A data processing method for a smart management platform

Abstract

The present invention relates to the technical field of data analysis and processing, and specifically to a data processing method for an intelligent management platform. The present invention first obtains the vibration signal and swing data of each tower crane in each operating cycle; then analyzes and obtains the vibration characteristic parameters of each vibration data, and then constructs a two-dimensional sample point in combination with the swing data to obtain a state reference cluster; according to the deviation of all two-dimensional sample points relative to each state reference cluster in the current operating cycle of the tower crane to be monitored, the tower crane to be monitored is subjected to stability monitoring and early warning. The present invention quantitatively evaluates the corresponding vibration characteristic parameters by analyzing the local vibration disorder of the vibration data, and obtains the state reference cluster in combination with the vibration characteristic parameters of the vibration data in the historical operating cycle of the tower crane and the swing data, which can help evaluate the abnormal conditions of the tower crane under complex working conditions, thereby improving the stability monitoring effect of the tower crane by the intelligent management platform.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis and processing, and in particular to a data processing method for an intelligent management platform. Background Art

[0002] In the wave of modernization and transformation of the construction industry, the smart management platform integrates cutting-edge technologies such as the Internet of Things, cloud computing, big data analysis, and artificial intelligence to build a comprehensive system covering the entire process of construction site data, becoming a key technology for improving project efficiency, ensuring construction safety, and achieving green construction. Tower cranes are a type of large-scale construction equipment commonly used on construction sites, and their stability is crucial to construction safety; traditional tower crane maintenance is usually carried out regularly, and it is difficult to monitor and warn of abnormal stability of tower cranes in a timely manner.

[0003] With the construction and operation of the smart site management platform, it is possible to evaluate the stability of the tower crane and other safety risks by determining whether the real-time collected data such as the vibration or swing of the tower crane exceeds the safety threshold. However, the operating conditions of the tower crane are complex, and the vibration and swing of the tower crane are not only affected by the hoisted objects, but also by external environmental factors such as wind. The setting of the safety threshold is limited by experience and cannot adapt to complex conditions, which in turn affects the stability monitoring effect of the tower crane. Summary of the invention

[0004] In order to solve the technical problem that the existing intelligent management platform has poor stability monitoring effect on tower cranes, the purpose of the present invention is to provide a data processing method for an intelligent management platform. The technical solution adopted is as follows:

[0005] Obtain the vibration signal and swing data of each tower crane in each operation cycle in the smart management platform;

[0006] According to the frequency characteristics of each vibration signal and the local fluctuation of each vibration data in the vibration signal, a reference period and a corresponding periodic signal segment of each vibration data are obtained; within the periodic signal segment of each vibration data, according to the change and distribution of all vibration data, combined with the deviation of each vibration data from the reference period of other vibration data, the vibration disorder degree of each vibration data is obtained;

[0007] Clustering all vibration data in each of the vibration signals according to the reference period to obtain all clusters; adjusting the vibration disorder degree corresponding to each vibration data in the cluster according to the discreteness of the reference period of all vibration data in each cluster and the difference between each vibration data and the reference period corresponding to the cluster center to obtain vibration characteristic parameters;

[0008] According to the vibration characteristic parameters of each vibration data and the corresponding swing data at the same time, a two-dimensional sample point is constructed; according to all the two-dimensional sample points in the historical operation cycles of all tower cranes, a preset number of state reference clusters are obtained; according to the deviation of all the two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to each state reference cluster, stability monitoring and early warning are performed on the tower crane to be monitored.

[0009] Furthermore, the method for obtaining the reference period and the periodic signal segment includes:

[0010] Based on the frequencies corresponding to all peaks in the frequency spectrum corresponding to each vibration signal, obtaining all vibration cycles of each vibration signal;

[0011] In each of the vibration signals, any vibration data is taken as target data, and twice the duration between the target data and the nearest extreme point in the time series is taken as the target period of the target data;

[0012] Among all the vibration cycles, the vibration cycle with the smallest period difference from the target cycle is used as the reference period of the target data; with the target data as the center, a local signal segment with the same duration as that of the reference cycle is obtained in the vibration signal as the periodic signal segment of the target data.

[0013] Furthermore, the method for obtaining the vibration disorder degree includes:

[0014] In the periodic signal segment of each vibration data, according to the range and skewness of all vibration data, obtaining the change characteristic parameters of the vibration data in the periodic signal segment;

[0015] In the reference period of each vibration data, according to the period length difference between each vibration data and all other vibration data corresponding to the reference period, a period disorder coefficient of the vibration data in the reference period is obtained; the period length difference is positively correlated with the period disorder coefficient;

[0016] The vibration disorder degree corresponding to the reference period is obtained according to the change characteristic parameter and the disorder coefficient, and the change characteristic parameter and the disorder coefficient are both positively correlated with the vibration disorder degree.

[0017] Furthermore, the method for obtaining the vibration characteristic parameters includes:

[0018] Normalizing the period difference between each vibration data in the cluster and the reference period corresponding to the cluster center, and using the normalization result as an adjustment parameter for the vibration disorder degree corresponding to each vibration data;

[0019] According to the discreteness of the reference period of each vibration data in each of the clusters, combined with the adjustment parameter, an adjustment weight of each vibration data is obtained;

[0020] The vibration disorder degree corresponding to the vibration data is weighted by using the adjustment weight, and the weighted result is used as a vibration characteristic parameter.

[0021] Furthermore, the method for obtaining the adjustment weight includes:

[0022] Calculating the variance of the reference period of all vibration data in each of the clusters, and using the normalized variance as the degree of dispersion;

[0023] If the discrete degree is greater than a preset threshold, the sum of the constant 1 and the adjustment parameter is used as the adjustment weight of the corresponding vibration data;

[0024] If the discrete degree is less than or equal to a preset threshold, a value obtained by subtracting the adjustment parameter from a constant 1 is used as an adjustment weight for the corresponding vibration data.

[0025] Furthermore, the method for acquiring the state reference cluster includes:

[0026] Obtaining the working condition vector of each tower crane in each operation cycle in the smart platform, wherein the vector elements in the working condition vector at least include the length of the operation cycle, the wind speed index, the temperature index, and the weight of the hoisted object;

[0027] A tower crane with the same model and specification as the tower crane to be monitored is selected as a preliminary tower crane; based on the similarity between the working condition vectors, all reference operating cycles of the current operating cycle of the tower crane to be monitored are screened out from all historical operating cycles of all the preliminary tower cranes;

[0028] All the two-dimensional sample points in all the reference operation cycles are clustered to obtain a preset number of state reference clusters.

[0029] Furthermore, the method of clustering all the two-dimensional sample points in all the reference operation cycles includes:

[0030] Based on a preset K value and a K-means algorithm, all the two-dimensional sample points in all the reference operation cycles are clustered; the preset K value is equal to the preset number.

[0031] Furthermore, the preset number is 2.

[0032] Furthermore, the method for performing stability monitoring and early warning on the tower crane to be monitored according to the deviation of all the two-dimensional sample points relative to each state reference cluster in the current operation cycle of the tower crane to be monitored includes:

[0033] The state reference cluster includes a first state reference cluster and a second state reference cluster;

[0034] Calculate the mean of the Euclidean distances of all the two-dimensional sample points in each state reference cluster relative to the corresponding cluster center, and obtain the first warning parameter and the second warning parameter respectively;

[0035] The mean value of the Euclidean distances of all the two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to the cluster center of the first state reference cluster is used as the first state parameter of the current operation cycle of the tower crane to be monitored; the mean value of the Euclidean distances of all the two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to the cluster center of the second state reference cluster is used as the second state parameter of the current operation cycle of the tower crane to be monitored;

[0036] According to the magnitude relationship between the first state parameter and the first warning parameter, and the magnitude relationship between the second state parameter and the second warning parameter, stability monitoring and early warning are performed on the monitored tower crane.

[0037] Furthermore, the method for performing stability monitoring and early warning on the tower crane to be monitored according to the magnitude relationship between the first state parameter and the first early warning parameter, and the magnitude relationship between the second state parameter and the second early warning parameter, includes:

[0038] If the first state parameter is less than or equal to the first warning parameter, and the second state parameter is less than or equal to the second warning parameter, the tower crane is determined to be normal; if the first state parameter is greater than the first warning parameter, or the second state parameter is greater than the second warning parameter, the tower crane is determined to be abnormal.

[0039] The present invention has the following beneficial effects:

[0040] The present invention first obtains the vibration signal and swing data of each tower crane in each operation cycle; obtains the reference period and corresponding periodic signal segment of each vibration data according to the frequency characteristics of each vibration signal and the local fluctuation of each vibration data in the vibration signal; roughly analyzes and obtains the vibration period of the tower crane at the corresponding moment of each vibration data from the time domain and the frequency domain, so as to subsequently analyze the change of the vibration period and evaluate the stability of the tower crane; in the periodic signal segment of each vibration data, according to the change and distribution of all vibration data and the deviation of each vibration data from the reference period of other vibration data, obtains the vibration disorder degree of each vibration data, and preliminarily quantifies the local complex vibration characteristics of each vibration data; because the vibration disorder degrees of vibration data corresponding to similar reference periods should be similar, all vibration data in each vibration signal are clustered according to the reference period. Classification, so as to adjust the vibration disorder degree according to the clustering results later; according to the discrete situation of the reference period of all vibration data in each cluster, combined with the difference between each vibration data and the corresponding reference period of the cluster center, adjust the vibration disorder degree of each vibration data in the corresponding cluster, so that the disordered vibration characteristics are easy to distinguish from the more regular periodic vibration characteristics, and obtain vibration characteristic parameters; according to the vibration characteristic parameters of each vibration data and the corresponding swing data at the same time, construct a two-dimensional sample point; according to all two-dimensional sample points in the historical operation cycle of all tower cranes, obtain a preset number of state reference clusters, and the state reference cluster roughly contains all vibration characteristic parameters and possible situations of swing data of the tower crane to be monitored; and then according to the deviation of all two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to each state reference cluster, the stability monitoring and early warning of the tower crane to be monitored is carried out. The present invention quantitatively evaluates the vibration characteristic parameters by analyzing the local vibration disorder of each vibration data, and then obtains the state reference cluster in combination with the vibration characteristic parameters and swing data of the vibration data in all historical operation cycles of all tower cranes, which can help evaluate the abnormal situation of the tower crane under complex working conditions, thereby improving the stability monitoring effect of the tower crane on the intelligent management platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. 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 creative work.

[0042] Figure 1 A flow chart of a data processing method for an intelligent management platform provided by an embodiment of the present invention;

[0043] Figure 2A flow chart of a method for obtaining a vibration disorder degree provided by an embodiment of the present invention;

[0044] Figure 3 A flowchart of a method for acquiring vibration characteristic parameters provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the data processing method of a smart management platform proposed by the present invention, its specific implementation method, structure, characteristics and effects are described in detail as follows 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. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[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 is a detailed description of a data processing method for a smart management platform provided by the present invention with reference to the accompanying drawings.

[0048] See also Figure 1 , which shows a flow chart of a data processing method of a smart management platform provided by an embodiment of the present invention, specifically comprising:

[0049] Step S1, obtaining the vibration signal and swing data of each tower crane hook in each operation cycle in the smart management platform.

[0050] In one embodiment of the present invention, the tower crane Internet of Things monitoring system in the smart management platform is first used to obtain in real time the vibration data and swing data of all tower cranes in each operation, i.e., each operation cycle, so as to monitor the abnormal operation status of the tower crane; specifically, a vibration sensor and a gyroscope are installed at the hook in the middle position of the tower crane's boom to collect the vibration data and swing data of the tower crane in real time during operation, and the swing data is the angular velocity; the acquisition frequency of the sensor is set to 1s / time, and the implementer can set it according to actual needs; further, a data curve is constructed according to the acquisition timing sequence to obtain vibration signals and swing signals for subsequent analysis of vibration disorder.

[0051] Step S2, obtaining the reference period and corresponding periodic signal segment of each vibration data according to the frequency characteristics of each vibration signal and the local fluctuation of each vibration data in the vibration signal; within the periodic signal segment of each vibration data, obtaining the vibration disorder degree of each vibration data according to the change and distribution of all vibration data and the deviation of each vibration data from the reference period of other vibration data.

[0052] When a tower crane is carrying objects, it usually uses a hook to lift objects for horizontal and vertical movements. The vertical movement includes the lifting or lowering of the hook, and the horizontal movement includes the luffing movement of the hook and the swing movement of the boom. During vertical movement, because the object is subjected to force in the vertical direction, the swing data is basically zero, and the vibration of the tower crane is less affected; but horizontal movement may pose a threat to the stability of the tower crane.

[0053] Because when the tower crane moves horizontally, no matter it is variable amplitude motion or rotary motion, the objects will be affected by horizontal force during the lifting process. Due to the inertia of the objects, the objects will swing periodically, and then react to the tower crane through the hook, causing the tower crane to swing and vibrate periodically, thus threatening the stability of the tower crane. In addition, during the operation of the tower crane, multiple motion states are superimposed, and external environmental factors such as wind will interfere with the regular vibration period of the objects, making the vibration changes of the tower crane irregular and disordered, thus interfering with the vibration analysis.

[0054] Therefore, the embodiment of the present invention first obtains the reference period and corresponding periodic signal segment of each vibration data according to the frequency characteristics of each vibration signal and the local fluctuation of each vibration data in the vibration signal. The reference period is based on the frequency characteristics and local fluctuations, and roughly analyzes and obtains the vibration period of the tower crane at the corresponding moment of each vibration data from the time domain and frequency domain of the complex vibration signal, so as to subsequently analyze the vibration period change and evaluate the stability of the tower crane; the periodic signal segment is the corresponding signal segment within the reference period of each vibration data, so as to subsequently analyze the vibration change characteristics within the reference period corresponding to each vibration data.

[0055] Preferably, in one embodiment of the present invention, considering that spectrum analysis can effectively capture all main frequency components in the vibration signal, thereby providing comprehensive vibration characteristics, the period of the vibration signal can be obtained based on the frequency domain diagram analysis; and considering that the local fluctuation of the vibration signal can provide instantaneous or short-term changes of the signal, it can also help analyze the period of the vibration signal; therefore, the method for obtaining the reference period and periodic signal segment includes:

[0056] Based on the frequencies corresponding to all peaks in the spectrum diagram corresponding to each vibration signal, all vibration periods of each vibration signal are obtained; in each vibration signal, any vibration data is taken as the target data, and twice the duration between the target data and the nearest extreme point in the time series is taken as the target period of the target data; among all vibration periods, the vibration period with the smallest period difference from the target period is taken as the reference period of the target data; with the target data as the center, a local signal segment with the same duration as the reference period is obtained in the vibration signal as the periodic signal segment of the target data.

[0057] As an example, firstly, Fourier transform is performed on the vibration signal to obtain the spectrum of the vibration signal, and then all peaks in the spectrum are obtained through the peak detection algorithm, and the inverse of the frequency corresponding to each peak is taken as a vibration period, so as to obtain all vibration periods of the vibration signal; then, any vibration data in the vibration signal is taken as the target data, and the duration between the extreme point closest to it is taken as a half-period, and the half-period is multiplied by 2 to obtain the target period of the target data; then, the target period obtained in time sequence is compared with all vibration periods obtained in the frequency domain, and the vibration period with the smallest period difference is taken as the reference period of the target data, so as to obtain the periodic signal segment of the target data corresponding to the reference period.

[0058] It should be noted that obtaining a spectrum diagram through Fourier transform, peak detection, and obtaining extreme value points are all well-known technologies and will not be described in detail here.

[0059] In another embodiment of the present invention, the implementer may first decompose the vibration signal to obtain several vibration components, further perform Fourier transform on each vibration component, obtain all corresponding frequency spectra and all peaks therein, and then obtain all vibration periods of the vibration signal; then, obtain the reference period and the corresponding period signal segment in combination with the target period of the target data, which will not be repeated here.

[0060] Taking into account that the working conditions at the corresponding moment of each vibration data may be different, such as different external environmental conditions or the operating status of the tower crane, the vibration of the tower crane is complex, which will lead to inconsistent vibration change characteristics of different vibration data; considering that in the absence of external environmental influences, the tower crane only performs a single state movement, such as only vertical operation or only horizontal movement, the force on the tower crane is simple, and it usually manifests as regular vibration; the vibration characteristics of the tower crane are inconsistent under different working conditions, and when the vibration is more complex, the stability monitoring of the tower crane is more complicated, and the possibility of abnormal stability is also greater.

[0061] Therefore, in the embodiment of the present invention, the vibration disorder degree of each vibration data is obtained according to the changes and distribution of all vibration data and the deviation of each vibration data from the reference period of other vibration data within the periodic signal segment of each vibration data. The vibration disorder degree quantifies the complex vibration characteristics of each vibration data within the local period segment corresponding to the reference period, so as to subsequently evaluate the stability of the tower crane.

[0062] Preferably, in one embodiment of the present invention, the method for obtaining the vibration disorder degree includes:

[0063] See also Figure 2 , which shows a flow chart of a method for obtaining the degree of vibration disorder provided by an embodiment of the present invention, specifically comprising:

[0064] Step S201 , in each periodic signal segment of vibration data, obtaining the variation characteristic parameters of the vibration data in the periodic signal segment according to the range and skewness of all vibration data.

[0065] Considering that the range of the data can quickly identify the overall fluctuation range of the vibration data, and the skewness can help understand the distribution characteristics of the vibration data within the reference period, the combination of the two can evaluate the change characteristic parameters of the vibration data within the periodic signal segment. The change characteristic parameters preliminarily quantify the complex vibration characteristics of the vibration data.

[0066] As an example, the product of the range and the skewness is mapped to the S-type function, and the mapping result is used as the change characteristic parameter. The calculation formula of the change characteristic parameter is:

[0067] ;in, For each vibration signal The characteristic parameters of the vibration data changes within the local segment of the vibration data cycle; is an S-type function; For each vibration signal The range of all vibration data within a local segment of the vibration data cycle; For each vibration signal The skewness of all vibration data within a local segment of the period of a vibration data.

[0068] In the calculation formula of the variation characteristic parameter, the value of skewness may be positive, negative or zero, representing different distribution characteristics of vibration data respectively; its product with the range may also be positive, negative or zero. By mapping the product to the S-type function, the final mapping result will be distributed between 0 and 1 so as to participate in subsequent calculations.

[0069] It should be noted that the calculation of range and skewness are well-known techniques and will not be elaborated here; in other examples, implementers may also use other data feature measurement parameters such as variance instead of range or skewness, or use other mapping functions, which will not be elaborated here.

[0070] Step S202, within the reference period of each vibration data, according to the period length difference between each vibration data and the reference periods corresponding to all other vibration data, obtain the period disorder coefficient of the vibration data within the reference period; the period length difference is positively correlated with the period disorder coefficient.

[0071] Considering that under normal circumstances, within the local signal segment of each vibration data, the reference period of its adjacent vibration data should be similar; if the difference in the reference period of the vibration data is greater, the corresponding period disorder coefficient is greater, which means that the vibration characteristics of the data are more complex.

[0072] So as an example, the calculation formula for the periodic disorder coefficient is: ;in, For each vibration signal The period disorder coefficient of the vibration data in the periodic local segment of the vibration data; For each vibration signal In the partial period of vibration data, except for the The serial number of other vibration data other than the vibration data; For each vibration signal The total number of vibration data in a partial segment of a vibration data cycle; For each vibration signal The period length of the reference period of the vibration data; For each vibration signal The period length of the reference period of the vibration data; is the absolute value symbol.

[0073] In other examples, the implementer may also directly calculate the variance of all vibration data corresponding to the reference period within the reference period of each vibration data, and use the variance to measure the periodic disorder coefficient; the standard deviation, range, etc. may also be used, which will not be elaborated here.

[0074] Step S203, obtaining the vibration disorder degree corresponding to the reference period according to the change characteristic parameter and the disorder coefficient, wherein the change characteristic parameter and the disorder coefficient are both positively correlated with the vibration disorder degree.

[0075] As an example, the product of the variation characteristic parameter and the disorder coefficient is mapped to In the function, the mapping result is used as the vibration disorder degree; in other examples, the implementer may also combine the two by basic mathematical operations such as addition or weighted summation or related mapping means to obtain the vibration disorder degree, which will not be elaborated here.

[0076] At this point, the vibration disorder degree of each vibration data in each vibration signal of each tower crane is obtained.

[0077] Step S3, clustering all vibration data in each vibration signal according to the reference period to obtain all clusters; according to the discreteness of the reference period of all vibration data in each cluster, combined with the difference between each vibration data and the reference period corresponding to the cluster center, adjusting the vibration disorder degree corresponding to each vibration data in the corresponding cluster to obtain the vibration characteristic parameters.

[0078] Taking into account that the vibration disorder degrees of vibration data corresponding to similar reference periods should be similar, the embodiment of the present invention first clusters all vibration data in each vibration signal according to the reference period to obtain all clusters; so as to subsequently adjust the vibration disorder degrees of all vibration data in each cluster to obtain vibration characteristic parameters. The vibration characteristic parameters further reflect the local change characteristics of each vibration data. The larger the vibration characteristic parameters, the lower the possibility of tower crane stability.

[0079] In one embodiment of the present invention, a K-means algorithm is specifically used, and the difference between reference periods is used as a measure of distance. The preset K value is obtained by the elbow method, and all vibration data are clustered to obtain all clusters; the reference period of the vibration data in each cluster should be relatively similar; it should be noted that the K-means algorithm and the elbow method are already well-known technologies and will not be described in detail here; implementers may also use other clustering algorithms.

[0080] Preferably, in one embodiment of the present invention, the method for obtaining vibration characteristic parameters includes:

[0081] See also Figure 3 , which shows a flow chart of a method for obtaining vibration characteristic parameters provided by an embodiment of the present invention, specifically comprising:

[0082] Step S301 : normalize the period difference between each vibration data in the cluster and the reference period corresponding to the cluster center, and use the normalization result as an adjustment parameter of the vibration disorder degree corresponding to each vibration data.

[0083] As the corresponding reference period of each vibration data in the cluster deviates more from the cluster center, it means that it deviates more from the abnormality relative to most vibration data, and it is more likely that the vibration data corresponding to the abnormal tower crane corresponds to the reference period;

[0084] As an example, within each cluster, the absolute value of the difference between the period of the vibration data and the reference period corresponding to the cluster center is linearly normalized, and the normalized result is used as an adjustment parameter for the degree of vibration disorder. The larger the period difference, the larger the corresponding adjustment parameter, so as to obtain the adjustment weight later.

[0085] Step S302 : obtaining an adjustment weight of each vibration data according to the discreteness of the reference period of each vibration data in each cluster in combination with the adjustment parameter.

[0086] In a preferred embodiment of the present invention, the method for obtaining the adjustment weight includes:

[0087] The variance of the reference period of all vibration data in each cluster is calculated, and the normalized variance is used as the degree of dispersion; if the degree of dispersion is greater than the preset threshold, the sum of the constant 1 and the adjustment parameter is used as the adjustment weight of the corresponding vibration data; if the degree of dispersion is less than or equal to the preset threshold, the value obtained by subtracting the adjustment parameter from the constant 1 is used as the adjustment weight of the corresponding vibration data.

[0088] As an example, the variance of the reference period of all vibration data in each cluster is linearly normalized so that the value range of the mapping result is 0 to 1 to obtain the discrete degree; the preset threshold is set to 0.5, and the implementer can also set it according to actual needs; when the discrete degree is greater than 0.5, the adjustment weight of the corresponding vibration data is ; When the discrete degree is less than or equal to 0.5, the adjustment weight of the corresponding vibration data is ;in, It is the adjustment parameter of the vibration data corresponding to the vibration disorder degree.

[0089] When the reference period of the vibration data in the cluster is more discrete and the reference period of the vibration data is larger than the reference period corresponding to the cluster center, the adjustment weight of each vibration data will be larger and greater than 1, so that the vibration characteristic parameter will be larger after weighting the vibration disorder degree; when the discreteness of the reference period of the vibration data in the cluster is relatively small, the adjustment weight of the vibration data will be smaller and less than 1, so that the vibration characteristic parameter will be smaller after weighting the vibration disorder degree; by adjusting the vibration disorder degree in the overall range of all vibration data of the vibration signal in combination with the reference period, the disordered vibration characteristics are easily distinguished from the more regular periodic vibration characteristics, so as to carry out subsequent tower crane stability warning.

[0090] In another embodiment of the present invention, the implementer may also multiply the variance of the reference period of all vibration data in each cluster by the period difference between each vibration data in the cluster and the reference period corresponding to the cluster center, and then linearly normalize the product to obtain the adjustment weight of the vibration disorder degree corresponding to each vibration data; the larger the variance of the reference period in the cluster and the more the reference period of the vibration data deviates from the cluster center, the greater the adjustment weight; thereby making it easy to distinguish between the disordered vibration characteristics and the more regular periodic vibration characteristics, so as to facilitate subsequent tower crane stability warning.

[0091] Step S303: weighting the vibration disorder degree of the corresponding vibration data by adjusting the weight, and taking the weighted result as the vibration characteristic parameter.

[0092] As an example, the adjustment weight is multiplied by the vibration disorder degree, and the product corresponding to the weighted result is used as the vibration characteristic parameter.

[0093] So far, the vibration characteristic parameters of each vibration data in each vibration signal of each tower crane have been obtained.

[0094] Step S4, constructing two-dimensional sample points according to the vibration characteristic parameters of each vibration data and the corresponding swing data at the same time; obtaining a preset number of state reference clusters according to all two-dimensional sample points in the historical operation cycles of all tower cranes; and performing stability monitoring and early warning on the tower crane to be monitored according to the deviation of all two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to each state reference cluster.

[0095] Considering the stability of the tower crane, it is necessary to consider not only the vibration characteristics but also its swing conditions; therefore, the embodiment of the present invention first constructs two-dimensional sample points according to the vibration characteristic parameters of each vibration data and the corresponding swing data at the same time. It should be noted that the construction of two-dimensional sample points is a technical means well known to those skilled in the art, that is, the vibration characteristic parameters and the swing data are sample points constructed in one dimension respectively, which will not be described in detail here.

[0096] Taking into account that for large-scale smart construction sites, multiple tower cranes are usually set up at the construction site for simultaneous construction; then each operating tower crane is under the same external environment and hoisting operation state, that is, under similar working conditions, the vibration and swing conditions of each tower crane are also similar; therefore, the embodiment of the present invention further obtains a preset number of state reference clusters, and each state reference cluster roughly contains all the vibration characteristic parameters of the tower crane and possible situations of the swing data.

[0097] It should be noted that the tower cranes involved in obtaining the state reference cluster usually have similar external environment and hoisting operation status as the tower crane to be monitored, that is, when different tower cranes are operating, the temperature, wind force and other external environmental factors are relatively small, and the weight of the hoisted objects and the duration of continuous operation are also relatively small; the reference tower crane to be monitored can be determined based on the historical data in the environmental monitoring Internet of Things monitoring system and the tower crane Internet of Things monitoring system integrated in the current smart management platform, so as to obtain the state reference cluster based on the two-dimensional sample points corresponding to all reference tower cranes.

[0098] In a preferred embodiment of the present invention, a method for acquiring a state reference cluster includes:

[0099] Obtain the working condition vector of each tower crane in each operating cycle in the smart platform, where the vector elements in the working condition vector include at least the operating cycle length, wind speed index, temperature index and weight of the hoisted object; select the tower crane with the same model and specification as the tower crane to be monitored as the preliminary tower crane; based on the similarity between the working condition vectors, select all reference operating cycles of the current operating cycle of the tower crane to be monitored from all historical operating cycles of all preliminary tower cranes; cluster all two-dimensional sample points in all reference operating cycles to obtain a preset number of state reference clusters.

[0100] As an example, firstly, the working condition vector of each tower crane in each operation cycle is constructed by combining the above-mentioned vector elements, and the preliminary selected tower cranes of the same model as the tower crane to be monitored are further selected; then, the absolute value of the difference between the working condition vectors corresponding to all historical operation cycles of the preliminary selected tower crane and the working condition vector of the current operation cycle of the tower crane to be monitored is calculated, and the absolute value of the difference is linearly normalized, and the working condition vector corresponding to the historical operation cycle whose normalization result is less than the preset similarity threshold value such as 0.5 is used as the reference operation cycle of the current operation cycle of the tower crane to be monitored; further, based on the preset K value and K-means algorithm, all two-dimensional sample points in all reference operation cycles are clustered to obtain all state reference clusters; the preset K value is equal to the preset number. Among them, considering that the movement state of the tower crane mainly includes horizontal and vertical, the preset number is 2, that is, two state reference clusters are obtained.

[0101] It should be noted that clustering based on a preset K value and a K-means algorithm is a well-known technology, and implementers may also use other vector similarity measurement methods to screen reference operating cycles, which will not be elaborated here.

[0102] After obtaining all state reference clusters, considering that the larger the vibration characteristic parameters and the larger the swing data, the worse the current stability of the tower crane is, and the greater the necessity of early warning is; and the state reference cluster roughly contains all the vibration characteristic parameters of the tower crane and the possible situations of the swing data, so according to the deviation of all two-dimensional sample points relative to each state reference cluster in the current operation cycle of the tower crane to be monitored, the monitored tower crane can be given an intelligent early warning.

[0103] Taking into account that the vibration signals within all reference operating cycles may contain two-dimensional sample points when the tower crane is normal and abnormal, but the possibility of the tower crane being abnormal within all reference operating cycles is extremely low, the early warning parameters of the tower crane can be preliminarily evaluated based on all two-dimensional sample points in all reference operating cycles; and then, the two-dimensional sample points in the current operating cycle can be combined to evaluate whether the tower crane is abnormal.

[0104] Preferably, in one embodiment of the present invention, according to the deviation of all two-dimensional sample points relative to each state reference cluster in the current operation cycle of the tower crane to be monitored, the method for stability monitoring and early warning of the tower crane to be monitored includes:

[0105] The state reference cluster includes a first state reference cluster and a second state reference cluster; the mean of the Euclidean distances of all two-dimensional sample points in each state reference cluster relative to the corresponding cluster center is calculated to obtain the first warning parameter and the second warning parameter respectively;

[0106] The mean value of the Euclidean distances of all two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to the cluster center of the first state reference cluster is used as the first state parameter of the current operation cycle of the tower crane to be monitored; the mean value of the Euclidean distances of all two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to the cluster center of the second state reference cluster is used as the second state parameter of the current operation cycle of the tower crane to be monitored;

[0107] According to the magnitude relationship between the first state parameter and the first warning parameter, and the magnitude relationship between the second state parameter and the second warning parameter, stability monitoring and early warning are performed on the monitored tower crane.

[0108] Among them, if the first state parameter is less than or equal to the first warning parameter, and the second state parameter is less than or equal to the second warning parameter, the tower crane is judged to be normal; if the first state parameter is greater than the first warning parameter, or the second state parameter is greater than the second warning parameter, the tower crane is judged to be abnormal.

[0109] In the stability monitoring and early warning method, the first early warning parameter and the second early warning parameter respectively reflect the approximate distribution level of the two-dimensional sample points in the corresponding state reference cluster; the smaller the parameter value, the more concentrated most of the two-dimensional sample points are in the corresponding state reference cluster, which means that the vibration and swing of the tower crane in the corresponding state cluster are similar, which reflects that the vibration and swing of the tower crane are more regular; on the contrary, the larger the parameter value, the more chaotic the vibration and swing of the tower crane are, and the greater the threat to the stability of the tower crane; the first state parameter and the second state parameter respectively reflect the distribution of all two-dimensional sample points relative to the corresponding cluster center in the current operation cycle of the tower crane to be monitored. If the first state parameter and the second state parameter are both less than or equal to the corresponding early warning parameter, it means that the vibration characteristics of the tower crane to be monitored in the current operation cycle are closer to the reference data in all reference operation cycles, that is, the possibility of abnormality is low; if any state parameter of the first state parameter or the second state parameter is greater than the corresponding early warning parameter, that is, the first state parameter is greater than the first early warning parameter, and the second state parameter is greater than the second early warning parameter, any one of the two is true or both are true, the more chaotic the vibration and swing of the tower crane are, and the tower crane is judged to be abnormal.

[0110] It should be noted that the two state reference clusters are a horizontal motion state cluster and a vertical motion state cluster, which will not be described in detail here.

[0111] At this point, the monitored tower crane can be subjected to stability monitoring and early warning. When the tower crane is abnormal, the smart management platform will issue an early warning prompt, and the tower crane will stop working and wait for relevant equipment maintenance personnel to perform maintenance. When the tower crane is normal, the tower crane can be put into the next operation cycle and only routine regular maintenance is required.

[0112] In summary, the embodiment of the present invention first obtains the vibration signal and swing data of each tower crane in each operation cycle; then obtains the reference cycle and periodic signal segment of each vibration data in each vibration signal; analyzes and obtains the vibration disorder degree of each vibration data; then clusters the vibration data according to the reference cycle, and then adjusts the vibration disorder degree corresponding to each vibration data in combination with the clustering result to obtain vibration characteristic parameters; constructs two-dimensional sample points according to the vibration characteristic parameters and swing data of the vibration data; obtains a state reference cluster according to all two-dimensional sample points in the historical operation cycle of all tower cranes; and performs stability monitoring and early warning on the tower crane to be monitored according to the deviation of all two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to each state reference cluster. The present invention quantitatively evaluates the vibration characteristic parameters by analyzing the local vibration disorder of each vibration data, and then obtains the state reference cluster in combination with the vibration characteristic parameters and swing data of the vibration data in all historical operation cycles of all tower cranes, which can help evaluate the abnormal conditions of the tower crane under complex working conditions, thereby improving the stability monitoring effect of the tower crane on the intelligent management platform.

[0113] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0114] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A data processing method for a smart management platform, characterized in that: The method comprises: Obtain the vibration signal and swing data of each tower crane in each operation cycle in the smart management platform; According to the frequency characteristics of each vibration signal and the local fluctuation of each vibration data in the vibration signal, a reference period and a corresponding periodic signal segment of each vibration data are obtained; within the periodic signal segment of each vibration data, according to the change and distribution of all vibration data, combined with the deviation of each vibration data from the reference period of other vibration data, the vibration disorder degree of each vibration data is obtained; Clustering all vibration data in each of the vibration signals according to the reference period to obtain all clusters; adjusting the vibration disorder degree corresponding to each vibration data in the cluster according to the discreteness of the reference period of all vibration data in each cluster and the difference between each vibration data and the reference period corresponding to the cluster center to obtain vibration characteristic parameters; According to the vibration characteristic parameters of each vibration data and the corresponding swing data at the same time, a two-dimensional sample point is constructed; according to all the two-dimensional sample points in the historical operation cycle of all tower cranes, a preset number of state reference clusters are obtained; according to the deviation of all the two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to each state reference cluster, stability monitoring and early warning are performed on the tower crane to be monitored; The method for obtaining the reference period and the periodic signal segment includes: Based on the frequencies corresponding to all peaks in the frequency spectrum corresponding to each vibration signal, obtaining all vibration cycles of each vibration signal; In each of the vibration signals, any vibration data is taken as target data, and twice the duration between the target data and the nearest extreme point in the time series is taken as the target period of the target data; Among all the vibration cycles, the vibration cycle with the smallest cycle difference from the target cycle is used as the reference cycle of the target data; with the target data as the center, a local signal segment with the same duration as that of the reference cycle is obtained in the vibration signal as the periodic signal segment of the target data; The method for obtaining the vibration disorder degree includes: In the periodic signal segment of each vibration data, according to the range and skewness of all vibration data, obtaining the change characteristic parameters of the vibration data in the periodic signal segment; In the reference period of each vibration data, according to the period length difference between each vibration data and all other vibration data corresponding to the reference period, a period disorder coefficient of the vibration data in the reference period is obtained; the period length difference is positively correlated with the period disorder coefficient; Obtaining the vibration disorder degree corresponding to the reference period according to the change characteristic parameter and the disorder coefficient, wherein both the change characteristic parameter and the disorder coefficient are positively correlated with the vibration disorder degree; The method for obtaining the vibration characteristic parameters includes: Normalizing the period difference between each vibration data in the cluster and the reference period corresponding to the cluster center, and using the normalization result as an adjustment parameter for the vibration disorder degree corresponding to each vibration data; According to the discreteness of the reference period of each vibration data in each of the clusters, combined with the adjustment parameter, an adjustment weight of each vibration data is obtained; weighting the vibration disorder degree of the corresponding vibration data by using the adjustment weight, and taking the weighted result as a vibration characteristic parameter; The method for acquiring the state reference cluster includes: Obtaining the working condition vector of each tower crane in each operation cycle in the smart platform, wherein the vector elements in the working condition vector at least include the length of the operation cycle, the wind speed index, the temperature index, and the weight of the hoisted object; A tower crane with the same model and specification as the tower crane to be monitored is selected as a preliminary tower crane; based on the similarity between the working condition vectors, all reference operating cycles of the current operating cycle of the tower crane to be monitored are screened out from all historical operating cycles of all the preliminary tower cranes; Clustering all the two-dimensional sample points in all the reference operation cycles to obtain a preset number of state reference clusters; The preset number is 2; The method for performing stability monitoring and early warning on the tower crane to be monitored according to the deviation of all the two-dimensional sample points relative to each state reference cluster in the current operation cycle of the tower crane to be monitored includes: The state reference cluster includes a first state reference cluster and a second state reference cluster; Calculate the mean of the Euclidean distances of all the two-dimensional sample points in each state reference cluster relative to the corresponding cluster center, and obtain the first warning parameter and the second warning parameter respectively; The mean value of the Euclidean distances of all the two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to the cluster center of the first state reference cluster is used as the first state parameter of the current operation cycle of the tower crane to be monitored; the mean value of the Euclidean distances of all the two-dimensional sample points in the current operation cycle of the tower crane to be monitored relative to the cluster center of the second state reference cluster is used as the second state parameter of the current operation cycle of the tower crane to be monitored; According to the magnitude relationship between the first state parameter and the first warning parameter, and the magnitude relationship between the second state parameter and the second warning parameter, stability monitoring and early warning are performed on the monitored tower crane.

2. The data processing method of a smart management platform according to claim 1, characterized in that: The method for obtaining the adjustment weight includes: Calculating the variance of the reference period of all vibration data in each of the clusters, and using the normalized variance as the degree of dispersion; If the discrete degree is greater than a preset threshold, the sum of the constant 1 and the adjustment parameter is used as the adjustment weight of the corresponding vibration data; If the discrete degree is less than or equal to a preset threshold, a value obtained by subtracting the adjustment parameter from a constant 1 is used as an adjustment weight for the corresponding vibration data.

3. The data processing method of a smart management platform according to claim 1, characterized in that: The method for clustering all the two-dimensional sample points in all the reference operation cycles includes: Based on a preset K value and a K-means algorithm, all the two-dimensional sample points in all the reference operation cycles are clustered; the preset K value is equal to the preset number.

4. The data processing method of a smart management platform according to claim 1, characterized in that: The method for performing stability monitoring and early warning on a tower crane to be monitored according to the magnitude relationship between the first state parameter and the first early warning parameter, and the magnitude relationship between the second state parameter and the second early warning parameter, comprises: If the first state parameter is less than or equal to the first warning parameter, and the second state parameter is less than or equal to the second warning parameter, the tower crane is determined to be normal; if the first state parameter is greater than the first warning parameter, or the second state parameter is greater than the second warning parameter, the tower crane is determined to be abnormal.

Images