Automatic BIM model error detection system using deep learning

Abstract

The present invention relates to the field of building information model, specifically to an automatic BIM model error detection system using deep learning, including a data reading and conversion module, which is used to read a first BIM geographic location model and a first BIM electrical model, and obtain a power load model, a first power supply model, a second power supply model, and a connection device model of each power supply unit model through model mapping, model segmentation, and model decomposition; a power supply continuity detection module, which is used to obtain a power supply continuity error identification mark; a power supply quantity detection module, which is used to obtain a power supply quantity error identification mark; and a detection result output module, which is used to obtain a BIM model error detection result according to a power supply continuity error identification mark and a power supply quantity error identification mark. The present invention realizes automatic model error detection by intelligently analyzing the connection relationship of the BIM electrical model of an intelligent building containing rooftop photovoltaics and the matching relationship of the equipment capacity.

Description

Technical Field

[0001] The present invention relates to the technical field of building information modeling, and in particular to an automatic BIM model error detection system using deep learning. Background Art

[0002] BIM model (Building Information Modeling), or building information model, is a model that digitally expresses the physical and functional characteristics of construction projects and facilities throughout their life cycle. BIM models can be divided into architectural models, structural models, water supply and drainage models, HVAC models, and electrical models according to professional classification. Among them, the BIM electrical model includes information such as power load, electrical wiring, equipment location, and distribution boxes, which can help designers and operation and maintenance personnel carry out full-cycle design and operation of building electrical systems. At present, BIM electrical models are gradually being applied to the design and operation and maintenance of smart buildings. Smart buildings often contain distributed power sources such as rooftop photovoltaics, while traditional BIM electrical models are mostly unidirectional flow models. When combined with distributed power models such as rooftop photovoltaics, incompatibility errors may occur. However, existing BIM model error detection technologies make it difficult to detect these errors. Summary of the invention

[0003] (1) Technical issues to be solved

[0004] The purpose of the present invention is to provide a BIM model error automatic detection system using deep learning to solve the problem of automatic detection of BIM electrical model errors in smart buildings including rooftop photovoltaics.

[0005] (2) Technical solution

[0006] To achieve the above object, the present invention provides a BIM model error automatic detection system using deep learning, the system comprising:

[0007] A data reading and conversion module is used to read a first BIM geographic location model and a first BIM electrical model; map the first BIM electrical model to a two-dimensional plane to obtain a second BIM electrical model; divide the second BIM electrical model into a power supply unit model; decompose the power supply unit model to obtain a power load model, a first power supply model, a second power supply model, and a connection device model of each power supply unit model;

[0008] A power supply continuity detection module, connected to the data reading and conversion module, is used to perform power supply continuity error judgment according to the power supply unit model and obtain a power supply continuity error identification mark;

[0009] A power supply detection module, connected to the data reading and conversion module, is used to perform power supply error judgment based on the second BIM electrical model and the first BIM geographic location model to obtain a power supply error identification mark;

[0010] The detection result output module is connected to the power supply continuity detection module and the power supply quantity detection module, and is used to obtain the BIM model error detection result according to the power supply continuity error identification mark and the power supply quantity error identification mark.

[0011] Furthermore, the data reading and conversion module includes:

[0012] A data reading module, used to read a first BIM geographic location model and a first BIM electrical model; the first BIM geographic location model includes a building longitude and a building latitude;

[0013] A model mapping module, connected to the data reading module, is used to map the first BIM electrical model to a two-dimensional plane to obtain a second BIM electrical model; the second BIM electrical model includes the power load, roof photovoltaic, power distribution equipment, mains interface in the first BIM electrical model and the first circuit connection relationship between the power load, roof photovoltaic, power distribution equipment and mains interface; the mains interface is the connection point between the second BIM electrical model and the external power grid;

[0014] The power supply unit segmentation module is connected to the model mapping module and is used to read the number of power loads N in the second BIM electrical model. L 、Number of rooftop photovoltaic D 、Number of power distribution equipment T 、Number of AC power interfaces G ; The N D Rooftop PV and N G Combine the N mains interfaces to obtain the first upper power supply set; L The power loads are respectively traced upward according to the first circuit connection relationship, and searched for the N L All the upper power sources that supply power to the i-th power load are used to obtain the second upper power source set of the i-th power load, i = 1, 2...N L ; The second upper power source set is a subset of the first upper power source set; establish a binary connection relationship between the i-th power load and the second upper power source set of the i-th power load, i = 1, 2...N L ; respectively combine the binary connection relationship between the i-th power load, the second upper power supply set of the i-th power load, and the i-th power load and the second upper power supply set of the i-th power load into a power supply unit model, i = 1, 2...N L ; The number of power supply unit models is N L ;

[0015] The model decomposition module is connected to the power supply unit segmentation module, and is used to decompose the power supply unit model to obtain the power load model, the first power supply model, the second power supply model, and the connection device model of each power supply unit model; the first power supply model is all the rooftop photovoltaics included in the power supply unit model; the second power supply model is all the AC power interfaces included in the power supply unit model.

[0016] Furthermore, the power supply continuity detection module includes:

[0017] A power supply statistics module, used to respectively count the number of mains power interfaces in the second power supply model in each of the power supply unit models;

[0018] A power supply continuity error identification module is connected to the power supply statistics module and is used to identify power supply continuity errors according to the number of AC power interfaces in the second power supply model in each of the power supply unit models; if the number of AC power interfaces in the second power supply model in at least one of the power supply unit models is 0, the power supply continuity error identification flag F0 is set to 0, otherwise the power supply continuity error identification flag F0 is set to 1.

[0019] Furthermore, the power supply detection module includes:

[0020] A data extraction module is used to extract the building longitude and latitude of the first BIM geographic location model; extract the N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND ; Extract the N L The rated power of each power load is P1, P2, ... NL ; Extract the N L Rated power factor of each electrical load Extract the N L The annual average utilization hours of each power load is T1, T2, ... T NL ; Extract the N L The simultaneous coefficient w of the power load; extract the N T The rated capacity of each distribution equipment is M1, M2, M NT ; Extract the N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT ;

[0021] The roof photovoltaic power supply calculation module is connected to the data extraction module and is used to calculate the power supply according to the building longitude and latitude, N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND, the pre-set photovoltaic neural network calculates the rooftop photovoltaic power Q;

[0022] The power supply error recognition module is connected to the roof photovoltaic power supply calculation module and is used to calculate the power Q and N according to the roof photovoltaic L The rated power of each power load is P1, P2, ... NL 、N L Rated power factor of each electrical load N L The annual average utilization hours of each power load is T1, T2, ... T NL 、N L The simultaneous coefficients w and N of the power loads T The rated capacity of each distribution equipment is M1, M2, M NT 、N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT , calculate and obtain the power supply error identification mark.

[0023] Furthermore, the rooftop photovoltaic power supply calculation module includes:

[0024] A light measurement module, used to obtain a first light intensity and a first light time according to the building longitude and building latitude of the first BIM geographical location model;

[0025] The photovoltaic power measurement module is connected to the light measurement module and is used to calculate the photovoltaic power according to the first light intensity, the first light time, and N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND And the pre-set photovoltaic neural network is used to calculate the rooftop photovoltaic power consumption Q.

[0026] Furthermore, the photovoltaic power measurement module includes:

[0027] The photovoltaic panel total area calculation module is used to convert the N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND Converted into the total area of ​​photovoltaic panels S through the area conversion formula sum , where the area conversion formula is:

[0028]

[0029] Among them, k lap is the pre-set overlap factor;

[0030] The neural network calculation module is connected to the photovoltaic panel total area calculation module and is used to calculate the total area S of the photovoltaic panel. sum, the first light intensity, and the first light time are input into a preset photovoltaic neural network to calculate the rooftop photovoltaic power consumption Q.

[0031] Furthermore, the power supply error identification module includes:

[0032] The first identification calculation module is used to convert the N L The rated power of each power load is P1, P2, ... NL 、N L Rated power factor of each electrical load The N L The simultaneous coefficient w of the power load, the N T The rated capacity of each distribution equipment is M1, M2, M NT , is converted into the first identifier F1 through the first identifier calculation formula, where the first identifier calculation formula is:

[0033]

[0034] The second identification calculation module is used to calculate the rooftop photovoltaic power Q, N L The rated power of each power load is P1, P2, ... NL 、N L Rated power factor of each electrical load N L The annual average utilization hours of each power load is T1, T2, ... T NL 、N T The rated capacity of each distribution equipment is M1, M2, M NT 、N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT , is converted into the second identifier F2 through the second identifier calculation formula, where the second identifier calculation formula is:

[0035]

[0036] The power supply error identification mark calculation module is connected to the first mark calculation module and the second mark calculation module, and is used to convert the first mark F1 and the second mark F2 into the power supply error identification mark F3 through the power supply error identification mark calculation formula, wherein the power supply error identification mark calculation formula is:

[0037]

[0038] Furthermore, the detection result output module includes:

[0039] The upper level identification and extraction module is used to extract the power supply continuity error identification mark F0 and the power supply quantity error identification mark F3;

[0040] The detection result calculation module is connected to the upper-level identification and extraction module, and is used to convert the power supply continuity error identification mark F0 and the power supply error identification mark F3 into the BIM model error detection mark F4 through the detection result calculation formula, wherein the detection result calculation formula is:

[0041]

[0042] The detection result output display module is connected to the detection result calculation module, and is used to convert the BIM model error detection identifier F4 into a BIM model error detection result; if the BIM model error detection identifier F4=0, "model error" is output; if the BIM model error detection identifier F4=1, "model correct" is output.

[0043] (3) Beneficial effects

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] Automatic detection of model errors can be achieved by intelligently analyzing the connection relationship of the BIM electrical model of the smart building containing rooftop photovoltaics and the matching relationship of equipment capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a module block diagram of an automatic BIM model error detection system using deep learning according to Example 1 of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Before giving examples, it is necessary to explain the application scenarios of the present invention. The system is used for automatic detection of errors in the BIM electrical model of smart buildings including rooftop photovoltaics.

[0049] Example 1: Figure 1 As shown, this embodiment provides a BIM model error automatic detection system using deep learning, and the system includes:

[0050] A data reading and conversion module is used to read a first BIM geographic location model and a first BIM electrical model; map the first BIM electrical model to a two-dimensional plane to obtain a second BIM electrical model; divide the second BIM electrical model into a power supply unit model; decompose the power supply unit model to obtain a power load model, a first power supply model, a second power supply model, and a connection device model of each power supply unit model;

[0051] It should be noted that the BIM model includes all kinds of information throughout the life cycle of the building. In order to automatically detect errors in the BIM electrical model of a smart building with rooftop photovoltaics, this embodiment only extracts the first BIM geographic location model and the first BIM electrical model in the BIM model. Among them, the first BIM geographic location model includes the longitude and latitude of the location of the smart building. The first BIM electrical model contains detailed information on all power loads, rooftop photovoltaics, power distribution equipment, and mains interfaces in the smart building, as well as their circuit connection relationships.

[0052] A power supply continuity detection module, connected to the data reading and conversion module, is used to perform power supply continuity error judgment according to the power supply unit model and obtain a power supply continuity error identification mark;

[0053] A power supply detection module, connected to the data reading and conversion module, is used to perform power supply error judgment based on the second BIM electrical model and the first BIM geographic location model to obtain a power supply error identification mark;

[0054] The detection result output module is connected to the power supply continuity detection module and the power supply quantity detection module, and is used to obtain the BIM model error detection result according to the power supply continuity error identification mark and the power supply quantity error identification mark.

[0055] Furthermore, the data reading and conversion module includes:

[0056] A data reading module, used to read a first BIM geographic location model and a first BIM electrical model; the first BIM geographic location model includes a building longitude and a building latitude;

[0057] A model mapping module, connected to the data reading module, is used to map the first BIM electrical model to a two-dimensional plane to obtain a second BIM electrical model; the second BIM electrical model includes the power load, roof photovoltaic, power distribution equipment, mains interface in the first BIM electrical model and the first circuit connection relationship between the power load, roof photovoltaic, power distribution equipment and mains interface; the mains interface is the connection point between the second BIM electrical model and the external power grid;

[0058] The power supply unit segmentation module is connected to the model mapping module and is used to read the number of power loads N in the second BIM electrical model. L 、Number of rooftop photovoltaic D 、Number of power distribution equipment T 、Number of AC power interfaces G ; The N D Rooftop PV and N G Combine the N mains interfaces to obtain the first upper power supply set; L The power loads are respectively traced upward according to the first circuit connection relationship, and searched for the N L All the upper power sources that supply power to the i-th power load are used to obtain the second upper power source set of the i-th power load, i = 1, 2...N L ; The second upper power source set is a subset of the first upper power source set; establish a binary connection relationship between the i-th power load and the second upper power source set of the i-th power load, i = 1, 2...N L ; respectively combine the binary connection relationship between the i-th power load, the second upper power supply set of the i-th power load, and the i-th power load and the second upper power supply set of the i-th power load into a power supply unit model, i = 1, 2...N L ; The number of power supply unit models is N L ;

[0059] The model decomposition module is connected to the power supply unit segmentation module, and is used to decompose the power supply unit model to obtain the power load model, the first power supply model, the second power supply model, and the connection device model of each power supply unit model; the first power supply model is all the rooftop photovoltaics included in the power supply unit model; the second power supply model is all the AC power interfaces included in the power supply unit model.

[0060] For example, the data reading module reads that the longitude of the building in the first BIM geographic location model is 101°08′ east longitude and the latitude of the building is 26°05′ north latitude. The data reading module reads the first BIM electrical model. The model mapping module obtains the second BIM electrical model. The power supply unit segmentation module reads the number of power loads N in the second BIM electrical model. L =98, Number of rooftop photovoltaics N D =10, Number of power distribution equipment N T =2, Number of AC power interfaces N G=1. The 10 rooftop photovoltaics and 1 mains interface are combined to obtain the first upper power source set: {rooftop photovoltaic 1, rooftop photovoltaic 2, rooftop photovoltaic 3, rooftop photovoltaic 4, rooftop photovoltaic 5, rooftop photovoltaic 6, rooftop photovoltaic 7, rooftop photovoltaic 8, rooftop photovoltaic 9, rooftop photovoltaic 10, mains interface 1}. The 98 power loads are traced upward according to the first circuit connection relationship, and all upper power sources supplying power to the 98 power loads are searched respectively to obtain the second upper power source set of the first power load {rooftop photovoltaic 7, mains interface 1}, the second upper power source set of the second power load {rooftop photovoltaic 8, mains interface 1}... the second upper power source set of the 98th power load {rooftop photovoltaic 2, rooftop photovoltaic 9, mains interface 1}. A binary connection relationship is established between the first power load and the second upper power source set of the first power load {roof photovoltaic 7, mains interface 1}, that is, {roof photovoltaic 7-first power load, mains interface 1-first power load}, which means that roof photovoltaic 7 can supply power to the first power load, and the mains can also supply power to the first power load. Similarly, the module also establishes a binary connection relationship between the second power load and the second upper power source set of the second power load...a binary connection relationship between the 98th power load and the second upper power source set of the 98th power load. It is worth noting that the process of tracing upward according to the first circuit connection relationship needs to consider the switch opening and closing state in the first circuit connection relationship. Based on the above data, 98 power supply unit models are obtained. Through the model decomposition module, the 98 power supply unit models are decomposed respectively. Exemplarily, the power load model in the power supply unit model including the first power load is the first power load model, the first power supply model is the rooftop photovoltaic 7, the second power supply model is the mains interface 1, and the connection device model is the connection relationship between the rooftop photovoltaic 7 and the first power load, and the connection relationship between the mains interface 1 and the first power load.

[0061] Furthermore, the power supply continuity detection module includes:

[0062] A power supply statistics module, used to respectively count the number of mains power interfaces in the second power supply model in each of the power supply unit models;

[0063] A power supply continuity error identification module is connected to the power supply statistics module and is used to identify power supply continuity errors according to the number of AC power interfaces in the second power supply model in each of the power supply unit models; if the number of AC power interfaces in the second power supply model in at least one of the power supply unit models is 0, the power supply continuity error identification flag F0 is set to 0, otherwise the power supply continuity error identification flag F0 is set to 1.

[0064] Exemplarily, the number of AC power interfaces in the second power supply model in the 98 power supply unit models is counted through the power supply statistics module. For example, the number of AC power interfaces in the second power supply model in the power supply unit model containing the first power load is 1. The power supply continuity error identification module is used to identify whether there is a power supply continuity error. If the number of AC power interfaces in the second power supply model in at least one power supply unit model in the 98 power supply unit models is 0, the power supply continuity error identification mark F0 is set to 0, otherwise the power supply continuity error identification mark F0 is set to 1. Exemplarily, since the number of AC power interfaces in the second power supply model in the 98 power supply unit models is 1, the power supply continuity error identification mark F0 is set to 1. The principle is that all power loads in the intelligent building must be connected to a stable power supply, otherwise they cannot work normally. In the BIM electrical model of the intelligent building containing rooftop photovoltaics, there are only two types of power supplies: rooftop photovoltaics and AC power interfaces. The power generation of rooftop photovoltaics is intermittent, and only the power supply of the mains interface is stable, so each power supply unit model must contain at least one mains interface, otherwise the BIM electrical model will be wrong.

[0065] Furthermore, the power supply detection module includes:

[0066] A data extraction module is used to extract the building longitude and latitude of the first BIM geographic location model; extract the N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND ; Extract the N L The rated power of each power load is P1, P2, ... NL ; Extract the N L Rated power factor of each electrical load Extract the N L The annual average utilization hours of each power load is T1, T2, ... T NL ; Extract the N L The simultaneous coefficient w of the power load; extract the N T The rated capacity of each distribution equipment is M1, M2, M NT ; Extract the N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT ;

[0067] The roof photovoltaic power supply calculation module is connected to the data extraction module and is used to calculate the power supply according to the building longitude and latitude, N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND , the pre-set photovoltaic neural network calculates the rooftop photovoltaic power Q;

[0068] The power supply error recognition module is connected to the roof photovoltaic power supply calculation module and is used to calculate the power Q and N according to the roof photovoltaic L The rated power of each power load is P1, P2, ... NL 、N L Rated power factor of each electrical load N L The annual average utilization hours of each power load is T1, T2, ... T NL 、N L The simultaneous coefficients w and N of the power loads T The rated capacity of each distribution equipment is M1, M2, M NT 、N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT , calculate and obtain the power supply error identification mark.

[0069] Furthermore, the rooftop photovoltaic power supply calculation module includes:

[0070] A light measurement module, used to obtain a first light intensity and a first light time according to the building longitude and building latitude of the first BIM geographical location model;

[0071] It should be noted that the module can be implemented by establishing an offline mapping relationship between the building longitude, building latitude, first light intensity, and first light time, or by accessing an online database. The required basic data can come from professional meteorological data platforms such as the Xihe Energy Meteorological Big Data Platform.

[0072] The photovoltaic power measurement module is connected to the light measurement module and is used to calculate the photovoltaic power according to the first light intensity, the first light time, and N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND And the pre-set photovoltaic neural network is used to calculate the rooftop photovoltaic power consumption Q.

[0073] Furthermore, the photovoltaic power measurement module includes:

[0074] The photovoltaic panel total area calculation module is used to convert the N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND Converted into the total area of ​​photovoltaic panels S through the area conversion formula sum , where the area conversion formula is:

[0075]

[0076] Among them, k lapis the pre-set overlap factor;

[0077] The neural network calculation module is connected to the photovoltaic panel total area calculation module and is used to calculate the total area S of the photovoltaic panel. sum , the first light intensity, and the first light time are input into a preset photovoltaic neural network to calculate the rooftop photovoltaic power consumption Q.

[0078] It should be noted that, theoretically, there is a positive proportional relationship between photovoltaic power generation and the total area of ​​photovoltaic panels, the first light intensity, and the first light time. However, its proportionality coefficient is related to the photovoltaic power generation efficiency and is difficult to set accurately. Therefore, the photovoltaic power measurement module overcomes the drawbacks of artificially setting the photovoltaic power generation efficiency by adopting a deep learning approach. It is necessary to pre-train the model with actual training samples to obtain a photovoltaic neural network. sum , the first light intensity, and the first light time are input into a preset photovoltaic neural network to calculate the rooftop photovoltaic power consumption Q.

[0079] Furthermore, the power supply error identification module includes:

[0080] The first identification calculation module is used to convert the N L The rated power of each power load is P1, P2, ... NL 、N L Rated power factor of each electrical load The N L The simultaneous coefficient w of the power load, the N T The rated capacity of each distribution equipment is M1, M2, M NT , is converted into the first identifier F1 through the first identifier calculation formula, where the first identifier calculation formula is:

[0081]

[0082] The second identification calculation module is used to calculate the rooftop photovoltaic power Q, N L The rated power of each power load is P1, P2, ... NL 、N L Rated power factor of each electrical load N L The annual average utilization hours of each power load is T1, T2, ... T NL 、N T The rated capacity of each distribution equipment is M1, M2, M NT 、N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT , is converted into the second identifier F2 through the second identifier calculation formula, where the second identifier calculation formula is:

[0083]

[0084] The power supply error identification mark calculation module is connected to the first mark calculation module and the second mark calculation module, and is used to convert the first mark F1 and the second mark F2 into the power supply error identification mark F3 through the power supply error identification mark calculation formula, wherein the power supply error identification mark calculation formula is:

[0085]

[0086] It should be noted that the first identification calculation module is used to determine whether the power load model matches the power distribution equipment model when the rooftop photovoltaic system is not working, that is, when the light intensity is 0. The second identification calculation module is used to determine whether the rooftop photovoltaic model, the power load model, and the power distribution equipment model match in terms of power. Exemplarily, F1=1 is obtained through the first identification calculation module, F2=1 is obtained through the second identification calculation module, and F3=1 is obtained through the power supply error identification calculation module.

[0087] Furthermore, the detection result output module includes:

[0088] The upper level identification and extraction module is used to extract the power supply continuity error identification mark F0 and the power supply quantity error identification mark F3;

[0089] The detection result calculation module is connected to the upper-level identification and extraction module, and is used to convert the power supply continuity error identification mark F0 and the power supply error identification mark F3 into the BIM model error detection mark F4 through the detection result calculation formula, wherein the detection result calculation formula is:

[0090]

[0091] The detection result output display module is connected to the detection result calculation module, and is used to convert the BIM model error detection identifier F4 into a BIM model error detection result; if the BIM model error detection identifier F4=0, "model error" is output; if the BIM model error detection identifier F4=1, "model correct" is output.

[0092] Exemplarily, since F0=1 and F3=1, the detection result output module obtains F4=1 and outputs "the model is correct".

[0093] Finally, it should be noted that: Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Automatic BIM model error detection system using deep learning, characterized by: The system comprises: A data reading and conversion module is used to read a first BIM geographic location model and a first BIM electrical model; map the first BIM electrical model to a two-dimensional plane to obtain a second BIM electrical model; divide the second BIM electrical model into a power supply unit model; decompose the power supply unit model to obtain a power load model, a first power supply model, a second power supply model, and a connection device model of each power supply unit model; A power supply continuity detection module, connected to the data reading and conversion module, is used to perform power supply continuity error judgment according to the power supply unit model to obtain a power supply continuity error identification mark F0; A power supply detection module, connected to the data reading and conversion module, is used to perform power supply error judgment according to the second BIM electrical model and the first BIM geographic location model to obtain a power supply error identification mark F3; The power supply quantity detection module comprises: A data extraction module is used to extract the building longitude and latitude of the first BIM geographic location model; extract N from the second BIM electrical model D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND ; Extract N L The rated power of each power load is P1, P2, ... NL ; Extract N L The rated power factor of each electrical load is φ1, φ2, ... NL ; Extract N L The annual average utilization hours of each power load is T1, T2, ... T NL ; Extract N L The simultaneous coefficient w of the power load; extract N T The rated capacity of each distribution equipment is M1, M2, ... NT ; Extract N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT ; Among them, N D Indicates the number of rooftop photovoltaics; N L Indicates the number of power loads; N T Indicates the number of power distribution equipment; The roof photovoltaic power supply calculation module is connected to the data extraction module and is used to calculate the power supply according to the building longitude and latitude, N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND , the pre-set photovoltaic neural network calculates the rooftop photovoltaic power Q; The power supply error recognition module is connected to the roof photovoltaic power supply calculation module and is used to calculate the power Q and N according to the roof photovoltaic L The rated power of each power load is P1, P2, ... NL 、N L The rated power factor of each electrical load is φ1, φ2, ... NL 、N L The annual average utilization hours of each power load is T1, T2, ... T NL 、N L The simultaneous coefficients w and N of the power loads T The rated capacity of each distribution equipment is M1, M2, ... NT 、N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT , calculate and obtain the power supply error identification mark F3; The power supply error identification module includes: The first identification calculation module is used to convert the N L The rated power of each power load is P1, P2, ... NL 、N L The rated power factor of each electrical load is φ1, φ2, ... NL 、N L The simultaneous coefficient w of the power load, the N T The rated capacity of each distribution equipment is M1, M2, ... NT , is converted into the first identifier F1 through the first identifier calculation formula, where the first identifier calculation formula is: ； The second identification calculation module is used to calculate the rooftop photovoltaic power Q, N L The rated power of each power load is P1, P2, ... NL 、N L The rated power factor of each electrical load is φ1, φ2, ... NL 、N L The annual average utilization hours of each power load is T1, T2, ... T NL 、N T The rated capacity of each distribution equipment is M1, M2, ... NT 、N T The average annual utilization hours of the power distribution equipment c1, c2, ... c NT , is converted into the second identifier F2 through the second identifier calculation formula, where the second identifier calculation formula is: ； The power supply error identification mark calculation module is connected to the first mark calculation module and the second mark calculation module, and is used to convert the first mark F1 and the second mark F2 into the power supply error identification mark F3 through the power supply error identification mark calculation formula, wherein the power supply error identification mark calculation formula is: ； The detection result output module is connected to the power supply continuity detection module and the power supply quantity detection module, and is used to obtain the BIM model error detection result according to the power supply continuity error identification mark and the power supply quantity error identification mark.

2. The BIM model error automatic detection system using deep learning as claimed in claim 1, characterized in that: The data reading and conversion module comprises: A data reading module, used to read a first BIM geographic location model and a first BIM electrical model; the first BIM geographic location model includes a building longitude and a building latitude; A model mapping module, connected to the data reading module, is used to map the first BIM electrical model to a two-dimensional plane to obtain a second BIM electrical model; the second BIM electrical model includes the power load, roof photovoltaic, power distribution equipment, mains interface in the first BIM electrical model and the first circuit connection relationship between the power load, roof photovoltaic, power distribution equipment and mains interface; the mains interface is the connection point between the second BIM electrical model and the external power grid; The power supply unit segmentation module is connected to the model mapping module and is used to read the number of power loads N in the second BIM electrical model. L 、Number of rooftop photovoltaic D 、Number of power distribution equipment T 、Number of AC power interfaces G ; The N D Rooftop PV and N G Combine the N mains interfaces to obtain the first upper power supply set; L The power loads are respectively traced upward according to the first circuit connection relationship, and searched for the N L All the upper power sources that supply power to the i-th power load are used to obtain the second upper power source set of the i-th power load, i = 1, 2...N L ; The second upper power source set is a subset of the first upper power source set; establish a binary connection relationship between the i-th power load and the second upper power source set of the i-th power load, i=1, 2...N L ; The binary connection relationship between the i-th power load, the second upper power supply set of the i-th power load, and the i-th power load and the second upper power supply set of the i-th power load are respectively combined into a power supply unit model, i=1, 2...N L ; The number of power supply unit models is N L ; The model decomposition module is connected to the power supply unit segmentation module, and is used to decompose the power supply unit model to obtain the power load model, the first power supply model, the second power supply model, and the connection device model of each power supply unit model; the first power supply model is all the rooftop photovoltaics included in the power supply unit model; the second power supply model is all the AC power interfaces included in the power supply unit model.

3. The BIM model error automatic detection system using deep learning as claimed in claim 1, characterized in that: The power supply continuity detection module comprises: A power supply statistics module, used to respectively count the number of mains power interfaces in the second power supply model in each of the power supply unit models; A power supply continuity error identification module is connected to the power supply statistics module and is used to identify power supply continuity errors according to the number of AC power interfaces in the second power supply model in each of the power supply unit models; if the number of AC power interfaces in the second power supply model in at least one of the power supply unit models is 0, the power supply continuity error identification flag F0 is set to 0, otherwise the power supply continuity error identification flag F0 is set to 1.

4. The BIM model error automatic detection system using deep learning as claimed in claim 1, characterized in that: The rooftop photovoltaic power supply calculation module includes: A light measurement module, used to obtain a first light intensity and a first light time according to the building longitude and building latitude of the first BIM geographical location model; The photovoltaic power measurement module is connected to the light measurement module and is used to calculate the photovoltaic power according to the first light intensity, the first light time, and N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND And the pre-set photovoltaic neural network is used to calculate the rooftop photovoltaic power consumption Q.

5. The BIM model error automatic detection system using deep learning as claimed in claim 4, characterized in that: The photovoltaic power measurement module includes: The photovoltaic panel total area calculation module is used to convert the N D The photovoltaic panel area of ​​each rooftop photovoltaic panel is S1, S2, ... S ND Converted into the total area of ​​photovoltaic panels S through the area conversion formula sum , where the area conversion formula is: ； Among them, k lap is the pre-set overlap factor; The neural network calculation module is connected to the photovoltaic panel total area calculation module and is used to calculate the total area S of the photovoltaic panel. sum , the first light intensity, and the first light time are input into a preset photovoltaic neural network to calculate the rooftop photovoltaic power consumption Q.

6. The BIM model error automatic detection system using deep learning as claimed in claim 1, characterized in that: The detection result output module comprises: The upper level identification and extraction module is used to extract the power supply continuity error identification mark F0 and the power supply quantity error identification mark F3; The detection result calculation module is connected to the upper-level identification and extraction module, and is used to convert the power supply continuity error identification mark F0 and the power supply error identification mark F3 into the BIM model error detection mark F4 through the detection result calculation formula, wherein the detection result calculation formula is: ； The detection result output display module is connected to the detection result calculation module, and is used to convert the BIM model error detection identifier F4 into a BIM model error detection result; if the BIM model error detection identifier F4=0, "model error" is output; if the BIM model error detection identifier F4=1, "model correct" is output.

Images