Method and device for identifying concrete pouring interface based on PSO-GA-BP neural network

Abstract

The invention discloses a method and device for identifying a concrete pouring interface based on a PSO-GA-BP neural network, and the method comprises the steps: building a multi-sensor data collection cloud platform, and obtaining the liquid level height, conductivity, turbidity, PH value and other data; secondly, carrying out dimensionless processing on the data, and constructing a BP neural network structure model by taking the PH value, the turbidity value and the conductivity as input; optimizing an initial weight and a threshold value of the BP neural network through a PSO-GA hybrid algorithm; carrying out clustering analysis by adopting a PSO-GA-BP model; according to the clustering analysis result, the interface position is judged, and the interface liquid level height is estimated. By means of the method, intelligent monitoring of the concrete pouring interface in the pile hole can be achieved, excessive pouring is avoided, and cost is saved. The method does not depend on the experience of an operator, can accurately judge the concrete slurry interface, and is time-saving, labor-saving, low in labor intensity and convenient and simple to operate.

Description

Technical Field

[0001] The invention relates to the technical field of bridge construction monitoring, and in particular to a method and a device for identifying a concrete pouring interface based on a PSO-GA-BP neural network. Background Art

[0002] Bored piles have been widely used in the foundations of highway, railway bridges and other structural engineering projects due to their mature construction technology, high bearing capacity and wide application range. However, since bored piles are a hidden project, the quality of the piles is related to the safety performance and quality control of the entire engineering structure. Among them, real-time and accurate monitoring and positioning of the concrete pouring interface in the pile hole is a key link in the quality control of the pile. Inaccurate interface positioning will easily lead to engineering defects such as pull-out and leakage, buried pipes, stuck pipes, hollow pile tops, and excessive over-pouring during the pipe pouring process, resulting in material waste and increased costs.

[0003] At present, a heavy hammer with a rope is often used to hang into the hole, so that it passes through the mud sedimentation layer and stays on the concrete surface (or 10 to 20 cm below the surface). The depth of the hammer shown by the measuring rope is used as the depth of the concrete surface. This method is simple and easy to use and is widely used. However, this method often requires operators to have rich engineering experience. At the same time, due to the different experience of operators, it is impossible to obtain a unified result. Not only can the concrete surface not be accurately judged, but it is also time-consuming, labor-intensive, and labor-intensive. In addition, the existence of a floating slurry layer between concrete and mud makes the measurement result error very large. In the actual construction process, the phenomenon of over-filling of concrete is common, and it has become a common problem in the process control of bored piles. The over-filled concrete removed after the excavation of the foundation pit must be transported to the suburbs for landfill, which will occupy land and damage the environment. It not only wastes concrete, but also increases the labor cost of breaking piles, adding unnecessary cost expenditures.

[0004] Therefore, realizing intelligent monitoring of the concrete pouring interface in the pile hole, strictly controlling over-pouring, avoiding excessive over-pouring, and saving costs are engineering and technical issues that have yet to be solved in pile foundation projects. Summary of the invention

[0005] In view of this, the present invention provides a method and device for identifying the concrete pouring interface based on the PSO-GA-BP neural network. The PSO-GA-BP neural network model is used to identify the pouring interface and calculate the position height, so as to intelligently monitor the concrete pouring interface. The method and device can overcome the limitations of traditional methods, realize intelligent automatic real-time and accurate monitoring, improve the work efficiency and safety of concrete pouring interface monitoring, and reduce costs.

[0006] To achieve the above object, the technical solution adopted by the present invention is:

[0007] In a first aspect, an embodiment of the present invention provides a method for identifying a concrete pouring interface based on a PSO-GA-BP neural network, the method comprising the following steps:

[0008] S1. Obtain relevant data of the mud layer and concrete layer in the bored pile, and perform dimensionless processing on the data;

[0009] S2, constructing a BP neural network model, and optimizing the initial weights and thresholds of the BP neural network through a PSO-GA hybrid algorithm;

[0010] S3. Input the processed data into the optimized BP neural network model for cluster analysis. According to the cluster analysis results, determine the position of the perfusion interface and calculate the height of the perfusion interface.

[0011] In an optional embodiment, in step S1, relevant data of the mud layer and the concrete layer in the bored pile are obtained through a multi-source data monitoring cloud platform, and the multi-source data monitoring cloud platform includes: a multi-sensor integration module, a data acquisition module, a communication module, a cloud server module and a terminal display module; wherein:

[0012] The multi-sensor integrated module includes: pH sensor, turbidity sensor, conductivity sensor, liquid level height sensor, which are used to obtain physical parameter data of pH value, turbidity value, conductivity, and liquid level height, and convert them into electrical signal data; the data acquisition module collects and stores the electrical signal data in real time, and transmits the electrical signal data to the cloud server module through the communication module; the cloud server module is used to process and calculate the electrical signal data, and display the processing and calculation results through the terminal display module.

[0013] In an optional implementation, in step S1, the dimensionless calculation formula is:

[0014]

[0015] in, is the original data x i (k), i∈(1,2,...,N), k∈(1,2,...,n); n is the number of reference sequences, and N is the number of comparison sequences.

[0016] In an optional embodiment, in step S2, the constructed BP neural network model is used for cluster analysis to identify whether the sensor data belongs to the mud layer or the concrete layer. The number of neurons in the input layer is 3, the number of neurons in the hidden layer is 8, and the number of neurons in the output layer is 2, that is, a 3-8-2 BP neural network model structure.

[0017] In an optional implementation, in step S2, the initial weights and threshold parameters of the BP neural network are optimized using a PSO-GA hybrid algorithm to obtain a global optimal solution as the initial values ​​of the BP neural network structure parameters, and optimization training is performed. The process includes:

[0018] ① Initialize the particle swarm population and the parameters required by the genetic algorithm;

[0019] ② Calculate the fitness value of the particle, find pbest and gbest, and use the fitness function to calculate the fitness function value of the particle;

[0020] ③ Update the position and speed of the particles, introduce genetic algorithms, select and cross-operate particle populations, update individual extreme values ​​and global extreme values, that is, obtain the fitness value of the new particle population and the fitness value of the extreme value of the population;

[0021] ④ End condition: When the maximum number of iterations or a better fitness value is reached, the PSO-GA algorithm ends, otherwise, return to ②;

[0022] ⑤ The connection weights and thresholds optimized by PSO-GA are used as parameters of the BP neural network, and then simulation experiments are carried out to obtain the prediction function output of the BP neural network.

[0023] In an optional embodiment, in step S3, the boundary between the mud layer and the concrete layer data is determined based on the cluster analysis results, the pouring interface position is determined to be between the liquid level heights corresponding to the two types of data, and the pouring interface height is calculated using the following formula:

[0024]

[0025] Where h represents the height of the injection interface, h1 represents the liquid level height corresponding to the mud layer data, and h2 represents the liquid level height corresponding to the mud layer data.

[0026] In a second aspect, an embodiment of the present invention further provides a device for identifying a concrete pouring interface based on a PSO-GA-BP neural network, comprising: a data acquisition device, a memory, and a processor, wherein:

[0027] The data acquisition device is used to acquire relevant data of the mud layer and the concrete layer in the cast-in-place pile;

[0028] The memory stores computer instructions;

[0029] The processor is data-connected to the data acquisition device and the memory. The processor executes the computer instructions to perform a method for identifying a concrete pouring interface based on a PSO-GA-BP neural network to identify the concrete pouring interface.

[0030] In a third aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method for identifying a concrete pouring interface based on a PSO-GA-BP neural network is implemented.

[0031] Compared with the prior art, the present invention has at least the following beneficial effects:

[0032] 1. The present invention provides a method for identifying the concrete pouring interface based on the PSO-GA-BP neural network. The PSO-GA-BP neural network model is used to identify the pouring interface and calculate the position height, so as to intelligently monitor the concrete pouring interface. The method can overcome the limitations of traditional methods, realize intelligent automatic real-time and accurate monitoring, improve the work efficiency and safety of concrete pouring interface monitoring, and reduce costs.

[0033] 2. The present invention does not rely on the operator's experience, can accurately determine the concrete slurry interface, saves time and effort, has low labor intensity, and is easy and convenient to operate.

[0034] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0035] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0038] Figure 1 A schematic flow chart of a method for identifying a concrete pouring interface based on a PSO-GA-BP neural network provided in an embodiment of the present invention.

[0039] Figure 2 Schematic diagram of a multi-source data monitoring cloud platform provided in an embodiment of the present invention.

[0040] Figure 3A schematic diagram of the process of optimizing the BP neural network using the PSO-GA algorithm provided in an embodiment of the present invention.

[0041] Figure 4 A schematic diagram of the BP neural network model structure provided in an embodiment of the present invention.

[0042] Figure 5 A schematic diagram of the change in fitness function value of the PSO-GA-BP neural network model provided in an embodiment of the present invention.

[0043] Figure 6 A schematic diagram of the error value change of the PSO-GA-BP neural network model provided in an embodiment of the present invention.

[0044] Figure 7 A schematic diagram of the structure of a device for identifying a concrete pouring interface based on a PSO-GA-BP neural network provided in an embodiment of the present invention.

[0045] Among them, 1-data acquisition device; 2-memory; 3-processor. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0047] In the description of the present invention, it should be noted that: in some processes described in the specification and drawings of this application, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be performed in the order in which they appear in this document or may be performed in parallel. In addition, various serial numbers are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0048] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] See also Figure 1 As shown, the present invention provides a method for identifying a concrete pouring interface based on a PSO-GA-BP neural network, which is used to identify a bored concrete pile interface. The method mainly includes the following steps:

[0050] S1. Obtain relevant data of the mud layer and concrete layer in the bored pile, and perform dimensionless processing on the data;

[0051] S2, constructing a BP neural network model, and optimizing the initial weights and thresholds of the BP neural network through a PSO-GA hybrid algorithm;

[0052] S3. Input the processed data into the optimized BP neural network model for cluster analysis. According to the cluster analysis results, determine the position of the perfusion interface and calculate the height of the perfusion interface.

[0053] The specific implementation methods of the method of the present invention are described in detail below:

[0054] See also Figure 2 As shown, in this embodiment, first build a multi-sensor acquisition cloud platform (i.e., a multi-source data monitoring cloud platform), which is composed of a multi-sensor integration module, a data acquisition module, a 4G communication module, a cloud server module and a terminal display module. The multi-sensor integration module is composed of four sensors integrated into a probe as small as possible so that different parameters can be measured simultaneously. During signal acquisition, the multi-sensor integrated probe is uniformly lowered into the bored pile, and the motor operation is controlled according to the set torque threshold, in order to ensure that the probe is always at the interface. The data such as pH value, turbidity value, conductivity, liquid level height are collected and transmitted to realize functions such as equipment control. The physical parameter signals such as pH value, turbidity value, conductivity, liquid level height are converted into electrical signals. The data acquisition module is integrated with a communication chip and a storage chip, and the electrical signal data is collected and stored in real time, and the data is transmitted to the cloud server module through the 4G communication module. The cloud server module uses the communication network layer to transmit and the cloud computing platform to process the information obtained by the acquisition layer. The communication network layer includes the Internet, wifi, broadband, telecommunications network and network management system, which realizes the functions of data acquisition, data transmission, equipment control and the like of the terminal equipment.

[0055] Furthermore, in order to unify the data dimensions and avoid large differences in magnitude, the data is dimensionless. Specifically:

[0056] Let X be a sequence set, X = [x i (k)] where i∈(1,2,...,N), k∈(1,2,...,n), n is the number of reference sequences, and N is the number of comparison sequences;

[0057]

[0058] in, For x i (k) Data processing results.

[0059] Furthermore, a BP neural network is constructed. The basic components of the BP neural network are neurons on each network layer. The number of neurons in each layer is determined according to the number of network input and output parameters and empirical formulas.

[0060] In this embodiment, the BP neural network model is used for cluster analysis, and there are two categories, namely the mud layer and the concrete. For the pouring interface recognition, it is necessary to identify whether the sensor is located in the mud layer or in the concrete layer, so the number of neurons in the output layer can be determined to be 2.

[0061] y=(y1,y2)

[0062] Wherein, y1 represents the classification result of the mud layer. When y1=1, it means that the sensor position is determined in the mud layer; when y1=0, it means that the sensor position is not in the mud layer. y2 represents the classification result of the concrete layer. When y2=1, it means that the sensor position is determined in the concrete layer; when y2=0, it means that the sensor position is not in the concrete layer.

[0063] When y = (1,0), it means it is in the mud layer; when y = (0,1), it means it is in the concrete layer. In the present invention, when the sensor module dives for measurement, the collected data are:

[0064] 1) pH value;

[0065] 2) Turbidity value;

[0066] 3) Conductivity.

[0067] Therefore, it can be determined that the number of neurons in the input layer is 3, which are pH value, turbidity value and conductivity data. That is:

[0068] x=(x1,x2,x3)

[0069] Therefore, it is determined that the BP network model structure is 3-8-2 type.

[0070] Further, see Figure 3 As shown, the present invention uses the PSO-GA algorithm to optimize the weights and thresholds of the BP neural network. The main steps of the PSO-GA algorithm are as follows:

[0071] ① Initialize the particle swarm population, the parameters required by the genetic algorithm and establish the BP neural network.

[0072] ② Calculate the fitness value of the particle, find pbest and gbest, and use the fitness function to calculate the fitness function value of the particle.

[0073] ③ Update the position and velocity of the particles, introduce genetic algorithms, select and cross-operate particle populations, and update individual extreme values ​​and global extreme values, that is, obtain the fitness value of the new particle population and the fitness value of the extreme value of the population.

[0074] ④ End condition: When the set maximum number of iterations or a better fitness value is reached, the PSO-GA algorithm ends, otherwise, return to step 2.

[0075] ⑤ The connection weights and thresholds optimized by PSO-GA are used as the parameters of the BP neural network.

[0076] Furthermore, the training set samples are used to train the set PSO-GA-BP neural network model. After the maximum iteration of the PSO-GA optimization algorithm, the global optimal fitness value and error are minimized, and the PSO-GA-BP neural network model is used for cluster analysis.

[0077] Furthermore, based on the cluster analysis results, the boundary between the two types of sequence data was determined, the position of the perfusion interface was determined, and the interface liquid level height was estimated. i is the data measured in the mud, identifying z i+1 is the measured data in concrete. It can be determined that the interface is located between the liquid level heights h1 and h2 corresponding to the two columns of data z1 and z2. The liquid level heights h1 and h2 are obtained by the liquid level sensor, and the interface height is estimated as:

[0078]

[0079] Where h is the interface height.

[0080] In a specific embodiment, the present invention takes a bored pile pouring construction site under construction in Hohhot as an example, uses the built multi-source data acquisition platform to collect pH, turbidity, conductivity and liquid level data, and collects 1200 sets of data from a bored pile pouring construction site under construction, of which 1000 sets of data are used for model training and the other 200 sets are used as test samples. The BP neural network training parameters are set as shown in Table 1.

[0081] Table 1 Model training parameters

[0082]

[0083] Furthermore, the training set samples are used to train the PSO-GA-BP neural network model. Figure 5 It can be seen that after 40 iterations of the PSO-GA optimization algorithm, the global optimal fitness value dropped from about 2.83 to 2.18, and the error of the neural network model became smaller and smaller, proving that the optimization effect of the GA-PSO optimization algorithm on the BP neural network model is very significant.

[0084] After 40 iterations of optimization, the final global optimal Gbest is set as the initial connection weight and threshold of the BP neural network model, and then the neural network model is trained. Figure 6It can be seen that after 13 training iterations, the error of the training set of the neural network model reached 0.00751, which is less than the target training error of 0.01.

[0085] After the PSO-GA-BP neural network model is trained, the test set is predicted and verified, and compared with the actual measured results. In this embodiment, it is determined that the interface is between the liquid level heights of 11.03m and 11.79m corresponding to the two columns of data measured in the mud and the data measured in the concrete. This liquid level height is a set of data in the test sample. The estimated interface height is: 11.41m; by comparison, it is found that the classification result is completely consistent with the actual situation. In 200 monitorings, there were only 12 prediction errors, and the total prediction accuracy was 94%.

[0086] From the description of the above embodiments, those skilled in the art can know that the present invention proposes a method for identifying a concrete pouring interface based on a PSO-GA-BP neural network, by monitoring data such as liquid level height, conductivity and pH value, using a concrete pouring interface recognition technology of the PSO-GA-BP neural network method, by monitoring data such as turbidity, conductivity and pH value, using a PSO-GA-BP neural network model to identify the pouring interface, calculate the position height, and achieve the purpose of intelligently monitoring the concrete pouring interface.

[0087] Furthermore, the present invention also provides a system for identifying a concrete pouring interface based on a PSO-GA-BP neural network, which uses a method for identifying a concrete pouring interface based on a PSO-GA-BP neural network in the above embodiment to identify a concrete pouring interface, and the system includes:

[0088] The data acquisition module is used to obtain the relevant data of the mud layer and concrete layer in the cast-in-place pile and perform dimensionless processing on the data;

[0089] Model building module, used to build BP neural network model and optimize the initial weights and thresholds of BP neural network through PSO-GA hybrid algorithm;

[0090] The interface recognition module is used to input the processed data into the optimized BP neural network model for cluster analysis, determine the position of the perfusion interface and calculate the height of the perfusion interface according to the cluster analysis results.

[0091] The implementation principle and technical effects of the system provided in the embodiment of the present invention are the same as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference can be made to the corresponding contents in the aforementioned method embodiment, which will not be repeated here.

[0092] In addition, the embodiment of the present invention also provides a device for identifying a concrete pouring interface based on a PSO-GA-BP neural network, which is used to identify a concrete pouring interface. The device can be implemented by software and / or hardware. Figure 7 As shown, the apparatus comprises a data acquisition device 1, a memory 2 and a processor 3. The data acquisition device 1, the memory 2 and the processor 3 may be connected via a bus or other means.

[0093] The data acquisition device 1 is used to obtain the relevant data of the mud layer and the concrete layer in the bored pile, and send it to the processor 3. The processor 3 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components and other chips, or a combination of the above chips.

[0094] The memory 2, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer executable programs and modules, such as programs or instructions corresponding to the method for identifying the concrete pouring interface based on the PSO-GA-BP neural network.

[0095] The processor 3 executes various functional applications and data processing of the processor by running the non-transient software program or instruction stored in the memory 2, that is, implements the method for identifying the concrete pouring interface based on the PSO-GA-BP neural network in the above method embodiment.

[0096] The memory 2 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created by the processor 2 and the like.

[0097] In addition, the memory 2 may include a high-speed random access memory and may also include a non-volatile memory. More specific examples of the memory 2 (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, and any suitable combination of the above.

[0098] In some aspects, the memory 2 optionally includes memories remotely located relative to the processor 3 , and these remote memories may be connected to the processor 3 via a network.

[0099] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, devices or computer program products, etc. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects.

[0100] It should be noted that the word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several distinct components, and by means of a suitably programmed computer.

[0101] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0102] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying concrete pouring interface based on PSO-GA-BP neural network, characterized in that: The method comprises the following steps: S1. Obtain relevant data of the mud layer and concrete layer in the bored pile, and perform dimensionless processing on the data; S2, constructing a BP neural network model, and optimizing the initial weights and thresholds of the BP neural network through a PSO-GA hybrid algorithm; S3. Input the processed data into the optimized BP neural network model for cluster analysis. According to the cluster analysis results, determine the position of the perfusion interface and calculate the height of the perfusion interface.

2. The method for identifying concrete pouring interface based on PSO-GA-BP neural network according to claim 1 is characterized in that: In step S1, relevant data of the mud layer and the concrete layer in the bored pile are obtained through a multi-source data monitoring cloud platform, and the multi-source data monitoring cloud platform includes: a multi-sensor integration module, a data acquisition module, a communication module, a cloud server module and a terminal display module; wherein: The multi-sensor integrated module includes: pH sensor, turbidity sensor, conductivity sensor, liquid level height sensor, which are used to obtain physical parameter data of pH value, turbidity value, conductivity, and liquid level height, and convert them into electrical signal data; the data acquisition module collects and stores the electrical signal data in real time, and transmits the electrical signal data to the cloud server module through the communication module; the cloud server module is used to process and calculate the electrical signal data, and display the processing and calculation results through the terminal display module.

3. The method for identifying concrete pouring interface based on PSO-GA-BP neural network according to claim 1 is characterized in that: In step S1, the calculation formula for dimensionless processing is: in, is the original data x i (k), i∈(1,2,...,N), k∈(1,2,...,n); n is the number of reference sequences, and N is the number of comparison sequences.

4. The method for identifying concrete pouring interface based on PSO-GA-BP neural network according to claim 1 is characterized in that: In step S2, the constructed BP neural network model is used for cluster analysis to identify whether the sensor data belongs to the mud layer or the concrete layer. The number of neurons in the input layer is 3, the number of neurons in the hidden layer is 8, and the number of neurons in the output layer is 2, which is a 3-8-2 type BP neural network model structure.

5. The method for identifying concrete pouring interface based on PSO-GA-BP neural network according to claim 1 is characterized in that: In step S2, the PSO-GA hybrid algorithm is used to optimize the initial weights and threshold parameters of the BP neural network, and the global optimal solution is obtained as the initial value of the BP neural network structure parameters, and optimization training is performed. The process includes: ① Initialize the particle swarm population and the parameters required by the genetic algorithm; ② Calculate the fitness value of the particle, find pbest and gbest, and use the fitness function to calculate the fitness function value of the particle; ③ Update the position and speed of the particles, introduce genetic algorithms, select and cross-operate particle populations, update individual extreme values ​​and global extreme values, that is, obtain the fitness value of the new particle population and the fitness value of the extreme value of the population; ④When the set maximum number of iterations or a better fitness value is reached, the PSO-GA algorithm ends, otherwise, it returns to ②; ⑤ The connection weights and thresholds optimized by PSO-GA are used as parameters of the BP neural network, and then simulation experiments are carried out to obtain the prediction function output of the BP neural network.

6. The method for identifying concrete pouring interface based on PSO-GA-BP neural network according to claim 4 is characterized in that: In step S3, according to the cluster analysis results, the boundary between the mud layer and the concrete layer data is determined, the pouring interface position is determined to be between the liquid level heights corresponding to the two types of data, and the pouring interface height is calculated. The calculation formula is: Where h represents the height of the injection interface, h1 represents the liquid level height corresponding to the mud layer data, and h2 represents the liquid level height corresponding to the mud layer data.

7. A device for identifying concrete pouring interface based on PSO-GA-BP neural network, characterized in that: include: A data acquisition device, a memory and a processor, wherein: The data acquisition device is used to acquire relevant data of the mud layer and the concrete layer in the cast-in-place pile; The memory stores computer instructions; The processor is data-connected to the data acquisition device and the memory. The processor executes the computer instructions to execute the method for identifying the concrete pouring interface based on the PSO-GA-BP neural network as described in any one of claims 1 to 6, thereby identifying the concrete pouring interface.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method for identifying a concrete pouring interface based on a PSO-GA-BP neural network as described in any one of claims 1 to 6 is implemented.

Images