Engineering operation intensity grade identification method and system, and storage medium

Abstract

The invention discloses an engineering operation intensity grade identification method and system and a storage medium, and belongs to the technical field of engineering machinery, and the method comprises the steps: obtaining real-time load data under current engineering operation; inputting the real-time load data into a pre-trained intensity grading evaluation model to obtain an engineering operation intensity grade of a corresponding time sequence; the acquisition method of the pre-trained strength grading evaluation model comprises the following steps: acquiring time domain load data of the engineering machinery under various engineering operation profiles; all kinds of load data are aligned in a time domain, and the aligned load data are sliced; training a gradient boosting decision tree model by using the sliced load data to obtain an intensity grading evaluation model, determining the intensity grading evaluation model by introducing the gradient boosting decision tree model, and autonomously determining the intensity grade number and boundary by the evaluation model, the problem that the operation intensity level of an existing engineering mechanical product is lacked is effectively solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of engineering machinery, and in particular relates to a method, system and storage medium for identifying the intensity level of an engineering operation. Background Art

[0002] How to accurately identify and define the operating profiles that engineering machinery products experience throughout their life cycle is of great significance to the failure mechanism analysis, failure improvement, reliability design and life prediction of the products. In practice, due to the complex and changeable operating conditions of engineering machinery products and the large randomness of operating loads, it is difficult to define the operating conditions with accurate load values. Qualitative definition methods are usually used, such as light load, heavy load, etc., which basically cannot provide valuable information for the technical research of the product. At the same time, the recording of operating conditions is a continuous process. Different parts may be in different operating condition levels under the same operating profile, and cannot be simply represented by the operating condition level of the whole machine. Therefore, the present invention provides a method for grading the operating conditions of engineering machinery products and a system having this method, which can achieve accurate division of operating condition levels and quantify each operating condition level.

[0003] At present, there are two main directions for determining the work intensity level: 1. Determination technology based on image recognition. This method requires the installation of image or video acquisition equipment on the construction site or machine, and the use of image processing algorithms such as target detection to measure the product's work process in real time. This method is not very real-time among users, and each vehicle needs to be installed with an image or video recording device, which is costly. At the same time, during the acquisition process, the amount of real-time data transmitted back is large, which places high demands on the storage system, and the image recognition focuses on the distinction of work materials and the recording of work trajectories. It is necessary to install more sensors on the product to assist in the use of image recognition results, and a lot of manpower and material resources are required in post-processing. 2. Determination technology based on work data analysis. This method is mainly based on the work load of typical parts of the product, such as the handle, pump, engine, etc., and uses different algorithms (such as statistics, machine learning) to form a regular model to obtain the classification result. This method does not make good use of all the work data, and is basically based on the principle of extracting data features, resulting in loss of working condition information and deviation in recognition results. Moreover, both of the above two technical methods cannot output the order of work intensity in the time dimension, resulting in low availability of work intensity levels. Summary of the invention

[0004] The present invention provides a method, system and storage medium for identifying the intensity level of engineering operations. By introducing a gradient boosting decision tree model in the rating of the working intensity of engineering machinery, an intensity grading evaluation model is determined. The evaluation model independently determines the number and boundaries of intensity levels, which effectively solves the current problems of missing operation intensity levels of engineering machinery products and the inability to characterize the operation intensity in a time series.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for identifying an intensity level of an engineering operation, comprising: Obtain real-time load data under current engineering operations; Inputting the real-time load data into a pre-trained strength grading evaluation model to obtain the engineering operation intensity level of the corresponding time sequence; The method for obtaining the pre-trained intensity grading evaluation model includes: Obtain time domain load data of various engineering operation profiles of construction machinery; Align various types of load data in the time domain, and slice the aligned load data; The sliced ​​load data is used to train the gradient boosting decision tree model to obtain the strength grading evaluation model.

[0006] Optionally, the obtaining of time domain load data of various types of engineering operation profiles of engineering machinery includes: obtaining discrete time point operation data of the engineering machinery operating under various typical working conditions, wherein the type and measurement points of the discrete time point operation data are selected according to the product characteristics of the engineering machinery.

[0007] Optionally, aligning various types of payload data in the time domain and slicing the aligned payload data includes: Align the time stamps of the time domain load data under each working condition, and use linear interpolation to supplement the time domain load data with low acquisition frequency; The moving window length and step length are determined, and the aligned load data of each working condition are sliced ​​into data frames of the same length according to the moving window length and step length.

[0008] Optionally, the moving window length and step size are determined according to the data collection frequency and the computing power of the vehicle-mounted terminal.

[0009] Optionally, the use of the sliced ​​load data to train a gradient boosting decision tree model to obtain a strength grading evaluation model includes: Classifying and labeling the working condition data frames according to the working intensity of the working condition; The labeled load data is input into the gradient boosting decision tree model and the label prediction value is output; The loss function value between the label prediction value and the label true value is calculated, and the model parameters are adjusted according to the loss function value to obtain the strength grading evaluation model.

[0010] Optionally, the classifying and labeling the working condition data frames according to the working intensity of the working condition includes: setting a label for the working intensity of each typical working condition according to experience, and obtaining a label for each working condition data frame ( x i , y i ),in, x i Indicates i A working condition data frame, y i Indicates i The typical working conditions include no-load, half-load, full-load and overload, which correspond to label values ​​1, 2, 3 and 4 respectively.

[0011] Optionally, the inputting of the real-time load data into the strength grading evaluation model to obtain the engineering operation strength level of the corresponding time sequence includes: Collect real-time load data of engineering operations through various sensors; Slicing the real-time load data according to a preset moving window length and step size and inputting the slices into the strength grading evaluation model to obtain the working condition grade of each data frame represented by a time series; The working condition level of each data frame is weighted and summed according to the preset weight ratio to obtain the current engineering operation intensity level.

[0012] In a second aspect, the present invention provides a system for identifying the intensity level of an engineering operation, comprising: Load data acquisition module: used to obtain real-time load data under the current engineering operation; Strength level result acquisition module: used to input the real-time load data into the pre-trained strength grading evaluation model to obtain the engineering operation strength level of the corresponding time sequence; The method for obtaining the pre-trained intensity grading evaluation model includes: Obtain time domain load data of various types of engineering operation profiles of construction machinery; Align various types of load data in the time domain, and slice the aligned load data; The sliced ​​load data is used to train the gradient boosting decision tree model to obtain the strength grading evaluation model.

[0013] In a third aspect, the present invention provides a computer storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for identifying the intensity level of an engineering operation as described in any step of the first aspect is implemented.

[0014] In a fourth aspect, the present invention provides an electronic terminal comprising a processor and a memory connected to the processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method for identifying the intensity level of an engineering operation as described in any step of the first aspect is executed.

[0015] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the present invention uses the gradient boosting decision tree model to establish an evaluation model for work intensity levels, making full use of work load data. In the face of high-dimensional and multi-category data, a more accurate determination method for product work intensity levels is explored. This method is completely driven by product work data, and the evaluation model independently determines the number and boundaries of intensity levels, eliminating human influence and having a good processing effect on high-dimensional data. At the same time, this method is incorporated into the enterprise data collection system TDM to directly call the product work load data, which can effectively solve the current problem of missing work intensity levels for engineering machinery products. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The figure is a flow chart of a method for identifying the intensity level of engineering work in one embodiment of the present invention;

[0017] Figure 2 Shown is a flow chart of a method for identifying the intensity level of engineering work in another embodiment of the present invention;

[0018] Figure 3 The figure shows a schematic diagram of serial connection of working condition data in one embodiment of the present invention;

[0019] Figure 4 The figure shows a schematic diagram of rating the working condition data frames frame by frame in one embodiment of the present invention;

[0020] Figure 5 The figure shows a schematic diagram of gradient boosting decision tree model training in one embodiment of the present invention;

[0021] Figure 6 Shown is a schematic diagram of the operating condition classification evaluation results in one embodiment of the present invention;

[0022] Figure 7 The figure shows an architecture diagram of an engineering operation intensity level identification system based on operation load data in one embodiment of the present invention;

[0023] Figure 8 Shown is a diagram of the internal structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0025] Example 1

[0026] like Figure 1 As shown, this embodiment provides a method for identifying the intensity level of an engineering operation, including:

[0027] Obtain real-time load data under current engineering operations;

[0028] Inputting the real-time load data into a pre-trained strength grading evaluation model to obtain the engineering operation intensity level of the corresponding time sequence;

[0029] The method for obtaining the pre-trained intensity grading evaluation model includes:

[0030] Obtain time domain load data of various types of engineering operation profiles of construction machinery;

[0031] Align various types of load data in the time domain, and slice the aligned load data;

[0032] The sliced ​​load data is used to train the gradient boosting decision tree model to obtain the strength grading evaluation model.

[0033] The present invention introduces a gradient boosting decision tree model, and trains the gradient boosting decision tree model through historical load data, thereby obtaining a strength grading evaluation model whose input is operation load data and outputs the intensity level of engineering operation. The evaluation model independently determines the number and boundaries of intensity levels for the current engineering operation status based on multiple historical experiences, so that the process of obtaining strength grading results is more efficient and the results are more accurate.

[0034] Example 2

[0035] Based on Example 1, this example also makes the following design.

[0036] like Figure 2 As shown in the figure, the identification method of engineering operation intensity level is divided into the following steps.

[0037] Step 1: Acquire real-time operating data under a typical life profile of engineering machinery products. The real-time operating condition data of engineering machinery products acquired include but are not limited to engine fuel consumption, speed, battery voltage, current, hydraulic circuit pressure, ambient temperature, humidity and working device status. Users can determine the measurement points according to product characteristics.

[0038] Step 2: Align the timestamps of all the operation data represented by time series; determine the moving window length and step size, and slice the working condition data into data frames of the same length. If the moving window length is too long, the dimension of the data frame will increase; if the step size is too small, the amount of data will increase, increasing the calculation cost; if the moving window length is too small or the step size (compared to the moving window length) is too large, the working condition information will be lost. In this embodiment, the moving window length and step size are determined by the data collection frequency and the computing power of the vehicle terminal.

[0039] Step 3: Train the gradient boosting decision tree model through the working condition data to obtain the strength classification evaluation model. The training steps are as follows:

[0040] (1) Collect the load data of construction machinery under all possible typical working conditions, such as no-load, half-load, full-load, and overload, that is, the data set composed of the data frames obtained in step 2. Experts assign label values ​​to the working intensity of each typical working condition. The label values ​​of the working conditions such as no-load, half-load, full-load, and overload are 1, 2, 3, 4, etc. from low to high. The label values ​​are only used for working condition classification, and the difference between the scores does not represent the difference in working condition intensity.

[0041] (2) Take the working condition data as the model input and the label value as the model output. i The working condition data frame is x i , and its corresponding working condition score is y i , the data set is divided into a training set and a test set by stratified sampling, so that the proportion of each typical working condition in the training set and the test set is the same;

[0042] (3) Selecting model hyperparameters on the training set through cross-validation, including the number of trees, learning rate, and loss function;

[0043] (4) Retrain the model on the training set using the hyperparameters determined in step (3), and record it as the final strength grading evaluation model f ( x ), after the working condition classification evaluation model is determined, the newly collected working condition data can be used f ( x ) Calculate its work intensity level.

[0044] Step 4: Input the data frame into the intensity grading evaluation model to calculate the work intensity level of the time window corresponding to the data frame;

[0045] Step 5: Process the intensity levels of all data frames in a weighted manner to obtain the construction machinery working condition level represented by a time series.

[0046] An example of applying the above steps is as follows:

[0047] S1. Obtain the operating load data of a certain type of engineering machinery product under various typical working conditions. This example includes temperature, wind speed, strain, cylinder pressure 1, cylinder pressure 2, and cylinder pressure 3. The attributes of various load data are shown in Table 1.

[0048] Table 1 Data types and attributes Data Item unit Data Types Collection frequency Value range temperature ℃ Numeric 1Hz -20~+45 Wind speed m / s Numeric 1Hz 0~20 strain μ Numeric 5Hz 0~180 Cylinder pressure 1 MPa Numeric 10Hz 0~40 Cylinder pressure 2 MPa Numeric 20Hz 0~40 Cylinder pressure 3 MPa Numeric 10Hz 0~40

[0049] S2. Align the working condition data according to time, determine the moving window length and step size, and slice the working condition data into data frames of the same length. The number of data frames is 10. The minimum data acquisition frequency is 1 Hz and the maximum is 20 Hz. In order to retain as much working condition data as possible, 20 Hz is selected as the alignment frequency. In this embodiment, the moving window length is set to 1 s and the step size is set to 1 s.

[0050] Generally speaking, the load data of working conditions are all low-frequency data, and the change trend between two measuring points is unchanged. When aligning data, for the lower-frequency data, the linear interpolation method is used to supplement the missing values.

[0051] In this example, the processed six working condition data are concatenated to form a new set of input data, such as Figure 3 As shown, this process does not require the elimination of dimension effects.

[0052] S3. Determine whether the operation intensity grading evaluation model has been determined. In this embodiment, the operation intensity grading evaluation model has not been determined yet, and the model needs to be trained first.

[0053] S3.1. Based on expert experience, the actual strength of the product is preset into five levels, and the corresponding label values ​​are 1, 2, 3, 4, and 5 respectively.

[0054] S3.2. Figure 4 The concatenated data is used as the model input, and the working condition level is used as the model output. i The working condition data frame is x i , and its corresponding working condition level is y i The data set is divided into a training set and a test set by stratified sampling, so that the proportion of each typical working condition in the training set and the test set is the same. The working level is evaluated frame by frame, such as Figure 4 shown.

[0055] S3.3. On the training set, the model hyperparameters were selected through five-fold cross validation and grid search. The hyperparameters of the gradient boosting decision tree include the number of trees, learning rate, and loss function. The final selected hyperparameters are 500 trees and a learning rate of 0.01.

[0056] S3.4, retrain the model on the training set according to the hyperparameters determined in step S3.3, the data of each working condition and their weight ratios, such as Figure 5 As shown in the figure, the final working condition classification evaluation model is recorded as f ( x ), ,in, n Indicates the number of working condition data frames, w i express x i The corresponding weight. f ( x ) has a classification accuracy of 0.99 on the test set.

[0057] S4, based on the strength grade evaluation model determined in S3, the real-time operation load data during the product operation process is input into this model, and then the strength grade evaluation model can be used. f ( X ) Output intensity level.

[0058] S5. Process all data frames to obtain the actual strength level and its corresponding ratio. Whether stored locally, uploaded to the cloud, or directly displayed on the terminal machine, the strength level data in this form has advantages over the original working condition data. Figure 6 As shown, after traversing the data to complete the evaluation of the work intensity level, the system calculates the proportion of each work level intensity in the current time period based on the weight. The weight of each data in the example is equal by default.

[0059] S6. When the amount of accumulated data increases, based on f ( x ) intensity grading model can also output the boundary range corresponding to each working level. Table 2 gives a reference quantitative definition range.

[0060] Table 2 Data range corresponding to each intensity level

[0061] Example 3

[0062] like Figure 7 As shown, this embodiment provides a system for identifying the intensity level of engineering operations, including:

[0063] Load data acquisition module: used to obtain real-time load data under the current engineering operation;

[0064] Strength level result acquisition module: used to input the real-time load data into the pre-trained strength grading evaluation model to obtain the engineering operation strength level of the corresponding time sequence;

[0065] The method for obtaining the pre-trained intensity grading evaluation model includes:

[0066] Obtain time domain load data of various types of engineering operation profiles of construction machinery;

[0067] Align various types of load data in the time domain, and slice the aligned load data;

[0068] The sliced ​​load data is used to train the gradient boosting decision tree model to obtain the strength grading evaluation model.

[0069] Example 4

[0070] This embodiment provides a computer storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for identifying the intensity level of an engineering operation as described in any step of Embodiment 2 is implemented.

[0071] Example 5

[0072] The embodiment of the present invention further provides a computer device, which may be a server, and its internal structure diagram may be as shown in FIG. Figure 8 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data obtained and generated in the method of the robot autonomously entering the packaging container. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the method of the aforementioned embodiment 2 is implemented.

[0073] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0074] The computer device provided in the embodiment of the present invention can execute the engineering operation intensity level identification method based on operation load data provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0075] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0076] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0077] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0078] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0079] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for identifying the intensity level of engineering work, characterized in that: include: Obtain real-time load data under current engineering operations; Inputting the real-time load data into a pre-trained strength grading evaluation model to obtain the engineering operation intensity level of the corresponding time sequence; The method for obtaining the pre-trained intensity grading evaluation model includes: Obtain time domain load data of various types of engineering operation profiles of construction machinery; Align various types of load data in the time domain, and slice the aligned load data; The sliced ​​load data is used to train the gradient boosting decision tree model to obtain the strength grading evaluation model.

2. The method for identifying the intensity level of engineering work according to claim 1, characterized in that: The obtaining of time domain load data of various types of engineering operation profiles of engineering machinery includes: obtaining discrete time point operation data of the engineering machinery under various typical operating conditions, wherein the type and measurement point of the discrete time point operation data are selected according to the product characteristics of the engineering machinery.

3. The method for identifying the intensity level of engineering work according to claim 1 or 2, characterized in that: The step of aligning various types of payload data in the time domain and slicing the aligned payload data includes: Align the time stamps of the time domain load data under each working condition, and use linear interpolation to supplement the time domain load data with low acquisition frequency; The moving window length and step length are determined, and the aligned load data of each working condition are sliced ​​into data frames of the same length according to the moving window length and step length.

4. The method for identifying the intensity level of engineering work according to claim 3, characterized in that: The moving window length and step size are determined according to the data collection frequency and the computing power of the vehicle-mounted terminal.

5. The method for identifying the intensity level of engineering work according to claim 4, characterized in that: The use of the sliced ​​load data to train the gradient boosting decision tree model to obtain a strength grading evaluation model includes: Classifying and labeling the working condition data frames according to the working intensity of the working condition; Input the labeled load data into the gradient boosting decision tree model and output the label prediction value; The loss function value between the label prediction value and the label true value is calculated, and the model parameters are adjusted according to the loss function value to obtain the strength grading evaluation model.

6. The method for identifying the intensity level of engineering work according to claim 5, characterized in that: The classifying and labeling the working condition data frames according to the working intensity of the working condition includes: setting labels for the working intensity of each typical working condition according to experience, and obtaining labels for each working condition data frame ( x i , y i ),in, x i Indicates i A working condition data frame, y i Indicates i The typical working conditions include no-load, half-load, full-load and overload, which correspond to label values ​​1, 2, 3 and 4 respectively.

7. The method for identifying the intensity level of engineering work according to claim 6, characterized in that: The step of inputting the real-time load data into the strength grading evaluation model to obtain the engineering operation strength level of the corresponding time sequence includes: Collect real-time load data of engineering operations through various sensors; Slicing the real-time load data according to a preset moving window length and step size and inputting the slices into the strength grading evaluation model to obtain the working condition grade of each data frame represented by a time series; The working condition level of each data frame is weighted and summed according to the preset weight ratio to obtain the current engineering operation intensity level.

8. A system for identifying the intensity level of engineering operations, characterized in that: include: Load data acquisition module: used to obtain real-time load data under the current engineering operation; Strength level result acquisition module: used to input the real-time load data into the pre-trained strength grading evaluation model to obtain the engineering operation strength level of the corresponding time sequence; The method for obtaining the pre-trained intensity grading evaluation model includes: Obtain time domain load data of various types of engineering operation profiles of construction machinery; Align various types of load data in the time domain, and slice the aligned load data; The sliced ​​load data is used to train the gradient boosting decision tree model to obtain the strength grading evaluation model.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying the intensity level of an engineering operation as described in any one of claims 1 to 7 is implemented. A computer storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the method for identifying the intensity level of an engineering operation based on operation load data as described in any one of claims 1 to 7 is implemented.

10. An electronic terminal, characterized in that: It comprises a processor and a memory connected to the processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the method for identifying the intensity level of engineering work as described in any one of claims 1 to 7 is executed.

Images