Method for real-time identification of the state of the technological process of equipment for building machines

Abstract

The building machine-oriented equipment process state real-time identification method relates to the construction engineering construction technical field.Aiming at the problems of lacking monitoring means of process flow and low accuracy of working state identification of the building machine of the existing high-rise and super high-rise building in the construction process, the building machine process flow is divided into four working states of stopping, construction, lifting and pumping, the acceleration time history signal on the main force component in the operation process of the building machine is obtained in real time by using a three-axis vibration sensor, the time domain and frequency domain characteristics are analyzed, the combined feature vector under the four working states of the building machine is trained by the method of machine learning, the current measured acceleration time history signal and the corresponding feature vector are obtained, and the classification model after training is compared in real time, so that the four key working states of stopping, construction, lifting and pumping of the current building machine are quickly and accurately identified.

Description

Technical Field

[0001] This invention relates to the field of building construction technology, and in particular to a method for real-time identification of the equipment process status of building construction machines. Background Technology

[0002] Building construction rigs are high-altitude work platforms for concrete structures widely used in the construction of high-rise and super high-rise buildings. They can carry tower cranes, concrete placing booms, construction hoist attachment devices, and other construction equipment, and climb in tandem with the vertical construction rhythm of the main building structure, resulting in good overall construction efficiency. However, the structure and process of building construction rigs integrating multiple devices are relatively complex, and there are significant safety risks during the climbing process, requiring effective safety monitoring measures for building construction rigs.

[0003] Currently, safety monitoring of building construction machines mainly focuses on the mechanical and environmental conditions of structural components, while safety monitoring of their construction processes is less common. For each floor of the concrete structure, the building construction machine undergoes multiple processes, including climbing, rebar tying, formwork opening and closing, and concrete pouring. Different processes require monitoring of specific parts and components. Relying on traditional modeling methods and information model update techniques cannot effectively reveal the real-time operational status of the building construction machine. Furthermore, due to numerous construction interference factors, the accuracy of its operational status identification is low. Summary of the Invention

[0004] The existing high-rise and super high-rise building construction machines suffer from a lack of monitoring methods for process flows and low accuracy in identifying operational status during construction. The purpose of this invention is to provide a real-time identification method for the process flow status of construction machines.

[0005] The technical solution adopted by this invention to solve its technical problem is: a method for real-time identification of equipment process status for building construction machines, the steps of which are as follows:

[0006] S1: Based on the actual process flow of the building construction machine, it is divided into four working states: shutdown, construction, lifting, and pumping. These are represented by S1. A S B S C S D It means that a triaxial vibration sensor is installed at the center of the main horizontal load-bearing layer of the building machine, and the triaxial vibration sensor is fixed in close contact with the main load-bearing component of the building machine.

[0007] S3: Under the four operating states of the building machine, the acceleration data of the triaxial vibration sensor is read in real time and an acceleration time history curve is established. The acceleration time history curve is divided into n segments according to the time interval Δt, so that n acceleration time history signals are obtained along the X, Y, and Z directions under each operating state. The acceleration time history signals in the three directions are ax , a y , a z ;

[0008] S4: For a working state, in each time period Δt, the triaxial vibration sensor has m readings in each direction, and the i-th reading is a xi , a yi , a zi , i = 1, 2, 3…m, and the three direction measured acceleration time history signals are combined into an acceleration time history signal R:

[0009]

[0010] R = {R i}, i = 1, 2, 3…m

[0011] S5: Extract the root mean square value R rms , the average value , and the peak-to-peak value R pp of each acceleration time history signal R;

[0012] S6: Fourier transform the data after removing the mean value of each acceleration time history signal R to obtain the frequency domain data Y R , and calculate the entropy value R H ;

[0013] S7: Obtain the combined feature vector F = {F k}, k = 1, 2, 3…n of the n acceleration time history signals R collected by the triaxial vibration sensor, as follows:

[0014]

[0015] According to the actual process flow of the building machine, the combined feature vectors F of the acceleration time history signals R under S A , S B , S C , S D four working states are obtained respectively;

[0016] S8: Train the combined feature vectors F under S A , S B , S C , S D four working states using a classifier, and deploy the trained classification model in the information monitoring system of the industrial computer;

[0017] S9: The building machine enters the real-time recognition stage, and reads an acceleration time history curve every time period Δt. According to steps S3-S7, the current measured acceleration time history signal R c and the corresponding feature vector F are obtainedc The current working state of the formwork machine is determined by comparing the real-time acceleration time history signal R c S A , S B , S C , S D four working state results are output respectively;

[0018] Wherein, n is the total number of acceleration time history curves measured by a single triaxial vibration sensor in a single direction segmented according to a time period Δt;

[0019] m is the number of sampling points of a single triaxial vibration sensor in a single direction within a time period Δt;

[0020] Δt is the sampling interval time period set by the triaxial vibration sensor, which is a constant;

[0021] a x , a y , a z are respectively the real-time measured acceleration time history signals of the single triaxial vibration sensor in X, Y, Z directions;

[0022] a xi , a yi , a zi are respectively the acceleration values measured by the triaxial vibration sensor in X, Y, Z directions at the ith moment within a time period Δt;

[0023] R is the combined acceleration time history signal of the triaxial vibration sensor in X, Y, Z directions within a time period Δt;

[0024] R rms is the root mean square value of R;

[0025] is the average value of R;

[0026] R pp is the peak-to-peak value of R;

[0027] R H is the entropy value of R;

[0028] R i is the acceleration value of the ith moment in R;

[0029] F is the combined feature vector of n acceleration time history signals R;

[0030] F c is the combined feature vector of the measured acceleration time history signal R c ;

[0031] R cThe acceleration time history signal measured in the operation state recognition process of the building machine.

[0032] The building machine-oriented equipment process state real-time recognition method of the application firstly divides the building machine process into four working states of stop, construction, lifting and pumping, uses a three-axis vibration sensor to obtain the acceleration time history signal on the main force-bearing component in the operation process of the building machine, analyzes the time domain and frequency domain characteristics, trains the combined feature vector under the four working states of the building machine through the method of machine learning, obtains the current measured acceleration time history signal and the corresponding feature vector, compares it with the trained classification model in real time, and then quickly and accurately recognizes the four key working states of the building machine, such as stop, construction, lifting and pumping, accurately reflects the dynamic changes of the process state of the building machine on site, enhances the autonomous discrimination ability of the working state of the building machine, improves the intelligent degree of the building machine, ensures the safety of the equipment operation, effectively solves the problems of lack of real-time monitoring means and low recognition accuracy in the construction process of the traditional formwork equipment, and the practice shows that the method can better identify the key working state of the building machine by using the acceleration time domain and frequency domain characteristic data collected by the limited three-axis vibration sensor, and can avoid the adverse effects of noise generated by surrounding artificial activities, and improve the recognition accuracy.

[0033] Further, the step S9 further includes a step S10 of setting a time amount ΔT, ΔT≥4Δt, acquiring the readings of the acceleration time history signal including the previous time period Δt of the current time every time period Δt, and if the number of M SA 、M SB 、M SC 、M SD working states are recognized in the time amount ΔT, output the working state in the next time amount ΔT by the following method:

[0034] When M SA >M SB ≥0, M SC =0, M SD =0, output state S A .

[0035] When M SB >M SA ≥0, M SC =0, M SD =0, output state S B .

[0036] When M SA ≥0, M SB ≥0, M SC ≥1, M SD =0, output state S C .

[0037] When M SA ≥ 0, M SB ≥ 0, M SC = 0, M SD ≥ 1, the output state S D ;

[0038] When M SA ≥ 0, M SB ≥ 0, M SC ≥ 1, M SD ≥ 1, the output state E;

[0039] Wherein, M SA , M SB , M SC , M SD are the number of S A , S B , S C , S D four kinds of output state in single triaxial vibration sensor in the time amount ΔT;

[0040] E is an abnormal state.

[0041] Further, the step S9 further comprises: setting a time amount ΔT, and ΔT ≥ (Δt + 4s), and the reading of acceleration time history signal including every second in the previous time period of the current time is obtained every time period Δt.

[0042] Further, the step S4 further comprises: setting an interference threshold ε, and ε > 0,

[0043] If max (︱a xi ∣,︱a yi ∣,︱a zi ∣)> ε,

[0044] The corresponding a x , a y , a z value in acceleration time history signal a xi , a yi , a zi is deleted.

[0045] Further, the step S7 further comprises: setting an offset characteristic value R cr , and setting an offset amount δ of acceleration time history signal R, that is Whenever (R i - δ) × (R i-1 - δ) < 0, R cr is increased once.

[0046] R cris an offset characteristic value of the acceleration time history signal R;

[0047] is an offset of the acceleration time history signal R;

[0048] k is a set offset factor, k ∈ (0, 1).

[0049] Further, in the step S1, N triaxial vibration sensors are arranged at different main stress positions of the building machine, and according to the steps S3-S7, the working state of each triaxial vibration sensor is judged and output as follows:

[0050] When N SA = N, N SB = 0, N SC = 0, N SD = 0, the output state S A ;

[0051] When N SA ≥ 0, N SB ≥ 1, N SC = 0, N SD = 0, the output state S B ;

[0052] When N SA ≥ 0, N SB ≥ 0, N SC ≥ 1, N SD = 0, the output state S C ;

[0053] When N SA ≥ 0, N SB ≥ 0, N SC = 0, N SD ≥ 1, the output state S D ;

[0054] When N SA ≥ 0, N SB ≥ 0, N SC ≥ 1, N SD ≥ 1, the output state E;

[0055] N is the total number of triaxial vibration sensors arranged at the main stress position of a single building machine, N ≥ 2;

[0056] N SA , N SB , N SC , N SD are respectively S A , S B , S C , S DThe number of sensors corresponding to the four output states.

[0057] Furthermore, in step S5, the root mean square value R of each acceleration time history signal R segment is... rms ,average value Peak-to-peak value R pp The calculation formula is as follows:

[0058]

[0059]

[0060] R pp =max(R) i )-min(R i )

[0061] Furthermore, step S6 includes the following steps:

[0062]

[0063] S601: Perform a Fourier transform (FFT) on the mean-removed acceleration time history signal R of each segment to obtain the frequency domain data Y. R And calculate the power spectral density S R :

[0064] S602: Power spectral density S obtained from step S601 R Calculate the probability density P at each frequency point i Then, based on the probability density P i Calculate the entropy value R H The calculation formulas are as follows:

[0065]

[0066]

[0067] Furthermore, the classifier in step S8 can be a KNN or SVM machine learning algorithm for classification. Attached Figure Description

[0068] Figure 1 This is a schematic diagram of a triaxial vibration sensor installed on the main load-bearing component of a building construction machine in one embodiment of the real-time identification method for equipment process status of a building construction machine according to the present invention.

[0069] Figures 2 to 5 S is an embodiment of the present invention A S B S C S DA schematic diagram of the X-axis acceleration time history data measured by the triaxial vibration sensor under four working conditions;

[0070] Figures 6 to 8 S is an embodiment of the present invention A S B S C S D Scatter plot comparing feature values ​​under four operating conditions;

[0071] Figure 9 S is an embodiment of the present invention A S B S C S D The confusion matrix for classification prediction of four working states. The labels in the figure are as follows:

[0072] 1. Building core tube; 2. Building machine; 3. Triaxial vibration sensor. Detailed Implementation

[0073] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the drawings are all in a very simplified form and use non-precise proportions, and are only used to facilitate and clarify the purpose of illustrating the embodiments of the present invention.

[0074] Combination Figures 1 to 9 The method for real-time identification of equipment process status for building construction machines according to the present invention includes the following specific steps:

[0075] S1: Based on the actual process flow of the building construction machine 2, it is divided into four working states: shutdown, construction, lifting, and pumping. These are represented by S1. A S B S C S D Indicates; among them, the work stoppage status S A The building construction machine 2 is stationary on the building core tube 1, and there are no personnel or equipment moving on the building construction machine 2; Construction state S B This is the normal operating state of the building construction machine, including tying steel bars and dismantling and assembling formwork; lifting state S C The pumping state refers to the upward or downward movement of the building construction machine through its own mechanical power system; D This refers to the operational state of a concrete placing boom deployed on a building construction machine, which delivers concrete to the floors via pumping equipment. A triaxial vibration sensor 3 is installed at the center of the main horizontal load-bearing layer of the building construction machine 2. The triaxial vibration sensor 3 is fixed in close contact with the main load-bearing component of the building construction machine 2. The sampling frequency f of the triaxial vibration sensor 3 is not less than 100Hz. Each triaxial vibration sensor measures the acceleration data in the X, Y, and Z directions of its own location in real time and sends the data to the monitoring information system in the industrial control computer through a data acquisition device.

[0076] S3: As Figures 2 to 5 As shown, in the S of the building machine 2 A S B S C S D Under four working conditions, the acceleration data of the triaxial vibration sensor 3 is read in real time and acceleration time history curves are established. The acceleration time history curves are divided into n segments according to the time interval Δt, so that n acceleration time history signals are obtained along the X, Y, and Z directions under each working condition. The acceleration time history signals in the three directions are a x a y a z ;

[0077] S4: For a given working state, within each time interval Δt, the triaxial vibration sensor 3 has m readings in each direction, and the i-th reading is a xi a yi a zi For i = 1, 2, 3…m, the measured acceleration time history signals in the three directions are combined into a single acceleration time history signal R:

[0078]

[0079] R = {R i}, i = 1, 2, 3...m

[0080] S5: Extract the root mean square value R of each acceleration time history signal R segment. rms ,average value Peak-to-peak value R pp Three time-domain characteristics;

[0081] S6: Perform a Fourier transform (FFT) on the mean-removed acceleration time history signal R to obtain the frequency domain data Y. R And calculate the entropy value R. H ;

[0082] S7: Obtain the combined feature vector F = {F_n} of the n-segment acceleration time history signals R collected by the triaxial vibration sensor 3. k ﹜, k=1,2,3…n, as follows:

[0083]

[0084] Based on the actual process flow of the building construction machine 2, S is derived respectively. A S B S C S D The combined feature vector F of the acceleration time history signal R under four operating conditions;

[0085] S8: Train S using a classifier A S B S C S D The combined feature vector F under four working states is used to deploy the trained classification model in the information monitoring system of the industrial control computer.

[0086] S9: The building machine enters the real-time identification stage, reading an acceleration time history curve every time interval Δt. Based on steps S3 to S7, the currently measured acceleration time history signal R is obtained. c and the corresponding eigenvector F c The current working state of the building machine 2 is determined by comparing it with the classification model trained in step S8 in real time, and the measured acceleration time history signal R is output. c S A S B S C S D Four working status results;

[0087] Where n is the total number of segments of the acceleration time history curve measured in a single direction by a single triaxial vibration sensor 3 according to the time period Δt;

[0088] m represents the number of sampling points in a single direction of a single triaxial vibration sensor within the time period Δt;

[0089] f is the sampling frequency of the triaxial vibration sensor;

[0090] Δt is the sampling interval time set by the triaxial vibration sensor 3, which is a constant, such as 10s, 20s, 30s, etc.

[0091] a x a y a z These are the real-time acceleration time-history signals measured by a single triaxial vibration sensor in the X, Y, and Z directions, respectively.

[0092] a xi a yi a zi These represent the acceleration values ​​measured by the triaxial vibration sensor in the X, Y, and Z directions at time i within the time interval Δt.

[0093] R is the combined acceleration time history signal of the triaxial vibration sensor in the X, Y, and Z directions within the time period Δt;

[0094] R rms Let R be the root mean square value;

[0095] The average value of R;

[0096] Rpp R is the peak-to-peak value of R;

[0097] R c Let R be the entropy value;

[0098] R cr The offset characteristic value of the acceleration time history signal R;

[0099] R i Let be the acceleration value in R at time i;

[0100] F is a combined feature vector of n acceleration time history signals R;

[0101] R c The acceleration time history signal measured during the operation status identification of the building construction machine;

[0102] F c The measured acceleration time history signal R c The combined feature vectors.

[0103] The present invention provides a real-time identification method for the equipment process status of a building construction machine. First, the process flow of the building construction machine 2 is divided into four working states: shutdown, construction, lifting, and pumping. A triaxial vibration sensor 3 is used to acquire the acceleration time-history signals on the main load-bearing components of the building construction machine 2 in real time during operation. The time-domain and frequency-domain characteristics are analyzed. Then, a machine learning method is used to train a combined feature vector for the four working states of the building construction machine 2, obtaining the currently measured acceleration time-history signal and the corresponding feature vector. This is then compared in real time with the trained classification model to quickly and accurately identify the current shutdown, construction, lifting, or pumping state of the building construction machine 2. The method transmits four key working states, accurately reflecting the dynamic changes in the process status of the building construction machine 2 on site. This enhances the machine's ability to autonomously identify its working state, improves its intelligence level, ensures the safety of equipment operation, and effectively solves the problems of lack of real-time monitoring methods and low identification accuracy in the construction process of traditional formwork equipment. Practice has shown that this method, using the acceleration time-domain and frequency-domain characteristic data collected by the limited triaxial vibration sensor 3, can better identify the key working states of the building construction machine 2 and avoid the adverse effects of noise generated by surrounding human activities, thus improving the accuracy of identification.

[0104] To improve the reliability of the building construction machine's operating status identification, step S10 is added after step S9, the details of which are as follows:

[0105] S10: Set a time quantity ΔT, where ΔT ≥ 4Δt. If Δt = 10 seconds, then ΔT must contain at least 40 seconds. The acceleration time history curve of the time quantity ΔT is divided into at least 4 segments according to the time interval Δt. Every 10 seconds, the readings of the acceleration time history signal, including the 10 seconds preceding the current moment, are acquired, and the readings of each time interval Δt do not overlap. If M is identified within the time quantity ΔT... SA M SB M SC M SD The number of working states is determined over the next time interval ΔT by outputting the working states as follows:

[0106] When M SA >M SB ≥0, M SC =0,M SD When = 0, output state S A ;

[0107] When M SB >M SA ≥0, M SC =0,M SD When = 0, output state S B ;

[0108] When M SA ≥0, M SB ≥0, M SC ≥1, M SD When = 0, output state S C ;

[0109] When M SA ≥0, M SB ≥0, M SC =0,M SD When ≥1, output state S D ;

[0110] When M SA ≥0, M SB ≥0, M SC ≥1, M SD When ≥1, output state E;

[0111] Among them, M SA M SB M SC M SD S represents the value of a single triaxial vibration sensor within a time interval ΔT. A S B S C S DThe number of four output states. Climbing and pouring are mutually exclusive states and will not occur simultaneously in practice. If both pouring and climbing have identification results within the time ΔT, then the output state is E, which is an abnormal output state.

[0112] Step S9 further includes setting a time quantity ΔT, where ΔT ≥ (Δt + 4s). If Δt = 10 seconds, then ΔT includes at least 14 seconds. Every 10 seconds, the readings of the acceleration time history signal for each second within the 10 seconds prior to the current moment are obtained, thereby determining the working status of the building machine per second. There is a 9-second overlap between the readings of the current time period and its previous time period, which greatly reduces the time quantity ΔT and improves the recognition efficiency.

[0113] Step S4 further includes: setting an interference threshold ε, where ε > 0.

[0114] If max(|a xi ∣,︱a yi ∣,︱a zi ∣)>ε,

[0115] Then delete the acceleration time history signal a x a y a z The corresponding a in xi a yi a zi This value helps to avoid the adverse effects of interference factors such as people walking near the triaxial vibration sensor.

[0116] Step S7 further includes: due to the asynchrony of the lifting cylinder, the measurement value of the triaxial vibration sensor 3 will be offset during the lifting and lowering state of the building construction machine 2. In order to improve the feature recognition rate of the lifting and lowering state of the building construction machine 2 and avoid miscalculation due to construction interference noise, an offset feature value R is set. cr It is the number of times the acceleration time history signal waveform intersects the time axis. Given an offset δ of an acceleration time history signal R, that is... For this acceleration time history signal R, whenever (R i -δ)×(R i-1 When -δ)<0, R cr Add one more time;

[0117] R cr The offset characteristic value of the acceleration time history signal R;

[0118] δ is the offset of the acceleration time history signal R;

[0119] k is a set value. Offset factor, k∈(0,1).

[0120] To improve the accuracy of sensing at different parts of the building machine frame, in step S1, N (N≥2) triaxial vibration sensors 3 are set at different main stress points of the building machine 2. According to steps S3 to S7, the corresponding working state of each triaxial vibration sensor 3 is determined, and the output state is as follows:

[0121] When N SA =N, N SB =0, N SC =0, N SD When = 0, output state S A ;

[0122] When N SA ≥0, N SB ≥1, N SC =0, N SD When = 0, output state S B ;

[0123] When N SA ≥0, N SB ≥0, N SC ≥1, N SD When = 0, output state S C ;

[0124] When N SA ≥0, N SB ≥0, N SC =0, N SD When ≥1, output state S D ;

[0125] When N SA ≥0, N SB ≥0, N SC ≥1, N SD When ≥1, output state E;

[0126] N represents the total number of triaxial vibration sensors 3 installed on the main stress-bearing parts of a single building machine;

[0127] N SA N SB N SC N SD There are N triaxial vibration sensors, 3 of which are S A S B S C S D The number of sensors corresponding to the four output states.

[0128] In step S5, the root mean square value R of each acceleration time history signal R segment is... rms ,average value Peak-to-peak value R pp The calculation formula is as follows:

[0129]

[0130]

[0131] R pp =max(R) i )-min(R i )

[0132] Step S6 includes the following steps:

[0133] S601: Perform a Fourier transform (FFT) on the mean-removed acceleration time history signal R of each segment to obtain the frequency domain data Y. R And calculate the power spectral density S R :

[0134]

[0135] S602: Power spectral density S obtained from step S601 R Calculate the probability density P at each frequency point i Then, based on the probability density P i Calculate the entropy value R H The calculation formulas are as follows:

[0136]

[0137]

[0138] The classifier in step S8 can be a machine learning algorithm such as KNN or SVM for classification.

[0139] like Figure 1 As shown, this embodiment takes a super high-rise construction project as an example. The project adopts a frame core tube structure system. The height of the building core tube 1 is 350.10m, and the above-ground part is 70 stories. The building core tube 1 is constructed using a building construction machine. A triaxial vibration sensor 3 is installed at the web position of the steel beam on the main stress plane of the building construction machine 2 to collect vibration acceleration data in real time. The sampling frequency of the triaxial vibration sensor 3 is f = 125, and Δt is taken as 30 seconds. Then, the number of sampling points m of a single triaxial vibration sensor 3 in one direction during the time period Δt is 3750.

[0140] Based on the standard floor construction process, the working state of the building machine 2 on a standard floor is divided into four situations: shutdown, construction, lifting, and pumping, each represented by S. A S B S C S D express. Figures 2 to 5 S is shownA S B S C S D Acceleration time history data a in the X-axis direction measured by triaxial vibration sensor 3 under four working conditions x It can be preliminarily seen that the amplitude and waveform characteristics of acceleration time-domain information vary greatly under different working conditions.

[0141] To compare the relevant data characteristics, n = 53 segments of acceleration time history signal were extracted for each working state. Through steps S3-S7, a combined feature vector F of 53 acceleration time history signals was obtained, where ε = 200 and k = 0.3. Table 1 below gives the S corresponding to the acceleration time history signals. A S B S C S D Calculation results of typical characteristic values ​​for four working states.

[0142] Table 1 Typical characteristic values ​​of the building construction machine under four working states

[0143]

[0144] Figures 6 to 8 The image shows a scatter plot comparing the characteristic values ​​of 53 acceleration time-history signals under each working state of the building machine. The four working states can be well distinguished by the root mean square value, average value, peak-to-peak value, entropy value, and offset characteristic value.

[0145] The combined feature vectors are trained using a KNN classifier (K=10 nearest neighbors), and the trained classification model is deployed in the information monitoring system of the industrial control computer to obtain the current measured acceleration time history signal and the corresponding feature vector. The model is then compared and verified in real time with the trained classification model to determine the current working status of the building machine.

[0146] In the actual construction process, based on the measured results of the four working states of the building machine 2, 2,017,500 data points were extracted for each state for verification, that is, there are n=538 acceleration time history curves for each state. Figure 9 The classification prediction confusion matrix for the verification results of four working states of the building construction machine is presented. The verification results show that the recall rates for the four working states—stop, construction, lifting, and pumping—are 99.4%, 100.0%, 84.9%, and 99.3%, respectively, with an overall recognition rate of 95.9%, meeting the accuracy requirements for identifying the working states of building construction machines in high-rise and super high-rise buildings. Real-world data examples demonstrate that using only the limited acceleration time-domain and frequency-domain feature data collected by a single triaxial vibration sensor 3 is sufficient to effectively identify the key working states of the building construction machine 2, while avoiding the adverse effects of noise generated by surrounding human activities.

[0147] The above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure shall fall within the scope of the claims.

Claims

1. A method for real-time identification of equipment process status for building construction machines, characterized in that, The steps are as follows: S1: Based on the actual process flow of the building construction machine, it is divided into four working states: shutdown, construction, lifting, and pumping. These are represented by S1. A S B S C S D It means that a triaxial vibration sensor is installed at the center of the main horizontal load-bearing layer of the building machine, and the triaxial vibration sensor is fixed in close contact with the main load-bearing component of the building machine. S3: Under the four working states of the building construction machine, the acceleration data of the triaxial vibration sensor is read in real time and an acceleration time history curve is established. The acceleration time history curve is then divided into time periods Δ. t Divided into n This segment ensures that each working state obtains its own value along the X, Y, and Z directions. n The acceleration time history signals for the segment are as follows: the acceleration time history signals in the three directions are respectively... a x 、a y 、a z ; S4: For a given work state, in each time period Δ t Inside, the triaxial vibration sensor has vibrations in each direction. m The first reading, the... i The readings are respectively a xi , a yi , a zi , i =1,2,3… m The measured acceleration time history signals in the three directions are combined into a single acceleration time history signal R: ； S5: Extract the acceleration time history signal for each segment. R root mean square value R rms ,average value Peak-to-peak value R pp Three time-domain characteristics; S6: Convert each acceleration time history signal R The frequency domain data is obtained by performing a Fourier transform on the mean-removed data. Y R And calculate the entropy value. R H ; S7: The data collected by the triaxial vibration sensor is obtained. n Segment acceleration time history signal R Combined feature vectors F =﹛F k ﹜, k =1,2,3… n ,as follows: ； Based on the actual construction process of the building machine, S is derived respectively. A S B S C S D Acceleration time history signals under four operating conditions R Combined feature vectors F ; S8: Train S using a classifier A S B S C S D The combined feature vector F under four working states is used to deploy the trained classification model in the information monitoring system of the industrial control computer; S9: The building machine enters the real-time recognition stage, and every time interval Δ... t Read a segment of the acceleration time history curve, and obtain the currently measured acceleration time history signal according to steps S3~S7. R c and the corresponding feature vectors F c The measured acceleration time history signal is then compared in real time with the classification model trained in step S8 to determine the current working state of the building machine, and outputs the measured acceleration time history signal. R c S A S B S C S D Four working status results; S10: Set a time quantity Δ T Δ T ≥4Δ t Δ t Get the previous time period Δ including the current time. t The reading of the acceleration time history signal, if the time quantity Δ T M was identified respectively within SA M SB M SC M SD The number of working states, in the following time Δ T The working status is output using the following method: When M SA >M SB ≥0, M SC =0, M SD When =0, output state S A ; When M SB >M SA ≥0, M SC =0, M SD When =0, output state S B ; When M SA ≥0, M SB ≥0, M SC ≥1, M SD When =0, output state S C ; When M SA ≥0, M SB ≥0, M SC =0, M SD When ≥1, output state S D ; When M SA ≥0, M SB ≥0, M SC ≥1, M SD When ≥1, output state E; Among them, M SA M SB M SC M SD They are time quantities Δ T S in a single triaxial vibration sensor A S B S C S D The number of the four output states; E indicates an abnormal state; in, n The acceleration time history curve measured in a single direction by a single triaxial vibration sensor is arranged according to the time interval Δ t The total number of segments; m For the time period Δ t The number of sampling points in a single direction for a single triaxial vibration sensor; Δ t The sampling interval time period set for the triaxial vibration sensor is a constant; a x 、a y 、a z These are the real-time acceleration time-history signals measured by a single triaxial vibration sensor in the X, Y, and Z directions, respectively. a xi 、a yi 、a zi They are time periods Δ t Internal triaxial vibration sensor, X, Y, Z directions i The acceleration value measured at time t; R For the time period Δ t The combined acceleration time history signals in the X, Y, and Z directions from the internal triaxial vibration sensor; R rms for R The root mean square value; for R The average value; R pp for R peak value; R H for R The entropy value; R i for R The Middle i The acceleration value at any given moment; F for n Segment acceleration time history signal R The combined feature vectors; F c The measured acceleration time history signal R c The combined feature vectors; R c The acceleration time history signal measured during the operation status identification of the building construction machine.

2. The real-time identification method for equipment process status of building construction machines according to claim 1, characterized in that, Step S9 further includes: setting a time quantity Δ T , and Δ T ≥(Δ t+ 4 s ), every time period Δ t Acquire readings of the acceleration time history signal per second for the previous time period, including the current moment.

3. The real-time identification method for equipment process status of building construction machines according to claim 1, characterized in that, Step S4 further includes: setting an interference threshold. ε ,and ε >0, If max(|a xi |, |a yi |, |a zi |) > ε, Then delete the acceleration time history signal. a x 、a y 、a z The corresponding a xi 、a yi 、a zi value.

4. The method for real-time identification of equipment process status for building construction machines according to claim 1, characterized in that, Step S7 further includes: setting offset feature values. R cr Set an acceleration time history signal R offset δ ,Right now δ= kR ,whenever( R i - δ )×( R i-1 - δ When ) < 0, R cr Add one more time; R cr For acceleration time history signal R The offset characteristic value; δ For acceleration time history signal R The offset; k For setting offset factor, k ∈(0,1).

5. The method for real-time identification of equipment process status for building construction machines according to claim 1, characterized in that, In step S1, N triaxial vibration sensors are installed at different main stress points of the building machine. Based on steps S3 to S7, the corresponding working state of each triaxial vibration sensor is determined, and the output state is as follows: When N SA =N,N SB =0, N SC =0, N SD When =0, output state S A ; When N SA ≥0, N SB ≥1, N SC =0, N SD When =0, output state S B ; When N SA ≥0, N SB ≥0, N SC ≥1, N SD When =0, output state S C ; When N SA ≥0, N SB ≥0, N SC =0, N SD When ≥1, output state S D ; When N SA ≥0, N SB ≥0, N SC ≥1, N SD When ≥1, output state E; N represents the total number of triaxial vibration sensors installed on the main stress-bearing parts of a single building machine, N≥2; N SA N SB N SC N SD S, representing N triaxial vibration sensors, respectively. A S B S C S D The number of sensors corresponding to the four output states.

6. The method for real-time identification of equipment process status for building construction machines according to claim 1, characterized in that: In step S5, each acceleration time history signal R root mean square value R rms ,average value R Peak-to-peak value R pp The calculation formula is as follows: 。 7. The method for real-time identification of equipment process status for building construction machines according to claim 1, characterized in that, Step S6 includes the following steps: S601: Convert each acceleration time history signal R The mean-removed data is subjected to a Fourier Transform (FFT) to obtain the frequency domain data. Y R And calculate the power spectral density. S R : ； S602: Power spectral density obtained from step S601 S R Calculate the probability density at each frequency point P i Then based on probability density P i Calculate the entropy value R H The calculation formulas are as follows: 。 8. The method for real-time identification of equipment process status for building construction machines according to claim 1, characterized in that: In step S8, the classifier is selected using either the KNN or SVM machine learning algorithm for classification.

Images