outdoor temperature, measuring time, floor heating setting temperature, floor heating working mode and floor heating switch condition.
3. The method of claim 1, wherein step two, obtaining historical usage data of the user for a predetermined period of time from the log data, comprises:
the method comprises the steps of obtaining floor heating setting state data of a preset time period and environment parameter data of the environment where the floor heating is located, wherein the starting time of the preset time period is not later than the last N hours in the recorded data, the floor heating setting state data comprise the set temperature, the working mode and the switching condition, and the environment parameter data comprise the indoor temperature, the outdoor temperature and the recorded time.
4. The method of claim 1, wherein step three, the method for preprocessing the user historical usage data to obtain user usage data containing environmental information comprises:
and performing set state data correction on the historical user use data so as to obtain user use data containing environment information, wherein the set state data correction comprises the step of expanding the floor heating set state data at the moment A forward by a time length B-A according to the change of the environment parameter data after the moment A starts and the moment B finishes, wherein the moment B is the moment when the indoor temperature data reaches the stability or the moment when the floor heating set state data changes again.
5. The method according to claim 1, wherein, in step four, the predetermined neural network is a simple long-term memory neural network, and the method for generating the personalized deep learning model through training learning comprises the following steps:
collecting historical user use data of the preset time period to form an original data set, wherein parameters in use of the original user comprise indoor temperature, outdoor temperature, time record, set temperature, working mode and switching condition;
executing set state data correction on the collected original data set to obtain the usage record data set containing the environmental information, and executing data serialization on the usage record data set containing the environmental information to generate a training data set; the set state data correction comprises the step of expanding the floor heating set state data at the moment A forward by a time length B-A according to the change of the environment parameter data after the moment A starts and the moment B finishes, wherein the moment B is the moment when the indoor temperature data is stable or the floor heating set state data is changed again; the data serialization comprises the step of converting the data in the usage record data set containing the environment information to obtain a training data set, wherein the training data set contains sample number, time step and attribute;
training the training data set by using a simple long-time memory neural network so as to generate the personalized deep learning model.
6. The method of claim 5, wherein the method of generating the personalized deep learning model further comprises:
before model training, presetting the number of hidden layers, the number of nodes in the hidden layers, a hidden layer activation function and a learning rate of the simple long-time and short-time memory neural network, and randomly initializing parameters of the simple long-time and short-time memory neural network, wherein the hidden layer activation function is a tanh function;
in the model training process, forward propagation is executed, the propagation loss is calculated by using a Mean Absolute Error (MAE) function, the learning rate is adjusted by using an Adam optimization algorithm, and the weight and the bias of the model are adjusted by using an error back propagation algorithm.
7. The method according to claim 1, wherein step five, the method for predicting the user usage data containing the environmental information by using the personalized deep learning model to obtain the predicted personalized user usage habit comprises:
collecting the user use data containing the environmental information in a preset time length to form a test data set, carrying out data serialization on the user use data containing the environmental information, and inputting the data into the personalized deep learning model for prediction to obtain the use habits of the personalized user; the data serialization includes transforming data in a test data set, the test data set including a sample number, a time step, and an attribute; the personalized user use habits comprise the set temperature, the working mode, the switch condition and the set time of the floor heating.
8. The method according to claim 1, characterized in that step six, based on the usage habits of the personalized users, the method for setting floor heating operation parameters in the next working time comprises the following steps:
and according to the personalized user habits, when the next working time reaches the set time, the floor heating operation parameters are modified to be the same as the set temperature, the working mode and the switch condition, wherein the floor heating operation parameters comprise the set temperature, the working mode and the switch condition.
9. The method of claim 1, wherein step seven, in the next working time using process, a user adjusts and produces new floor heating operation parameters for the floor heating operation parameters, and the new floor heating operation parameters are added into the recorded data as new recorded data, and the method comprises the following steps:
if the floor heating operation parameters are found to be not in line with expectations at the moment in the using process of a user, the floor heating operation parameters can be manually adjusted to generate new floor heating operation parameters, wherein the new floor heating operation parameters comprise set temperature, working modes and switch conditions; and generating new floor heating setting state data according to the new floor heating operation parameters, acquiring the environmental parameter data of the environment where the floor heating is located at the moment, and combining the environmental parameter data with the new floor heating setting state data to generate new recorded data.
10. The method of claim 1, wherein steps one through seven are repeated until the user no longer adjusts floor heating operating parameters in step seven.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010695337.4A CN111797980A (en) | 2020-07-20 | 2020-07-20 | Self-adaptive learning method for personalized floor heating use habits |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010695337.4A CN111797980A (en) | 2020-07-20 | 2020-07-20 | Self-adaptive learning method for personalized floor heating use habits |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111797980A true CN111797980A (en) | 2020-10-20 |
Family
ID=72807798
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010695337.4A Pending CN111797980A (en) | 2020-07-20 | 2020-07-20 | Self-adaptive learning method for personalized floor heating use habits |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111797980A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112255928A (en) * | 2020-10-30 | 2021-01-22 | 北京金山云网络技术有限公司 | Smart home control method, device and system and electronic equipment |
| CN113343893A (en) * | 2021-06-24 | 2021-09-03 | 武汉虹信技术服务有限责任公司 | Equipment face recognition control method based on personnel passing rule and application thereof |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101251873B1 (en) * | 2012-12-20 | 2013-04-10 | 동국대학교 산학협력단 | Method for controlling building air conditioning system by using intelligent control algorithm |
| CN108694484A (en) * | 2018-08-30 | 2018-10-23 | 广东工业大学 | A kind of photovoltaic power generation power prediction method |
| CN109471370A (en) * | 2018-11-23 | 2019-03-15 | 珠海格力电器股份有限公司 | A kind of behavior prediction and control method based on exhaust fan operation data, system |
| CN109491253A (en) * | 2017-09-11 | 2019-03-19 | 安徽师范大学 | A kind of on-line study type individualized intelligent house system and its control method |
| CN109959122A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of Energy consumption forecast for air conditioning method based on shot and long term memory Recognition with Recurrent Neural Network |
| CN109961177A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of general water cooled central air conditioner energy consumption prediction technique based on shot and long term memory Recognition with Recurrent Neural Network |
| CN110059801A (en) * | 2019-03-20 | 2019-07-26 | 青岛海尔空调器有限总公司 | Energy efficiency of air conditioner control method neural network based |
| CN110456647A (en) * | 2019-07-02 | 2019-11-15 | 珠海格力电器股份有限公司 | A kind of intelligent home furnishing control method and intelligent home control device |
| CN110578994A (en) * | 2018-06-11 | 2019-12-17 | 珠海格力电器股份有限公司 | operation method and device |
| CN110686377A (en) * | 2019-09-20 | 2020-01-14 | 珠海格力电器股份有限公司 | Control method for air conditioner temperature self-adaptive adjustment, computer readable storage medium and air conditioner |
| CN110836525A (en) * | 2019-11-19 | 2020-02-25 | 珠海格力电器股份有限公司 | Self-adaptive adjusting method and device for air conditioner running state |
| CN110864414A (en) * | 2019-10-30 | 2020-03-06 | 郑州电力高等专科学校 | Air conditioner power utilization load intelligent control scheduling method based on big data analysis |
| CN111365828A (en) * | 2020-03-06 | 2020-07-03 | 上海外高桥万国数据科技发展有限公司 | Model prediction control method for realizing energy-saving temperature control of data center by combining machine learning |
-
2020
- 2020-07-20 CN CN202010695337.4A patent/CN111797980A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101251873B1 (en) * | 2012-12-20 | 2013-04-10 | 동국대학교 산학협력단 | Method for controlling building air conditioning system by using intelligent control algorithm |
| CN109491253A (en) * | 2017-09-11 | 2019-03-19 | 安徽师范大学 | A kind of on-line study type individualized intelligent house system and its control method |
| CN110578994A (en) * | 2018-06-11 | 2019-12-17 | 珠海格力电器股份有限公司 | operation method and device |
| CN108694484A (en) * | 2018-08-30 | 2018-10-23 | 广东工业大学 | A kind of photovoltaic power generation power prediction method |
| CN109471370A (en) * | 2018-11-23 | 2019-03-15 | 珠海格力电器股份有限公司 | A kind of behavior prediction and control method based on exhaust fan operation data, system |
| CN109959122A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of Energy consumption forecast for air conditioning method based on shot and long term memory Recognition with Recurrent Neural Network |
| CN109961177A (en) * | 2019-03-11 | 2019-07-02 | 浙江工业大学 | A kind of general water cooled central air conditioner energy consumption prediction technique based on shot and long term memory Recognition with Recurrent Neural Network |
| CN110059801A (en) * | 2019-03-20 | 2019-07-26 | 青岛海尔空调器有限总公司 | Energy efficiency of air conditioner control method neural network based |
| CN110456647A (en) * | 2019-07-02 | 2019-11-15 | 珠海格力电器股份有限公司 | A kind of intelligent home furnishing control method and intelligent home control device |
| CN110686377A (en) * | 2019-09-20 | 2020-01-14 | 珠海格力电器股份有限公司 | Control method for air conditioner temperature self-adaptive adjustment, computer readable storage medium and air conditioner |
| CN110864414A (en) * | 2019-10-30 | 2020-03-06 | 郑州电力高等专科学校 | Air conditioner power utilization load intelligent control scheduling method based on big data analysis |
| CN110836525A (en) * | 2019-11-19 | 2020-02-25 | 珠海格力电器股份有限公司 | Self-adaptive adjusting method and device for air conditioner running state |
| CN111365828A (en) * | 2020-03-06 | 2020-07-03 | 上海外高桥万国数据科技发展有限公司 | Model prediction control method for realizing energy-saving temperature control of data center by combining machine learning |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112255928A (en) * | 2020-10-30 | 2021-01-22 | 北京金山云网络技术有限公司 | Smart home control method, device and system and electronic equipment |
| CN113343893A (en) * | 2021-06-24 | 2021-09-03 | 武汉虹信技术服务有限责任公司 | Equipment face recognition control method based on personnel passing rule and application thereof |
| CN113343893B (en) * | 2021-06-24 | 2022-08-05 | 武汉虹信技术服务有限责任公司 | Equipment face recognition control method based on personnel passing rule and application thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111260030B (en) | A-TCN-based power load prediction method and device, computer equipment and storage medium | |
| JP5888640B2 (en) | Photovoltaic power generation prediction apparatus, solar power generation prediction method, and solar power generation prediction program | |
| US10565955B2 (en) | Display status adjustment method, display status adjustment device and display device | |
| Liu et al. | Forecast of water demand in Weinan City in China using WDF-ANN model | |
| CN110267292B (en) | Cellular network flow prediction method based on three-dimensional convolutional neural network | |
| CN108876021B (en) | Medium-and-long-term runoff forecasting method and system | |
| CN111797980A (en) | Self-adaptive learning method for personalized floor heating use habits | |
| CN110007613B (en) | Warming prediction method and system for heat storage type electric heater and storage medium | |
| CN102946253B (en) | Data sampling method and system and the application process in parameter identification thereof and system | |
| CN112532746B (en) | Cloud edge cooperative sensing method and system | |
| Wichaipanich et al. | A comparison of neural network-based predictions of foF2 with the IRI-2012 model at conjugate points in Southeast Asia | |
| CN116321620B (en) | Intelligent lighting switch control system and method thereof | |
| CN114330935B (en) | New energy power prediction method and system based on multiple combination strategies integrated learning | |
| CN110836525A (en) | Self-adaptive adjusting method and device for air conditioner running state | |
| CN116722545B (en) | Photovoltaic power generation prediction method based on multi-source data and related equipment | |
| CN107730045B (en) | Baseline load thermal inertia correction method based on discrete inertia dynamics system | |
| CN109447843B (en) | Photovoltaic output prediction method and device | |
| CN115860232A (en) | Steam load prediction method, system, electronic device and medium | |
| CN110009132A (en) | A kind of short-term electric load fining prediction technique based on LSTM deep neural network | |
| CN110322063B (en) | Power consumption simulation prediction method and storage medium | |
| CN111007721B (en) | Method for dynamically selecting sampling period in linear steady system | |
| Tiwari et al. | Phase space structure, attractor dimension, Lyapunov exponent and nonlinear prediction from Earth’s atmospheric angular momentum time series | |
| Rioual et al. | Reinforcement learning strategies for energy management in low power iot | |
| CN110806903A (en) | Configuration parameter determining method and device of electric cooker | |
| Li et al. | A Q-learning-based solar energy prediction algorithm with energy data association |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20201020 |