Building load prediction method, training method and system considering random thermal disturbances

Abstract

The present invention provides a building load prediction method, training method and system considering random thermal disturbances, which can be applied to the technical field of building energy supply system control. The method includes: in response to receiving a building load prediction requirement for a target time period, acquiring historical operation information, attribute parameter information and historical environmental factor information of a building radiant energy supply system for a target building corresponding to the target time period; constructing a first model according to the historical operation information, attribute parameter information and historical environmental factor information; performing continuous-time processing on the variable parameters of the first model to obtain target parameter information; processing the target parameter information according to a target model to obtain a thermal disturbance prediction value of the target building, the target model being constructed by coupling the first model and a second model; and inputting the thermal disturbance prediction value of the target building into the first model to obtain a load prediction value of the target building.

Description

Technical Field

[0001] The present invention relates to the technical field of building energy supply system control, and particularly to a method, a training method and a system for predicting thermal disturbances in a building. Background Art

[0002] A building radiant energy supply system embeds heat / cold carrier fluid pipes inside the building envelope structure, and uses these pipes for energy storage and transfer, so as to achieve the control of the indoor environment of the building. Based on the building energy supply terminal with heat / cold carrier fluid pipes embedded inside, it has great energy storage potential, and the heat storage / dissipation of the envelope structure is adjustable, which is conducive to the implementation of flexible building energy use. However, the thermal disturbances in the building radiant energy supply system usually affect the indoor environment and energy consumption.

[0003] In the process of implementing the concept of the present invention, the inventors found that in the related art, the dynamic characteristics of thermal disturbances under the lack of measurement conditions are usually not considered, and the influence of different thermal disturbances on the building load is often simplified by integrating thermal disturbance terms, resulting in low prediction accuracy of the target building load. Furthermore, the control strategy is difficult to adapt to the rapidly changing environment or complex system requirements, making the overall performance of the building radiant energy supply system in actual operation lower than expected. Summary of the Invention

[0004] In view of the above problems, the present invention provides a method, a training method and a system for predicting building load considering random thermal disturbances.

[0005] According to a first aspect of the present invention, there is provided a method, a training method and a system for predicting building load considering random thermal disturbances, the method comprising: in response to receiving a building load prediction requirement for a target period, acquiring historical operation information, attribute parameter information and historical environmental factor information of the building radiant energy supply system of a target building corresponding to the target period; constructing a first model according to the historical operation information, the attribute parameter information and the historical environmental factor information, wherein the first model includes variable parameters; performing continuous-time processing on the variable parameters of the first model to obtain target parameter information; processing the target parameter information according to a target model to obtain a thermal disturbance prediction value of the target building, wherein the target model is constructed by coupling the first model and a second model, and the second model is obtained by training an initial model with the thermal disturbance prediction value as a label, using sample historical operation information, sample historical environmental factor information and sample time parameter information; inputting the thermal disturbance prediction value of the target building into the first model to obtain a load prediction value of the target building.

[0006] According to an embodiment of the present invention, processing target parameter information according to a target model to obtain a thermal disturbance prediction value of a target building includes: coupling a first model and a second model to obtain a target model; in the target model, determining a predicted indoor temperature value for a target time period according to an initial disturbance prediction value output by the second model; determining a loss function value according to the predicted indoor temperature value for the target time period and the actual indoor temperature value for the target time period; and obtaining a thermal disturbance prediction value of the target building when the loss function value is less than a preset loss function value.

[0007] According to an embodiment of the present invention, determining a loss function value according to the predicted indoor temperature value for a target time period and the actual indoor temperature value for the target time period includes: determining a target loss function according to the predicted indoor temperature value for the target time period and the actual indoor temperature value for the target time period; and determining a loss function value according to the target loss function.

[0008] According to an embodiment of the present invention, the first model further includes: constant parameters, wherein coupling the first model and the second model to obtain a target model includes: determining target parameter information according to the constant parameters and variable parameters of the first model; and coupling the first model and the second model according to the target parameter information and the initial disturbance prediction value to obtain a target model.

[0009] According to an embodiment of the present invention, the historical operation information includes the temperature information of the building envelope, the temperature information of the upper surface of the floor, the temperature information of the lumped nodes of the embedded pipeline plane, the supply water temperature information, the return water temperature information, and the water flow rate and heat supply; the attribute parameter information includes the heat capacity of the building envelope, the heat capacity of the indoor air, the heat capacity of the floor, the heat capacity of the embedded pipeline, the thermal resistance between the outdoor air and the building envelope, the thermal resistance between the indoor air and the building envelope, the thermal resistance of the outer window, the thermal resistance between the embedded pipeline and the floor, the thermal resistance between the embedded pipeline and the soil, the heat capacity of the water, the conversion coefficient of solar radiation, and the area of the outer window; the historical environmental factor information includes the indoor temperature information and the outdoor temperature information; wherein, according to the historical operation information, the attribute parameter information, and the historical environmental factor information, a first model is constructed, including: constructing a first balance equation according to the temperature information of the building envelope, the indoor temperature information, the heat capacity of the building envelope, the thermal resistance between the outdoor air and the building envelope, the thermal resistance between the indoor air and the building envelope, and the first process disturbance; constructing a second balance equation according to the temperature information of the building envelope, the indoor temperature information, the outdoor temperature information, the temperature information of the upper surface of the floor, the heat capacity of the indoor air, the thermal resistance between the indoor air and the building envelope, the thermal resistance between the indoor air and the floor, the thermal resistance of the outer window, the first thermal disturbance parameter, and the second process disturbance; constructing a third balance equation according to the indoor temperature information, the temperature information of the upper surface of the floor, the temperature information of the lumped nodes of the embedded pipeline plane, the area of the outer window, the conversion coefficient of solar radiation, the heat capacity of the floor, the thermal resistance between the indoor air and the floor, the thermal resistance between the embedded pipeline and the floor, the second thermal disturbance parameter, and the third process disturbance; constructing a fourth balance equation according to the temperature information of the soil, the temperature information of the lumped nodes of the embedded pipeline plane, the temperature information of the upper surface of the floor, the heat capacity of the embedded pipeline, the thermal resistance between the embedded pipeline and the floor, the thermal resistance between the embedded pipeline and the soil, the heat supply, and the fourth process disturbance; constructing a fifth balance equation according to the supply water temperature information, the return water temperature information, the heat capacity of the water, and the water flow rate and heat supply; and constructing a first model according to the first balance equation, the second balance equation, the third balance equation, the fourth balance equation, and the fifth balance equation.

[0010] According to an embodiment of the present invention, continuous-time processing is performed on the variable parameters of the first model to obtain target parameter information, including: performing continuous-time processing on the variable parameters of the first model to obtain a continuous-time first model; inputting the historical energy supply information, the attribute parameter information, and the historical environmental factor information into the continuous-time first model to obtain the target parameter information.

[0011] The second aspect of the present invention provides a training method for a target model based on a building radiant energy supply system, including: obtaining training samples, where the training samples include sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample disturbance parameter information; using the target model to process the sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample disturbance parameter information, and outputting target parameter information; when the target parameter information is determined, using a feedforward neural network to train the target model to obtain the trained target model.

[0012] The third aspect of the present invention provides a building load prediction system considering random heat disturbances, including: a first acquisition module, a construction module, a first obtaining module, a second obtaining module, and a third obtaining module.

[0013] The first acquisition module is configured to, in response to receiving a heat disturbance prediction requirement for a target time period, obtain the historical operation information, attribute parameter information, and historical environmental factor information of the building radiant energy supply system of the target building corresponding to the target time period.

[0014] The construction module is configured to construct a first model according to the historical operation information, attribute parameter information, and historical environmental factor information, where the first model includes variable parameters.

[0015] The first obtaining module is configured to perform continuous-time processing on the variable parameters of the first model to obtain target parameter information.

[0016] The second obtaining module is configured to process the target parameter information according to the target model to obtain a heat disturbance prediction value of the target building, where the target model is constructed by coupling the first model and the second model, and the second model is trained with the heat disturbance prediction value as a label for the historical operation information, historical environmental factor information, and time parameter information.

[0017] The third obtaining module is configured to input the heat disturbance prediction value of the target building into the first model to obtain a load prediction value of the target building.

[0018] The fourth aspect of the present invention provides a training device for a target model based on a building radiant energy supply system, including: a second acquisition module, an output module, and a training module.

[0019] The second acquisition module is configured to obtain training samples, where the training samples include sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample disturbance parameter information.

[0020] The output module is configured to use the target model to process the sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample disturbance parameter information, and output target parameter information.

[0021] A training module, configured to train a target model by using a feedforward neural network to obtain a trained target model when target parameter information is determined.

[0022] A fifth aspect of the present invention provides an electronic device, including: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0023] A sixth aspect of the present invention further provides a computer-readable storage medium, on which a computer program or instruction is stored, and when the computer program or instruction is executed by a processor, the steps of the above method are implemented.

[0024] A seventh aspect of the present invention further provides a computer program product, including a computer program or instruction, and when the computer program or instruction is executed by a processor, the steps of the above method are implemented.

[0025] According to the building load prediction method, training method and system considering random thermal disturbances provided by the present invention, by constructing a first model according to the obtained historical operation information, attribute parameter information and historical environmental factor information, and performing continuous-time processing on the variable parameters of the first model, target parameter information can be obtained. By using the target model to process the target parameter information, a thermal disturbance prediction value of the target building can be obtained. Since the target model is constructed by fusing the first model and the second model, it is possible to accurately predict multiple uncertain thermal disturbances under limited measurement conditions, improve the accuracy of the thermal disturbance prediction value of the target building, and further improve the accuracy of the load prediction value of the target building, so that the overall performance of the building radiant energy supply system during actual operation meets the expectations. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Through the following description of the embodiments of the present invention with reference to the drawings, the above content and other objects, features and advantages of the present invention will become clearer. In the drawings:

[0027] Figure 1 An application scenario diagram of the building load prediction method considering random thermal disturbances according to an embodiment of the present invention is shown.

[0028] Figure 2 A flowchart of the building load prediction method considering random thermal disturbances according to an embodiment of the present invention is shown.

[0029] Figure 3 A schematic diagram of the first model according to an embodiment of the present invention is shown.

[0030] Figure 4 A flowchart of the building load prediction method considering random thermal disturbances according to another embodiment of the present invention is shown.

[0031] Figure 5 Shows the construction flow chart of the target model according to an embodiment of the present invention.

[0032] Figure 6A Shows the comparison chart of the prediction results of personnel occupancy thermal disturbance according to an embodiment of the present invention.

[0033] Figure 6B Shows the comparison chart of the prediction results of solar radiation thermal disturbance according to an embodiment of the present invention.

[0034] Figure 7 Shows the flow chart of the training method of the target model based on the building radiant energy supply system according to an embodiment of the present invention.

[0035] Figure 8 Shows the structural block diagram of the building load prediction system considering random thermal disturbance according to an embodiment of the present invention.

[0036] Figure 9 Shows the structural block diagram of the training device of the target model based on the building radiant energy supply system according to an embodiment of the present invention.

[0037] Figure 10 Shows the block diagram of the electronic device suitable for implementing the building load prediction method considering random thermal disturbance according to an embodiment of the present invention. Detailed implementation manners

[0038] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In the following detailed description, for the sake of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present invention. However, obviously, one or more embodiments can also be implemented without these specific details. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concepts of the present invention.

[0039] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "including", "comprising" and the like used herein indicate the presence of the described features, steps, operations and / or components, but do not exclude the presence or addition of one or more other features, steps, operations or components.

[0040] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0041] In the case of using expressions such as "at least one of A, B, and C, etc.", generally, it should be interpreted according to the meaning that those skilled in the art usually understand this expression (for example, "a system having at least one of A, B, and C" should include but not be limited to a system having only A, only B, only C, having A and B, having A and C, having B and C, and / or having A, B, and C, etc.).

[0042] In the technical solution of the present invention, the user information involved (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties. And the processing of relevant data, such as collection, storage, use, processing, transmission, provision, disclosure, and application, all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0043] In the process of implementing the present invention, it is found that in the related art, the dynamic characteristics of thermal disturbances under lack of measurement conditions are usually not considered. Often, the influence of different thermal disturbances on building loads is simplified by integrating thermal disturbance terms, resulting in low prediction accuracy of the target building load, and further causing the overall performance of the building radiant energy supply system in actual operation to be lower than expected. Or a machine learning method is used to predict building thermal disturbances, but its prediction accuracy has a strong dependence on historical data.

[0044] In view of this, an embodiment of the present invention provides a method for predicting building load considering random thermal disturbances, including: in response to receiving a thermal disturbance prediction demand for a target time period, obtaining historical operation information, attribute parameter information, and historical environmental factor information of the building radiant energy supply system of a target building corresponding to the target time period; constructing a first model according to the historical operation information, attribute parameter information, and historical environmental factor information, where the first model includes variable parameters; performing continuous-time processing on the variable parameters of the first model to obtain target parameter information; processing the target parameter information according to a target model to obtain a thermal disturbance prediction value of the target building, where the target model is constructed by coupling the first model and a second model, and the second model is obtained by training an initial model with the thermal disturbance prediction value as a label, using sample historical operation information, sample historical environmental factor information, and sample time parameter information, and inputting the thermal disturbance prediction value of the target building into the first model to obtain a load prediction value of the target building.

[0045] Figure 1 The application scenario diagram of the method for predicting building load considering random thermal disturbances according to an embodiment of the present invention is shown.

[0046] AsFigure 1 As shown in Figure 1 , the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links among the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0047] Users can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for examples).

[0048] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices with a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, and desktop computers, etc.

[0049] The server 105 may be a server providing various services, such as a background management server (only for example) that supports the websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process data such as received user requests, and feedback the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0050] It should be noted that the building load prediction method considering random thermal disturbances provided by the embodiments of the present invention can generally be executed by the server 105. Correspondingly, the building load prediction system considering random thermal disturbances provided by the embodiments of the present invention can generally be set in the server 105. The building load prediction method considering random thermal disturbances provided by the embodiments of the present invention can also be executed by a server or a server cluster different from the server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or the server 105. Correspondingly, the building load prediction system considering random thermal disturbances provided by the embodiments of the present invention can also be set in a server or a server cluster different from the server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or the server 105.

[0051] It should be understood, Figure 1The numbers of the first terminal device, the second terminal device, the third terminal device, the network, and the server in [ ] are merely illustrative. According to the implementation requirements, there can be any number of the first terminal device, the second terminal device, the third terminal device, the network, and the server.

[0052] Figure 2 The flowchart of the building load prediction method considering random thermal disturbance according to an embodiment of the present invention is shown.

[0053] As Figure 2 shown, the building load prediction method 200 considering random thermal disturbance in this embodiment includes operations S210 to S250.

[0054] In operation S210, in response to receiving the building load prediction requirement for the target time period, obtain the historical operation information, attribute parameter information, and historical environmental factor information of the building radiant energy supply system of the target building corresponding to the target time period.

[0055] In operation S220, construct the first model according to the historical operation information, attribute parameter information, and historical environmental factor information.

[0056] In operation S230, perform continuous-time processing on the variable parameters of the first model to obtain the target parameter information.

[0057] In operation S240, process the target parameter information according to the target model to obtain the thermal disturbance prediction value of the target building.

[0058] In operation S250, input the thermal disturbance prediction value of the target building into the first model to obtain the load prediction value of the target building.

[0059] According to an embodiment of the present invention, the building radiant energy supply system of the target building can represent a system in which water pipes through which cold water / hot water flows are embedded inside the building envelope of the target building.

[0060] According to an embodiment of the present invention, the target time period can represent a certain period of time in the future. For example, the target time period can be the first month of the third quarter of the next year. Another example is that the target time period can be 24 hours within a day next Friday. Thermal disturbance can represent the disturbance of temperature change or heat transfer caused by external or internal factors. Thermal disturbance can be instantaneous or continuous, periodic or non-periodic, and will cause changes in the state or parameters of the building radiant energy supply system. Since thermal disturbance may be instantaneous, random, and non-periodic, it is difficult to accurately predict the thermal disturbance prediction value. Thermal disturbance can be multivariate. For example, it can be the thermal disturbance caused by personnel occupancy. Another example is that it can also be the solar radiation thermal disturbance.

[0061] According to an embodiment of the present invention, the historical operation information can characterize the operation information during the operation of the building radiant energy supply system in a previous period of time. For different target time periods, the historical operation information can be different. The attribute parameter information can characterize the attribute information of each part in the building radiant energy supply system, and the attribute information can be a fixed value. The historical environmental factor information can characterize the environmental temperature information in a previous period of time.

[0062] According to an embodiment of the present invention, a first model can be constructed based on the historical operation information, the attribute parameter information, and the historical environmental factor information. The first model can also be referred to as a gray box model. The first model can be used to predict the load of the target building.

[0063] According to an embodiment of the present invention, the first model can include variable parameters, and the variable parameters can characterize the variable parameters in the first model.

[0064] According to an embodiment of the present invention, continuous-time processing can be performed on the variable parameters of the first model, so as to obtain target parameter information, and there can be multiple pieces of target parameter information. For example, the target parameter information can include 3 pieces.

[0065] According to an embodiment of the present invention, the target model can be constructed by coupling the first model and the second model. The second model can be obtained by training an initial model using the sample historical operation information, the sample historical environmental factor information, and the sample time parameter information with the predicted value of the thermal disturbance as the label. The second model can include a neural network model.

[0066] According to an embodiment of the present invention, by processing the target parameter information using the target model, the predicted value of the disturbance of the target building can be obtained. For example, by processing the target parameter information using the target model, the predicted value of the disturbance of the target building within the next day can be obtained.

[0067] According to an embodiment of the present invention, the predicted value of the thermal disturbance of the target building can be input into the first model, and by processing the predicted value of the thermal disturbance of the target building through the first model, the predicted value of the load of the target building can be obtained.

[0068] According to an embodiment of the present invention, by constructing a first model based on the obtained historical operation information, attribute parameter information, and historical environmental factor information, and performing continuous-time processing on the variable parameters of the first model, target parameter information can be obtained. Using the target model to process the target parameter information, a predicted value of the thermal disturbance of the target building can be obtained. Since the target model is constructed by fusing the first model and the second model, it is possible to accurately predict multivariate uncertain thermal disturbances under limited measurement conditions, improve the accuracy of the predicted value of the thermal disturbance of the target building, and further improve the accuracy of the predicted value of the load of the target building, so that the overall performance of the building radiant energy supply system during the actual operation process meets the expectations.

[0069] Figure 3 FIG. shows a schematic diagram of the first model according to an embodiment of the present invention.

[0070] As Figure 3 shown, the first model based on thermal resistance and heat capacity (abbreviated as RC) is constructed, and it can be used to describe the simplified model of the thermal dynamics of the target building. The first model includes the indoor side of the target building, the envelope structure of the target building, and the outdoor side of the target building. The first model is divided into an above-ground part and a below-ground part by the ground, and the below-ground part belongs to an embedded pipe structure.

[0071] According to an embodiment of the present invention, the historical operation information may include the temperature information of the envelope structure, the temperature information of the upper surface of the floor, the temperature information of the lumped nodes of the embedded pipe plane, the return water temperature information, and the water flow rate and heat supply. The attribute parameter information may include the heat capacity of the envelope structure, the heat capacity of the indoor air, the heat capacity of the floor, the heat capacity of the embedded pipe, the thermal resistance between the outdoor air and the envelope structure, the thermal resistance between the indoor air and the envelope structure, the thermal resistance of the outer window, the thermal resistance between the embedded pipe and the floor, the thermal resistance between the embedded pipe and the soil, the heat capacity of water, the conversion coefficient of solar radiation, and the area of the outer window. The historical environmental factor information may include the indoor temperature information and the outdoor temperature information.

[0072] According to an embodiment of the present invention, a first model is constructed based on historical operation information, attribute parameter information, and historical environmental factor information, including: constructing a first balance equation according to the temperature information of the building envelope, indoor temperature information, heat capacity of the building envelope, thermal resistance between outdoor air and the building envelope, thermal resistance between indoor air and the building envelope, and a first process disturbance; constructing a second balance equation according to the temperature information of the building envelope, indoor temperature information, outdoor temperature information, temperature information of the upper surface of the floor, heat capacity of indoor air, thermal resistance between indoor air and the building envelope, thermal resistance between indoor air and the floor, thermal resistance between indoor air and the floor, thermal resistance of the external window, a first thermal disturbance parameter, and a second process disturbance; constructing a third balance equation according to the indoor temperature information, temperature information of the upper surface of the floor, temperature information of the lumped nodes of the embedded pipeline plane, area of the external window, conversion coefficient of solar radiation, heat capacity of the floor, thermal resistance between indoor air and the floor, thermal resistance between the embedded pipeline and the floor, a second thermal disturbance parameter, and a third process disturbance; constructing a fourth balance equation according to the temperature information of the soil, temperature information of the lumped nodes of the embedded pipeline plane, temperature information of the upper surface of the floor, heat capacity of the embedded pipeline, thermal resistance between the embedded pipeline and the floor, thermal resistance between the embedded pipeline and the soil, heating supply, and a fourth process disturbance; constructing a fifth balance equation according to the supply water temperature information, return water temperature information, heat capacity of water, water flow rate, and heating supply; and constructing the first model according to the first balance equation, the second balance equation, the third balance equation, the fourth balance equation, and the fifth balance equation.

[0073] According to an embodiment of the present invention, a first balance equation can be constructed according to the temperature information of the building envelope, indoor temperature information, heat capacity of the building envelope, thermal resistance between outdoor air and the building envelope, thermal resistance between indoor air and the building envelope, and a first process disturbance, as shown in the following formula (1).

[0074] (1);

[0075] Wherein, represents the temperature information of the building envelope of the target building, represents the outdoor temperature information; represents the indoor temperature information, represents the heat capacity of the building envelope, represents the thermal resistance between outdoor air and the building envelope, represents the thermal resistance between indoor air and the building envelope, represents the first process disturbance.

[0076] According to an embodiment of the present invention, a second balance equation can be constructed based on the temperature information of the enclosure structure, the indoor temperature information, the outdoor temperature information, the temperature information of the upper surface of the floor, the heat capacity of the indoor air, the thermal resistance between the indoor air and the enclosure structure, the thermal resistance between the indoor air and the floor, the thermal resistance of the external window, the first thermal disturbance parameter, and the second process disturbance, as shown in the following formula (2).

[0077] (2);

[0078] Wherein, represents the temperature information of the upper surface of the floor, represents the first thermal disturbance parameter; represents the heat capacity of the indoor air, represents the thermal resistance of the external window, represents the thermal resistance between the indoor air and the floor, represents the second process disturbance.

[0079] According to an embodiment of the present invention, a third balance equation can be constructed based on the indoor temperature information, the temperature information of the upper surface of the floor, the temperature information of the lumped nodes of the embedded pipeline plane, the area of the external window, the conversion coefficient of solar radiation, the heat capacity of the floor, the thermal resistance between the indoor air and the floor, the thermal resistance between the embedded pipeline and the floor, the second thermal disturbance parameter, and the third process disturbance, as shown in the following formula (3).

[0080] (3);

[0081] Wherein, represents the temperature information of the lumped nodes of the embedded pipeline plane, represents the area of the external window, represents the conversion coefficient of solar radiation, represents the second thermal disturbance parameter; represents the heat capacity of the floor, represents the thermal resistance between the embedded pipeline and the floor, represents the third process disturbance.

[0082] According to an embodiment of the present invention, the first thermal disturbance parameter can be a personnel occupancy thermal disturbance parameter. The second thermal disturbance parameter can be a solar radiation thermal disturbance parameter.

[0083] According to an embodiment of the present invention, a fourth balance equation can be constructed based on the temperature information of the soil, the temperature information of the lumped nodes of the embedded pipeline plane, the temperature information of the upper surface of the floor, the heat capacity of the embedded pipeline, the thermal resistance between the embedded pipeline and the floor, the thermal resistance between the embedded pipeline and the soil, the heating supply, and the fourth process disturbance, as shown in the following formula (4).

[0084] (4);

[0085] Among them, represents the soil temperature information, represents the heat supply of the building radiant energy supply system, represents the thermal resistance between the embedded pipeline and the soil, represents the heat capacity of the embedded pipeline, represents the fourth process disturbance.

[0086] According to an embodiment of the present invention, a fifth balance equation can be constructed based on the water supply temperature information, the return water temperature information, the heat capacity of water, the water flow rate, and the heat supply, as shown in the following formula (5).

[0087] (5);

[0088] Among them, represents the water supply temperature information, represents the return water temperature information, c represents the heat capacity of water, and m represents the water flow rate.

[0089] According to an embodiment of the present invention, a heat pump can be used for water supply. The first process disturbance, the second process disturbance, the third process disturbance, and the fourth process disturbance can be random process disturbances, and their influence on the building radiant energy supply system can be ignored.

[0090] According to an embodiment of the present invention, a first model can be further constructed based on the first balance equation, the second balance equation, the third balance equation, the fourth balance equation, and the fifth balance equation constructed above.

[0091] According to an embodiment of the present invention, continuous-time processing is performed on the variable parameters of the first model to obtain target parameter information, including: performing continuous-time processing on the variable parameters of the first model to obtain a continuous-time first model; inputting historical energy supply information, attribute parameter information, and historical environmental factor information into the continuous-time first model to obtain target parameter information.

[0092] According to an embodiment of the present invention, by performing continuous-time processing on the variable parameters of the first model, a continuous-time first model can be obtained, as shown in the following formula (6).

[0093] (6);

[0094] Among them, represents the state vector, , represents the control input vector, ; represents the disturbance vector, , represents a random disturbance of independent normal distribution.

[0095] According to an embodiment of the present invention, the random perturbation of an independent normal distribution can be negligible in a building radiant energy supply system. By inputting historical energy supply information, attribute parameter information, and historical environmental factor information into a continuous-time first model for processing, target parameter information can be obtained. The target parameter information may include first target parameter information, second target parameter information, and third target parameter information.

[0096] According to an embodiment of the present invention, the first target parameter information A can be expressed as the following formula (7).

[0097] (7);

[0098] According to an embodiment of the present invention, the second target parameter information B can be expressed as the following formula (8).

[0099] (8);

[0100] According to an embodiment of the present invention, the third target parameter information E can be expressed as the following formula (9).

[0101] (9);

[0102] According to an embodiment of the present invention, a first model can be constructed based on a first balance equation, a second balance equation, a third balance equation, a fourth balance equation, and a fifth balance equation. By performing continuous-time processing on the variable parameters of the first model, a continuous-time first model can be obtained. Using the continuous-time first model to process historical energy supply information, attribute parameter information, and historical environmental factor information, target parameter information can be obtained, improving the accuracy of the target parameter information.

[0103] According to an embodiment of the present invention, processing the target parameter information according to a target model to obtain a predicted value of the thermal perturbation of a target building includes: coupling the first model and the second model to obtain a target model; in the target model, determining a predicted value of the indoor temperature at a target time period according to an initial perturbation predicted value output by the second model; determining a loss function value according to the predicted value of the indoor temperature at the target time period and the actual value of the indoor temperature at the target time period; and obtaining a predicted value of the thermal perturbation of the target building when the loss function value is less than a preset loss function value.

[0104] According to an embodiment of the present invention, a target model can be obtained by coupling the first model and the second model.

[0105] According to an embodiment of the present invention, the above formula (7) can be discretized using the zero-order hold method, that is, the control signal remains unchanged within the sampling time period span and is converted from a digital signal to an analog signal. For example, the sampling time can be 1 hour, that is, Ts = 1h. Interval , can represent a time period. Keeping the control signal unchanged within the sampling time span can characterize that the heat supply remains unchanged within the sampling time span. The initial moment of the sampling period and the control signal are as shown in the following formulas (8) and (9).

[0106] (8);

[0107] (9);

[0108] According to an embodiment of the present invention, k represents the number of the initial moment in the sampling period, and the value range of k is greater than or equal to 0.

[0109] According to an embodiment of the present invention, after discretizing the above formula (7), the following formulas (10) to (15) can be obtained.

[0110] (10);

[0111] (11);

[0112] (12);

[0113] (13);

[0114] (14);

[0115] (15);

[0116] Wherein, represents the discretized state vector at the (k + 1)-th moment, the discretized state vector at the k-th moment, represents the discretized control signal at the k-th moment, represents the discretized disturbance vector at the k-th moment, represents the discretized random disturbance at the k-th moment, represents the observation matrix, represents the discretized first target parameter information, represents the discretized second target parameter information, represents the discretized third target parameter information, represents the indoor temperature information at the k-th moment.

[0117] According to an embodiment of the present invention, when the target parameter information is determined, a target model can be constructed based on a first model and a second model. The state space equations are as shown in the above formulas (10) and (11), and the input , , and are used, and then can be obtained.

[0118] According to an embodiment of the present invention, the initial disturbance prediction value can be output by the second model. In the target model, the indoor temperature prediction value for the target time period can be determined based on the initial disturbance prediction value. That is, the initial disturbance prediction value is input into the following formula (16), and the initial disturbance prediction value and are used to replace the thermal disturbance prediction value and , so as to determine the indoor temperature prediction value for the target time period, as shown in the following formula (17).

[0119] (16);

[0120] (17);

[0121] Wherein, represents the first initial disturbance prediction value, , represents the second initial disturbance prediction value, , represents the indoor temperature prediction value at the k-th moment.

[0122] According to an embodiment of the present invention, determining the loss function value based on the indoor temperature prediction value for the target time period and the actual indoor temperature value for the target time period includes: determining a target loss function based on the indoor temperature prediction value for the target time period and the actual indoor temperature value for the target time period; and determining the loss function value based on the target loss function.

[0123] According to an embodiment of the present invention, the target loss function L can be determined based on the indoor temperature prediction value for the target time period and the actual indoor temperature value for the target time period, as shown in the following formula (18).

[0124] (18);

[0125] According to an embodiment of the present invention, when the target loss function is determined, the loss function value can be determined.

[0126] According to an embodiment of the present invention, the preset loss function value can be set according to experience or according to requirements, and the setting method of the preset loss function is not limited herein. When the loss function value is less than the preset loss function value, the thermal disturbance prediction value of the target building can be obtained.

[0127] According to an embodiment of the present invention, the target model can be constructed by coupling the first model and the second model, and according to the initial disturbance value output by the second model, the indoor temperature prediction value can be determined. Thus, the target loss function can be determined in combination with the actual indoor temperature value in the target period, and then the loss function value can be determined. When the loss function value is less than the preset loss function value, the thermal disturbance prediction value of the target building can be obtained, which improves the accuracy of the thermal disturbance prediction value of the target building. And when the target model is constructed, the prediction process of thermal disturbance does not depend on historical data.

[0128] Figure 4 The flowchart of a building load prediction method considering random thermal disturbance according to another embodiment of the present invention is shown.

[0129] As Figure 4 shown, the building load prediction method 400 considering random thermal disturbance in this embodiment includes operations S410 to S490.

[0130] In operation S410, in response to receiving a building load prediction demand for a target period, the historical operation information, attribute parameter information, and historical environmental factor information of the building radiant energy supply system for the target building corresponding to the target period are acquired.

[0131] In operation S420, a first model is constructed according to the historical operation information, attribute parameter information, and historical environmental factor information.

[0132] In operation S430, continuous time processing is performed on the variable parameters of the first model to obtain target parameter information.

[0133] In operation S440, the first model is coupled with the second model to obtain a target model.

[0134] In operation S450, in the target model, according to the initial disturbance prediction value output by the second model, the indoor temperature prediction value for the target period is determined.

[0135] In operation S460, a target loss function is determined according to the indoor temperature prediction value for the target period and the actual indoor temperature value for the target period.

[0136] In operation S470, the loss function value is determined according to the target loss function.

[0137] In operation S480, when the loss function value is less than a preset loss function value, a thermal disturbance prediction value of the target building is obtained;

[0138] In operation S490, the thermal disturbance prediction value of the target building is input into the first model to obtain a load prediction value of the target building.

[0139] According to an embodiment of the present invention, coupling the first model and the second model to obtain a target model includes: determining target parameter information according to the constant parameters of the first model; coupling the first model and the second model according to the target parameter information and an initial disturbance prediction value to obtain the target model.

[0140] According to an embodiment of the present invention, the first model may further include constant parameters, and the constant parameters of the first model may represent invariant parameters in the first model. For example, the area of the outer ring, etc.

[0141] According to an embodiment of the present invention, the target parameter information may be determined by using the above formulas (7) to (9) according to the constant parameters and variable parameters of the first model.

[0142] According to an embodiment of the present invention, the first model and the second model may be coupled by using the above formulas (16) and (17) according to the target parameter information and an initial disturbance prediction value to obtain the target model.

[0143] According to an embodiment of the present invention, by coupling the first model and the second model, a target model can be constructed, and the accuracy of the thermal disturbance prediction value of the target building can be improved by using the target model.

[0144] According to an embodiment of the present invention, the building load prediction method considering random thermal disturbance can predict a kind of difficult-to-predict thermal disturbance, and then predict the building load. For example, it can predict the thermal disturbance caused by human occupancy. For another example, it can predict the thermal disturbance caused by solar radiation.

[0145] Figure 5 The flowchart of constructing the target model according to an embodiment of the present invention is shown.

[0146] As Figure 5As shown, the system of the target model includes an embedded tube building 510 and a thermal disturbance predictor 520. The thermal disturbance predictor 520 is a device applied to a building load prediction method considering random thermal disturbances. The embedded tube building 510 is also the above-mentioned building radiant energy supply system. The historical operation information, attribute parameter information, and historical environmental factor information of the embedded tube building 510 can be used to construct a first model, and the variable parameters and constant parameters of the first model can be determined. By performing continuous-time processing on the variable parameters of the first model, target parameter information can be obtained. Inputting the target parameter information into the thermal disturbance predictor 520, the first model and the second model can be coupled to obtain the target model. The target model processes the target parameter information to obtain the thermal disturbance prediction value.

[0147] Figure 6A Fig. shows a comparison chart of the predicted results of human occupancy thermal disturbances according to an embodiment of the present invention. Figure 6B Fig. shows a comparison chart of the predicted results of solar radiation thermal disturbances according to an embodiment of the present invention.

[0148] As Figure 6A and 6B shown, the building load prediction method considering random thermal disturbances can achieve accurate predictions for both human occupancy thermal disturbances and solar radiation thermal disturbances.

[0149] Figure 7 Fig. shows a flowchart of a training method for a target model based on a building radiant energy supply system according to an embodiment of the present invention.

[0150] As Figure 7 shown, the training method 700 for the target model based on the building radiant energy supply system includes operations S710 to S730.

[0151] In operation S710, training samples are obtained.

[0152] In operation S720, the target model processes the sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample disturbance parameter information, and outputs target parameter information.

[0153] In operation S730, when the target parameter information is determined, the target model is trained using a feedforward neural network to obtain the trained target model.

[0154] According to an embodiment of the present invention, the training samples may include sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample disturbance parameter information.

[0155] According to an embodiment of the present invention, when the target parameter information is determined, the target model can be trained using a feedforward neural network to obtain the trained target model. The training of the target model includes forward propagation and backward propagation.

[0156] According to an embodiment of the present invention, the forward propagation process includes the following formulas (19) to (21).

[0157] (19);

[0158] (20);

[0159] (21);

[0160] Wherein, x represents the input state vector, n represents the number of state vectors, represents the set of state vectors, represents the hidden layer is the input of node j in represents the hidden layer is the output of node j in, l represents the hidden layer index, represents the number of nodes in the (l-1)th layer, represents from the layer of nodes to the layer of nodes weight, represents the bias term, represents the selected activation function.

[0161] According to an embodiment of the present invention, the target loss function L can be determined according to the above formula (18), and the backpropagation process includes the following formulas (22) to (24).

[0162] (22);

[0163] (23);

[0164] (24);

[0165] Wherein, represents the derivative of the activation function, represents the learning rate, represents the index term.

[0166] According to an embodiment of the present invention, the input state vector may include outdoor temperature information, time, date type, occupancy situation of personnel, solar radiation time series characteristics, etc. Among them, the date type may be whether it is a weekend. The update method of each weight and bias is as in (22) to (24), and all samples in the sample are traversed.

[0167] Based on the above building load prediction method considering random thermal disturbances, the present invention also provides a building load prediction system considering random thermal disturbances. The following will be combined with Figure 8 to describe the system in detail.

[0168] Figure 8 The structural block diagram of the building load prediction system considering random thermal disturbances according to an embodiment of the present invention is shown.

[0169] As Figure 8 shown, the building load prediction system 800 considering random thermal disturbances in this embodiment includes a first acquisition module 810, a construction module 820, a first obtaining module 830, a second obtaining module 840, and a third obtaining module 850.

[0170] The first acquisition module 810 is configured to, in response to receiving a building load prediction requirement for a target time period, acquire historical operation information, attribute parameter information, and historical environmental factor information of the building radiant energy supply system for the target building corresponding to the target time period. In one embodiment, the first acquisition module 810 may be configured to perform the operation S210 described above, which will not be elaborated here.

[0171] The construction module 820 is configured to construct a first model according to the historical operation information, attribute parameter information, and historical environmental factor information, where the first model includes variable parameters. In one embodiment, the construction module 820 may be configured to perform the operation S220 described above, which will not be elaborated here.

[0172] The first obtaining module 830 is configured to perform continuous-time processing on the variable parameters of the first model to obtain target parameter information. In one embodiment, the first obtaining module 830 may be configured to perform the operation S230 described above, which will not be elaborated here.

[0173] The second obtaining module 840 is configured to process the target parameter information according to the target model to obtain a thermal disturbance prediction value of the target building, where the target model is constructed by coupling the first model and the second model, and the second model is trained with the thermal disturbance prediction value as a label using the historical operation information, historical environmental factor information, and time parameter information. In one embodiment, the second obtaining module 840 may be configured to perform the operation S240 described above, which will not be elaborated here.

[0174] The third obtaining module 850 is configured to input the thermal disturbance prediction value of the target building into the first model to obtain a load prediction value of the target building. In one embodiment, the third obtaining module 850 may be configured to perform the operation S250 described above, which will not be elaborated here.

[0175] According to an embodiment of the present invention, the second obtaining module 840 includes: a first obtaining sub-module, a second obtaining sub-module, a third obtaining sub-module, and a fourth obtaining sub-module.

[0176] The first obtaining sub-module is configured to couple the first model and the second model to obtain a target model.

[0177] The second obtaining sub-module is configured to determine a predicted value of the indoor temperature at a target time period in the target model according to an initial disturbance predicted value output by the second model.

[0178] The third obtaining sub-module is configured to determine a loss function value according to the predicted value of the indoor temperature at the target time period and the actual value of the indoor temperature at the target time period.

[0179] The fourth obtaining sub-module is configured to obtain a predicted value of the thermal disturbance of the target building when the loss function value is less than a preset loss function value.

[0180] According to an embodiment of the present invention, the third obtaining sub-module includes: a first obtaining unit and a second obtaining unit.

[0181] The first obtaining unit is configured to determine a target loss function according to the predicted value of the indoor temperature at the target time period and the actual value of the indoor temperature at the target time period.

[0182] The second obtaining unit is configured to determine a loss function value according to the target loss function.

[0183] According to an embodiment of the present invention, the first model further includes: a constant parameter, and the first obtaining sub-module includes: a third obtaining unit and a fourth obtaining unit.

[0184] The third obtaining unit is configured to determine target parameter information according to the constant parameter of the first model.

[0185] The fourth obtaining unit is configured to couple the first model and the second model according to the target parameter information and the initial disturbance predicted value to obtain a target model.

[0186] According to an embodiment of the present invention, the historical operation information includes the temperature information of the building envelope, the temperature information of the upper surface of the floor, the temperature information of the lumped nodes of the embedded pipeline plane, the supply water temperature information, the return water temperature information, and the water flow rate and heat supply; the attribute parameter information includes the heat capacity of the building envelope, the heat capacity of the indoor air, the heat capacity of the floor, the heat capacity of the embedded pipeline, the thermal resistance between the outdoor air and the building envelope, the thermal resistance between the indoor air and the building envelope, the thermal resistance of the exterior window, the thermal resistance between the embedded pipeline and the floor, the thermal resistance between the embedded pipeline and the soil, the heat capacity of water, the conversion coefficient of solar radiation, and the area of the exterior window; the historical environmental factor information includes the indoor temperature information and the outdoor temperature information; wherein, the construction module 820 includes: a first construction sub-module, a second construction sub-module, a third construction sub-module, a fourth construction sub-module, a fifth construction sub-module, and a sixth construction sub-module.

[0187] The first construction sub-module is configured to construct a first balance equation according to the temperature information of the building envelope, the indoor temperature information, the heat capacity of the building envelope, the thermal resistance between the outdoor air and the building envelope, the thermal resistance between the indoor air and the building envelope, and the first process disturbance.

[0188] The second construction sub-module is configured to construct a second balance equation according to the temperature information of the building envelope, the indoor temperature information, the outdoor temperature information, the temperature information of the upper surface of the floor, the heat capacity of the indoor air, the thermal resistance between the indoor air and the building envelope, the thermal resistance between the indoor air and the floor, the thermal resistance of the exterior window, the first thermal disturbance parameter, and the second process disturbance.

[0189] The third construction sub-module is configured to construct a third balance equation according to the indoor temperature information, the temperature information of the upper surface of the floor, the temperature information of the lumped nodes of the embedded pipeline plane, the area of the exterior window, the conversion coefficient of solar radiation, the heat capacity of the floor, the thermal resistance between the indoor air and the floor, the thermal resistance between the embedded pipeline and the floor, the second thermal disturbance parameter, and the third process disturbance.

[0190] The fourth construction sub-module is configured to construct a fourth balance equation according to the temperature information of the soil, the temperature information of the lumped nodes of the embedded pipeline plane, the temperature information of the upper surface of the floor, the heat capacity of the embedded pipeline, the thermal resistance between the embedded pipeline and the floor, the thermal resistance between the embedded pipeline and the soil, the heat supply, and the fourth process disturbance.

[0191] The fifth construction sub-module is configured to construct a fifth balance equation according to the supply water temperature information, the return water temperature information, the heat capacity of water, and the water flow rate and heat supply.

[0192] The sixth construction sub-module is configured to construct a first model according to the first balance equation, the second balance equation, the third balance equation, the fourth balance equation, and the fifth balance equation.

[0193] According to an embodiment of the present invention, the first obtaining module 830 includes: a fifth obtaining sub-module and a sixth obtaining sub-module.

[0194] The fifth obtaining sub-module is configured to perform continuous-time processing on the variable parameters of the first model to obtain a continuous-time first model.

[0195] The sixth obtaining sub-module is configured to input historical energy supply information, attribute parameter information, and historical environmental factor information into the continuous-time first model to obtain target parameter information.

[0196] Based on the above training method of the target model for a building radiant energy supply system, the present invention also provides a training device for the target model of a building radiant energy supply system. The following will be combined with Figure 9 to describe this device in detail.

[0197] Figure 9 The structural block diagram of the training device for the target model of a building radiant energy supply system according to an embodiment of the present invention is shown.

[0198] As Figure 9 shown, the training device 900 for the target model of the building radiant energy supply system in this embodiment includes a second acquisition module 910, an output module 920, and a training module 930.

[0199] The second acquisition module 910 is configured to acquire training samples, where the training samples include sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample perturbation parameter information. In one embodiment, the second acquisition module 910 can be used to perform the operation S710 described above, which will not be elaborated here.

[0200] The output module 920 is configured to process the sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample perturbation parameter information by using the target model, and output target parameter information. In one embodiment, the output module 920 can be used to perform the operation S720 described above, which will not be elaborated here.

[0201] The training module 930 is configured to, when the target parameter information is determined, train the target model by using a feedforward neural network to obtain a trained target model. In one embodiment, the training module 930 can be used to perform the operation S730 described above, which will not be elaborated here.

[0202] According to an embodiment of the present invention, any number of modules among modules, sub-modules, units, and sub-units can be combined and implemented in one module, or any one of them can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of modules, sub-modules, units, and sub-units can be at least partially implemented as a hardware circuit, such as a field-programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by any other reasonable means such as hardware or firmware through circuit integration or packaging, or can be implemented in any one of the three implementation manners of software, hardware, and firmware, or in an appropriate combination of any several of them. Alternatively, at least one of modules, sub-modules, units, and sub-units can be at least partially implemented as a computer program module, and when the computer program module is run, corresponding functions can be executed.

[0203] Figure 10 A block diagram of an electronic device suitable for implementing a building load prediction method considering random thermal disturbances according to an embodiment of the present invention is shown.

[0204] As Figure 10 shown, the electronic device 1000 according to an embodiment of the present invention includes a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage section 1008 into a RAM (Random Access Memory). The processor 1001 can include, for example, a general microprocessor (such as a CPU), an instruction set processor, and / or a related chipset, and / or a dedicated microprocessor (such as an application-specific integrated circuit (ASIC)), etc. The processor 1001 can also include on-board memory for caching purposes. The processor 1001 can include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0205] In the RAM 1003, various programs and data required for the operation of the electronic device 1000 are stored. The processor 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. The processor 1001 performs various operations of the method flow according to the embodiments of the present invention by executing the programs in the ROM 1002 and / or the RAM 1003. It should be noted that the programs may also be stored in one or more memories other than the ROM 1002 and the RAM 1003. The processor 1001 may also perform various operations of the method flow according to the embodiments of the present invention by executing the programs stored in the one or more memories.

[0206] According to an embodiment of the present invention, the electronic device 1000 may further include an input / output (I / O) interface 1005, and the I / O interface 1005 is also connected to the bus 1004. The electronic device 1000 may further include one or more of the following components connected to the I / O interface 1005: an input portion 1006 including a keyboard, a mouse, etc.; an output portion 1007 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage portion 1008 including a hard disk, etc.; and a communication portion 1009 including a network interface card such as a LAN card, a modem, etc. The communication portion 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed so that a computer program read from it can be installed into the storage portion 1008 as needed.

[0207] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist separately without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the one or more programs are executed, the above method according to the embodiments of the present invention is implemented.

[0208] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present invention, the computer-readable storage medium may be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or apparatus. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the ROM 1002 and / or RAM 1003 described above and / or one or more memories other than the ROM 1002 and RAM 1003.

[0209] An embodiment of the present invention further includes a computer program product, which includes a computer program that contains program code for executing the method shown in the flowchart. When the computer program product runs in a computer system, the program code is used to enable the computer system to implement the building load prediction method considering random thermal disturbances provided by the embodiments of the present invention.

[0210] When the computer program is executed by the processor 1001, it executes the above functions defined in the system / apparatus of the embodiments of the present invention. According to an embodiment of the present invention, the above-described systems, apparatuses, modules, units, etc. can be implemented by computer program modules.

[0211] In one embodiment, the computer program can rely on tangible storage media such as optical storage devices and magnetic storage devices. In another embodiment, the computer program can also be transmitted and distributed in the form of a signal on a network medium, and be downloaded and installed through the communication part 1009, and / or be installed from the removable medium 1011. The program code contained in the computer program can be transmitted by any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0212] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1009, and / or be installed from the removable medium 1011. When the computer program is executed by the processor 1001, it executes the above functions defined in the system of the embodiments of the present invention. According to an embodiment of the present invention, the above-described systems, devices, apparatuses, modules, units, etc. can be implemented by computer program modules.

[0213] According to embodiments of the present invention, program code for executing the computer programs provided by the embodiments of the present invention can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages include, but are not limited to, such as Java, C++, Python, the "C" language, or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., by connecting through the Internet using an Internet service provider).

[0214] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and combinations of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0215] Those skilled in the art can understand that the features described in the various embodiments of the present invention can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention can be combined and / or combined in various ways. All such combinations and / or combinations fall within the scope of the present invention.

[0216] The above describes the embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although the embodiments are described separately above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. Without departing from the scope of the present invention, those skilled in the art can make various substitutions and modifications, and all such substitutions and modifications should fall within the scope of the present invention.

Claims

1. A building load prediction method considering random thermal disturbances, characterized in that: The method comprises: In response to receiving a building load forecast demand for a target period, acquiring historical operation information, attribute parameter information, and historical environmental factor information of a building radiation energy supply system for a target building corresponding to the target period; A first model is constructed according to the historical operation information, the attribute parameter information and the historical environmental factor information, wherein the first model includes variable parameters, the historical operation information includes temperature information of the enclosure structure, temperature information of the upper and lower surfaces of the floor, temperature information of the lumped nodes of the embedded pipe plane, water supply temperature information, return water temperature information and water flow and heating capacity; the attribute parameter information includes heat capacity of the enclosure structure, heat capacity of indoor air, heat capacity of the floor, heat capacity of the embedded pipe, thermal resistance between outdoor air and the enclosure structure, thermal resistance between indoor air and the enclosure structure, thermal resistance of the external window, thermal resistance between the embedded pipe and the floor, thermal resistance between the embedded pipe and the soil, heat capacity of water, conversion coefficient of solar radiation and area of ​​the external window; the historical environmental factor information includes indoor temperature information and outdoor temperature information; the variable parameters represent the variable parameters in the first model; Performing continuous time processing on the variable parameters of the first model to obtain target parameter information includes: performing continuous time processing on the variable parameters of the first model to obtain a continuous time first model; inputting the historical operation information, the attribute parameter information and the historical environmental factor information into the continuous time first model to obtain target parameter information; Processing the target parameter information according to the target model to obtain the thermal disturbance prediction value of the target building includes: coupling the first model with the second model to obtain the target model; in the target model, determining the indoor temperature prediction value of the target time period according to the initial disturbance prediction value output by the second model; determining the loss function value according to the indoor temperature prediction value of the target time period and the actual indoor temperature value of the target time period; and obtaining the thermal disturbance prediction value of the target building when the loss function value is less than a preset loss function value, wherein the second model is obtained by training the initial model using the thermal disturbance prediction value as a label and using sample historical operation information, sample historical environmental factor information and sample time parameter information; The thermal disturbance prediction value of the target building is input into the first model to obtain the load prediction value of the target building.

2. The method according to claim 1, characterized in that The step of determining the loss function value according to the indoor temperature prediction value of the target time period and the indoor temperature actual value of the target time period comprises: Determining a target loss function according to the predicted value of the indoor temperature during the target period and the actual value of the indoor temperature during the target period; According to the target loss function, the loss function value is determined.

3. The method according to claim 1, characterized in that: The first model further includes: a constant parameter, wherein coupling the first model with the second model to obtain the target model includes: Determining the target parameter information according to the constant parameters of the first model and the variable parameters; According to the target parameter information and the initial disturbance prediction value, the first model and the second model are coupled to obtain a target model.

4. The method according to claim 1, characterized in that: The constructing of a first model according to the historical operation information, the attribute parameter information and the historical environmental factor information includes: Constructing a first equilibrium equation according to the temperature information of the enclosure structure, the indoor temperature information, the heat capacity of the enclosure structure, the thermal resistance between the outdoor air and the enclosure structure, the thermal resistance between the indoor air and the enclosure structure, and the first process disturbance; Constructing a second equilibrium equation according to the temperature information of the enclosure structure, the indoor temperature information, the outdoor temperature information, the temperature information of the floor surface, the heat capacity of the indoor air, the thermal resistance between the indoor air and the enclosure structure, the thermal resistance between the indoor air and the floor, the thermal resistance of the external window, the first thermal disturbance parameter and the second process disturbance; A third equilibrium equation is constructed according to the indoor temperature information, the temperature information of the upper surface of the floor, the temperature information of the lumped node of the embedded pipe plane, the area of ​​the external window, the conversion coefficient of the solar radiation, the heat capacity of the floor, the thermal resistance between the indoor air and the floor, the thermal resistance between the embedded pipe and the floor, the second thermal disturbance parameter and the third process disturbance; Constructing a fourth equilibrium equation according to the temperature information of the soil, the temperature information of the lumped nodes of the embedded pipe plane, the temperature information of the upper surface of the floor, the heat capacity of the embedded pipe, the thermal resistance between the embedded pipe and the floor, the thermal resistance between the embedded pipe and the soil, the heat supply and the fourth process disturbance; constructing a fifth balance equation according to the water supply temperature information, the return water temperature information, the heat capacity of the water, the water flow rate and the heat supply; The first model is constructed according to the first equilibrium equation, the second equilibrium equation, the third equilibrium equation, the fourth equilibrium equation and the fifth equilibrium equation.

5. A training method for a target model based on a building radiation energy supply system, characterized in that: The method comprises: Acquire a training sample, wherein the training sample includes sample historical operation information, sample attribute parameter information, sample historical environmental factor information, and sample disturbance parameter information; Processing the sample historical operation information, the sample attribute parameter information, the sample historical environmental factor information and the sample disturbance parameter information using the target model, and outputting target parameter information; When the target parameter information is determined, the target model is trained using a feedforward neural network to obtain a trained target model, wherein the trained target model is the target model described in any one of claims 1 to 4.

6. A building load prediction system considering random thermal disturbances, characterized in that: The system comprises: A first acquisition module is used to, in response to receiving a building load forecast demand for a target period, acquire historical operation information, attribute parameter information and historical environmental factor information of a building radiation energy supply system for a target building corresponding to the target period; A construction module is used to construct a first model according to the historical operation information, the attribute parameter information and the historical environmental factor information, wherein the first model includes variable parameters, the historical operation information includes temperature information of the enclosure structure, temperature information of the upper surface of the floor, temperature information of the lumped node of the embedded pipe plane, water supply temperature information, return water temperature information and water flow and heating capacity; the attribute parameter information includes heat capacity of the enclosure structure, heat capacity of indoor air, heat capacity of the floor, heat capacity of the embedded pipe, thermal resistance between outdoor air and enclosure structure, thermal resistance between indoor air and enclosure structure, thermal resistance of external window, thermal resistance between embedded pipe and floor, thermal resistance between embedded pipe and soil, heat capacity of water, conversion coefficient of solar radiation and area of ​​external window; the historical environmental factor information includes indoor temperature information and outdoor temperature information; the variable parameters represent the variable parameters in the first model; A first obtaining module is used to perform continuous time processing on the variable parameters of the first model to obtain target parameter information; The first obtaining module includes: a fifth obtaining submodule and a sixth obtaining submodule; The fifth obtaining submodule is used for performing continuous time processing on the variable parameters of the first model to obtain a continuous time first model; The sixth obtaining submodule is used to input the historical operation information, the attribute parameter information and the historical environmental factor information into the continuous time first model to obtain target parameter information; A second obtaining module is used to process the target parameter information according to a target model to obtain a thermal disturbance prediction value of the target building, wherein the target model is constructed by coupling the first model and the second model, and the second model is obtained by training the historical operation information, the historical environmental factor information and the time parameter information with the thermal disturbance prediction value as a label; The second obtaining module includes: a first obtaining submodule, a second obtaining submodule, a third obtaining submodule and a fourth obtaining submodule; The first obtaining submodule is used to couple the first model with the second model to obtain the target model; The second obtaining submodule is used to determine the indoor temperature prediction value of the target time period in the target model according to the initial disturbance prediction value output by the second model; The third obtaining submodule is used to determine the loss function value according to the indoor temperature prediction value of the target period and the indoor temperature actual value of the target period; The fourth obtaining submodule is used to obtain the thermal disturbance prediction value of the target building when the loss function value is less than a preset loss function value; The third obtaining module is used to input the thermal disturbance prediction value of the target building into the first model to obtain the load prediction value of the target building.

7. A training device based on a target model of a building radiation energy supply system, characterized in that: The device comprises: A second acquisition module is used to acquire training samples, wherein the training samples include sample historical operation information, sample attribute parameter information, sample historical environmental factor information and sample disturbance parameter information; An output module, used to process the sample historical operation information, the sample attribute parameter information, the sample historical environmental factor information and the sample disturbance parameter information using the target model, and output target parameter information; A training module is used to train the target model using a feedforward neural network when the target parameter information is determined, so as to obtain a trained target model, wherein the trained target model is the target model described in any one of claims 1 to 4.

8. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 4 or 5.

Images