Building internal temperature prediction method applied to new energy base planning
The prediction of the internal temperature of the new energy base building through generalized regression neural networks has solved the problem of inaccurate internal temperature prediction in the existing technology, improved the prediction accuracy, and optimized the energy utilization rate.
Patent Information
- Application Number
- CN202411377918.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2025-02-11
AI Technical Summary
The existing technology is difficult to accurately predict the internal temperature of the building in the planning of new energy bases, especially when considering the planning of multi-type buildings and large-scale distributed energy equipment, there is a lack of effective research on the response characteristics of the building.
The generalized regression neural network method is used to obtain temperature data and external factor data at multiple different points in the building (such as light intensity, external temperature and time), and then a pre-trained generalized regression neural network is constructed to predict the internal temperature of the building.
Improve the accuracy of temperature prediction in the building, ensure the basic requirements for plant growth in the building, provide data support for flexibly adjusting building loads based on the indoor temperature of the building, and provide more accurate for the large-scale distributed energy equipment planning of new energy bases. data optimizes energy utilization.
Smart Images
Figure CN119416932A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of new energy base planning, and in particular to a method for predicting the internal temperature of a building applied to new energy base planning. Background Art
[0002] The Northwest region is rich in scenic and solar resources. In recent years, the rise of various parks has significantly increased the demand for various energy sources in various regions. Such a huge, complex and diverse energy demand has brought great challenges to the power dispatching of new energy bases and also led to some security issues. The desert Gobi desert area in the northwest is remote and the local power grid structure is weak. Further disorderly planning on this fragile power grid will lead to escalating security threats. In the existing technology, the research on the planning of new energy bases mainly focuses on how to reasonably dispatch various types of renewable energy power generation and improve the comprehensive utilization efficiency of energy. On the premise of ensuring the reliability of power supply, there is little research on planning large-scale distributed energy equipment and taking the response characteristics of buildings into account in the planning process.
[0003] The response characteristics of the building include changes in the internal temperature of the building due to external factors under certain energy supply conditions, including seasons, light intensity, external temperature, etc. Therefore, accurately predicting the internal temperature of buildings, especially the internal temperature of buildings in new energy bases containing multiple types of buildings (such as buildings, cold storage or residential buildings), plays a very important role in the rational planning of large-scale distributed energy equipment.
[0004] Therefore, how to accurately predict the internal temperature of buildings in new energy bases is an urgent problem to be solved. Summary of the invention
[0005] The present application provides a method for predicting the internal temperature of a building for use in the planning of a new energy base. The method predicts the indoor temperature of a building in a new energy base under different light intensity, external temperature and time conditions through a generalized regression neural network, thereby improving the accuracy of the prediction of the indoor temperature of the building and providing more accurate data for the rational planning of large-scale distributed energy equipment.
[0006] In order to achieve the above object, the present invention adopts the following technical scheme:
[0007] In a first aspect, the present invention provides a method for predicting internal temperature of a building for use in new energy base planning, the method comprising:
[0008] Acquire temperature data, where the temperature data is temperature values at multiple different points in the building;
[0009] Acquire external factor data corresponding to each temperature in the temperature data, and use the temperature data and the corresponding external factor data as training samples, wherein the external factor data includes light intensity, external temperature and time;
[0010] Inputting the training samples into a generalized regression neural network, determining an optimal smoothing factor using a leave-one-out method, obtaining a probability density function based on the optimal smoothing factor, and constructing a pre-trained generalized regression neural network based on the probability density function;
[0011] The external factor data is input into the pre-trained generalized regression neural network to obtain the predicted temperature corresponding to different external factor data.
[0012] In a preferred example of the present application, it can be further configured that the probability density function is obtained based on the optimal smoothing factor, including:
[0013] Define the probability density function of the independent variable x and the dependent variable y as f(x,y), and the conditional mean of the probability density function under the given independent variable X is The expression is:
[0014]
[0015] Where dy represents the integral of y, and E[y|X] represents the expected value of y when X takes a preset value;
[0016] The estimated value of the probability density Substituting f(X,y) in the above formula, simplifying the integral operation to obtain the conditional mean The expression is:
[0017]
[0018] Then, we get the estimated value The expression is:
[0019]
[0020] Where n is the number of training samples, i is the serial number of the training sample, p is the dimension of the independent variable, the independent variable is the external factor data, X i , Y i are sample observations of random variables x and y, and δ is the smoothing factor.
[0021] In a preferred example of the present application, it can be further configured that the generalized regression neural network includes an input layer, a pattern layer, a summation layer and an output layer, and the dimension of the input vector of the input layer is 3.
[0022] In a preferred example of the present application, it can be further set that the number of neurons in the pattern layer is equal to the number of neurons in the input layer, and the transfer function followed by the information transmission between the neurons in the pattern layer and the neurons in the input layer is:
[0023]
[0024] Among them, X is the input variable, that is, the external factor data, X i is the learning sample of the i-th neuron.
[0025] In a preferred example of the present application, it can be further configured that the summation layer uses two different types of neurons to perform weighted summation on the data output by the neurons in the pattern layer, wherein one type of neuron in the summation layer uses a direct summation method to sum the data output by the neurons in the pattern layer:
[0026]
[0027] Another neuron in the summation layer performs weighted summation on the data output by the neurons in the pattern layer using a weighted summation method:
[0028]
[0029] And S N and S D Output to the output layer.
[0030] In a preferred example of the present application, it can be further configured that the output layer is based on S N and S D Get the estimated results The calculation formula for the estimated results is:
[0031]
[0032] In a second aspect, the present application provides a device for predicting internal temperature of a building for use in new energy base planning, the device comprising:
[0033] A data acquisition module is used to obtain temperature data, wherein the temperature data is the temperature values of multiple different points in the building;
[0034] Acquire external factor data corresponding to each temperature in the temperature data, and use the temperature data and the corresponding external factor data as training samples, wherein the external factor data includes light intensity, external temperature and time;
[0035] A training module, used for inputting the training samples into a generalized regression neural network, determining an optimal smoothing factor using a leave-one-out method, obtaining a probability density function based on the optimal smoothing factor, and constructing a pre-trained generalized regression neural network based on the probability density function;
[0036] The output module is used to input the external factor data into the pre-trained generalized regression neural network to obtain the predicted temperature corresponding to different external factor data.
[0037] In a preferred example of the present application, it can be further configured that the probability density function is obtained based on the optimal smoothing factor, including:
[0038] Define the probability density function of the independent variable x and the dependent variable y as f(x,y), and the conditional mean of the probability density function under the given independent variable X is The expression is:
[0039]
[0040] Where dy represents the integral of y, and E[y|X] represents the expected value of y when X takes a preset value;
[0041] The estimated value of the probability density Substituting f(X,y) in the above formula, simplifying the integral operation to obtain the conditional mean The expression is:
[0042]
[0043] Then, we get the estimated value The expression is:
[0044]
[0045] Where n is the number of training samples, i is the serial number of the training sample, p is the dimension of the independent variable, the independent variable is the external factor data, X i , Y i are sample observations of random variables x and y, and δ is the smoothing factor.
[0046] In a preferred example of the present application, it can be further configured that the generalized regression neural network includes an input layer, a pattern layer, a summation layer and an output layer, and the dimension of the input vector of the input layer is 3.
[0047] In a preferred example of the present application, it can be further set that the number of neurons in the pattern layer is equal to the number of neurons in the input layer, and the transfer function followed by the information transmission between the neurons in the pattern layer and the neurons in the input layer is:
[0048]
[0049] Among them, X is the input variable, that is, the external factor data, X i is the learning sample of the i-th neuron.
[0050] In a preferred example of the present application, it can be further configured that the summation layer uses two different types of neurons to perform weighted summation on the data output by the neurons in the pattern layer, wherein one type of neuron in the summation layer uses a direct summation method to sum the data output by the neurons in the pattern layer:
[0051]
[0052] Another neuron in the summation layer performs weighted summation on the data output by the neurons in the pattern layer using a weighted summation method:
[0053]
[0054] And S N and S D Output to the output layer.
[0055] In a preferred example of the present application, it can be further configured that the output layer is based on S N and S D Get the estimated results The calculation formula for the estimated results is:
[0056]
[0057] In a third aspect, the present application provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for predicting internal temperature of a building applied to new energy base planning as described in any one of the above items are implemented.
[0058] In a fourth aspect, the present application provides a computer-readable storage medium having a program stored thereon, wherein when the program is executed by a processor, a method for predicting internal temperature of a building applied to new energy base planning as described in any one of the above items is implemented.
[0059] In a fifth aspect, the present application provides a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the method for predicting internal temperature of a building applied to new energy base planning as described in any of the above items.
[0060] In summary, compared with the prior art, the technical solution provided in the embodiment of the present application has at least the following beneficial effects:
[0061] The present application provides a method for predicting the internal temperature of a building for use in the planning of a new energy base. The method takes into account the effects of external temperature, light intensity and time changes on the internal temperature of a building, and predicts the indoor temperature of the building through a generalized regression neural network, thereby improving the accuracy of the prediction of the indoor temperature of the building. The method not only ensures the basic requirements for plant growth in the building, but also provides data support for flexibly adjusting the building load according to the indoor temperature of the building, provides more accurate data for the rational planning of large-scale distributed energy equipment in new energy bases, and optimizes energy utilization. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] Figure 1 A flowchart of a method for predicting internal temperature of a building applied to new energy base planning is provided as an embodiment of the present application.
[0063] Figure 2 A generalized neural network structure diagram of a method for predicting internal building temperature applied to new energy base planning provided in one embodiment of the present application.
[0064] Figure 3 A building prediction temperature field diagram of a building internal temperature prediction method applied to new energy base planning is provided in one embodiment of the present application.
[0065] Figure 4 A module diagram of a device for predicting internal temperature of a building for use in new energy base planning, provided as an embodiment of the present application. DETAILED DESCRIPTION
[0066] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0067] In one embodiment of the present application, a method for predicting the internal temperature of a building for use in new energy base planning is provided. Figure 1 As shown, the method includes:
[0068] S100: Acquire temperature data, where the temperature data is temperature values at multiple different points in the building;
[0069] Specifically, the temperature values at multiple different points in the building are the temperature values at different locations in the building. By collecting temperature data at multiple points, extreme values are avoided and the prediction effect of the model is improved.
[0070] S200: Acquire external factor data corresponding to each temperature in the temperature data, and use the temperature data and the corresponding external factor data as training samples, wherein the external factor data includes light intensity, external temperature and time;
[0071] Specifically, when each temperature in the temperature data is collected, the light intensity outside the building, the outside temperature and the collection time are recorded, that is, the corresponding external factor data, and then the real-time external factor data is obtained. A temperature value and its corresponding set of light intensity, outside temperature and time form a training sample. The training sample includes multiple training samples, and the temperature data is collected at different points indoors in the building, which is conducive to avoiding extreme values.
[0072] S300: inputting the training sample into a generalized regression neural network, using a leave-one-out method to determine an optimal smoothing factor, obtaining a probability density function based on the optimal smoothing factor, and constructing a pre-trained generalized regression neural network based on the probability density function;
[0073] Specifically, the generalized regression neural network consists of a four-layer structure, such as Figure 2 As shown, it includes an input layer, a pattern layer, a summation layer and an output layer, wherein the dimensions of the input variables include light intensity, outside temperature and time, and the outside temperature is the outdoor temperature of the building.
[0074] In a preferred embodiment, first, define the probability density function of variables x and y as f(x, y), and calculate the expression of the probability density function of the dependent variable under the condition of the given independent variable X, that is, the conditional mean of the probability density function The expression is as follows:
[0075]
[0076] Where dy represents the integral of y, and E[y|X] represents the expected value of y when X takes a preset value;
[0077] Because the probability density function The estimated value of is obtained by non-parametric estimation of the training samples, so the expression is:
[0078]
[0079] Where n is the number of training samples, i is the serial number of the training sample, p is the dimension of the independent variable, the independent variable is the external factor data, X i , Y i are sample observations of random variables x and y, and δ is the smoothing factor.
[0080] The estimated value of the probability density Substituting f(X,y) in the above formula, simplifying the integral operation to obtain the conditional mean The expression is:
[0081]
[0082] The independent variable X is external factor data.
[0083] In a preferred embodiment, the independent variables have three dimensions: light intensity, external temperature and time, and the number of neurons in the input layer is defined as 3, that is, the dimension of the input vector of the input layer is 3.
[0084] In a preferred embodiment, the input layer transfers the temperature data to the pattern layer via a Gaussian transfer function. The pattern layer trains the samples, the number of neurons in the pattern layer is equal to the number of neurons in the input layer, and the transfer function followed by the information transfer between neurons is:
[0085]
[0086] Wherein, X is the input variable of the generalized regression neural network, i.e., the outside temperature data, X i is the learning sample of the i-th neuron.
[0087] In a preferred embodiment, the summation layer uses two different types of neurons to perform weighted summation on the neuron data of the pattern layer, first performing direct summation:
[0088]
[0089] Then do the weighted summation:
[0090]
[0091] Next, the summation result S N and S D Output to the output layer.
[0092] In a preferred embodiment, the output layer divides the summation result to obtain an estimated result The expression is:
[0093]
[0094] The estimated results That is, a set of predicted temperatures corresponding to the external factor data.
[0095] S400: Inputting the external factor data into the pre-trained generalized regression neural network to obtain predicted temperatures corresponding to different external factor data.
[0096] Specifically, a set of external factor data, namely, a set of light intensity, external temperature and time, is input into the pre-trained model to obtain the predicted indoor temperature of the building corresponding to the set of external factor data.
[0097] After obtaining the predicted results of the building temperature, the building temperature field can be fitted by the cubic spline interpolation method. Then, according to the predicted temperature data at different points of the building cross section, the comprehensive temperature distribution of the entire cross section can be modeled using MATLAB software to obtain the following: Figure 3 The temperature field shown, Figure 3 (a) Figure 3 (b) Figure 3 (c) and Figure 3 (d) The predicted building temperature field at 7:00, 12:00, 17:00 and 22:00 respectively. By modeling the indoor temperature, the indoor temperature visualization effect is achieved.
[0098] In this embodiment, taking into account the impact of external temperature, light intensity and time changes on the internal temperature of the building, the indoor temperature of the building is predicted by a generalized regression neural network, which improves the accuracy of the indoor temperature prediction of the building. It not only ensures the basic requirements for plant growth in the building, but also provides data support for flexibly adjusting the building load according to the indoor temperature of the building, provides more accurate data for the rational planning of large-scale distributed energy equipment in new energy bases, and optimizes energy utilization.
[0099] This application also provides a building internal temperature prediction device for use in new energy base planning. Figure 4 As shown, the device comprises:
[0100] The data acquisition module 100 is used to obtain temperature data, wherein the temperature data is the temperature values of multiple different points in the building; obtain the external factor data corresponding to each temperature in the temperature data, and use the temperature data and the corresponding external factor data as training samples, wherein the external factor data includes light intensity, external temperature and time;
[0101] A training module 200 is used to input the training samples into a generalized regression neural network, determine an optimal smoothing factor using a leave-one-out method, obtain a probability density function based on the optimal smoothing factor, and construct a pre-trained generalized regression neural network based on the probability density function;
[0102] The output module 300 is used to input the external factor data into the pre-trained generalized regression neural network to obtain the predicted temperature corresponding to different external factor data.
[0103] The functional implementation of each module in the above-mentioned building internal temperature prediction device applied to new energy base planning corresponds to the steps in the above-mentioned building internal temperature prediction method embodiment applied to new energy base planning, and its functions and implementation processes will not be repeated here one by one.
[0104] The present application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for predicting internal temperature of a building applied to new energy base planning as described in any of the above embodiments are implemented.
[0105] The present application also provides a computer-readable storage medium, on which a program is stored, wherein the computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive and / or a memory stick, etc., and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The working process, working details and technical effects of the computer-readable storage medium provided in this embodiment can be found in the above embodiment of a method for predicting internal temperature of a building applied to new energy base planning, and will not be repeated here.
[0106] The application also provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the method for predicting internal temperature of a building applied to new energy base planning as described in any of the above embodiments.
[0107] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0108] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. The above-mentioned embodiments only express several implementation methods of the present application, and their descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present application, several deformations and improvements can be made, which all belong to the protection scope of the present application. Therefore, the protection scope of the patent of this application shall be based on the attached claims.
Claims
1. A method for predicting internal temperature of a building applied to new energy base planning, characterized in that: include: Acquire temperature data, where the temperature data is temperature values at multiple different points in the building; Acquire external factor data corresponding to each temperature in the temperature data, and use the temperature data and the corresponding external factor data as training samples, wherein the external factor data includes light intensity, external temperature and time; Inputting the training samples into a generalized regression neural network, determining an optimal smoothing factor using a leave-one-out method, obtaining a probability density function based on the optimal smoothing factor, and constructing a pre-trained generalized regression neural network based on the probability density function; The external factor data is input into the pre-trained generalized regression neural network to obtain the predicted temperature corresponding to different external factor data.
2. The method for predicting internal temperature of a building for use in new energy base planning according to claim 1, characterized in that: The obtaining of a probability density function based on the optimal smoothing factor comprises: Define the probability density function of the independent variable x and the dependent variable y as f(x,y), and the conditional mean of the probability density function under the given independent variable X is The expression is: Where dy represents the integral of y, and E[y|X] represents the expected value of y when X takes a preset value; The estimated value of the probability density Substituting f(X,y) in the above formula, simplifying the integral operation to obtain the conditional mean The expression is: Get an estimate The expression is: Where n is the number of training samples, i is the serial number of the training sample, p is the dimension of the independent variable, the independent variable is the external factor data, X i , Y i are sample observations of random variables x and y, and δ is the smoothing factor.
3. The method for predicting internal building temperature for new energy base planning according to claim 1, characterized in that: The generalized regression neural network includes an input layer, a pattern layer, a summation layer and an output layer, and the dimension of the input vector of the input layer is 3.
4. The method for predicting internal building temperature for new energy base planning according to claim 3 is characterized in that: The number of neurons in the pattern layer is equal to the number of neurons in the input layer, and the transfer function followed by the information transmission between the neurons in the pattern layer and the neurons in the input layer is: Among them, X is the input variable, that is, the external factor data, X i is the learning sample of the i-th neuron.
5. The method for predicting internal temperature of a building for use in new energy base planning according to claim 4, characterized in that: The summation layer uses two different types of neurons to perform weighted summation on the data output by the neurons in the pattern layer, wherein one type of neuron in the summation layer uses a direct summation method to sum the data output by the neurons in the pattern layer: Another neuron in the summation layer performs weighted summation on the data output by the neurons in the pattern layer using a weighted summation method: And S N and S D Output to the output layer.
6. The method for predicting internal temperature of a building for use in new energy base planning according to claim 5, characterized in that: The output layer is based on S N and S D Get the estimated results The calculation formula for the estimated results is:
7. A building internal temperature prediction device applied to new energy base planning, characterized in that: include: A data acquisition module is used to obtain temperature data, wherein the temperature data is the temperature values of multiple different points in the building; Acquire external factor data corresponding to each temperature in the temperature data, and use the temperature data and the corresponding external factor data as training samples, wherein the external factor data includes light intensity, external temperature and time; A training module, used for inputting the training samples into a generalized regression neural network, determining an optimal smoothing factor using a leave-one-out method, obtaining a probability density function based on the optimal smoothing factor, and constructing a pre-trained generalized regression neural network based on the probability density function; The output module is used to input the external factor data into the pre-trained generalized regression neural network to obtain the predicted temperature corresponding to different external factor data.
8. The building internal temperature prediction device for new energy base planning according to claim 7, characterized in that: The obtaining of a probability density function based on the optimal smoothing factor comprises: Define the probability density function of the independent variable x and the dependent variable y as f(x,y), and the conditional mean of the probability density function under the given independent variable X is The expression is: Where dy represents the integral of y, and E[y|X] represents the expected value of y when X takes a preset value; The estimated value of the probability density Substituting f(X,y) in the above formula, simplifying the integral operation to obtain the conditional mean The expression is: Then, we get the estimated value The expression is: Where n is the number of training samples, i is the serial number of the training sample, p is the dimension of the independent variable, the independent variable is the external factor data, X i , Y i are sample observations of random variables x and y, and δ is the smoothing factor.
9. The device for predicting internal temperature of a building for use in new energy base planning according to claim 8, characterized in that: The generalized regression neural network includes an input layer, a pattern layer, a summation layer and an output layer, and the dimension of the input vector of the input layer is 3.
10. The building internal temperature prediction device for new energy base planning according to claim 9, characterized in that: The number of neurons in the pattern layer is equal to the number of neurons in the input layer, and the transfer function followed by the information transmission between the neurons in the pattern layer and the neurons in the input layer is: Among them, X is the input variable, that is, the external factor data, X i is the learning sample of the i-th neuron.
11. The building internal temperature prediction device for new energy base planning according to claim 10, characterized in that: The summation layer uses two different types of neurons to perform weighted summation on the data output by the neurons in the pattern layer, wherein one type of neuron in the summation layer sums the data output by the neurons in the pattern layer by direct summation: Another neuron in the summation layer performs weighted summation on the data output by the neurons in the pattern layer using a weighted summation method: And S N and S D Output to the output layer.
12. The building internal temperature prediction device for new energy base planning according to claim 11, characterized in that: The output layer is based on S N and S D Get the estimated results The calculation formula for the estimated results is:
13. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for predicting internal temperature of a building applied to new energy base planning are implemented as described in any one of claims 1 to 6.
14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, wherein when the program is executed by a processor, the method for predicting internal temperature of a building applied to new energy base planning according to any one of claims 1 to 6 is implemented.
15. A computer program product comprising computer instructions, characterized in that: The computer instructions implement the steps of the method according to claims 1 to 6 when executed by a processor.