An intelligent layout method and system for building energy-consuming equipment

Abstract

The present invention discloses an intelligent layout method and system for building energy-using equipment, which relates to the field of building energy-using equipment layout. Specifically, each group of virtual scene training data includes building scene information of each virtual building scene, equipment information of multiple building energy-using equipment in each virtual building scene, and building energy-using equipment layout data; a comprehensive reward function is defined based on the energy efficiency function, spatial distribution function, equipment maintenance function, and spatial illumination function of the virtual building scene; a pre-trained initial virtual scene training model is selected; the initial virtual scene training model is trained with multiple groups of virtual scene training data, and the initial virtual scene training model is optimized based on the comprehensive reward function to obtain a virtual scene training model; a layout plan for the to-be-tested scene data is obtained through the virtual scene training model. The present invention can achieve an ideal layout effect when facing complex and changeable building environments, and improves the design efficiency.

Description

Technical Field

[0001] The present invention relates to the field of building energy - using equipment layout, and particularly to an intelligent layout method and system for building energy - using equipment. Background Art

[0002] In modern architectural design, the reasonable layout and positioning of building energy - using equipment (such as ventilation systems, air - conditioning systems, lighting equipment, and power distribution systems, etc.) are crucial, directly affecting the energy efficiency and living comfort of buildings. Traditional design methods rely on designers' experience and manual operations, which are not only time - consuming and laborious, with low design efficiency, but also prone to design errors and unreasonable layouts. Currently, using intelligent algorithms to automatically layout and position building energy - using equipment has become a research hotspot. However, when facing complex and changeable building environments, existing algorithms often fail to achieve ideal layout effects, resulting in low design efficiency and difficulty in meeting actual needs. Therefore, at present, an intelligent layout method and system for building energy - using equipment are needed, which can meet the actual complex requirements of building energy - using equipment layout, achieve ideal layout effects, and improve design efficiency. Summary of the Invention

[0003] The technical problem to be solved by the present invention is that it is difficult to achieve an ideal layout effect when facing complex and changeable building environments, resulting in low design efficiency. The purpose is to provide an intelligent layout method and system for building energy - using equipment, which solves the above - mentioned technical problems.

[0004] The present invention is achieved through the following technical solutions:

[0005] An intelligent layout method for building energy - using equipment includes:

[0006] Collecting multiple groups of virtual scene training data according to multiple virtual building scenes; each group of the virtual scene training data includes the building scene information of each virtual building scene, the equipment information of multiple building energy - using equipment in each virtual building scene, and building energy - using equipment layout data;

[0007] Defining a comprehensive reward function based on the energy - efficiency function, space - distribution function, equipment - maintenance function, space - utilization function, and space - illumination function of the virtual building scene;

[0008] Selecting a pre - trained initial virtual scene training model; training the initial virtual scene training model with multiple groups of the virtual scene training data, and optimizing the trained initial virtual scene training model based on the comprehensive reward function to obtain a virtual scene training model;

[0009] Layout the building energy - using equipment through the layout scheme of the to - be - tested scene data obtained by the virtual scene training model.

[0010] The layout data of the building energy-consuming equipment includes the orientation, position, and connection type between each piece of the building energy-consuming equipment and the corresponding virtual building scene; the connection type includes the number of connections, the specification of the connection parts, the distributable space, and the fixing method.

[0011] Obtain the actual production capacity and actual power consumption of each piece of the building energy-consuming equipment through the equipment information of each piece of the building energy-consuming equipment; define the actual energy efficiency objective function of each building energy-consuming equipment according to the ratio of the actual production capacity and the actual power consumption of each piece of the building energy-consuming equipment; use the maximization of the actual energy efficiency objective function to obtain the optimal actual energy efficiency.

[0012] Analyze the type weight through the layout data of multiple pieces of the building energy-consuming equipment; calculate the equipment weight according to the type weight and the actual power consumption; obtain the energy efficiency function of the virtual scene training model according to the optimal actual energy efficiency and the equipment weight of each piece of the building energy-consuming equipment.

[0013] The analysis of the type weight through the layout data of multiple pieces of the building energy-consuming equipment includes:

[0014] Obtain the equipment type of each piece of the building energy-consuming equipment through the equipment information of multiple pieces of the building energy-consuming equipment; score one or more of the orientation, position, and connection type according to the preset scoring rules of the equipment type of each piece of the building energy-consuming equipment, and obtain the type weight of the building energy-consuming equipment according to the scoring result.

[0015] The space distribution function is obtained through the following steps: obtain the equipment coverage area of each piece of the building energy-consuming equipment through the equipment information of multiple pieces of the building energy-consuming equipment; obtain the distribution distance between different pieces of the building energy-consuming equipment through the layout method of multiple pieces of the building energy-consuming equipment; define the space distribution weight objective function for evaluating the space distribution weight of each piece of the building energy-consuming equipment according to the equipment coverage area of each piece of the building energy-consuming equipment and the distribution distance between different pieces of the building energy-consuming equipment; use the space distribution weight objective function to calculate the optimal space distribution weight, and obtain the space distribution function of the virtual scene training model according to the optimal space distribution weight.

[0016] The equipment maintenance function is obtained through the following steps: obtain the installation position of each piece of the building energy-consuming equipment through the layout method of multiple pieces of the building energy-consuming equipment; obtain the occupied space of each piece of the building energy-consuming equipment through the equipment information of multiple pieces of the building energy-consuming equipment; define the equipment maintenance difficulty objective function for evaluating the equipment maintenance difficulty according to the installation position and the occupied space between each piece of the building energy-consuming equipment; use the equipment maintenance difficulty objective function to obtain the optimal equipment maintenance difficulty, and obtain the equipment maintenance function of the virtual scene training model according to the optimal equipment maintenance difficulty.

[0017] The space utilization rate function is obtained through the following steps: obtaining the installation positions of the building energy-using devices through the layout modes of the plurality of building energy-using devices; obtaining the occupied spaces of the building energy-using devices through the device information of the plurality of building energy-using devices; defining a device space utilization objective function for evaluating the space utilization rate of each building energy-using device according to the ratio of the occupied space of each building energy-using device to the space of the corresponding virtual building scene; obtaining the optimal space utilization rate by using the device space utilization objective function, and obtaining the space utilization rate function of the virtual scene training model according to the optimal space utilization rate.

[0018] The space illumination function is obtained through the following steps: the building energy-using devices include one or more lighting devices; obtaining the space illumination function of each lighting device according to the rated luminous flux and the lighting area of each lighting device and the preset standard illumination of the corresponding virtual building scene.

[0019] The building scene information of each virtual building scene includes: building space information and building scene type; the device information of the plurality of building energy-using devices includes: any one or more of size, shape, quantity, installation method, accessory specification and operating parameters.

[0020] An intelligent layout system for building energy-using devices includes:

[0021] A training data acquisition module: used for acquiring multiple groups of virtual scene training data according to multiple virtual building scenes; each group of the virtual scene training data includes the building scene information of each virtual building scene, the device information of the plurality of building energy-using devices in each virtual building scene, and building energy-using device layout data.

[0022] A reward function definition module: used for defining a comprehensive reward function based on the energy efficiency function, space distribution function, device maintenance function, space utilization rate function and space illumination function of the virtual building scene.

[0023] A scene model training module: used for selecting a pre-trained initial virtual scene training model; training the initial virtual scene training model through multiple groups of the virtual scene training data, and optimizing the trained initial virtual scene training model based on the comprehensive reward function to obtain a virtual scene training model.

[0024] A device scheme layout module: used for laying out the building energy-using devices through the layout scheme of the to-be-tested scene data obtained by the virtual scene training model.

[0025] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0026] The present invention provides an intelligent layout method for building energy - using equipment. By collecting building scene information, multiple building energy - using equipment, the performance of multiple building energy - using equipment, and the layout methods of multiple said building energy - using equipment according to multiple virtual building scenes, a virtual scene training model for the scene layout of building energy - using equipment is trained, so as to output the layout methods of multiple building energy - using equipment for the to - be - detected virtual scene data. Moreover, by using a comprehensive reward function defined by the energy efficiency function, space distribution function, equipment maintenance function, and space illumination function of the virtual building scene, the virtual scene training model is optimized. Through the optimized virtual scene training model, a layout plan for the to - be - tested scene data is obtained, enabling the model to quickly and accurately obtain a building energy - using equipment layout that better meets the actual requirements of the scene. The present invention solves the problem that in the face of a complex and changeable building environment, it is difficult to achieve an ideal layout effect, resulting in low design efficiency and difficulty in meeting actual requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts. In the appended

[0028] In the drawings:

[0029] Figure 1 is a flowchart of the intelligent layout method for building energy - using equipment in Embodiment 1 of the present invention;

[0030] Figure 2 is a schematic diagram of the intelligent layout system for building energy - using equipment in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] To make the objectives, technical solutions, and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with the embodiments and the drawings. The illustrative embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0032] Embodiment 1

[0033] As Figure 1 shown, the embodiment of the present application provides an intelligent layout method for building energy - using equipment, including:

[0034] Collecting multiple groups of virtual scene training data according to multiple virtual building scenes; each group of the virtual scene training data includes the building scene information of each of the virtual building scenes, the equipment information of multiple said building energy - using equipment in each of the virtual building scenes, and the building energy - using equipment layout data;

[0035] Define a comprehensive reward function based on the energy efficiency function, spatial distribution function, equipment maintenance function, space utilization function, and spatial illuminance function of the virtual building scene;

[0036] Select a pre-trained initial virtual scene training model; train the initial virtual scene training model with multiple groups of the virtual scene training data, and optimize the trained initial virtual scene training model based on the comprehensive reward function to obtain a virtual scene training model;

[0037] Layout the building energy-consuming equipment according to the layout plan of the to-be-tested scene data obtained by the virtual scene training model.

[0038] The building scene information of each virtual building scene includes: building space information and building scene type; the equipment information of multiple building energy-consuming equipment includes: any one or more of size, shape, quantity, installation method, fitting specifications, and operating parameters.

[0039] The multiple virtual building scenes for collecting multiple groups of virtual scene training data can be created by different client users through Revit software, including parametric families containing various building virtual scenes and equipment information and equipment layout data of different building energy-consuming equipment.

[0040] Building energy-consuming equipment such as ventilation systems, air-conditioning systems, lighting equipment, power distribution systems, etc.; equipment information includes information such as the size, shape, and performance parameters of the equipment. The building energy-consuming equipment layout data includes the orientation, position, and connection type between each building energy-consuming equipment and the corresponding virtual building scene. The connection type includes connection quantity, connector specifications, distributable space, fixing method, etc. For example: the connection type of the air-conditioning system includes connection quantity (number of air vents), connector specifications (duct size), distributable space (installation position restrictions caused by equipment characteristics), fixing method (wall-mounted air conditioner is fixed with a drilled bracket, floor-standing air conditioner is placed on the ground for fixing), etc.

[0041] The building energy-consuming equipment layout data includes the orientation, position, and connection type between each building energy-consuming equipment and the corresponding virtual building scene; the connection type includes connection quantity, connector specifications, distributable space, and fixing method.

[0042] The building space information may include building geometric information and building environment information; the building geometric information may include the length, width, height, and area of one or more rooms in the building, the relative positions between different rooms, the positions of walls and doors / windows, etc., and the building environment information may include the building location, building orientation, weather, outdoor brightness, and air visibility, etc. The virtual building scenes can be classified according to whether the living functions are the same, such as types like residential, office, entertainment, etc.; or can be classified according to room types, such as types like living room, bedroom, office, meeting room, etc. Multiple building energy-consuming devices may include a ventilation system, an air-conditioning system, lighting equipment, and a power distribution system, etc. The above-mentioned fitting specifications may include the size, quantity, installation position, and installation method of the fittings, etc.

[0043] The energy efficiency function is obtained through the following steps:

[0044] Obtain the actual production capacity and actual power consumption of each of the building energy-consuming devices through the device information of each of the building energy-consuming devices; define the actual energy efficiency target function of each building energy-consuming device according to the ratio of the actual production capacity and actual power consumption of each of the building energy-consuming devices; use the maximization of the actual energy efficiency target function to obtain the optimal actual energy efficiency;

[0045] Analyze the type weights through the building energy-consuming device layout data of multiple building energy-consuming devices; calculate the device weights according to the type weights and the actual power consumption; obtain the energy efficiency function of the virtual scene training model according to the optimal actual energy efficiency and device weights of each of the building energy-consuming devices.

[0046] The analyzing the type weights through the building energy-consuming device layout data of multiple building energy-consuming devices includes:

[0047] Obtain the device type of each of the building energy-consuming devices through the device information of multiple building energy-consuming devices; score one or more of the orientation, position, and connection type according to the preset scoring rules of the device type of each of the building energy-consuming devices, and obtain the type weight of the building energy-consuming device according to the scoring results.

[0048] The performance of multiple building energy-consuming devices includes the model number, device name, and device use for distinguishing the device type, etc. The device type can be identified. Then score one or more corresponding data in the building energy-consuming device layout data according to the scoring rules of this type, so as to calculate the type weight according to the size of the score. The type scoring can adopt the following rules:

[0049] For example, the 10-point scoring mechanism for each performance parameter can be set as follows: 10 points indicates full compliance with requirements; 7-9 points indicates basic compliance with requirements; 4-6 points indicates partial compliance with requirements; 1-3 points indicates serious non-compliance with requirements. The above specific judgment criteria can be adjusted with reference to the relevant technical standard requirements of the equipment type. Taking the cabinet air conditioner A as an example, the following parameters can be selected to evaluate the equipment type weight. The number of evaluation air vents is 4 (8 points), the duct diameter is 100 mm (9 points), it can be distributed in two corners by the window in the living room (6 points), and the fixing method is directly placed on the ground (6 points), with a comprehensive score of 29 points. The above scoring system is only for example. It can be scored manually. When a large number of building energy-consuming equipment needs to be scored, based on a large amount of experimental data of the relevant equipment types, the actual impact of various parameters on energy efficiency can be verified to obtain the corresponding type weight value.

[0050] Optionally, normalize the type scores of multiple building energy-consuming equipment of the same equipment type in the current virtual building scene, and comprehensively calculate the type weight, which can be expressed as:

[0051]

[0052] Among them, Y i represents the equipment type weight, and C i represents the type scores of each building energy-consuming equipment of the same type; n is the number of building energy-consuming equipment of the same type.

[0053] The weight of the equipment type can be evaluated according to different threshold ranges of the actual power consumption, so as to calculate the final equipment weight based on the equipment type weight and the actual power consumption of each building energy-consuming equipment.

[0054] Optionally, the calculation formula of the equipment weight is expressed as:

[0055]

[0056] Among them, w i represents the equipment weight of the i-th building energy-consuming equipment, ρ is the equipment weight calculation coefficient, which is set as a corresponding constant according to the equipment type; P i represents the actual power consumption of the i-th building energy-consuming equipment; Y j is the type weight.

[0057] The spatial distribution function is obtained through the following steps: obtaining the equipment coverage area of each building energy-consuming equipment through the equipment information of multiple building energy-consuming equipment; obtaining the distribution distance between different building energy-consuming equipment through the layout of multiple building energy-consuming equipment; defining a spatial distribution weight objective function for evaluating the spatial distribution weight of each building energy-consuming equipment according to the equipment coverage area of each building energy-consuming equipment and the distribution distance between different building energy-consuming equipment; calculating the optimal spatial distribution weight by using the spatial distribution weight objective function, and obtaining the spatial distribution function of the virtual scene training model according to the optimal spatial distribution weight.

[0058] Among them, the equipment coverage area is the occupied area of the building energy-consuming equipment on the two-dimensional horizontal plane, which can be obtained according to the maximum area in the circumferential direction of the building energy-consuming equipment; the occupied space is the occupied volume of the building energy-consuming equipment in the three-dimensional space.

[0059] Define the distribution function: The distribution function should reflect the reasonable distribution of equipment in the building space. For example, the distribution function can be defined according to the distance and coverage range between equipment:

[0060]

[0061] Among them, D represents the distribution rationality, and d ij represents the distance between the i-th equipment and the j-th equipment. The goal of the distribution function is to minimize D.

[0062] Distribution scoring: Calculate the distribution score of the equipment according to the distribution function and incorporate it into the comprehensive reward function. The formula for the distribution score is:

[0063]

[0064] Among them, S dist represents the distribution score, and k dist is the distribution rationality reward coefficient.

[0065] The equipment maintenance function is obtained through the following steps: obtaining the installation position of each building energy-consuming equipment through the layout of multiple building energy-consuming equipment; obtaining the occupied space of each building energy-consuming equipment through the equipment information of multiple building energy-consuming equipment; defining an equipment maintenance difficulty objective function for evaluating the equipment maintenance difficulty according to the installation position and occupied space between each building energy-consuming equipment; obtaining the optimal equipment maintenance difficulty by using the equipment maintenance difficulty objective function, and obtaining the equipment maintenance function of the virtual scene training model according to the optimal equipment maintenance difficulty.

[0066] The equipment maintenance difficulty objective function can be defined according to the installation position and occupied space of the equipment, so as to reflect the difficulty of equipment maintenance:

[0067] Among them, M represents maintenance convenience, and m i represents the maintenance space of the i-th building energy-consuming equipment in the virtual building scenario. The optimal maintenance difficulty is obtained by minimizing the equipment maintenance difficulty objective function M.

[0068] The space utilization rate function is obtained through the following steps: obtaining the installation positions of the building energy-consuming equipment through the layout methods of the multiple building energy-consuming equipment; obtaining the occupied space of each building energy-consuming equipment through the equipment information of the multiple building energy-consuming equipment; defining an equipment space utilization objective function for evaluating the space utilization rate of each building energy-consuming equipment according to the ratio of the occupied space of each building energy-consuming equipment to the space of the corresponding virtual building scenario; obtaining the optimal space utilization rate by using the equipment space utilization objective function, and obtaining the space utilization rate function of the virtual scene training model according to the optimal space utilization rate.

[0069] Define the space utilization function: The space utilization function should reflect the ratio of the space occupied by the equipment to the building space. For example, the space utilization function can be defined according to the volume of the equipment and the volume of the building space:

[0070]

[0071] Among them, U represents the space utilization rate, and V i represents the volume of the i-th equipment, and V total represents the total volume of the building space. The goal of the space utilization function is to minimize U.

[0072] Space utilization score: Calculate the space utilization score of the equipment according to the space utilization function and incorporate it into the comprehensive reward function. The formula for the space utilization score is:

[0073]

[0074] Among them, S space represents the space utilization score, and k space is the space utilization reward coefficient, and the preset values are different for different equipment types.

[0075] The space illuminance function is obtained through the following steps: The building energy-consuming equipment includes one or more lighting devices; according to the rated luminous flux and lighting area of each lighting device, and the preset standard illuminance of the corresponding virtual building scenario, the space illuminance function of each lighting device is obtained.

[0076] Define the space illuminance function: The space illuminance function should reflect the rationality of the lighting that the lighting device can provide in the building space. The lighting intensity should not be too large or too small and should be within a reasonable range.

[0077]

[0078] Wherein: S lx represents the spatial illuminance score of the lighting device, and k lx represents the spatial illuminance reward coefficient, and a preset value can be obtained according to the model of the corresponding lighting device. represents the rated luminous flux of the lamp, S represents the area of the region, and E std represents the standard illuminance corresponding to the following illuminance standard table under different spatial styles.

[0079] The illuminance standard table is shown in Table 1 below:

[0080] Spatial pattern Standard illuminance Bedroom 75lx Bathroom 100lx Kitchen 100lx Living room 100lx Dining room 150lx

[0081] Table 1

[0082] The building scene information of each of the virtual building scenes includes: building space information and building scene types; the device information of multiple building energy-consuming devices includes: any one or more of size, shape, quantity, installation method, accessory specifications, and operating parameters.

[0083] When optimizing data through the comprehensive reward function, for example, if the cooling capacity of a certain model of air conditioner is 4500W and the power consumption is 1500W, then its energy efficiency is:

[0084]

[0085] Assume that there are three air conditioning systems in the current building, and their energy efficiencies are 3, 2.5, and 2.8 respectively, and the rated powers are 1500w, 1000w, and 1000w. The type scores are 29 points, 32 points, and 36 points respectively. The device type weights Y1, Y2, and Y3 are calculated comprehensively:

[0086]

[0087] Take the device weight calculation coefficient ρ of this device type as 3.04, and comprehensively consider the device power and type weight to calculate the device weights w1, w2, and w3:

[0088]

[0089] The energy efficiency optimization reward score can be calculated through the optimal actual energy efficiency and device weight of each building energy-consuming device, and then the optimization reward score is incorporated into the reward function of the virtual scene training model.

[0090] The optimization reward score can be expressed as:

[0091]

[0092] Among them, S eff represents the energy efficiency optimization reward score, w i represents the device weight of the i-th device, η i represents the optimal actual energy efficiency of the i-th building energy-consuming device, k eff is the energy efficiency optimization reward coefficient, which is a constant obtained according to the corresponding device type, and n is the number of building energy-consuming devices.

[0093] Taking the energy efficiency optimization reward coefficient as 1.0, the energy efficiency optimization reward score is:

[0094] S eff = 1.0 * (0.39·3 + 0.287·2.5 + 0.322·2.8) = 2.79;

[0095] The spatial distribution weight objective function reflects the reasonable distribution of devices in the building space, and the spatial distribution weight objective function is defined according to the distance and coverage range between devices:

[0096]

[0097] Among them, D represents the distribution rationality, d ij represents the distribution distance between the i-th building energy-consuming device and the j-th building energy-consuming device, and n is the number of building energy-consuming devices. The optimal spatial distribution weight is obtained by minimizing the spatial distribution weight objective function D.

[0098] Optionally, calculate the distribution score of the device according to the optimal spatial distribution weight and incorporate it into the reward function of the virtual scene training model. The formula for the distribution score is:

[0099]

[0100] Among them, S dist represents the distribution score; k dist is the distribution rationality reward coefficient, which is a constant obtained according to the building scene type of the virtual building scene.

[0101] For example, if the distances between three air conditioning systems are 5m, 6m, and 7m respectively, then the spatial distribution weight objective function is:

[0102]

[0103] Taking the distribution rationality reward coefficient as 1.0, calculate the distribution score:

[0104]

[0105] Optionally, calculate the maintenance score of the device according to the device maintenance difficulty objective function and incorporate it into the reward function of the virtual scene training model. The formula for the maintenance score is:

[0106] where S main represents the maintenance score; k main is the maintenance convenience reward coefficient, which is a constant obtained according to the device type of the building energy-using device; M is the device maintenance difficulty objective function.

[0107] For example: The occupied spaces for maintaining three air-conditioning systems are 2m 2 , 2.5m 2 and 3m 2 , respectively. Then the device maintenance difficulty objective function is:

[0108]

[0109] Take the maintenance convenience reward coefficient as 1.0 and calculate the maintenance score:

[0110]

[0111] Calculate the ratio of the volume of each air-conditioning system to the volume of the building space: For example, the volumes of the three air-conditioning systems are 0.12m 3 , 0.15m 3 and 0.18m 3 , respectively, and the total volume of the building space is 100m 3 . Then the space utilization function is:

[0112]

[0113] Take the space utilization reward coefficient as 0.01 and calculate the space utilization score:

[0114]

[0115] During the learning process of the virtual scene training model, through multiple iterations, continuously adjust and optimize the layout method until the optimal solution is found. Optionally, by adjusting the weight coefficients of the above energy efficiency optimization reward score, distribution score, maintenance score, space utilization score and space illuminance, according to the sum of the results of each score * its respective score weight coefficient, the comprehensive reward function of the virtual scene training model can be obtained, and the virtual scene training model can be optimized in multiple aspects by integrating all score results.

[0116] After the virtual scene training model outputs the layout modes of multiple building energy - using devices for the virtual scene data to be detected, according to the building scene information of the virtual building scene, the virtual building scene can be generated by Revit software according to the final layout mode. Through the API interface of Revit, the layout scheme is converted into a visual building model, which is convenient for designers to make further modifications and adjustments.

[0117] Embodiment 2

[0118] As Figure 2 shown, an intelligent layout system for building energy - using devices provided by an embodiment of the present application includes:

[0119] A training data acquisition module: used to collect multiple groups of virtual scene training data according to multiple virtual building scenes; each group of the virtual scene training data includes the building scene information of each virtual building scene, the device information of multiple building energy - using devices in each virtual building scene, and the building energy - using device layout data;

[0120] A reward function definition module: used to define a comprehensive reward function based on the energy efficiency function, space distribution function, device maintenance function, space utilization function, and space illuminance function of the virtual building scene;

[0121] A scene model training module: used to select a pre - trained initial virtual scene training model; train the initial virtual scene training model with multiple groups of the virtual scene training data, and optimize the trained initial virtual scene training model based on the comprehensive reward function to obtain a virtual scene training model;

[0122] A device scheme layout module: used to layout the building energy - using devices through the layout scheme of the to - be - tested scene data obtained by the virtual scene training model.

[0123] In summary, an embodiment of the present application provides an intelligent layout method and system for building energy - using devices:

[0124] Collect building scene information, multiple building energy-consuming devices, the performance of multiple building energy-consuming devices, and the layout modes of multiple said building energy-consuming devices according to multiple virtual building scenes, and train a virtual scene training model for the scene layout of building energy-consuming devices, so as to output the layout modes of multiple building energy-consuming devices for the virtual scene data to be detected; and, use the comprehensive reward function defined by the energy efficiency function, spatial distribution function, equipment maintenance function, and spatial illuminance function of the virtual building scene to optimize the virtual scene training model; obtain the layout plan of the scene data to be tested through the optimized virtual scene training model, so that the model can quickly and accurately obtain the layout of building energy-consuming devices that better meets the actual requirements of the scene. The present invention solves the problem that it is difficult to achieve an ideal layout effect in the face of complex and changeable building environments, resulting in low design efficiency and difficulty in meeting actual requirements.

[0125] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent layout method for building energy-consuming equipment, characterized in that, Including: Collecting multiple groups of virtual scene training data according to multiple virtual building scenes; Each group of the virtual scene training data includes building scene information of each of the virtual building scenes, device information of multiple building energy-consuming devices in each of the virtual building scenes, and building energy-consuming device layout data; Defining a comprehensive reward function based on the energy efficiency function, spatial distribution function, device maintenance function, space utilization function, and space illumination function of the virtual building scene; Selecting a pre-trained initial virtual scene training model; training the initial virtual scene training model with multiple groups of the virtual scene training data, and optimizing the trained initial virtual scene training model based on the comprehensive reward function to obtain a virtual scene training model; Laying out the building energy-consuming devices through the layout scheme of the to-be-tested scene data obtained by the virtual scene training model.

2. The intelligent layout method for building energy equipment according to claim 1, characterized in that The building energy-consuming device layout data includes the orientation, position, and connection type between each building energy-consuming device and the corresponding virtual building scene; the connection type includes the number of connections, connector specifications, distributable space, and fixing method.

3. The intelligent layout method of building energy equipment according to claim 2, characterized in that, The energy efficiency function is obtained through the following steps: Obtaining the actual production capacity and actual power consumption of each building energy-consuming device through the device information of each building energy-consuming device; defining the actual energy efficiency target function of each building energy-consuming device according to the ratio of the actual production capacity and actual power consumption of each building energy-consuming device; obtaining the optimal actual energy efficiency by maximizing the actual energy efficiency target function; Analyzing the type weights through the building energy-consuming device layout data of multiple building energy-consuming devices; Calculating the device weights according to the type weights and the actual power consumption; obtaining the energy efficiency function of the virtual scene training model according to the optimal actual energy efficiency and the device weights of each building energy-consuming device.

4. The intelligent layout method of building energy-using equipment according to claim 3, characterized in that The analyzing the type weights through the building energy-consuming device layout data of multiple building energy-consuming devices includes: Obtaining the device type of each building energy-consuming device through the device information of multiple building energy-consuming devices; scoring one or more of the orientation, position, and connection type according to the preset scoring rules of the device type of each building energy-consuming device, and obtaining the type weight of the building energy-consuming device according to the scoring result.

5. The intelligent layout method of an energy-consuming device for buildings according to claim 1, wherein The spatial distribution function is obtained through the following steps: obtaining the device coverage area of each building energy-consuming device through the device information of multiple building energy-consuming devices; obtaining the distribution distance between different building energy-consuming devices through the layout method of multiple building energy-consuming devices; defining a spatial distribution weight target function for evaluating the spatial distribution weight of each building energy-consuming device according to the device coverage area of each building energy-consuming device and the distribution distance between different building energy-consuming devices; calculating the optimal spatial distribution weight by using the spatial distribution weight target function, and obtaining the spatial distribution function of the virtual scene training model according to the optimal spatial distribution weight.

6. The intelligent layout method for building energy-consuming equipment according to claim 1, characterized in that, The device maintenance function is obtained through the following steps: obtaining the installation positions of the building energy-consuming devices through the layout modes of the multiple building energy-consuming devices; obtaining the occupied spaces of the building energy-consuming devices through the device information of the multiple building energy-consuming devices; defining a device maintenance difficulty objective function for evaluating the device maintenance difficulty according to the installation positions and the occupied spaces among the building energy-consuming devices; obtaining the optimal device maintenance difficulty by using the device maintenance difficulty objective function, and obtaining the device maintenance function of the virtual scene training model according to the optimal device maintenance difficulty.

7. The intelligent layout method for building energy equipment according to claim 1, characterized in that, The space utilization rate function is obtained through the following steps: obtaining the installation positions of the building energy-consuming devices through the layout modes of the multiple building energy-consuming devices; obtaining the occupied spaces of the building energy-consuming devices through the device information of the multiple building energy-consuming devices; defining a device space utilization objective function for evaluating the space utilization rate of each building energy-consuming device according to the ratio of the occupied space of each building energy-consuming device to the space of the corresponding virtual building scene; obtaining the optimal space utilization rate by using the device space utilization objective function, and obtaining the space utilization rate function of the virtual scene training model according to the optimal space utilization rate.

8. The intelligent layout method of an energy-consuming device for buildings according to claim 1, characterized in that, The space illumination function is obtained through the following steps: the building energy-consuming devices include one or more lighting devices; obtaining the space illumination function of each lighting device according to the rated luminous flux and the lighting area of each lighting device and the preset standard illumination of the virtual building scene.

9. The intelligent layout method of an energy-consuming device for construction according to claim 1, characterized in that, The building scene information of each virtual building scene includes: building space information and building scene type; the device information of the multiple building energy-consuming devices includes: any one or more of size, shape, quantity, installation method, accessory specifications and operating parameters.

10. An intelligent layout system for building energy-consuming equipment, characterized in that, Including: A training data acquisition module: used for acquiring multiple groups of virtual scene training data according to multiple virtual building scenes. Each group of the virtual scene training data includes the building scene information of each virtual building scene, the device information of the multiple building energy-consuming devices in each virtual building scene, and the building energy-consuming device layout data. A reward function definition module: used for defining a comprehensive reward function based on the energy efficiency function, the space distribution function, the device maintenance function, the space utilization rate function and the space illumination function of the virtual building scene. A scene model training module: used for selecting a pre-trained initial virtual scene training model; training the initial virtual scene training model through multiple groups of the virtual scene training data, and optimizing the trained initial virtual scene training model based on the comprehensive reward function to obtain a virtual scene training model. A device scheme layout module: used for laying out the building energy-consuming devices through the layout scheme of the to-be-tested scene data obtained by the virtual scene training model.

Images