A method and device for identifying building abnormality based on intelligent algorithm

Abstract

The application relates to the technical field of building management, in particular to a building abnormal change identification method and device based on an intelligent algorithm, which comprises a building abnormal change identification center, an abnormal change analysis unit, an abnormal change division unit, a cause analysis unit, an abnormal change prediction unit and a visual response unit; the application preliminarily analyzes point cloud data at different time points to determine whether abnormal changes occur in a building, meanwhile, the abnormal building is distinguished in a color marking mode, so that the building which has abnormal changes in a target area can be directly and intuitively understood, the abnormal building is analyzed in an information progressive mode, so that it can be directly and intuitively understood whether the abnormal change of the abnormal building is caused by a sudden event or a slow long-term change accumulation, and the sudden abnormal change and the long-term accumulated abnormal change are analyzed, so that the management efficiency of the sudden abnormal change of the target building is improved, and the structure state in the target building is understood based on the strain change trend, so that advanced early warning management is facilitated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of building management, and in particular to a building abnormality identification method and device based on intelligent algorithms. BACKGROUND

[0002] With the acceleration of urbanization, the number and scale of buildings are increasing. During the use of buildings, due to changes in geological conditions, surrounding construction influences, and aging of the structure itself, different degrees of abnormality may occur, such as settlement, inclination, and crack expansion. If these abnormalities cannot be discovered and handled in time, they may lead to structural safety problems of the building, and even cause major accidents.

[0003] However, at present, traditional building abnormality monitoring methods mainly include manual measurement and simple sensor monitoring. Manual measurement such as leveling and theodolite measurement has high accuracy, but low efficiency and long measurement period, making it difficult to monitor the abnormality of the building in real time. At the same time, it is difficult to analyze the trend of the internal structure of the building based on the abnormality of the building, resulting in reduced safety of the building, and the causes of the abnormality of the building cannot be understood in time, reducing the management efficiency of the building.

[0004] In view of the above technical defects, a solution is proposed. SUMMARY

[0005] The present application aims to provide a building abnormality identification method and device based on intelligent algorithms to solve the above technical defects. The present application preliminarily analyzes the point cloud data at different time points to determine whether the building has abnormality. The abnormal buildings are distinguished by color marking to intuitively understand the buildings with abnormality in the target area. The abnormal buildings are analyzed by information progression to intuitively understand whether the abnormality of the building is a sudden event or a slow and long-term change accumulation. Analysis from both sudden abnormality and long-term cumulative abnormality helps to intuitively understand whether the sudden abnormality of the target building is caused by environmental wind or surrounding construction interference to improve the management efficiency of the sudden abnormality of the target building. On the other hand, based on the strain change trend of the target building, the internal structure state of the target building is understood, and the risk of abnormality of the target building is prewarned and managed.

[0006] The purpose of the present application can be achieved by the following technical solution: a building abnormality identification device based on intelligent algorithms, comprising a building abnormality identification center, an abnormality analysis unit, an abnormality division unit, a cause analysis unit, an abnormality prediction unit, and a visual response unit.

[0007] The building abnormality identification center is used to call initial point cloud information of each building in a target area, and send the initial point cloud information to an abnormality analysis unit for point cloud abnormality supervision evaluation analysis, to obtain abnormal points and normal points, and a visual response unit marks the abnormal points and the normal points in a corresponding set three-dimensional model of the building, marks the abnormal points as red, and marks the normal points as green.

[0008] The abnormality division unit is used to divide and analyze the Euclidean distance of the collected abnormal points to obtain a burst signal or an accumulated signal.

[0009] When the burst signal is generated, the cause analysis unit is used to analyze external causes of the building abnormality based on environmental data of an environment where the target building is located, to obtain an external interference signal or a warning signal, and when the accumulated signal is generated, the abnormality prediction unit is used to analyze the safety prediction risk of the building abnormality based on strain data of the target building, to obtain a steady-state signal or a dynamic warning signal.

[0010] Preferably, the point cloud abnormality supervision evaluation analysis process is as follows:

[0011] A monitoring period is set, and the monitoring period is set as a time threshold, initial point cloud information of each building in the entire target area within the time threshold is obtained by means of aerial photography of a UAV, the point cloud data includes a time stamp and three-dimensional coordinate parameters, the point cloud data is preprocessed, the preprocessing includes cleaning and filtering, and the initial point cloud information of the building after preprocessing is set as reference point cloud information.

[0012] Point cloud data of each building in the entire target area within the time threshold is obtained again by means of aerial photography of a UAV, and the point cloud data of the building after preprocessing is set as initial point cloud information.

[0013] Preferably, points in the reference point cloud information are set as reference points, and points in the initial point cloud information are set as target points Qi, the target point Qi in the initial point cloud information is searched for a nearest neighbor reference point Pj in the reference point cloud information, wherein i is a natural number greater than zero, j is a natural number greater than zero, the Euclidean distance D of the point pair Qi, Pj is calculated, a standard Euclidean distance is called, the Euclidean distance D is compared and analyzed with the standard Euclidean distance, and abnormal points and normal points are obtained.

[0014] Preferably, the building abnormality type division analysis process is as follows:

[0015] obtaining the Euclidean distance of the abnormal point within a time threshold, collecting the Euclidean distance of the abnormal point every t1 time length, t1 being greater than zero, obtaining the ratio between the Euclidean distance and t1 time length, setting the ratio between the Euclidean distance and t1 time length as the abnormal change rate, constructing the change curve of the abnormal change rate of the abnormal point based on the time sequence, and setting the change curve of the abnormal change rate of the abnormal point as the abnormal change rate curve;

[0016] obtaining the value obtained by subtracting the previous abnormal change rate from the next abnormal change rate based on the abnormal change rate curve, and setting the value as the burst abnormal rate, comparing and analyzing the maximum value in the burst abnormal rate with the preset burst abnormal rate peak value to obtain the burst signal or the accumulated signal.

[0017] Preferably, the external cause analysis process of the building abnormality is as follows: obtaining the time period corresponding to the burst abnormal rate corresponding to the burst signal being greater than or equal to the preset burst abnormal rate threshold, and setting the time period as the abnormal time period, obtaining the building corresponding to the abnormal point in the abnormal time period, and setting the building as the target building, obtaining the environmental data of the environment in which the target building is located in the abnormal time period, the environmental data including wind grade and construction vibration source, wherein the construction vibration source includes piling and blasting.

[0018] Preferably, the wind grade is discriminated, if the wind grade is greater than or equal to the preset wind grade threshold, a wind disturbance signal is generated, and if the wind grade is less than the preset wind grade threshold, a normal signal is generated.

[0019] The construction vibration source is discriminated, if the construction vibration source exists, a construction disturbance signal is generated, and if the construction vibration source does not exist, a non-construction signal is generated.

[0020] If the wind disturbance signal and the construction disturbance signal are generated or the wind disturbance signal and the non-construction signal are generated or the normal signal and the construction disturbance signal are generated, an external interference signal is obtained, and if the normal signal and the non-construction signal are generated, an early warning signal is obtained.

[0021] Preferably, the building abnormality safety prediction risk analysis process is as follows:

[0022] obtaining the strain data of the abnormal point in the target building corresponding to different time points within a time threshold, the strain data representing the strain value, obtaining the time interval T between adjacent two time points, obtaining the strain data between adjacent two time points, and setting them as YB1 and YB2 respectively, setting the value obtained by subtracting YB1 from YB2 as the strain change amount YP.

[0023] The ratio of the strain change amount YP to the time interval T is obtained, and the ratio of the strain change amount YP to the time interval T is set as a strain change rate, a change curve of the strain change rate is constructed based on a time sequence, and is set as a strain change rate curve; meanwhile, a plurality of sets of strain change rate curves are called, fitting processing is performed on the strain change rate curve, a linear regression model after fitting is obtained, and is set as a strain prediction regression model; the next time point is input into the strain prediction regression model, a strain change rate prediction value output by the strain prediction regression model is obtained, and the strain change rate prediction value is discriminated to obtain a steady-state signal or a dynamic early warning signal.

[0024] The beneficial effects of the present application are as follows:

[0025] (1) The present application preliminarily analyzes point cloud data at different time points to determine whether a building has moved, and distinguishes the moved building by color marking, so as to intuitively understand the moved building in the target area, and then facilitate the distinction and rational management of the building. The moved building is analyzed by information progression, so as to intuitively understand whether the movement of the moved building is a sudden event or a slow and long-term change accumulation, so as to rationally manage the building where the moved point is located.

[0026] (2) The present application analyzes from two aspects of sudden movement and long-term accumulated movement, which helps to intuitively understand whether the sudden movement of the target building is caused by environmental wind force or surrounding construction interference, and then make a reasonable response to the sudden movement of the target building, so as to improve the management efficiency of the sudden movement of the target building. On the other hand, based on the strain change trend of the target building, the state of the structure in the target building is understood, and then the target building movement risk is prewarned and managed. BRIEF DESCRIPTION OF DRAWINGS

[0027] The present application will be further described below with reference to the accompanying drawings;

[0028] Fig. 1 is a system flowchart of the present application;

[0029] Fig. 2 is a method reference diagram of embodiment three of the present application;

[0030] Fig. 3 is an analysis reference diagram of embodiment one of the present application. DETAILED DESCRIPTION

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments;

[0033] Example 1:

[0034] Please see Figs. 1-2 As shown, the present invention is a building anomaly identification device based on intelligent algorithms, including a building anomaly identification center, an anomaly analysis unit, an anomaly classification unit, a cause analysis unit, an anomaly prediction unit, and a visual response unit. The building anomaly identification center and the anomaly analysis unit are connected in a one-way communication. The anomaly analysis unit is connected in a one-way communication with the anomaly classification unit, the cause analysis unit, and the visual response unit. The anomaly classification unit and the anomaly prediction unit are connected in a two-way communication. The anomaly classification unit is connected in a one-way communication with the cause analysis unit and the visual response unit. The cause analysis unit and the visual response unit are connected in a one-way communication.

[0035] The building anomaly identification center is used to retrieve the initial point cloud information of each building in the target area and send the initial point cloud information to the anomaly analysis unit for point cloud anomaly monitoring and assessment analysis to determine whether a building has undergone anomalies. At the same time, it distinguishes anomaly buildings by color marking. The specific point cloud anomaly monitoring and assessment analysis process is as follows:

[0036] Set a monitoring period and set the monitoring period as a time threshold. Use drone aerial photography to obtain the initial point cloud information of each building in the entire target area within the time threshold. The point cloud data includes timestamps, three-dimensional coordinate parameters, etc. The point cloud data is preprocessed, including cleaning and filtering, and the initial point cloud information of the preprocessed buildings is set as the reference point cloud information.

[0037] The point cloud data of each building in the entire target area within the time threshold is obtained again by drone aerial photography, and the preprocessed point cloud data of the buildings is set as the initial point cloud information.

[0038] Reference point cloud information in the point is set as a reference point, and an initial point cloud information point is set as a target point Qi;

[0039] The target point Qi in the initial point cloud information is acquired, and the nearest neighboring reference point Pj of the target point Qi in the reference point cloud information is searched, wherein i is a natural number greater than zero, j is a natural number greater than zero, the Euclidean distance D of the point pair Qi and Pj is calculated, the standard Euclidean distance is called, the Euclidean distance D is compared with the standard Euclidean distance, if the Euclidean distance D is greater than or equal to the standard Euclidean distance, the corresponding target point is set as an abnormal point, if the Euclidean distance D is less than the standard Euclidean distance, the corresponding target point is set as a normal point, and the visual response unit is used to respond to the abnormal point and the normal point. The abnormal point and the normal point are marked in the corresponding set building three-dimensional model, the abnormal point is marked as red, and the normal point is marked as green, so that the building in the target area which has an abnormality can be intuitively understood, and the building can be conveniently distinguished and reasonably managed.

[0040] The abnormal point is used for building abnormal type division analysis of the collected Euclidean distance, so that the abnormality of the abnormal point is intuitively understood, whether it is a sudden event or a slow long-term change accumulation, so that the building where the abnormal point is located can be reasonably managed. The specific building abnormal type division analysis process is as follows:

[0041] The Euclidean distance of the abnormal point within a time threshold is acquired, the Euclidean distance of the abnormal point is collected every t1 time length, t1 is greater than zero, the ratio between the Euclidean distance and t1 time length is acquired, the ratio between the Euclidean distance and t1 time length is set as an abnormal change rate, a change curve of the abnormal change rate of the abnormal point is constructed based on time sequence, and the change curve of the abnormal change rate of the abnormal point is set as an abnormal change rate curve.

[0042] Based on the abnormal change rate curve, the value obtained by subtracting the previous abnormal change rate from the next abnormal change rate is acquired, the value obtained by subtracting the previous abnormal change rate from the next abnormal change rate is set as a sudden abnormal rate, and the maximum value in the sudden abnormal rate is compared with a preset sudden abnormal rate peak value. If the maximum value in the sudden abnormal rate is greater than or equal to the preset sudden abnormal rate peak value, a sudden signal is generated, that is, it is considered that the change is caused by a rapid sudden event, if the maximum value in the sudden abnormal rate is less than the preset sudden abnormal rate peak value, an accumulation signal is generated, and it is considered that the change is caused by a slow long-term change accumulation. The visual response unit is used to respond to the sudden signal or the accumulation signal, and the preset warning text corresponding to the sudden signal or the accumulation signal is immediately displayed, so that the abnormality of the abnormal point is intuitively understood, whether it is a sudden event or a slow long-term change accumulation, so that the building where the abnormal point is located can be reasonably managed.

[0043] Embodiment two:

[0044] When the burst signal is generated, the cause analysis unit is used to analyze the external causes under the building abnormality of the collected environmental data of the environment where the target building is located, so as to intuitively understand whether the target building burst abnormality is caused by environmental wind or surrounding construction interference, and then make reasonable response to the target building burst abnormality, so as to improve the management efficiency of the target building burst abnormality. The specific external cause analysis process under the building abnormality is as follows:

[0045] The burst abnormality rate corresponding to the burst signal is obtained, which is greater than or equal to the preset burst abnormality rate threshold value corresponding to the time period, and is set as the abnormality time period. The target building corresponding to the abnormality point in the abnormality time period is obtained, and is set as the target building. The environmental data of the environment where the target building is located in the abnormality time period is obtained, including wind grade and construction vibration source, wherein the construction vibration source includes piling, blasting and the like.

[0046] The wind grade is discriminated and processed. If the wind grade is greater than or equal to the preset wind grade threshold value, the wind interference signal is generated. If the wind grade is less than the preset wind grade threshold value, the normal signal is generated.

[0047] The construction vibration source is discriminated and processed. If the construction vibration source exists, the construction interference signal is generated. If the construction vibration source does not exist, the non-construction signal is generated.

[0048] If the wind interference signal and the construction interference signal are generated, or the wind interference signal and the non-construction signal are generated, or the normal signal and the construction interference signal are generated, the external interference signal is obtained.

[0049] If the normal signal and the non-construction signal are generated, the early warning signal is obtained. The visual response unit is used to respond to the external interference signal or the early warning signal, and immediately displays the preset early warning words corresponding to the external interference signal or the early warning signal, so as to intuitively understand whether the target building burst abnormality is caused by environmental wind or surrounding construction interference, and then make reasonable response to the target building burst abnormality, so as to improve the management efficiency of the target building burst abnormality.

[0050] When the accumulation signal is generated, the abnormality prediction unit is used to analyze the safety prediction risk of the strain data of the abnormality point in the target building collected, so as to analyze from the strain value change trend of the target building, so as to make early warning to the strain risk of the target building, and then make early warning management to the abnormality risk of the target building, so as to improve the safety of the target building. The specific safety prediction risk analysis process of building abnormality is as follows:

[0051] The strain data of the abnormality point in the target building corresponding to different time points in the time threshold value is obtained, and the strain data represents the strain value.

[0052] Obtaining the time interval T between two adjacent time points, for example, if the time for the first time to collect the strain data of the abnormal point in the target building is s1, the time for the second time to collect the strain data of the abnormal point in the target building is s2, then the time interval T = s2-s1;

[0053] Obtaining the strain data between two adjacent time points, and setting YB1 and YB2 respectively, and setting the value obtained by subtracting YB1 from YB2 as the strain change YP;

[0054] Obtaining the ratio of strain change YP and time interval T, and setting the ratio of strain change YP and time interval T as the strain change rate, constructing the change curve of strain change rate based on time sequence, and setting it as the strain change rate curve, meanwhile, calling multiple groups of strain change rate curves, fitting the strain change rate curve, obtaining the linear regression model after fitting, and setting it as the strain prediction regression model, inputting the next time point into the strain prediction regression model, obtaining the strain change rate prediction value output by the strain prediction regression model, and discriminating the strain change rate prediction value:

[0055] If the strain change rate prediction value is less than the preset strain change rate prediction value threshold, a steady state signal is generated;

[0056] If the strain change rate prediction value is greater than or equal to the preset strain change rate prediction value threshold, a dynamic early warning signal is generated, and the visual response unit is used to respond to the steady state signal or the dynamic early warning signal, and immediately display the preset warning words corresponding to the steady state signal or the dynamic early warning signal, so as to understand the structure state in the target building based on the strain change trend of the target building, and then to carry out early warning management on the abnormal risk of the target building, so as to improve the safety of the structure in the target building.

[0057] Embodiment three:

[0058] A building abnormality identification method based on intelligent algorithm, comprising the following steps:

[0059] Step one: point cloud abnormality supervision evaluation analysis based on point cloud information analysis, obtaining the Euclidean distance D and discriminating the normal points and abnormal points;

[0060] Step two: building abnormality type division analysis of abnormal points through information progressive way, discriminating the maximum value in the burst abnormality rate, if the burst signal is obtained, step three is entered, if the accumulation signal is obtained, step four is entered;

[0061] Step three: external cause analysis of building abnormality based on environmental data of burst signal, discriminating the wind grade and construction vibration source, and outputting feedback of external interference signal or early warning signal;

[0062] Step four: based on the strain data under the accumulated signal, the building abnormal change safety prediction risk analysis is carried out, the strain change rate prediction value is discriminated, and the steady state signal or dynamic early warning signal is outputted and fed back;

[0063] To sum up, the application preliminarily analyzes the point cloud data at different time points to determine whether the building has abnormal change, and distinguishes the abnormal building through color marking, so as to intuitively understand the abnormal building in the target area, and then facilitate the differentiation and rational management of the building. The abnormal building is analyzed by the information progressive mode, so as to intuitively understand whether the abnormal change of the abnormal building is caused by a sudden event or a slow long-term change accumulation, so as to reasonably manage the building where the abnormal point is located. At the same time, the sudden abnormal change and the long-term cumulative abnormal change are analyzed, which helps to intuitively understand whether the sudden abnormal change of the target building is caused by environmental wind force or surrounding construction interference, and then make reasonable response to the sudden abnormal change of the target building, so as to improve the management efficiency of the sudden abnormal change of the target building. On the other hand, based on the strain change trend of the target building, the structure state in the target building is understood, and the target building abnormal risk is prewarned and managed.

[0064] The size of the threshold is set for easy comparison. The size of the threshold depends on the number of sample data and the base number set by the person skilled in the art for each group of sample data. As long as it does not affect the proportional relationship between the parameter and the quantized value, it is acceptable.

[0065] The above is only the preferred embodiment of the application, but the protection scope of the application is not limited to this. Any person skilled in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the application within the technical range disclosed by the application, which should be covered within the protection scope of the application.

Claims

1. A device for identifying building transaction based on intelligent algorithm, characterized in that, The building abnormality identification center, the abnormality analysis unit, the abnormality division unit, the cause analysis unit, the abnormality prediction unit, and the visual response unit are included. The building abnormality identification center is configured to call initial point cloud information of each building in a target area, and send the initial point cloud information to the abnormality analysis unit for point cloud abnormality supervision evaluation analysis to obtain abnormal points and normal points. The visual response unit marks the abnormal points and the normal points in a corresponding set three-dimensional model of the building, marks the abnormal points as red, and marks the normal points as green. The abnormality division unit is configured to perform building abnormality type division analysis on the Euclidean distance of the collected abnormal points to obtain a burst signal or an accumulated signal. When the burst signal is generated, the cause analysis unit is configured to perform external cause analysis on environmental data of an environment in which the target building is located under building abnormality to obtain an external interference signal or a warning signal. When the accumulated signal is generated, the abnormality prediction unit is configured to perform building abnormality safety prediction risk analysis on strain data of the abnormal points in the target building to obtain a steady-state signal or a dynamic warning signal. The point cloud abnormality supervision evaluation analysis process is as follows: A monitoring period is set, and the monitoring period is set as a time threshold. Initial point cloud information of each building in the entire target area within the time threshold is obtained by aerial photography of a drone. The point cloud data includes a timestamp and three-dimensional coordinate parameters. The point cloud data is preprocessed, including cleaning and filtering. The preprocessed initial point cloud information of the building is set as reference point cloud information. The point cloud data of each building in the entire target area within the time threshold is obtained again by aerial photography of the drone. The preprocessed point cloud data of the building is set as initial point cloud information. A point in the reference point cloud information is set as a reference point, and a point in the initial point cloud information is set as a target point Qi. The target point Qi in the initial point cloud information is searched for a nearest neighboring reference point Pj in the reference point cloud information. i is a natural number greater than zero, and j is a natural number greater than zero. The Euclidean distance D of the point pair Qi and Pj is calculated. The standard Euclidean distance is called. The Euclidean distance D and the standard Euclidean distance are compared and analyzed to obtain abnormal points and normal points.

2. The device for identifying building transaction based on intelligent algorithm according to claim 1, characterized in that, The building abnormality type division analysis process is as follows: The Euclidean distance of the abnormal points within the time threshold is obtained. The Euclidean distance of the abnormal points is collected every t1 length. t1 is greater than zero. The ratio between the Euclidean distance and t1 length is obtained. The ratio between the Euclidean distance and t1 length is set as an abnormal change rate. A change curve of the abnormal change rate of the abnormal points is constructed based on time series. The change curve of the abnormal change rate of the abnormal points is set as an abnormal change rate curve. The value obtained by subtracting the previous abnormal change rate from the next abnormal change rate based on the abnormal change rate curve is set as a burst abnormality rate. The maximum value in the burst abnormality rate is compared with a preset burst abnormality rate peak value to obtain a burst signal or an accumulated signal.

3. The device for identifying building transaction based on intelligent algorithm according to claim 1, characterized in that, The external cause analysis process under the building abnormality is as follows: the corresponding burst abnormality rate of the burst signal is greater than or equal to the preset burst abnormality rate threshold value, and the time period is set as the abnormality time period, the abnormality point in the abnormality time period is obtained, and the building corresponding to the abnormality point is set as the target building, the environment data of the environment where the target building is located in the abnormality time period is obtained, and the environment data includes wind grade and construction vibration source, wherein the construction vibration source includes piling, blasting.

4. The device for identifying building transaction based on intelligent algorithm according to claim 3, characterized in that, The wind grade is discriminated, if the wind grade is greater than or equal to the preset wind grade threshold value, a wind disturbance signal is generated, if the wind grade is less than the preset wind grade threshold value, a normal signal is generated; The construction vibration source is discriminated, if the construction vibration source exists, a construction disturbance signal is generated, if the construction vibration source does not exist, a non-construction signal is generated; If the wind disturbance signal and the construction disturbance signal or the wind disturbance signal and the non-construction signal or the normal signal and the construction disturbance signal are generated, an external interference signal is obtained, if the normal signal and the non-construction signal are generated, a warning signal is obtained.

5. The intelligent algorithm based building transaction identification device as claimed in claim 1, wherein, The building abnormality safety prediction risk analysis process is as follows: The strain data of the abnormality point in the target building corresponding to different time points in the time threshold value is obtained, the strain data represents the strain value, the time interval T between the adjacent two time points is obtained, the strain data between the adjacent two time points is obtained, and is respectively set as YB1 and YB2, the value obtained by subtracting YB1 from YB2 is set as the strain change amount YP; The ratio of the strain change amount YP to the time interval T is obtained, and the ratio of the strain change amount YP to the time interval T is set as the strain change rate, a change curve of the strain change rate is constructed based on the time sequence, and is set as the strain change rate curve, a plurality of groups of strain change rate curves are called, the strain change rate curve is fitted, a fitted linear regression model is obtained, and is set as the strain prediction regression model, the next time point is input into the strain prediction regression model, the strain change rate prediction value output by the strain prediction regression model is obtained, and the strain change rate prediction value is discriminated to obtain a steady state signal or a dynamic warning signal.

6. A method for identifying building transaction based on intelligent algorithm, the method is applied to the building transaction identification device based on intelligent algorithm in any one of claims 1-5, characterized in that, The following steps are included: Step one: point cloud abnormality supervision evaluation analysis based on point cloud information analysis, the obtained Euclidean distance D is discriminated to obtain normal points and abnormal points; Step two: the abnormal point is divided into building abnormality type by the way of information progression, the maximum value in the burst abnormality rate is discriminated, if the burst signal is obtained, step three is entered, if the accumulation signal is obtained, step four is entered; Step three: external cause analysis of building abnormality based on environmental data under burst signal, the wind grade and construction vibration source are discriminated, and the external interference signal or warning signal is outputted and fed back; Step four: building abnormality safety prediction risk analysis based on strain data under accumulation signal, the strain change rate prediction value is discriminated, and the steady state signal or dynamic warning signal is outputted and fed back; The point cloud abnormality supervision evaluation analysis process is as follows: A monitoring period is set and the monitoring period is set as a time threshold, initial point cloud information of each building in the entire target area within the time threshold is obtained by means of aerial photography of a UAV, the point cloud data includes a time stamp and a three-dimensional coordinate parameter, the point cloud data is preprocessed, the preprocessing includes cleaning and filtering, and the initial point cloud information of the building after preprocessing is set as reference point cloud information; Point cloud data of each building in the entire target area within the time threshold is obtained again by means of aerial photography of a UAV, and the point cloud data of the building after preprocessing is set as initial point cloud information; Points in the reference point cloud information are set as reference points, points in the initial point cloud information are set as target points Qi, the target point Qi in the initial point cloud information is searched for a nearest neighboring reference point Pj in the reference point cloud information, wherein i is a natural number greater than zero, j is a natural number greater than zero, the Euclidean distance D of the point pair Qi, Pj is calculated, a standard Euclidean distance is called, the Euclidean distance D is compared with the standard Euclidean distance, and abnormal points and normal points are obtained; The building abnormality type classification and analysis process is as follows: The ratio between the Euclidean distance and the t1 length is obtained, the ratio between the Euclidean distance and the t1 length is set as an abnormal change rate, the value obtained by subtracting the previous abnormal change rate from the next abnormal change rate is set as a sudden abnormal change rate, the maximum value in the sudden abnormal change rate is compared with a preset sudden abnormal change rate peak value: if the maximum value in the sudden abnormal change rate is greater than or equal to the preset sudden abnormal change rate peak value, a sudden signal is generated, that is, it is considered that the change is caused by a rapid sudden event, and if the maximum value in the sudden abnormal change rate is less than the preset sudden abnormal change rate peak value, an accumulation signal is generated, and it is considered that the change is caused by slow long-term accumulation.

Images