A method and system for collecting building construction data based on BIM

Abstract

The present invention relates to the technical field of sensor data denoising, and particularly to a BIM-based construction data acquisition method and system. First, based on the characteristic that noise influence usually does not affect the overall trend, combined with the characteristic that noise signals usually have a high signal-to-noise ratio and are usually randomly distributed, the significance of the denoising effect is calculated, and the decomposed denoised signal and noise signal are determined accordingly; then, considering that environmental noise usually affects data of various data types synchronously, so that noise signals of different data types usually show similar noise influences at the same time, the decomposition accuracy of each data type is calculated; finally, the possibility of stopping iteration is comprehensively determined based on the denoising effect significance and decomposition accuracy of the denoised signal, so that the denoising effect of the final denoised signal obtained after the iteration is stopped is better, thereby improving the accuracy of BIM model construction, and thus achieving higher accuracy in construction data acquisition.

Description

Technical Field

[0001] The present invention relates to the technical field of sensor data denoising, and in particular to a BIM-based building construction data acquisition method and system. Background Art

[0002] BIM stands for Building Information Modeling, a digital tool used in engineering design, construction, and management. The construction of BIM building models relies on sensors monitoring real-time data. By seamlessly integrating real-time data with BIM models, intelligent building management is achieved. By integrating digital and information-based building models and sharing and transmitting them throughout the project planning, operation, and maintenance lifecycle, engineering and technical personnel can correctly understand and efficiently respond to building information. BIM technology can also monitor usage progress in real time, predict and adjust construction sequences, and allocate resources rationally. Therefore, the application of BIM technology in the construction industry is an inevitable trend in digital transformation and development.

[0003] Because construction sites are subject to complex factors and construction weather, the sensor monitoring data used for BIM model construction contains a lot of noise. It is necessary to separate the monitoring signals to obtain structural information data that accurately reflects the real construction related information. Existing technologies generally use time domain denoising methods such as the moving average method to reduce the impact of noise by smoothing the signal. However, because traditional time domain denoising methods usually only focus on the monitoring data of a single sensor, the denoising of the data still mainly relies on the changes in the time series data. As a result, it is impossible to accurately determine whether the data anomalies are due to abnormal conditions in the actual construction building or noise caused by the influence of the construction environment. As a result, the denoising effect is usually poor, thereby reducing the accuracy of construction data collection. Summary of the Invention

[0004] In order to solve the technical problem that traditional time-domain denoising methods cannot accurately determine whether data anomalies are caused by abnormal conditions in real construction buildings or noise caused by the influence of the construction environment, resulting in generally low denoising accuracy, the purpose of the present invention is to provide a construction data acquisition method and system based on BIM. The technical solutions adopted are as follows:

[0005] In a first aspect, the present application provides a method for collecting building construction data based on BIM, comprising:

[0006] Collecting a construction data time series signal of each data type during a construction process; dividing the construction data time series signal into at least two local signal segments;

[0007] Iteratively decompose the construction data time series signal through ICA decomposition, determine the significance of the denoising effect of the iterative decomposition signal based on the similarity between the iterative decomposition signal after each iterative decomposition and the construction data time series signal, as well as the signal-to-noise ratio fluctuation and data fluctuation of each local signal segment in the iterative decomposition signal; and screen out the denoised signal and noise signal after each iterative decomposition for each data type based on the significance of the denoising effect;

[0008] In each iterative decomposition, the decomposition accuracy of each data type is determined based on the amplitude trend similarity between the noise signal of each data type and the noise signals of other data types and the overall amplitude difference distribution; under all data types, the probability of stopping the corresponding iteration after each iterative decomposition is determined based on the denoising effect significance of the denoised signal and the decomposition accuracy;

[0009] The final denoised signals corresponding to all data types after the iteration stops are determined by ICA decomposition combined with the iteration stopping possibility; and the BIM model is constructed using BIM software based on all the final denoised signals.

[0010] Furthermore, the process of obtaining the significance of the denoising effect includes:

[0011] Obtaining two iterative decomposition signals obtained after each iterative decomposition for each data type; determining the possibility of denoising based on the similarity of signal value distribution between the iterative decomposition signals and the construction data time series signal;

[0012] On the iteratively decomposed signal, calculating the signal-to-noise ratio of each local signal segment and the amplitude standard deviation of all data points therein; determining the noise performance value of each local signal segment based on the product of the signal-to-noise ratio and the amplitude standard deviation; and determining the noise possibility of the iteratively decomposed signal based on the stability of the noise performance values ​​of each local signal segment and the overall size of the noise performance values;

[0013] A positive correlation mapping is performed on the product of the negative correlation mapping value of the noise possibility and the denoising possibility to determine the significance of the denoising effect of the iterative decomposition signal.

[0014] Furthermore, the process of obtaining the denoising possibility includes:

[0015] The DTW distance between the iterative decomposition signal and the construction data timing signal is calculated using a dynamic time warping algorithm; the negative correlation mapping value of the DTW distance is normalized to determine the denoising possibility.

[0016] Furthermore, the process of obtaining the noise possibility includes:

[0017] The product of the noise performance value mean and the noise performance value standard deviation of all local signal segments on the iterative decomposition signal is normalized to determine the noise possibility of the iterative decomposition signal.

[0018] Furthermore, the process of obtaining the denoised signal and the noise signal of each data type after each iterative decomposition according to the significance of the denoising effect includes:

[0019] After each iterative decomposition of each data type, the iterative decomposition signal corresponding to the maximum denoising effect significance is used as the denoising signal; the iterative decomposition signal corresponding to the maximum denoising effect significance is used as the noise signal.

[0020] Furthermore, the process of obtaining the decomposition accuracy includes:

[0021] Each data type is taken as the target type in turn; other data types other than the target type are taken as the corresponding comparison types; wherein, the construction data timing signals of all data types are synchronously and iteratively decomposed;

[0022] Under each iterative decomposition, the signal value difference between the noise signal of each comparison type and the noise signal of the target type at each sampling time is calculated; the standard deviation of the signal value difference at all sampling times corresponding to each comparison type is negatively correlated to determine the signal difference consistency between each comparison type and the target type; the mean of the signal value difference at all sampling times is negatively correlated to determine the interference degree consistency between each comparison type and the target type; the noise similarity between each comparison type and the target type is determined based on the product of the signal difference consistency and the interference degree consistency;

[0023] The decomposition accuracy of the target type under each iterative decomposition is determined based on the mean of the noise similarity between all comparison types and the target type.

[0024] Furthermore, the process of obtaining the possibility of iterative stop includes:

[0025] In each iterative decomposition, the iterative stop weight of each data type is determined according to the denoising effect significance of the denoised signal of each data type and the decomposition accuracy; the accumulated values ​​of the iterative stop weights of all data types are normalized to determine the corresponding iterative stop possibility after each iterative decomposition.

[0026] Furthermore, the process of obtaining the final denoised signal includes:

[0027] If there is an iteration stop possibility greater than a preset stop threshold within the preset number of iterations, the iterative decomposition is stopped when the iteration stop possibility greater than the preset stop threshold first appears, and the denoised signals of all data types obtained when the iterative decomposition is stopped are used as the final denoised signals;

[0028] If there is no iteration stop possibility greater than the preset stop threshold within the preset number of iterations, the denoised signals of all data types obtained under the iterative decomposition corresponding to the largest iteration stop possibility are used as the final denoised signals.

[0029] Furthermore, the process of constructing a BIM model using BIM software according to all the final denoised signals includes:

[0030] After uploading the final denoised signals of all data types to the cloud, the project requirement model coordinates are established in the BIM software and imported into the building model to form a BIM model.

[0031] In a second aspect, the present application provides a BIM-based construction data acquisition system, the system comprising:

[0032] A data acquisition and preprocessing module is used to collect the construction data time series signal of each data type during the construction process; and divide the construction data time series signal into at least two local signal segments;

[0033] A first determination module is configured to iteratively decompose the construction data time series signal through ICA decomposition, determine the significance of the denoising effect of the iterative decomposition signal based on the similarity between the iterative decomposition signal after each iterative decomposition and the construction data time series signal, and the signal-to-noise ratio fluctuation and data fluctuation of each local signal segment in the iterative decomposition signal; and screen out the denoised signal and noise signal of each data type after each iterative decomposition based on the significance of the denoising effect;

[0034] The second determination module is configured to determine the decomposition accuracy of each data type based on the amplitude trend similarity and overall amplitude difference distribution between the noise signal of each data type and the noise signals of other data types in each iterative decomposition; and determine the likelihood of stopping the iteration corresponding to each iterative decomposition based on the denoising effect significance of the denoised signal and the decomposition accuracy for all data types.

[0035] The BIM model construction module is used to determine the final denoised signals corresponding to all data types after the iteration stops by combining the ICA decomposition with the iteration stop possibility; and construct the BIM model through BIM software based on all the final denoised signals.

[0036] In a third aspect, the present application provides a computer device comprising a memory and a processor. The memory is configured to store computer program code, and the processor is configured to call and execute the computer program code from the memory to perform the method of the first aspect or any embodiment of the first aspect of the present application.

[0037] In a fourth aspect, the present application provides a computer program product, comprising a computer program code. When the computer program code is executed, the method of the first aspect or any embodiment of the first aspect of the present application is performed.

[0038] In a fifth aspect, the present application provides a computer-readable storage medium, which stores computer program code. When the computer program code is executed, it performs the method of the first aspect of the present application or any embodiment of the first aspect.

[0039] This application has the following beneficial effects:

[0040] This application first calculates the significance of the denoising effect based on the fact that noise influence usually does not affect the overall trend, so that the denoised construction data should be similar to the construction data collected by the sensor in terms of change trend, combined with the fact that noise signals usually have a high signal-to-noise ratio and noise is usually randomly distributed, and thereby determines the decomposed denoised signal and noise signal; then, it is necessary to consider that environmental noise usually affects data of various data types synchronously, so that noise signals of different data types usually show similar noise influences at the same time, and calculate the decomposition accuracy of each data type; finally, the possibility of stopping the iteration is comprehensively determined based on the denoising effect significance and decomposition accuracy of the denoised signal, so that the denoising effect of the final denoised signal obtained after the iteration is stopped is better, thereby improving the accuracy of BIM model construction, and thus making the accuracy of construction data collection higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0042] Figure 1 A flow chart of a method for collecting construction data based on BIM provided by one embodiment of the present invention;

[0043] Figure 2 A structural diagram of a BIM-based construction data acquisition system provided by one embodiment of the present invention;

[0044] Figure 3 A schematic diagram of the structure of a computer device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following is a detailed description of a BIM-based construction data acquisition method and system proposed by the present invention, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment, and the specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0046] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0047] The following describes in detail a BIM-based construction data collection method and system provided by the present invention with reference to the accompanying drawings.

[0048] This application embodiment provides a method for collecting building construction data based on BIM. Figure 1 , which shows a flow chart of a method for collecting construction data based on BIM according to an embodiment of the present invention, the method comprising:

[0049] Step S101: collecting construction data time series signals of each data type during the construction process; dividing the construction data time series signals into at least two local signal segments.

[0050] Since it is necessary to obtain construction data of multiple dimensions of the construction building in the construction of the BIM model, the embodiment of the present invention obtains construction data time series signals of multiple data types; in a specific implementation of the embodiment of the present invention, the data types include structural stress changes, displacement changes, vibration changes and inclination changes, and the corresponding sensors are placed at the same plane position on the construction site to collect construction data. The data can be adjusted according to the construction data monitoring needs in the specific implementation environment. No further details are given here; in addition, it should be noted that in the embodiment of the present invention, the construction data of all data types are collected synchronously, the sampling frequency is set to collect once per second, and the sampling time is set to 10 minutes. That is, this application analyzes the construction data time series signal of 10 minutes, and all the collected construction time series data are subjected to curve fitting to determine the corresponding construction data time series signal. The implementer can adjust the time according to the specific implementation environment. In addition, in order to facilitate subsequent analysis, this application divides the construction data time series signal of each data type into at least two local signal segments, so that the local characteristics of the signal can be combined in the subsequent analysis; in a specific implementation of the embodiment of the present invention, the division of the local signal segment is performed every 30 seconds, which can be adjusted. It should be noted that curve fitting is a technical means well known to those skilled in the art and will not be further defined or elaborated upon here. Furthermore, it should be noted that, to avoid the influence of different dimensions, the signal values ​​of all construction data timing signals in the embodiments of the present invention are normalized values ​​and will not be further elaborated upon subsequently.

[0051] Step S102: Iteratively decompose the construction data time series signal through ICA decomposition, and determine the significance of the denoising effect of the iterative decomposition signal based on the similarity between the iterative decomposition signal and the construction data time series signal after each iterative decomposition, as well as the signal-to-noise ratio fluctuation and data fluctuation of each local signal segment in the iterative decomposition signal; and screen out the denoised signal and noise signal of each data type after each iterative decomposition based on the significance of the denoising effect.

[0052] Blind source separation (BBS) is an effective signal separation technology, which mainly performs ICA decomposition on the signal to achieve signal denoising, that is, the noise signal is separated from the original signal through ICA decomposition to obtain the required denoised signal. The data collected from construction is mainly affected by the building itself, and the interference signal of the construction environment may come from a variety of noise signal sources, but the number of signal sources is difficult to determine, and the signals to be collected are not necessarily independent of each other. It is possible that in a certain construction environment, the construction signal obtained has a certain correlation with the noise signal of environmental interference, which makes the ICA decomposition algorithm inaccurate for the real construction signal data collection. Therefore, it is necessary to correct the stopping condition of ICA decomposition, and it is not possible to only consider the independence of the two signals to achieve signal decomposition; therefore, the main purpose of this application is to determine the conditions for iterative stopping based on ICA decomposition. In each iterative decomposition, the signal is decomposed into two iterative decomposition signals; wherein, ICA decomposition is a technical means well known to those skilled in the art, and its specific decomposition process and iterative decomposition process are not further defined or elaborated here.

[0053] Compared to the signals corresponding to real data, construction environment noise signals have a smaller impact on sensor detection data. Therefore, the denoised data should be similar to the changing trends of the construction data collected by the sensor. In other words, the denoised signal and the construction data time series signal are more similar in terms of changing trends than the noise signal. In addition, environmental noise can cause the stress data signal to fluctuate erratically on an otherwise stable basis. For example, in the absence of other interference, the stress changes caused by a stable load in a certain part of the structure should be relatively smooth and predictable according to mechanical principles. However, if environmental noise is present, the signal curve will occasionally exhibit abrupt small peaks or valleys, deviating from the normal stress change trend. The time intervals and amplitudes of these fluctuations are often random. If the stability of the obtained construction signal is disrupted due to abnormal construction, the building will still change accordingly according to mechanical principles without any obvious randomness. For example, if the monitored building shows abnormal characteristics of tilting and collapsing, the monitored stress data, inclination data, etc. at this time have the characteristics of a changing trend of building collapse, such as the inclination data increases and the amplitude of change increases; the stress data drops rapidly until it returns to zero after the collapse; therefore, the noise signal has a high signal-to-noise ratio in each local signal segment, and the difference between the signal-to-noise ratios in different local signal segments changes significantly; while the denoised signal, that is, the normal signal, usually maintains stability, and the signal-to-noise ratio performance is usually not obvious; and due to the randomness of the noise, the noise signal usually appears as a chaotic and disordered distribution in the signal value distribution, and its signal value size is random, that is, the signal value fluctuation characteristics of the noise signal are relatively more significant. Therefore, this application first determines the significance of the denoising effect of the iterative decomposition signal based on the similarity between the iterative decomposition signal and the construction data time series signal after each iterative decomposition, as well as the signal-to-noise ratio fluctuation and data fluctuation of each local signal segment in the iterative decomposition signal, so that the greater the significance of the denoising effect, the more it conforms to the characteristics of the denoised signal after denoising, and the less likely it is to be a noise signal.

[0054] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the denoising effect significance includes:

[0055] Obtain two iterative decomposition signals for each data type after each iterative decomposition; determine the denoising possibility based on the similarity of the signal value distribution between the iterative decomposition signal and the construction data timing signal; wherein, the process of obtaining the denoising possibility includes: calculating the DTW distance between the iterative decomposition signal and the construction data timing signal through the dynamic time warping algorithm; normalizing the negative correlation mapping value of the DTW distance to determine the denoising possibility. It should be noted that the dynamic time warping algorithm is a technical means well known to those skilled in the art and will not be further defined or elaborated here.

[0056] According to the definition of the dynamic time warping algorithm, the smaller the DTW distance between the iterative decomposition signal and the construction data timing signal, the more similar the two signals are as a whole; therefore, the DTW is negatively correlated and mapped so that the greater the denoising possibility obtained, the more it conforms to the characteristics of higher similarity between the denoised signal and the construction data timing signal, and the more likely the corresponding iterative decomposition signal is to belong to the denoised signal, and the more significant its denoising effect is. In other possible implementation methods of the embodiments of the present invention, the normalized value of the Pearson correlation coefficient between the iterative decomposition signal and the construction data timing signal is used as the denoising possibility; that is, the signal similarity is calculated by replacing the dynamic time warping with the Pearson correlation coefficient, which will not be further elaborated here.

[0057] Furthermore, on the iterative decomposition signal, the signal-to-noise ratio of each local signal segment and the amplitude standard deviation of all data points therein are calculated; based on the product of the signal-to-noise ratio and the amplitude standard deviation, the noise performance value of each local signal segment is determined; based on the stability of the noise performance value of each local signal segment and the overall size of the noise performance value, the noise possibility of the iterative decomposition signal is determined; wherein, the process of obtaining the noise possibility includes: normalizing the product of the mean of the noise performance value and the standard deviation of the noise performance value of all local signal segments on the iterative decomposition signal to determine the noise possibility of the iterative decomposition signal. It should be noted that the method for calculating the signal-to-noise ratio is a technical means well known to those skilled in the art, and the amplitude standard deviation is the standard deviation of all signal values ​​on the local signal segment, which will not be further elaborated here.

[0058] The larger the noise performance value, the more significant the signal-to-noise ratio and data fluctuation characteristics that characterize the noise characteristics; and due to the randomness of noise, different local signal segments usually exhibit different noise performance; therefore, the larger the mean noise performance value of all local signal segments on the iterative decomposition signal and the larger the standard deviation of the noise performance value, the more significant the corresponding noise characteristics, that is, the greater the noise possibility of the corresponding iterative decomposition signal, and the smaller the significance of the denoising effect.

[0059] Finally, the product of the negative correlation mapping value of the noise possibility and the denoising possibility is positively mapped to determine the significance of the denoising effect of the iterative decomposition signal; the smaller the noise possibility, the greater the denoising possibility, the greater the possibility that the iterative decomposition signal belongs to the denoised signal, and the better the significance of its denoising effect; therefore, of the two iterative decomposition signals iteratively decomposed, the one with the larger denoising effect significance is usually the denoised signal, and the one with the smaller one is usually the noise signal; therefore, further, according to the denoising effect significance, the denoised signal and the noise signal of each data type after each iterative decomposition are screened out, and the acquisition process includes: after each iterative decomposition of each data type, the iterative decomposition signal corresponding to the largest denoising effect significance is used as the denoised signal; the iterative decomposition signal corresponding to the largest denoising effect significance is used as the noise signal.

[0060] In a specific implementation of the embodiment of the present invention, the process of obtaining the noise performance value is expressed by the formula: Z g =X g ×σ g Among them, Z g is the noise performance value of the local signal segment g; X g is the signal-to-noise ratio of the local signal segment g.

[0061] In a specific implementation of the embodiment of the present invention, the process of obtaining the significance of the denoising effect is expressed as follows: Among them, R k,h,r is the significance of the denoising effect of the r-th iterative decomposition signal of the h-th data type after the k-th iterative decomposition; D k,h,r is the DTW distance between the rth iterative decomposition signal of the hth data type after the kth iterative decomposition and the corresponding construction data time series signal; exp() is an exponential function with a natural constant as the base; Norm() is a linear normalization function; Norm(exp(-D k,h,r )) is the denoising possibility of the r-th iterative decomposition signal of the h-th data type after the k-th iterative decomposition; S is the mean value of the noise performance of all local signal segments on the r-th iterative decomposition signal of the h-th data type after the k-th iterative decomposition; h,k,r The standard deviation of the noise performance values ​​of all local signal segments on the r-th iterative decomposition signal of the h-th data type after the k-th iterative decomposition; is the noise probability of the rth iterative decomposition signal of the hth data type after the kth iterative decomposition; α is a preset adjustment parameter used to prevent the denominator from being zero, with an empirical value of 0.1. By subtracting a constant 1 from the denominator of the denoising probability ratio, the greater the difference between the denoising effect significance and 0, the stronger the significance of the decomposed iterative decomposition signal, with a negative value indicating a noise signal and a positive value indicating a denoised signal. Thus, the noise signal and denoised signal corresponding to each data type after each iterative decomposition are screened.

[0062] Step S103: In each iterative decomposition, the decomposition accuracy of each data type is determined based on the similarity of the amplitude trends between the noise signal of each data type and the noise signals of other data types and the overall amplitude difference distribution; in all data types, the possibility of stopping the iteration after each iterative decomposition is determined based on the denoising effect significance and decomposition accuracy of the denoised signal.

[0063] Different monitoring construction signals, such as inclination, displacement, and stress, represent different physical characteristics of construction, and are associated with the building structure itself and the external environment in different ways. For example, the inclination signal mainly reflects changes in the spatial angle of the building structure and is more susceptible to factors such as vibration from nearby uneven ground and foundation settlement; while the displacement signal focuses more on the movement of the building structure in plane space, and the impact of temperature, wind load, etc. on this signal is more obvious. Therefore, the degree of impact of environmental noise on building monitoring signals of different data types varies. At this time, since the sensors used to monitor construction data are fixed in position, this means that each sensor is exposed to the same construction environment noise field at the same time. Sensors for monitoring different signals such as inclination, displacement, and stress are installed at the same plane position on a certain floor of a construction site. When a pile driver starts working nearby, the vibration noise, airborne noise, etc. generated will be transmitted to the location of these sensors at the same time, so that they receive the interference of this environmental noise at the same time. Therefore, for different monitoring signals, the impact of construction environmental noise starts and changes synchronously, and there is no difference in the time sequence; that is, if there is environmental noise that interferes with construction data, the time when the noise is generated is synchronous; therefore, the construction environmental noise has different effects on different construction monitoring signals due to different physical properties and response modes, and presents the characteristics of synchronization of influence and similar change trends due to the fixed position of the sensor; therefore, noise signals of different data types usually show similar noise effects at the same time. Based on this, this application determines the decomposition accuracy of each data type according to the similarity of the amplitude trend between the noise signal of each data type and the noise signals of other data types and the overall amplitude difference distribution; and further evaluates the ICA iterative decomposition by calculating the decomposition accuracy.

[0064] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the decomposition accuracy includes:

[0065] Each data type is taken as the target type in turn; other data types except the target type are taken as the corresponding comparison types; wherein, the construction data time series signals of all data types are synchronously iteratively decomposed; in each iterative decomposition, the signal value difference between the noise signal of each comparison type and the noise signal of the target type at each sampling moment is calculated; the standard deviation of the signal value difference at all sampling moments corresponding to each comparison type is negatively correlated and mapped to determine the consistency of the signal difference between each comparison type and the target type; the mean of the signal value difference at all sampling moments is negatively correlated and mapped to determine the consistency of the interference degree between each comparison type and the target type; wherein, the signal value difference is the absolute value of the signal value difference.

[0066] For the target type and any comparison type, the closer the difference in signal values ​​at each sampling moment is, the more consistent the impact of the environmental noise of the construction on the two data types is, which means that the decomposed noise signal is more consistent with the characteristic of noise influence, and the corresponding decomposition accuracy is higher; therefore, for the target type, the greater the consistency of the signal difference between it and each comparison type, the higher the decomposition accuracy; in addition, the smaller the difference in signal value, the more it can be explained that the noise signal generated by the construction environment has the same interference on the signals of the two data types, and the decomposed noise signal is more consistent with the characteristic of noise influence, and the corresponding decomposition accuracy is higher; therefore, for the target type, if its The greater the consistency of the interference degree between each contrast type, the higher the decomposition accuracy; therefore, finally, the noise similarity between each contrast type and the target type is determined based on the product of the signal difference consistency and the interference degree consistency; the decomposition accuracy of the target type under each iterative decomposition is determined based on the mean of the noise similarity between all contrast types and the target type; the decomposition accuracy characteristics represented by the decomposition accuracy characterized by the noise similarity combined with the signal difference consistency and the interference degree consistency are used to determine a more accurate decomposition accuracy, so that the subsequent evaluation of the iterative stopping condition is more accurate; that is, the greater the decomposition accuracy, the more accurate the separated noise signal is, and the better the iterative decomposition effect represented at the noise level.

[0067] In a specific implementation of the embodiment of the present invention, the process of obtaining the decomposition accuracy is expressed by the formula: H k,h N is the decomposition accuracy of the h-th data type after the k-th iteration decomposition; h is the number of comparison types corresponding to the hth data type; P k,h,i is the standard deviation of the signal value differences between the hth data type and the corresponding i-th comparison type at all sampling moments after the k-th iterative decomposition; exp() is an exponential function with a natural constant as the base; exp(-P k,h,i ) is the signal difference consistency between the h-th data type and the corresponding i-th comparison type after the k-th iterative decomposition; Q k,h,i is the mean of the signal value differences between the hth data type and the corresponding i-th comparison type at all sampling moments after the k-th iterative decomposition; exp(-Q k,h,i ) is the consistency of interference between the h-th data type and the corresponding i-th comparison type after the k-th iterative decomposition; exp(-P k,h,i )×exp(-Q k,h,i ) is the noise similarity between the h-th data type and the corresponding i-th comparison type after the k-th iterative decomposition.

[0068] The greater the decomposition accuracy, the better the iterative decomposition effect represented at the noise level; if the denoising effect of the denoised signal obtained after iterative decomposition is more significant, it means that the reliability of the denoised signal obtained is higher and the denoising effect is greater; therefore, the greater the denoising effect of the denoised signal is, the better the iterative decomposition effect represented at the denoising level; and the better the iterative decomposition effect, the higher the ICA decomposition accuracy of the corresponding number of times, and at this time, it is more likely that the separation of the construction environment noise signal is achieved, and the ICA decomposition should be stopped; therefore, preferably, in some possible implementation methods of the embodiments of the present invention, the acquisition process of the iterative stop possibility includes: under each iterative decomposition, according to the denoising effect significance and decomposition accuracy of the denoised signal of each data type, determining the iterative stop weight of each data type; normalizing the cumulative value of the iterative stop weight of all data types, and determining the corresponding iterative stop possibility after each iterative decomposition. That is, when the denoising effect significance and decomposition accuracy of all data types are greater as a whole, the better the signal separation or decomposition effect under the corresponding number of iterations, the more it is necessary to stop the iteration.

[0069] In a specific implementation of the embodiment of the present invention, the process of obtaining the possibility of iterative stop is expressed by the formula: Among them, Y k is the probability of stopping the iteration after the k-th iteration decomposition; Norm() is the linear normalization function; M is the number of data types; H k,m The decomposition accuracy of the mth data type for the kth iteration; R ′ k,m The significance of the denoising effect of the denoised signal of the mth data type decomposed for the kth iteration; H k,m ×R ′ k,m Iteration stopping weight for factoring the m-th data type for the k-th iteration.

[0070] Step S104: Determine the final denoised signals corresponding to all data types after the iteration stops by combining ICA decomposition with the possibility of iterative stopping; and construct a BIM model using BIM software based on all the final denoised signals.

[0071] After obtaining the iteration stopping possibility of each iterative decomposition, it is necessary to select the final denoised signal obtained after the iteration stopping according to the iteration stopping possibility to construct the BIM model for building construction data collection.

[0072] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the final denoised signal includes:

[0073] If there is an iteration stop probability greater than the preset stop threshold within the preset number of iterations, the iterative decomposition is stopped when the iteration stop probability greater than the preset stop threshold first appears, and the denoised signals of all data types obtained when the iterative decomposition is stopped are used as the final denoised signals; if there is no iteration stop probability greater than the preset stop threshold within the preset number of iterations, the denoised signals of all data types obtained under the iterative decomposition corresponding to the largest iteration stop probability are used as the final denoised signals. In a specific implementation of an embodiment of the present invention, the preset number of iterations is set to 20. By setting the preset number of iterations, an infinite loop of iterations can be prevented; the preset stop threshold is set to 0.9, taking into account that the iteration stop probability is a normalized value; the purpose of setting the preset stop threshold to be larger is to ensure the effect of the ICA algorithm decomposition.

[0074] After obtaining the required final denoised signal with high decomposition accuracy and good denoising effect, the BIM model is finally constructed to collect construction data. In a specific implementation of the embodiment of the present invention, the process of constructing a BIM model using BIM software based on all the final denoised signals includes:

[0075] After uploading the final denoised signals of all data types to the cloud, the project's required model coordinates are established in BIM software, and the building model is imported to form a BIM model. Revit or Archicad can be used as BIM software, and this application uses Revit for BIM modeling. It should be noted that the BIM modeling process is well known to those skilled in the art and will not be further elaborated here.

[0076] To sum up, the present application proposes a method for collecting construction data based on BIM. First, based on the fact that noise influence usually does not affect the overall trend, the denoised construction data should be similar to the construction data collected by the sensor in terms of change trend. Combined with the fact that noise signals usually have a high signal-to-noise ratio and noise is usually randomly distributed, the denoising effect significance is calculated, and the decomposed denoised signal and noise signal are determined accordingly. Then, it is necessary to consider that environmental noise usually affects data of various data types synchronously, so that noise signals of different data types usually show similar noise influences at the same time, and the decomposition accuracy of each data type is calculated. Finally, the possibility of stopping the iteration is comprehensively determined based on the denoising effect significance and decomposition accuracy of the denoised signal, so that the denoising effect of the final denoised signal obtained after the iteration is stopped is better, thereby improving the accuracy of BIM model construction, and making the accuracy of construction data collection higher.

[0077] This application also provides a BIM-based building construction data acquisition system, see Figure 2, which shows a structural diagram of a BIM-based construction data acquisition system provided by an embodiment of the present invention. The system includes: a data acquisition preprocessing module 201, a first determination module 202, a second determination module 203 and a BIM model construction module 204.

[0078] The data acquisition and preprocessing module 201 is used to collect the construction data time series signal of each data type during the construction process; divide the construction data time series signal into at least two local signal segments;

[0079] The first determination module 202 is configured to iteratively decompose the construction data time series signal through ICA decomposition, determine the significance of the denoising effect of the iterative decomposition signal based on the similarity between the iterative decomposition signal after each iterative decomposition and the construction data time series signal, as well as the signal-to-noise ratio fluctuation and data fluctuation of each local signal segment in the iterative decomposition signal, and screen out the denoised signal and noise signal of each data type after each iterative decomposition based on the denoising effect significance;

[0080] The second determination module 203 is configured to determine the decomposition accuracy of each data type based on the amplitude trend similarity and overall amplitude difference distribution between the noise signal of each data type and the noise signals of other data types in each iterative decomposition; and determine the likelihood of stopping the iteration after each iterative decomposition based on the denoising effect significance and decomposition accuracy of the denoised signal for all data types.

[0081] The BIM model construction module 204 is used to determine the final denoised signals corresponding to all data types after the iteration stops by combining ICA decomposition with the possibility of iterative stopping; and construct the BIM model through BIM software based on all the final denoised signals.

[0082] It should be noted that the system provided in the above embodiment is merely an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the BIM-based construction data acquisition system and the BIM-based construction data acquisition method provided in the above embodiment are based on the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0083] The present application also provides a computer device. Figure 3, which shows a schematic diagram of the structure of a computer device provided by an embodiment of the present invention, the computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, wherein when the processor 302 executes the computer program 303, the computer device can execute any one of the BIM-based construction data collection methods introduced above.

[0084] An embodiment of the present application also provides a computer program product. When the computer program product is run on a computer device, the computer device can execute any one of the BIM-based construction data collection methods introduced above.

[0085] An embodiment of the present application also provides a computer-readable storage medium, in which computer program code is stored. When the computer program code runs on a computer device, the computer device can execute any one of the BIM-based construction data collection methods introduced above.

[0086] In the embodiments provided in the present application, it should be understood that the provided computer devices, computer program products and computer-readable storage media are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the methods provided above and will not be repeated here.

[0087] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0088] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for collecting construction data based on BIM, characterized in that: The method comprises: Collecting a construction data time series signal of each data type during a construction process; dividing the construction data time series signal into at least two local signal segments; Iteratively decompose the construction data time series signal through ICA decomposition, determine the significance of the denoising effect of the iterative decomposition signal based on the similarity between the iterative decomposition signal after each iterative decomposition and the construction data time series signal, as well as the signal-to-noise ratio fluctuation and data fluctuation of each local signal segment in the iterative decomposition signal; and screen out the denoised signal and noise signal after each iterative decomposition for each data type based on the significance of the denoising effect; In each iterative decomposition, the decomposition accuracy of each data type is determined based on the amplitude trend similarity between the noise signal of each data type and the noise signals of other data types and the overall amplitude difference distribution; under all data types, the probability of stopping the corresponding iteration after each iterative decomposition is determined based on the denoising effect significance of the denoised signal and the decomposition accuracy; The final denoised signals corresponding to all data types after the iteration stops are determined by ICA decomposition combined with the iteration stopping possibility; and the BIM model is constructed using BIM software based on all the final denoised signals.

2. The method for collecting construction data based on BIM according to claim 1, characterized in that: The process of obtaining the significance of the denoising effect includes: Obtaining two iterative decomposition signals obtained after each iterative decomposition for each data type; determining the possibility of denoising based on the similarity of signal value distribution between the iterative decomposition signals and the construction data time series signal; On the iteratively decomposed signal, calculating the signal-to-noise ratio of each local signal segment and the amplitude standard deviation of all data points therein; determining the noise performance value of each local signal segment based on the product of the signal-to-noise ratio and the amplitude standard deviation; and determining the noise possibility of the iteratively decomposed signal based on the stability of the noise performance values ​​of each local signal segment and the overall size of the noise performance values; A positive correlation mapping is performed on the product of the negative correlation mapping value of the noise possibility and the denoising possibility to determine the significance of the denoising effect of the iterative decomposition signal.

3. The method for collecting construction data based on BIM according to claim 2, characterized in that: The process of obtaining the denoising possibility includes: The DTW distance between the iterative decomposition signal and the construction data timing signal is calculated using a dynamic time warping algorithm; the negative correlation mapping value of the DTW distance is normalized to determine the denoising possibility.

4. The method for collecting construction data based on BIM according to claim 2, characterized in that: The process of obtaining the noise possibility includes: The product of the noise performance value mean and the noise performance value standard deviation of all local signal segments on the iterative decomposition signal is normalized to determine the noise possibility of the iterative decomposition signal.

5. The method for collecting construction data based on BIM according to claim 1, characterized in that: The process of obtaining the denoised signal and the noise signal of each data type after each iterative decomposition according to the significance of the denoising effect includes: After each iterative decomposition of each data type, the iterative decomposition signal corresponding to the maximum denoising effect significance is used as the denoising signal; the iterative decomposition signal corresponding to the maximum denoising effect significance is used as the noise signal.

6. The method for collecting construction data based on BIM according to claim 1, characterized in that: The process of obtaining the decomposition accuracy includes: Each data type is taken as the target type in turn; other data types other than the target type are taken as the corresponding comparison types; wherein, the construction data timing signals of all data types are synchronously and iteratively decomposed; Under each iterative decomposition, the signal value difference between the noise signal of each comparison type and the noise signal of the target type at each sampling time is calculated; the standard deviation of the signal value difference at all sampling times corresponding to each comparison type is negatively correlated to determine the signal difference consistency between each comparison type and the target type; the mean of the signal value difference at all sampling times is negatively correlated to determine the interference degree consistency between each comparison type and the target type; the noise similarity between each comparison type and the target type is determined based on the product of the signal difference consistency and the interference degree consistency; The decomposition accuracy of the target type under each iterative decomposition is determined based on the mean of the noise similarity between all comparison types and the target type.

7. The method for collecting construction data based on BIM according to claim 1, characterized in that: The process of obtaining the possibility of iterative stopping includes: In each iterative decomposition, the iterative stop weight of each data type is determined according to the denoising effect significance of the denoised signal of each data type and the decomposition accuracy; the accumulated values ​​of the iterative stop weights of all data types are normalized to determine the corresponding iterative stop possibility after each iterative decomposition.

8. The method for collecting construction data based on BIM according to claim 1, characterized in that: The process of obtaining the final denoised signal includes: If there is an iteration stop possibility greater than a preset stop threshold within the preset number of iterations, the iterative decomposition is stopped when the iteration stop possibility greater than the preset stop threshold first appears, and the denoised signals of all data types obtained when the iterative decomposition is stopped are used as the final denoised signals; If there is no iteration stop possibility greater than the preset stop threshold within the preset number of iterations, the denoised signals of all data types obtained under the iterative decomposition corresponding to the largest iteration stop possibility are used as the final denoised signals.

9. The method for collecting construction data based on BIM according to claim 1, characterized in that: The process of constructing a BIM model using BIM software based on all final denoised signals includes: After uploading the final denoised signals of all data types to the cloud, the project requirement model coordinates are established in the BIM software and imported into the building model to form a BIM model.

10. A BIM-based construction data acquisition system, characterized in that: The system comprises: A data acquisition and preprocessing module is used to collect the construction data time series signal of each data type during the construction process; and divide the construction data time series signal into at least two local signal segments; A first determination module is configured to iteratively decompose the construction data time series signal through ICA decomposition, determine the significance of the denoising effect of the iterative decomposition signal based on the similarity between the iterative decomposition signal after each iterative decomposition and the construction data time series signal, and the signal-to-noise ratio fluctuation and data fluctuation of each local signal segment in the iterative decomposition signal; and screen out the denoised signal and noise signal of each data type after each iterative decomposition based on the significance of the denoising effect; The second determination module is configured to determine the decomposition accuracy of each data type based on the amplitude trend similarity and overall amplitude difference distribution between the noise signal of each data type and the noise signals of other data types in each iterative decomposition; and determine the likelihood of stopping the iteration corresponding to each iterative decomposition based on the denoising effect significance of the denoised signal and the decomposition accuracy for all data types. The BIM model construction module is used to determine the final denoised signals corresponding to all data types after the iteration stops by combining the ICA decomposition with the iteration stop possibility; and construct the BIM model through BIM software based on all the final denoised signals.

Images