Thermal load prediction method based on artificial intelligence

Abstract

The invention provides a thermal load prediction method based on artificial intelligence, and the method comprises the steps: carrying out the decomposition processing of a heat supply structure of a standardized temporary building, and obtaining a plurality of heat supply regions and a thermal space relation, and the thermal space relation at least comprises a mutual heating relation and a mutual heat preservation relation; based on the heat space relation of the heat supply areas, the environment temperature is input into a heat load prediction model to obtain the heat load demand of each heat supply area; and based on the thermal load demand and the temporary building structure, building and calculating heat supply tree nodes of the multi-stage heat supply pipeline to obtain control parameters in the multi-stage heat supply pipeline.

Description

Technical Field

[0001] The present invention relates to data processing technologies, and in particular, to a heat load prediction method based on artificial intelligence. Background Art

[0002] In a cold environment, in order to ensure the comfort inside a temporary building, heating is required. However, due to the structural characteristics of the temporary building and the complex heat conduction and heat preservation relationships between different regions, how to accurately predict the heat load demand of each heating region, rationally design the heating pipeline system, and achieve efficient and energy-saving heating has become an urgent problem to be solved.

[0003] In the prior art, for the heating management of temporary buildings, empirical methods are mostly adopted, lacking intelligent and automated processing of heat loads, unable to accurately consider the relationships between different heating regions, resulting in inaccurate heat load calculations, causing energy waste or insufficient heating.

[0004] Therefore, how to intelligently predict the heat load demand of the heating region by combining building structure data and improve the accuracy of calculating heat load demand data has become an urgent problem to be solved. Summary of the Invention

[0005] An embodiment of the present invention provides a heat load prediction method based on artificial intelligence, which can intelligently predict the heat load demand of the heating region by combining building structure data and improve the accuracy of calculating heat load demand data.

[0006] In a first aspect of an embodiment of the present invention, a heat load prediction method based on artificial intelligence is provided, including: Performing decomposition processing on the heating structure of a standardized temporary building to obtain a plurality of heating regions and heat space relationships, where the heat space relationships at least include a mutual heating relationship and a mutual heat preservation relationship; Inputting the heat space relationship and the environmental temperature of the heating region into a heat load prediction model to obtain the heat load demand of each heating region; Based on the heat load demand and the temporary building structure, constructing and calculating the heating tree nodes of a multi-stage heating pipeline to obtain the control parameters in the multi-stage heating pipeline.

[0007] Optionally, the performing decomposition processing on the heating structure of a standardized temporary building to obtain a plurality of heating regions and heat space relationships includes: Decomposing the main structure of the temporary building to obtain a plurality of sub-structures, and taking the sub-structure with a heating module as the heating structure; Determining the heating region corresponding to each heating structure, and obtaining the heat space relationship according to the position of the heating region.

[0008] Optionally, the sub-structure is a standard modular plate structure, and each sub-structure corresponds to a modular room.

[0009] Optionally, the determining the heating area corresponding to each heating structure includes: Traverse each heating area in turn as the first heating area, and determine the adjacent areas of the first heating area. The adjacent areas include the areas corresponding to the 6 faces of the space formed by the first heating area; If the adjacent area corresponds to the heating area, then use the adjacent area as the second heating area corresponding to the first heating area.

[0010] Optionally, the obtaining the thermal space relationship according to the position of the heating area includes: If it is determined that the second heating area is located above or below the first heating area, then it is determined that the second heating area and the first heating area have a mutual heating relationship and a mutual heat preservation relationship; If it is determined that the second heating area is not located above or below the first heating area, then it is determined that the second heating area and the first heating area have a mutual heat preservation relationship.

[0011] Optionally, the thermal load requirements of each heating area are obtained by inputting the thermal space relationship and the environmental temperature of the heating area into the thermal load prediction model. The thermal space relationship includes an adjacent relationship, and includes: The thermal load prediction model obtains a set of heating relationships under corresponding temperature conditions based on the environmental temperature and the standard temperature. The set of heating relationships contains the thermal load demand values of heating areas with different thermal space relationships; Determine the thermal load demand value based on the thermal space relationship of each first heating area.

[0012] Optionally, the construction and calculation of the heating tree nodes for the multi-stage heating pipeline are performed based on the thermal load demand and the temporary building structure, and the control parameters in the multi-stage heating pipeline are obtained, including: Construct a space structure tree corresponding to the space of the temporary building structure. The space structure tree includes different layers and successively connects a first node, a second node, and a third node; According to the first heating area and the pipeline relationship of the multi-stage heating pipeline, perform transformation processing on the space structure tree to obtain a heating tree. The first node corresponds to the heat source primary pipeline, the second node corresponds to the secondary pipeline, and the third node corresponds to the heating module of each heating area.

[0013] Optionally, the constructing the space structure tree corresponding to the space of the temporary building structure includes: Construct a first node corresponding to the temporary building structure; Obtain the number of floors and columns of the temporary building, and construct second nodes corresponding to each floor and each column; Determine the sub-structures corresponding to each layer and each column, and construct the third nodes corresponding to each sub-structure.

[0014] Optionally, the transformation process of the spatial structure tree according to the first heating area and the pipeline relationship of the multi-stage heating pipelines to obtain the heating tree includes: If it is determined that there is 1 heat source, then the first node is corresponding to the primary pipeline of the heat source; If it is determined that there are multiple heat sources, then a parent node is newly created and the first nodes corresponding to other heat sources are newly created. After each first node is corresponding to the primary pipeline of the heat source, they are respectively connected to the parent node; Split the corresponding second nodes and connect them to the newly created first nodes.

[0015] Optionally, the construction and calculation of the heating tree nodes for the multi-stage heating pipelines based on the heat load demand and the temporary building structure to obtain the control parameters in the multi-stage heating pipelines includes: Statistically sum up the heat load demand values of all the third nodes corresponding to each second node to obtain the pipeline heat load value; Statistically sum up the heat load demand values of the second nodes corresponding to each third node to obtain the heat source heat load value; Based on the temperature of the heat medium produced by the current heat source, the heat source heat load value, and the pipeline heat load value, calculate the corresponding flow rates of the primary pipeline and the secondary pipeline of the heat source in sequence.

[0016] Optionally, the control parameters at least include the heat source temperature and the flow rate of the pipeline.

[0017] In the second aspect of the embodiments of the present invention, a training method for a heat load prediction model applicable to the first aspect is provided, including: Determine the first sub-structure to be tested and the second sub-structure for cooperative testing; Assemble the first sub-structure and the second sub-structure in different modes in sequence to obtain the combined structure states of different thermal space relationships of the first sub-structure; Adjust the environmental temperature, the heat source temperature, and the flow rate of the first sub-structure or the second sub-structure to obtain the heating relationship set of the first sub-structure under different combined structure states.

[0018] Optionally, the adjustment of the environmental temperature, the heat source temperature, and the flow rate of the first sub-structure or the second sub-structure to obtain the heating relationship set of the first sub-structure under different combined structure states includes: Fix the environmental temperature, the internal space of the first sub-structure, and the test temperature of the second sub-structure; If it is determined that the combined structure state does not have the second sub-structure, then directly open the flow valve of the first sub-structure; After the first sub-structure reaches the test temperature and stabilizes, the heat source temperature and flow rate under the corresponding combined structure state are obtained, and the heat load demand value is calculated based on the heat source temperature and flow rate.

[0019] Optionally, it further includes: If it is determined that there is a second sub-structure in the combined structure state, the flow valve corresponding to the second sub-structure is opened, and after the second sub-structure reaches the test temperature and stabilizes, the flow valve of the first sub-structure is opened.

[0020] Optionally, if it is determined that the first sub-structure or the second sub-structure does not reach the test temperature after the flow valve reaches the preset opening value, the heat source temperature is increased by a preset temperature value. Beneficial effects

[0021] In the present invention, by carefully decomposing the heating structure of the standardized temporary building, the main structure of the temporary building is first decomposed into multiple sub-structures, and the sub-structures with heating modules are selected as the heating structures. Then, the heating areas corresponding to each heating structure are determined, and the positional relationships of the heating areas are comprehensively analyzed to obtain the thermal space relationships (including mutual heating relationships, mutual heat preservation relationships, etc.). Based on these accurate heating area and thermal space relationship data, combined with the environmental temperature, it is input into the heat load prediction model. The heat load prediction model generates a heating relationship set based on the environmental temperature and the standard temperature, and then determines the heat load demand value according to the thermal space relationship of each heating area, realizing the accurate prediction of the heat load demand of each heating area. For heating areas with different positional relationships (such as there are mutual heating and mutual heat preservation relationships in the up and down positions, and there are mutual heat preservation relationships in non-up and down positions), the model can accurately consider these factors and obtain a more realistic heat load demand. It can intelligently predict the heat load demand of the heating area combined with the building structure data, improving the accuracy of calculating the heat load demand data.

[0022] Based on the accurate heat load demand prediction and the understanding of the temporary building structure, the present invention constructs and calculates the heating tree nodes for the multi-stage heating pipeline. First, a spatial structure tree corresponding to the temporary building structure space is constructed, including a first node, a second node, and a third node. Then, according to the pipeline relationship between the first heating area and the multi-stage heating pipeline, it is transformed into a heating tree, so that the first node corresponds to the heat source primary pipeline, the second node corresponds to the secondary pipeline, and the third node corresponds to the heating module of each heating area. The pipeline heat load value is obtained by statistically summing the heat load demand values of all the third nodes corresponding to each second node, and the heat source heat load value is obtained by statistically summing the heat load demand values of the second nodes corresponding to each third node. Furthermore, control parameters such as the flow rate of the heat source primary pipeline and the secondary pipeline are calculated based on the temperature of the heat medium produced by the heat source, the heat source heat load value, and the pipeline heat load value. When there are multiple heat sources, the heating tree structure is improved by newly building a parent node and multiple first nodes and reasonably connecting them, as well as splitting and connecting the second nodes.

[0023] The training method of the heat load prediction model provided by the present invention determines the first sub-structure to be tested and the second sub-structure for cooperative testing, and sequentially assembles them in different modes to obtain the combined structure states with different thermal spatial relationships of the first sub-structure. Adjust the ambient temperature, heat source temperature, and the flow rate of the first sub-structure or the second sub-structure. Fix the ambient temperature and the test temperature, control the opening sequence of the flow valves according to whether the second sub-structure exists in the combined structure state. After reaching the test temperature and stabilizing, obtain the heat source temperature and the flow rate, and calculate the heat load demand value, so as to obtain the set of heating relationships of the corresponding first sub-structure under different combined structure states. This training method comprehensively considers various possible thermal spatial relationships and heating situations, and can effectively improve the accuracy and adaptability of the heat load prediction model. During the training process, various positional relationships (such as up and down, left and right, front and back, etc.) between the second sub-structure and the first sub-structure and different numbers of the second sub-structure are tested, enabling the model to better handle various complex temporary building structures and environmental conditions, providing strong support for the intelligent control of the heating system, enabling the heating system to more flexibly and accurately adjust the heating parameters according to the actual situation, and improving the heating quality and energy utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a schematic flow chart of a heat load prediction method based on artificial intelligence provided by an embodiment of the present invention; Figure 2 is a schematic diagram of a temporary building provided by an embodiment of the present invention; Figure 3 is a schematic diagram of a heating tree provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0026] See Figure 1 , which is a schematic flow chart of a heat load prediction method based on artificial intelligence provided by an embodiment of the present invention. The method includes: S1, decomposing the heating structure of the standardized temporary building to obtain multiple heating areas and thermal spatial relationships, where the thermal spatial relationships at least include mutual heating relationships and mutual insulation relationships.

[0027] For standardized temporary buildings, the server decomposes their heating structures, refining the heating structures of the temporary buildings into multiple heating zones and clarifying the thermal space relationships between the heating zones. The thermal space relationships include mutual heating relationships and mutual heat insulation relationships, etc. This step is the basis for subsequent analysis and optimization of the heating systems of temporary buildings. By obtaining accurate heating zones and thermal space relationships, it can provide a data basis for the prediction of heat loads.

[0028] In some embodiments, the decomposition process of the heating structure of the standardized temporary building to obtain multiple heating zones and thermal space relationships includes: S11, decompose the main structure of the temporary building to obtain multiple sub-structures, and use the sub-structures with heating modules as the heating structures.

[0029] See Figure 2 , the sub-structure is a standard modular panel structure, and each sub-structure corresponds to a modular room.

[0030] It should be noted that temporary buildings can be, for example, buildings for people to live in scenarios such as construction sites. The server decomposes the main structure of the temporary building to obtain multiple sub-structures. These sub-structures are all standard modular panel structures, and each sub-structure corresponds to a modular room. Among these sub-structures, the server filters out the sub-structures with heating modules. In this way, the specific parts of the temporary building that need heating are clarified.

[0031] S12, determine the heating zone corresponding to each heating structure, and obtain the thermal space relationship according to the position of the heating zone.

[0032] After determining the heating structure, the server further determines the heating zone corresponding to each heating structure and obtains the thermal space relationship based on the position of the heating zone. This process can clarify the interaction relationships between different heating zones.

[0033] Among them, the determination of the heating zone corresponding to each heating structure includes: S121, traverse each heating zone in turn as the first heating zone, and determine the adjacent zones of the first heating zone. The adjacent zones include the zones corresponding to the 6 faces of the space formed by the first heating zone.

[0034] The server traverses each heating zone in turn, taking the currently traversed heating zone as the first heating zone. For the first heating zone, the server determines the zones corresponding to the 6 faces of the space it forms as the adjacent zones. In this way, the surrounding environment of each heating zone is comprehensively determined, laying a foundation for subsequent judgment of the thermal space relationship.

[0035] It is worth mentioning that the adjacent regions corresponding to different subspaces are different. For example, there are various situations as follows. The adjacent space of the subspace can be one of the situations such as above, below, left, right, front, back, above and below, left and right, front and back, upper left, upper and lower left, upper and lower right, etc.

[0036] S122. If the adjacent region corresponds to the heating region, then the adjacent region is used as the second heating region corresponding to the first heating region.

[0037] The server determines the adjacent region of the first heating region. If the adjacent region corresponds to the heating region (that is, the adjacent region is also a region that needs heating), then the adjacent region is used as the second heating region corresponding to the first heating region. By determining the second heating region, the association between different heating regions can be further clarified, so as to accurately obtain the thermal space relationship.

[0038] Among them, obtaining the thermal space relationship according to the position of the heating region includes: S123. If it is determined that the second heating region is located above or below the first heating region, then it is determined that the second heating region and the first heating region have a mutual heating relationship and a mutual heat preservation relationship.

[0039] The server determines the spatial positions of the determined first heating region and the second heating region. When the server detects that the second heating region is above or below the first heating region, considering the floor heating layer structure factor in the actual heating scenario, heat can be transferred between the heating regions in the upper and lower positions. Based on this, the server determines that there is both a mutual heating relationship and a mutual heat preservation relationship between the second heating region and the first heating region. This mutual heating relationship means that heat can flow bidirectionally between the two regions, and the mutual heat preservation relationship means that the two regions can reduce heat loss from each other, thus affecting each other's heat load requirements.

[0040] S124. If it is determined that the second heating region is not located above or below the first heating region, then it is determined that the second heating region and the first heating region have a mutual heat preservation relationship.

[0041] When the server determines that the second heating area is not above or below the first heating area, that is, when they are in other relative position relationships such as left - right, front - back, etc. In this case, since there is no special heat transfer structure (such as a floor heating layer) like the up - down position, the main heat interaction between the two heating areas is manifested as heat exchange and retention through structures such as walls. Therefore, the server determines that there is only a mutual heat insulation relationship between the second heating area and the first heating area. This mutual heat insulation relationship means that heat loss to the external environment can be reduced between the two areas through structures such as walls, which in turn affects the heat load requirements of each area. The server accurately identifying this thermal space relationship helps to more precisely analyze the heat load situation of the heating area.

[0042] S2. Based on the thermal space relationship of the heating area and the ambient temperature, input them into the heat load prediction model to obtain the heat load requirements of each heating area; The server inputs the previously determined thermal space relationship of the heating area (including mutual heating relationship, mutual heat insulation relationship, adjacent relationship, etc.) and the current ambient temperature data into the heat load prediction model together.

[0043] The heat load prediction model is an intelligent model trained and optimized with a large amount of data. It can comprehensively consider various factors and accurately predict the heat load requirements of each heating area. By taking the thermal space relationship and the ambient temperature as input parameters, the model can simulate the heat transfer and consumption conditions of the heating area in different situations, so as to obtain the heat load requirements of each heating area. There are embodiments for training the heat load prediction model later.

[0044] In some embodiments, for the step of inputting the thermal space relationship of the heating area and the ambient temperature into the heat load prediction model to obtain the heat load requirements of each heating area, where the thermal space relationship includes an adjacent relationship, it includes: S21. The heat load prediction model obtains a heating relationship set corresponding to the temperature condition based on the ambient temperature and the standard temperature. The heating relationship set contains the heat load demand values of heating areas with different thermal space relationships.

[0045] The heat load prediction model first obtains the current ambient temperature data and combines it with a pre - set standard temperature (such as 24 degrees). Based on these two temperature parameters, the model generates a heating relationship set corresponding to the temperature condition. This heating relationship set is a database containing the heat load demand values of heating areas with different thermal space relationships (such as mutual heating, mutual heat insulation, adjacent, etc.). These heat load demand values are obtained through the simulation and analysis of a large number of actual heating scenarios, and there are relevant elaborations in subsequent embodiments. The heating relationship set provides basic data and a reference basis for determining the heat load demand value based on the thermal space relationship later.

[0046] S22. Determine the heat load demand value based on the heat space relationship of each first heating area.

[0047] The server extracts the corresponding heat load demand value from the heating relationship set based on the specific heat space relationship of each first heating area.

[0048] Since different heat space relationships will result in different heat transfer and consumption situations between heating areas, the heat space relationship of each first heating area corresponds to a specific heat load demand value in the heating relationship set. Through matching, the server can determine the heat load demand value of each first heating area, ensuring that the heating system can provide precise heating according to the actual demand, improving energy utilization efficiency and heating effect.

[0049] S3. Based on the heat load demand and the temporary building structure, construct and calculate the heating tree nodes of the multi-level heating pipeline to obtain the control parameters in the multi-level heating pipeline.

[0050] After the server obtains the heat load demand of each heating area and masters the specific structural information of the temporary building, it starts to construct the heating tree nodes of the multi-level heating pipeline.

[0051] By constructing reasonable heating tree nodes and calculating the control parameters in the multi-level heating pipeline, such as heat source temperature, pipeline flow rate, etc., the effective control and optimization of the heating pipeline server can be achieved. The heating system can provide precise heating according to the actual demand of each heating area, improve energy utilization efficiency, and at the same time ensure the stability and reliability of heating.

[0052] In some embodiments, the constructing and calculating the heating tree nodes of the multi-level heating pipeline based on the heat load demand and the temporary building structure to obtain the control parameters in the multi-level heating pipeline includes: S31. Construct a space structure tree corresponding to the space of the temporary building structure, and the space structure tree includes different layers and successively connects a first node, a second node, and a third node.

[0053] See Figure 3 , the server constructs a space structure tree in order to more clearly present the relationship between the temporary building structure and the heating pipeline server. This space structure tree is composed of nodes at different levels, and the nodes are successively connected to form a tree-like structure with distinct levels and clear structure.

[0054] Among them, the constructing the space structure tree corresponding to the space of the temporary building structure includes: S311. Construct a first node corresponding to the temporary building structure.

[0055] The server first creates a first node corresponding to the overall structure of the temporary building. This first node can be regarded as the root node of the entire spatial structure tree. It represents the overall structure of the temporary building and is the basis and starting point for constructing other nodes later.

[0056] S312. Obtain the number of floors and columns of the temporary building, and construct second nodes corresponding to each floor and column.

[0057] The server obtains the information about the number of floors and columns of the temporary building. Based on this information, the server constructs a corresponding second node for each floor and column of the temporary building. Each second node corresponds to a specific floor and column position in the temporary building. There is 1 heating pipeline corresponding to each floor and column. For example, there is 1 heating pipeline corresponding to the 1st floor and 1st column, and 1 heating pipeline corresponding to the 2nd floor and 2nd column. By constructing the second nodes, the server can accurately map the spatial hierarchical structure of the temporary building in the spatial structure tree.

[0058] S313. Determine the sub-structures corresponding to each floor and column, and construct third nodes corresponding to each sub-structure.

[0059] The server determines the sub-structures corresponding to each floor and column (i.e., the modular rooms and the like obtained by previous decomposition). For each sub-structure, the server constructs a corresponding third node. The third nodes are at the bottom layer of the spatial structure tree. They correspond to the specific heating areas (sub-structures) and represent the individual specific spatial units in the temporary building.

[0060] S32. According to the first heating area and the pipeline relationship of the multi-level heating pipelines, perform transformation processing on the spatial structure tree to obtain a heating tree, where the first node corresponds to the heat source primary pipeline, the second node corresponds to the secondary pipeline, and the third node corresponds to the heating module of each heating area.

[0061] The server performs transformation processing on the previously constructed spatial structure tree according to the distribution of the first heating area and the pipeline relationship between the multi-level heating pipelines, so as to obtain a heating tree. In the heating tree, each node corresponds to a different part of the heating pipeline server, forming a tree-like structure related to the actual heating system.

[0062] Among them, the performing transformation processing on the spatial structure tree according to the first heating area and the pipeline relationship of the multi-level heating pipelines to obtain a heating tree includes: S321. If it is determined that there is 1 heat source, then correspond the first node to the heat source primary pipeline.

[0063] When the server determines that there is only 1 heat source in the heating system of the temporary building, the server connects the first node to the heat source primary pipeline. This correspondence clarifies the connection between the heat source inlet of the heating system and the overall structure of the temporary building.

[0064] S322. If it is determined that there are multiple heat sources, a parent node is newly created and first nodes corresponding to other heat sources are newly created. After each first node is corresponded to a heat source primary pipeline, they are respectively connected to the parent node.

[0065] When the server determines that there are multiple heat sources in the heating system of a temporary building, the server will newly create a parent node and create a corresponding first node for each heat source. Then, the server corresponds and connects each first node to the corresponding heat source primary pipeline, and connects these first nodes to the parent node respectively. By newly creating a parent node and multiple first nodes, the server can effectively manage multiple heat sources.

[0066] S323. Split the corresponding second node and connect it to the newly created first node.

[0067] The server splits the corresponding second node and connects the split second node to the newly created first node. Through this connection method, the server further improves the structure of the heating tree, enabling the secondary pipeline (represented by the second node) of the heating pipeline server to be effectively connected and coordinated with the heat source primary pipeline (represented by the first node). Among them, the basis for splitting needs to be combined with the actual situation of the building. For example, if second node 1 and second node 2 correspond to first node 1, then second node 1 and second node 2 are connected to first node 1.

[0068] In some embodiments, constructing and calculating the heating tree nodes for the multi-level heating pipeline based on the heat load demand and the temporary building structure to obtain the control parameters in the multi-level heating pipeline includes: S33. Statistically calculate the sum of the heat load demand values of all third nodes corresponding to each second node to obtain the pipeline heat load value.

[0069] For the constructed heating tree structure, the server traverses and statistically calculates all third nodes (corresponding to the heating modules in each heating area) connected to each second node (corresponding to the secondary pipeline). In this process, the server sequentially obtains the heat load demand values corresponding to each third node, and these heat load demand values are calculated previously by the heat load prediction model based on factors such as the heat space relationship and environmental temperature in the heating area. The server accumulates and sums the heat load demand values of all third nodes corresponding to each second node, and the obtained total is the pipeline heat load value corresponding to the second node. Through such statistical calculation methods, the server can determine the heat load amount that each pipeline needs to bear.

[0070] S34. Statistically calculate the sum of the heat load demand values of the second nodes corresponding to each third node to obtain the heat source heat load value.

[0071] The server analyzes each third node in the heating tree to determine the corresponding second node for each third node. Then, the server obtains the heat load demand value of each third node, and accumulatively sums up the heat load demand values of the second nodes corresponding to each third node. The obtained total is the heat source heat load value. This heat source heat load value reflects the total heat that the heat source needs to provide in the entire heating process. It comprehensively considers the heat load demands of each heating area and the connection relationship of the heating pipelines.

[0072] S35. Based on the temperature of the heat medium produced by the current heat source, the heat source heat load value, and the pipeline heat load value, calculate the corresponding flow rates of the primary pipeline and the secondary pipeline of the heat source in sequence.

[0073] After the server obtains the key data such as the temperature of the heat medium produced by the current heat source, the heat source heat load value, and the pipeline heat load value, it starts to calculate the flow rates of the primary pipeline and the secondary pipeline of the heat source.

[0074] Combining the heat source heat load value and the heat medium temperature, first calculate the required flow rate of the primary pipeline of the heat source to ensure that the heat source can output sufficient heat according to the heat load demand. Then, the server further calculates the corresponding flow rate of the secondary pipeline based on parameters such as the pipeline heat load value and the flow rate of the primary pipeline, so that the heat can be reasonably distributed in the secondary pipeline and transported to each heating area.

[0075] By calculating the flow rates of each level of pipeline in this way, the server can achieve precise control of the heating pipeline server, ensure the stable operation of the heating system, and meet the heat load demands of each heating area.

[0076] Among them, the control parameters at least include the heat source temperature and the flow rate of the pipeline. During the entire calculation process, the heat source temperature and the flow rate of the pipeline are important control parameters, which directly affect the operation efficiency and heating effect of the heating system. By accurately calculating and reasonably adjusting these control parameters, the server can optimize the performance of the heating system, improve the energy utilization efficiency, and at the same time ensure that each heating area in the temporary building can obtain a stable and comfortable heating environment.

[0077] The embodiment of the present invention provides a training method for a heat load prediction model applicable to a heat load prediction energy-saving method based on artificial intelligence, including: A1. Determine the first sub-structure to be tested and the second sub-structure for cooperative testing.

[0078] The first step in the server's work of training the heat load prediction model is to determine the first sub-structure to be tested and the second sub-structure for cooperative testing. These sub-structures simulate the modular rooms in the temporary building, including their thermal characteristics and spatial relationships. With the selected first and second sub-structures, various spatial combinations can be constructed subsequently to simulate the actual heating scenarios and provide data support for model training.

[0079] A2. Assemble the first sub-structure and the second sub-structure in different modes in sequence to obtain the combined structure states of different thermal spatial relationships of the first sub-structure.

[0080] Among them, each mode corresponds to a combination situation. Since there are many situations for the position of the second sub-structure relative to the first sub-structure, such as above, below, left, right, front, back, above and below, left and right, front and back, upper left, upper and lower left, upper and lower right, etc. Due to the existence of heat insulation and mutual heating, the heating flow required by the first sub-structure is different in different situations. It can be understood that when there is a second sub-structure around the first sub-structure, the heating flow required will be reduced compared to the situation where there is no second sub-structure around.

[0081] The server assembles the two sub-structures in different modes in sequence according to the various possibilities of the position of the second sub-structure relative to the first sub-structure. Due to the heat insulation and mutual heating effects between the sub-structures, the heating flow required by the first sub-structure is different under different position relationships. When there is a second sub-structure around, the heating flow of the first sub-structure is relatively reduced. Through the assembly, the server obtains the combined structure states of different thermal spatial relationships of the first sub-structure, simulating the complex thermal spatial distribution in the temporary building and providing diverse training data for the model.

[0082] A3. Adjust the environmental temperature, heat source temperature, and the flow rate of the first sub-structure or the second sub-structure to obtain the heating relationship set of the corresponding first sub-structure under different combined structure states.

[0083] The server adjusts the environmental temperature, heat source temperature, and the flow rates of the first and second sub-structures, collects the heating relationship data of the first sub-structure under different combined structure states, and forms a heating relationship set, providing a basis for the training and optimization of the heat load prediction model.

[0084] Among them, the heating relationship set includes the heating flow rate data in various situations such as above, below, left, right, front, back, above and below, left and right, front and back, upper left, upper and lower left, upper and lower right, etc. of the second sub-structure relative to the first sub-structure. One heating relationship corresponds to one heating data.

[0085] In some embodiments, the adjustment of the environmental temperature, heat source temperature, and the flow rate of the first sub-structure or the second sub-structure to obtain the heating relationship set of the corresponding first sub-structure under different combined structure states includes: A31. Fix the ambient temperature, the temperature inside the first sub-structure, and the test temperature of the second sub-structure.

[0086] The server sets and maintains the ambient temperature, the temperature inside the first sub-structure, and the test temperature of the second sub-structure at specific values. For example, set the ambient temperature to 0 degrees and the test temperature to 24°C. Fixing these temperatures can ensure that the tests are carried out under the same standard conditions, making the test data under different combined structure states comparable and improving the accuracy and reliability of the training data.

[0087] A32. If it is determined that the combined structure state does not have a second sub-structure, directly open the flow valve of the first sub-structure.

[0088] If the server determines that the combined structure state does not have a second sub-structure, it directly opens the flow valve of the first sub-structure. This operation simulates the scenario of the first sub-structure providing heating independently. By adjusting the heating flow of the first sub-structure alone, the heating data in this state can be obtained for training the model's heat load prediction ability for a single heating area.

[0089] A33. After the first sub-structure reaches the test temperature and stabilizes, obtain the heat source temperature and flow rate under the corresponding combined structure state, and calculate the heat load demand value based on the heat source temperature and flow rate.

[0090] After the temperature of the first sub-structure reaches the test temperature and stabilizes, the server records the heat source temperature and flow rate data at this time. Based on these data, calculate the heat load demand value of the first sub-structure under this combined structure state.

[0091] In the above embodiments, it further includes: If it is determined that the combined structure state has a second sub-structure, open the flow valve corresponding to the second sub-structure, and after the temperature inside the second sub-structure reaches the test temperature and stabilizes, open the flow valve of the first sub-structure.

[0092] When the server determines that the combined structure state has a second sub-structure, since the number and position of the second sub-structures have various changes, the server first opens the flow valve corresponding to the second sub-structure to make the temperature inside the second sub-structure reach the test temperature and remain stable. This simulates the actual scenario where multiple heating areas affect each other. After the second sub-structure reaches a stable temperature, heat transfer and heat exchange will occur between it and the first sub-structure. At this time, opening the flow valve of the first sub-structure can obtain the heating data under their interaction.

[0093] If it is determined that after the flow valve reaches the preset opening value, the first sub-structure or the second sub-structure does not reach the test temperature, increase the heat source temperature by a preset temperature value.

[0094] If the server detects that after the flow valve reaches the preset opening value, the temperature of the first sub-structure or the second sub-structure still does not reach the test temperature, it indicates that the heat provided by the current heat source is insufficient. To make the sub-structure reach the test temperature, the server increases the heat source temperature by a preset temperature value and continues the test until the sub-structure temperature reaches the stable test temperature, so as to obtain complete and accurate heating data for training and optimizing the heat load prediction model.

[0095] The present invention also provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided by the above various embodiments.

[0096] Among them, the storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium facilitating the transmission of a computer program from one place to another. The computer storage medium can be any available medium accessible by a general or special purpose computer. For example, the storage medium is coupled to the processor, so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an Application Specific Integrated Circuits (ASIC). In addition, the ASIC can be located in the user equipment. Of course, the processor and the storage medium can also exist as discrete components in the communication device. The storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0097] The present invention also provides a program product, which includes execution instructions stored in a storage medium. At least one processor of the device can read the execution instructions from the storage medium, and the execution of the execution instructions by at least one processor causes the device to implement the methods provided by the above various embodiments.

[0098] In the above embodiments of the terminal or the server, it should be understood that the processor can be a Central Processing Unit (CPU), and can also be other general purpose processors, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. The general purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the present invention can be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. The heat load prediction method based on artificial intelligence is characterized by: include: Decomposing the heating structure of the standardized temporary building to obtain a plurality of heating areas and thermal space relationships, wherein the thermal space relationship at least includes a mutual heating relationship and a mutual insulation relationship; Based on the thermal space relationship and ambient temperature of the heating area, the heat load demand of each heating area is obtained by inputting into the heat load prediction model; Based on the heat load demand and temporary building structure, the heating tree nodes of the multi-stage heating pipeline are constructed and calculated to obtain the control parameters in the multi-stage heating pipeline.

2. The method according to claim 1, characterized in that: The decomposition process of the heating structure of the standardized temporary building obtains multiple heating areas and thermal space relationships, including: Decomposing the main structure of the temporary building into multiple substructures, and taking the substructure with the heating module as the heating structure; The heating area corresponding to each heating structure is determined, and the thermal space relationship is obtained according to the location of the heating area.

3. The method according to claim 2, characterized in that The substructure is a standard modular plate structure, and each substructure corresponds to a modular room.

4. The method according to claim 2, characterized in that: The step of determining the heating area corresponding to each heating structure includes: Traversing each heating area in turn as a first heating area, determining an adjacent area of ​​the first heating area, wherein the adjacent area includes areas corresponding to six surfaces of the space formed by the first heating area; If the adjacent area corresponds to the heating area, the adjacent area is used as the second heating area corresponding to the first heating area.

5. The method according to claim 4, characterized in that The step of obtaining the thermal space relationship according to the location of the heating area includes: If it is determined that the second heating area is located above or below the first heating area, then it is determined that the second heating area and the first heating area are in a mutual heating relationship and a mutual heat preservation relationship; If it is determined that the second heating area is not located above or below the first heating area, it is determined that the second heating area and the first heating area are in a mutual heat preservation relationship.

6. The method according to claim 1, characterized in that The heat load demand of each heating area is obtained by inputting the heat-spatial relationship and the ambient temperature of the heating area into the heat load prediction model, wherein the heat-spatial relationship includes an adjacent relationship, including: The heat load prediction model obtains a set of heating relations under corresponding temperature conditions based on the ambient temperature and the standard temperature, wherein the set of heating relations has heat load demand values ​​of heating areas with different thermal space relations; A heating load demand value is determined based on the thermal spatial relationship of each first heating zone.

7. The method according to claim 6, characterized in that The construction and calculation of the heating tree nodes of the multi-stage heating pipeline based on the heat load demand and the temporary building structure to obtain the control parameters in the multi-stage heating pipeline include: Constructing a space structure tree corresponding to the space of the temporary building structure, wherein the space structure tree includes different layers and sequentially connects a first node, a second node, and a third node; The spatial structure tree is transformed according to the pipeline relationship of the first heating area and the multi-stage heating pipeline to obtain a heating tree, in which the first node corresponds to the first-level pipeline of the heat source, the second node corresponds to the second-level pipeline, and the third node corresponds to the heating module of each heating area.

8. The method according to claim 7, characterized in that The step of constructing a space structure tree corresponding to the space of the temporary building structure includes: constructing a first node corresponding to the temporary building structure; Get the number of floors and columns of the temporary building, and construct the second node corresponding to each floor and column; Determine the substructure corresponding to each layer and each column, and construct a third node corresponding to each substructure.

9. The method according to claim 7, characterized in that: The step of converting the spatial structure tree according to the pipeline relationship between the first heating area and the multi-stage heating pipeline to obtain the heating tree includes: If it is determined that there is only one heat source, the first node is matched to the first-level pipeline of the heat source; If it is determined that there are multiple heat sources, a new mother node is created and new first nodes corresponding to other heat sources are created. Each first node is matched with a first-level pipeline of the heat source and then connected to the mother node respectively. The corresponding second node is split and connected to the newly created first node.

10. The method according to claim 7, characterized in that The construction and calculation of the heating tree nodes of the multi-stage heating pipeline based on the heat load demand and the temporary building structure to obtain the control parameters in the multi-stage heating pipeline include: The heat load value of the pipeline is obtained by calculating the sum of the heat load demand values ​​of all third nodes corresponding to each second node; The heat load value of the heat source is obtained by calculating the sum of the heat load demand values ​​of the second node corresponding to each third node; Based on the temperature of the heat medium produced by the current heat source, the heat load value of the heat source, and the heat load value of the pipeline, the flow rates corresponding to the first-level pipeline and the second-level pipeline of the heat source are calculated in sequence.

11. The method according to claim 10, characterized in that The control parameters include at least the heat source temperature and the flow rate of the pipeline.

12. A training method for a heat load prediction model according to any one of claims 1 to 11, characterized in that: include: Determine a first substructure to be tested and a second substructure to be matched with the test; The first substructure and the second substructure are assembled in sequence according to different modes to obtain a combined structural state with different thermal-spatial relationships of the first substructure; The ambient temperature, the heat source temperature, the flow rate of the first substructure or the second substructure are adjusted to obtain a set of heating relationships of the corresponding first substructures under different combination structure states.

13. The training method according to claim 12, characterized in that: The adjustment of the ambient temperature, the heat source temperature, the flow of the first substructure or the second substructure to obtain a heating relationship set of the corresponding first substructure in different combination structure states includes: The ambient temperature, the space inside the first substructure, and the test temperature of the second substructure are fixed; If it is determined that the second substructure does not exist in the combined structure state, the flow valve of the first substructure is directly opened; After the first substructure reaches the test temperature and stabilizes, the heat source temperature and flow rate under the corresponding combined structure state are obtained, and the heat load demand value is calculated based on the heat source temperature and flow rate.

14. The training method according to claim 13, characterized in that: Also includes: If it is determined that the combined structure state has a second substructure, the flow valve corresponding to the second substructure is opened, and the flow valve of the first substructure is opened after the test temperature is reached and stabilized in the second substructure.

15. The training method according to claim 14, characterized in that: If it is determined that the first substructure or the second substructure does not reach the test temperature after the flow valve reaches the preset opening value, the temperature of the heat source is increased by the preset temperature value.

Images