Method, system and device for judging digital construction effect of poling machine

Abstract

The invention provides a pole erecting machine digital construction effect judgment method, system and device, and relates to the technical field of digital construction, and the method comprises the steps: obtaining construction data, and constructing a sample pool; samples in the sample pool are extracted and classified; selecting samples by using an active learning model to construct a training set; initializing a judgment model and performing model training in combination with the training set; outputting the optimized judgment model for prediction; and inputting prediction data characteristics to obtain a construction quality type. Compared with the prior art, the method for judging the digital construction effect of the poling machine, provided by the invention, is proposed based on the construction data of multiple physical quantities, and the construction category in the actual construction process is efficiently combined with the active learning model algorithm; through the training, the proposed judgment model can replace manual work to carry out efficient digital evaluation on the construction effect of the electric pole distributed in a single-pole and / or multi-pole state, so that the purpose of improving the construction efficiency is fully achieved.

Description

Technical Field

[0001] The present application belongs to the field of digital construction technology, and more specifically, relates to a method, system and device for judging the effectiveness of digital construction of a pole erector. Background Art

[0002] Overhead cable poles carry important power transmission functions and are crucial to the stability and safety of power supply. Through a comprehensive assessment of their construction quality, various problems that may exist during construction can be effectively prevented and resolved to ensure the stability of the pole structure and the safe operation of the cables.

[0003] Combined with the specific process of digital construction of pole erectors, the current digital construction method of pole erectors has shortcomings in judging the construction effectiveness, and the evaluation time is long and the efficiency is not high. For example, the patent "A pole assembly device and construction method for landslide sections" with publication number WO2021223446A1 only improves the construction method, and cannot take into account the evaluation of post-construction quality. It still requires a lot of manpower and time for manual verification; the patent "A distribution network pole erection engineering mechanized operation system and method" with publication number CN116167577A, although it can evaluate the effectiveness by calculating the construction data of the poles under construction one by one, but as the number of poles increases, it will also consume a lot of time and cost for calculation and evaluation work, which will affect the construction progress; the patent "distribution network grounding fault location method, system, electronic equipment and storage medium" with publication number CN118150943A, in actual engineering, each additional training sample will greatly increase the amount of calculation, and so on.

[0004] At this point, how to ensure the evaluation accuracy and efficiency of the construction effectiveness of the pole erector is the key to the current digital construction effectiveness evaluation of the pole erector. Summary of the invention

[0005] The purpose of the embodiments of the present application is to provide a method, system and device for judging the effectiveness of digital construction of a pole erector, so as to at least solve the technical problem of low efficiency in judging the effectiveness of digital construction of a pole erector in the prior art.

[0006] To achieve the above objectives, in a first aspect, the present application provides a method for determining the effectiveness of digital construction of a pole erector, the method comprising at least the following steps:

[0007] Obtain construction data and build a sample pool;

[0008] Extracting samples from the sample pool and classifying them;

[0009] Use active learning models to select samples and construct training sets;

[0010] Initializing the judgment model and performing model training in combination with the training set;

[0011] Output the optimized judgment model for prediction;

[0012] Input the prediction data features to get the construction quality type.

[0013] In a further embodiment of the present application, samples in the sample pool are extracted and classified, including single-pole construction quality judgment:

[0014] Determine whether the burial depth meets the requirements, and directly classify samples that do not meet the requirements as unqualified;

[0015] Calculate the deflection angle based on the front and rear deflection angle and the left and right deflection angle of the pole;

[0016] Calculate the pole tip offset based on the pole length and buried depth;

[0017] Outputs the single-pole construction quality type.

[0018] In a further embodiment of the present application, extracting samples from the sample pool and classifying them also includes multi-rod construction quality judgment:

[0019] Establish the relative rectangular coordinate system between the pole and the pole erector and determine the offset angle of the pole;

[0020] Performing coordinate system conversion according to the relative rectangular coordinate system to convert the relative angle into an absolute angle of the geodetic coordinate system;

[0021] Calculate the lateral offset, longitudinal offset and pile spacing of each pole relative to the baseline;

[0022] Output multi-rod construction quality type.

[0023] In a further embodiment of the present application, selecting samples using an active learning model to construct a training set includes:

[0024] S100: constructing an initial training set, and constructing and training a learning model based on the training set;

[0025] S101: using the learning model to predict samples in the sample pool according to the most uncertain criterion, and selecting the most valuable samples;

[0026] S102: labeling the most valuable samples, and adding and updating the training set;

[0027] S103: training and optimizing the learning model according to the updated training set;

[0028] S104: Repeat step S101 according to the optimized learning model until the learning model no longer selects samples;

[0029] S105: Output the training set.

[0030] In a further embodiment of the present application, constructing the initial training set includes: extracting some samples from the sample pool, and performing single-pole construction quality judgment and multi-pole construction quality judgment on each of the extracted samples to construct the initial training set.

[0031] In a further embodiment of the present application, initializing the judgment model and performing model training in combination with the training set includes:

[0032] The judgment model is trained based on the training set, and the accuracy of the judgment model is corrected using the proportional verification method until the proportion of qualified samples in the training set exceeds ρ (ρ is a positive number greater than zero).

[0033] In a further embodiment of the present application, the judgment model uses a random forest model.

[0034] In a further embodiment of the present application, obtaining construction data and building a sample pool includes:

[0035] Monitor the location information of the pole erector and poles, and obtain the front and rear, left and right offsets and height of the poles based on computer vision automatic recognition technology;

[0036] Construct a feature vector P representing the construction quality, where:

[0037] P = [P1, P2, P3, P4, P5]

[0038] In the formula, P1 represents the longitude and latitude coordinates of the pole, P2 represents the longitude and latitude coordinates of the pole erector, P3 represents the front and rear offset angle of the pole, P4 represents the left and right offset angle of the pole, and P5 represents the buried depth;

[0039] Repeat the above steps to complete the sample pool construction.

[0040] In a second aspect, the present application also provides a pole erector digital construction effectiveness judgment system, including a processor, a user interface, a network interface, a communication bus and a memory, wherein the processor, the user interface, the network interface and the memory communicate with each other via the communication bus;

[0041] Memory, used to store computer programs;

[0042] The processor is used to execute the above-mentioned method for judging the effectiveness of digital construction of the pole erecting machine by running the program stored in the memory.

[0043] In a third aspect, the present application also provides a device for judging the effectiveness of digital construction of a pole erecting machine, comprising:

[0044] Data acquisition module, used to obtain construction data;

[0045] A calculation and analysis module, used to construct a sample pool according to the construction data, extract samples from the sample pool and classify them, actively learn to construct and optimize a training set, and use the training set to optimize and output a judgment model;

[0046] The identification module is used to determine the construction quality category according to the judgment model.

[0047] The beneficial effects of the method for judging the effectiveness of digital construction of a pole erector provided by the present application are:

[0048] Compared with the prior art, the method for judging the digital construction effectiveness of the pole erector provided in the present application is proposed based on the construction data of multiple physical quantities, and efficiently combines the construction categories in the actual construction process with the active learning model algorithm. The judgment model trained and proposed in this way can replace manual work to efficiently conduct digital evaluation of the construction effectiveness of electric poles distributed in a single-pole and / or multi-pole state, thereby fully achieving the purpose of improving construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0050] Figure 1 A computer hardware structure block diagram provided for an embodiment of the present application;

[0051] Figure 2 A schematic diagram of the process of determining the effectiveness of digital construction of a pole erecting machine provided in an embodiment of the present application;

[0052] Figure 3 A schematic diagram of a process for forming a random forest training set based on an active learning convolutional neural network provided in an embodiment of the present application;

[0053] Figure 4 A three-dimensional schematic diagram of the construction quality of a pole group provided in an embodiment of the present application;

[0054] Figure 5 A structural block diagram of a device for judging the digital construction effectiveness of a pole erecting machine provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] It should be noted that when an element is referred to as being "fixed to" or "disposed on" another element, it can be directly on the other element or indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or indirectly connected to the other element.

[0057] It should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0058] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0059] Example 1

[0060] The method embodiment provided in the first embodiment of the present application can be executed in a mobile phone, computer, tablet or similar computing device. For example, the method embodiment provided in the first embodiment of the present application can be executed in a mobile phone, computer, tablet or similar computing device. Figure 1 , Figure 1 This is a computer hardware structure block diagram provided by the embodiment of the present application. Figure 1 As shown, the computer may include one or more ( Figure 1 Only one is shown) a processor 1001 (the processor 1001 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 1005 for storing data. Optionally, the computer may also include an interface 1003 for communication functions, a network interface 1004, a communication bus 1002, and input and output devices (such as a keyboard, a mouse, etc.).

[0061] The memory 1005 provided in the embodiment of the present application can be used to store computer programs, for example, application programs and modules of application software, such as a computer program corresponding to a method for judging the effectiveness of digital construction of a pole erector in the embodiment of the present application, so as to execute various functional applications and data processing, that is, to implement the above method. The memory 1005 may further include remote memories remotely arranged relative to the processor 1001, and these remote memories may be connected to the computer via a network. On this basis, examples of the network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0062] In this implementation, a method for judging the effectiveness of digital construction of a pole erector is provided. Figure 2 , Figure 3 as well as Figure 4 , the method comprises at least the following steps:

[0063] Step S10: Obtain construction data and build a sample pool;

[0064] The construction data involved in this embodiment may be actual construction data of the pole erector and all poles acquired by a monitoring device equipped with a positioning system (eg, Beidou micro system) and / or a visual inspection system.

[0065] Step S20: extracting some samples from the sample pool and classifying them;

[0066] The sample extraction in this embodiment is proposed based on stratified sampling technology (such as LHD multi-layer neural network model).

[0067] Step S30: Select samples using the active learning model to construct a training set;

[0068] The active learning model in this embodiment is preferably a convolutional neural network (CNN). As a deep neural network model for processing large images, it can greatly reduce the actual construction data samples that need to be trained through the characteristics of weight sharing, and efficiently retain effective information that is beneficial to model training.

[0069] Based on this, the CNN model here can be used to efficiently train the initial training set formed by extracting a small number of samples, and then use its own characteristics to predict, label and select the samples in the sample pool one by one, and then update the training set and use it as the basis for its own learning iterations to further improve the effectiveness of the constructed training set.

[0070] Step S40: Initialize the judgment model and perform model training in combination with the training set;

[0071] Step S50: outputting the optimized judgment model for prediction;

[0072] The judgment model in this embodiment is preferably a random forest model. The random selection characteristics of the judgment model constructed in this way in both sample selection and feature selection can cover the universality of actual construction data as comprehensively as possible, thereby improving the accuracy and efficiency of judging actual construction results.

[0073] Step S60: Input the predicted data features to obtain the construction quality type.

[0074] In summary, this embodiment is proposed based on the construction data of multiple physical quantities, and efficiently combines the construction categories in the actual construction process (i.e., the single-pole construction quality and the multi-pole construction quality) with the active learning model algorithm. The judgment model trained and proposed in this way can replace manual work to efficiently conduct digital evaluation of the construction results of electric poles distributed in a single-pole and / or multi-pole state, thereby fully achieving the purpose of improving construction efficiency.

[0075] In one implementation of this embodiment, please refer to Figure 2 and Figure 4 , extract samples from the sample pool and classify them, including single-pole construction quality judgment:

[0076] S11: Determine whether the buried depth of a single pole (corresponding to a single sample) meets the requirement, and directly classify the sample that does not meet the requirement as unqualified;

[0077] S12: On the basis of the buried depth reaching the requirement, the deflection angle γ is calculated according to the front and rear deflection angles and the left and right deflection angles of the pole. The calculation formula includes:

[0078]

[0079] R=X(α)*Y(β)*Z(γ)

[0080] In the above formula, α is the front and rear deflection angle, β is the left and right deflection angle, γ is the deflection angle in space, R X (α) is the rotation matrix of the pole along the X-axis, Ry(β) is the rotation matrix of the pole along the Y-axis, and Rz(γ) is the rotation matrix of the pole along the Z-axis.

[0081] S13: Calculate the pole tip offset P based on the length of the pole and the buried depth. The calculation formula is:

[0082] P=L*sinγ

[0083] In the above formula, L is the difference between the length of the pole and the buried depth, and γ is the deflection angle of the pole.

[0084] S14: Output single-pole construction quality type.

[0085] It can be understood that the single-pole construction quality types in this embodiment include a qualified single-pole construction quality type and an unqualified single-pole construction quality type.

[0086] In one implementation of this embodiment, please refer to Figure 2 and Figure 4 , extract samples from the sample pool and classify them, and also include multi-pole construction quality judgment:

[0087] S15: Establishing a relative rectangular coordinate system between the electric pole and the pole erector and determining the offset angle of the electric pole;

[0088] S16: performing coordinate system conversion according to the established relative rectangular coordinate system, converting the relative angle into an absolute angle of the geodetic coordinate system;

[0089] S17: Calculate the lateral offset, longitudinal offset and pile spacing of each pole relative to the baseline;

[0090] The reference line involved in this embodiment is Figure 4 The connection line between the pile foundations at the first and last ends of the pole group shown in .

[0091] S18: Output multi-rod construction quality type.

[0092] It can be understood that the multi-rod construction quality types here include a qualified multi-rod construction quality type and an unqualified multi-rod construction quality type.

[0093] In one implementation of this embodiment, please refer to Figure 2 , Figure 3 and Figure 4 , use the active learning model to select samples to build a training set, including:

[0094] S100: construct an initial training set, and build and train a learning model based on the training set;

[0095] The construction of the initial training set in this embodiment further includes: extracting some samples from the sample pool, and performing single-rod construction quality judgment and multi-rod construction quality judgment on each extracted sample to construct the initial training set.

[0096] S101: using the learning model to predict samples in the sample pool according to the most uncertain criterion, and selecting the most valuable samples;

[0097] S102: labeling the most valuable samples selected, and adding and updating the training set;

[0098] The sample annotation involved in this embodiment requires matching the obtained data with the quality situation to obtain a set of training samples. Obviously, by repeating this step, multiple sets of training samples for adding to the training set can be obtained.

[0099] Furthermore, sample annotation is a construction quality judgment formed by combining single-pole construction quality judgment and multi-pole construction quality judgment. The function can be selected:

[0100] R(p)=Type

[0101] In the above formula, R represents the optimal kernel function type judgment for the feature vector P, and Type represents the type of construction effect. In this paper, the following three types of effect types are selected, Type = 1, 2, 3. The purpose is to match different effect types for different samples. For example, 1 means that both the single-pole construction quality category and the multi-pole construction quality category meet the construction requirements (meeting the construction requirements is qualified), 2 means that the single-pole construction quality category meets the requirements and the multi-pole construction quality category does not meet the requirements (not meeting the requirements is unqualified), and 3 means that both the single-pole construction quality category and the multi-pole construction quality category do not meet the construction requirements.

[0102] S103: training and optimizing the learning model according to the updated training set;

[0103] S104: Repeat step S101 according to the optimized learning model until the learning model no longer selects samples;

[0104] S105: Output the training set.

[0105] For example, the construction process of the training set in this embodiment is supplemented as follows:

[0106] That is, the initial state training set L0 is constructed with the initial sample I, and then the initial convolutional neural network model M0 is obtained by training with the initial training set L0; based on the trained convolutional neural network model M0, all focal elements (samples) in the focal element pool U (i.e., the sample pool constructed in step S10) are predicted, and the most valuable samples are selected through the most uncertain index. Then, the most valuable samples selected are labeled through the single-pole construction quality judgment and the multi-pole construction quality judgment, and they are added to and updated in the training set, and the number of samples in the training set is increased by one. Then, the convolutional neural network model M0 is trained with the updated training set to obtain an updated convolutional neural network model M0. When the updated convolutional neural network model M0 no longer selects unlabeled samples from the focal element pool U, the construction of the training set is completed and output, otherwise the initial convolutional neural network model M0 is retrained and optimized.

[0107] Obviously, the training set construction method provided in the embodiment of the present application is constructed based on the active learning and iterative characteristics of the active learning model, and does not require the initial training of the convolutional neural network model by combining all sample numbers, which greatly saves the training time required in the actual construction process and effectively ensures the effectiveness and universality of the training set.

[0108] In a further embodiment of the present application, initializing the judgment model and performing model training in combination with the training set includes:

[0109] The judgment model (i.e., random forest model) is trained based on the training set, and the accuracy of the random forest model is corrected using the proportional verification method until the proportion of qualified samples in the training set exceeds ρ (ρ is a positive number greater than zero).

[0110] Optionally, precision trimming can be performed using the function position:

[0111]

[0112] In the above formula, ρ is the proportion of qualified samples n in the random forest training sample set N. When the proportion of qualified samples exceeds 1 / 3, the model accuracy meets the requirements; if the proportion does not meet the requirements, the model training is not completed, and the random forest training sample set needs to be adjusted until it meets the requirements.

[0113] Optionally, the proportion of qualified samples may also be set to be more than 1 / 2, 1 / 4, etc.

[0114] In one implementation of this embodiment, obtaining construction data and building a sample pool includes:

[0115] Monitor the location information of the pole erector and poles, and obtain the front and rear, left and right offsets and height of the poles based on computer vision automatic recognition technology;

[0116] Construct a feature vector P representing the construction quality, where:

[0117] P = [P1, P2, P3, P4, P5]

[0118] In the formula, P1 represents the longitude and latitude coordinates of the pole, P2 represents the longitude and latitude coordinates of the pole erector, P3 represents the front and rear offset angle of the pole, P4 represents the left and right offset angle of the pole, and P5 represents the buried depth;

[0119] Repeat the above steps to complete the sample pool construction.

[0120] Obviously, the sample pool constructed in this way covers multiple physical quantities of the pole erector and the pole during the digital construction of the pole erector. The method for judging the effectiveness of the digital construction of the pole erector based on this has innovatively proposed a digital construction method based on multiple physical quantities, which is conducive to improving the efficiency of intelligent evaluation of construction effectiveness, fully improving the phenomenon of low efficiency of manual judgment of effectiveness, and can further improve the construction efficiency of the digital construction of the pole erector.

[0121] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that makes outstanding contributions to the prior art can be embodied in the form of a software product, and the computer software product exists in a storage medium (such as ROM / RAM, disk, hard disk, optical disk, etc.), including a number of instructions for a terminal device (which can be a mobile phone, computer, server or network device, etc.) to execute the methods described in each embodiment of the present application.

[0122] Example 2

[0123] In this embodiment, a device for judging the effectiveness of digital construction of a pole erector is also provided, which is used to implement the above-mentioned embodiments and preferred implementations, and the descriptions that have been made will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0124] See also Figure 5 , Figure 5 is a structural block diagram of a device for judging the effectiveness of digital construction of a pole erector provided in an embodiment of the present application, such as Figure 5 As shown, the digital construction effectiveness judgment device for a pole erecting machine of this embodiment includes:

[0125] A data acquisition module 10 is used to obtain construction data;

[0126] The calculation and analysis module 20 is used to construct a sample pool according to the construction data, extract samples from the sample pool and classify them, actively learn to construct and optimize the training set, and use the training set to optimize and output the judgment model;

[0127] The identification module 30 is used to determine the construction quality category according to the judgment model.

[0128] Optionally, the identification module 30 includes: an identification unit, which is used to input the data features of the pole construction that needs to be predicted or judged into the judgment model to judge the construction quality category. That is, in the case of inputting data features, the device feedback can be that both the single-pole construction quality category and the multi-pole construction quality category meet the construction requirements, or the single-pole construction quality category meets the requirements and the multi-pole construction quality category does not meet the requirements, or the single-pole construction quality category and the multi-pole construction quality category do not meet the construction requirements. In this way, the effectiveness judgment result can be quickly proposed and the actual on-site construction can be guided to quickly make corresponding rectifications to ensure the construction quality.

[0129] Obviously, the digital construction effectiveness judgment device for the pole erector proposed in this embodiment fully combines the characteristics of the digital construction effectiveness judgment method for the pole erector, and can obtain more accurate construction effectiveness judgment results with less construction data, which is beneficial to improving construction efficiency and reducing the labor intensity of technicians.

[0130] It should be further explained that the above modules can be implemented by software or hardware, and for the latter, it can be implemented in the following ways, but not limited to: the above modules are all in the same processor; or the above modules are located in different processors in any combination.

[0131] Example 3

[0132] In this embodiment, a storage medium is also provided, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when it is run. A digital construction effectiveness judgment system for a pole erecting machine can also be directly installed, the system includes a processor, a user interface, a network interface, a communication bus and a memory, wherein the processor, the user interface, the network interface and the memory communicate with each other via the communication bus;

[0133] Memory, used to store computer programs;

[0134] The processor is used to execute the above-mentioned method for judging the effectiveness of digital construction of the pole erecting machine by running the program stored in the memory.

[0135] Optionally, in this embodiment, the above storage medium or system may be configured to store a computer program for performing the following steps:

[0136] S1: Obtain construction data and build a sample pool;

[0137] S2: extracting samples from the sample pool and classifying them;

[0138] S3: Use active learning model to select samples to construct training set;

[0139] S4: Initialize the judgment model and perform model training in combination with the training set;

[0140] S5: Output the optimized judgment model for prediction;

[0141] S6: Input the predicted data features to obtain the construction quality type.

[0142] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0143] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0144] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0145] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0146] S1: Obtain construction data and build a sample pool;

[0147] S2: extracting samples from the sample pool and classifying them;

[0148] S3: Use active learning model to select samples to construct training set;

[0149] S4: Initialize the judgment model and perform model training in combination with the training set;

[0150] S5: Output the optimized judgment model for prediction;

[0151] S6: Input the predicted data features to obtain the construction quality type.

[0152] The above descriptions are merely preferred embodiments of the present application, and the descriptions of various embodiments have different focuses. For parts that are not described in detail in a particular embodiment, please refer to the relevant descriptions of other embodiments.

[0153] In several embodiments provided in the present application, the device embodiments described are merely illustrative and are not intended to limit the present application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for judging the effectiveness of digital construction of a pole erector, characterized in that: The method comprises at least the following steps: Obtain construction data and build a sample pool; Extracting samples from the sample pool and classifying them; Use active learning models to select samples and construct training sets; Initializing the judgment model and performing model training in combination with the training set; Output the optimized judgment model for prediction; Input the predicted data features to obtain the construction quality type.

2. The method for judging the effectiveness of digital construction of a pole erecting machine as claimed in claim 1, characterized in that: Samples from the sample pool are extracted and classified, including single-pole construction quality judgment: Determine whether the burial depth meets the requirements, and directly classify samples that do not meet the requirements as unqualified; Calculate the deflection angle based on the front and rear deflection angle and the left and right deflection angle of the pole; Calculate the pole tip offset based on the pole length and buried depth; Outputs the single-rod construction quality type.

3. The method for judging the effectiveness of digital construction of a pole erecting machine as claimed in claim 2, characterized in that: The samples in the sample pool are extracted and classified, and the multi-rod construction quality judgment is also included: Establish the relative rectangular coordinate system between the pole and the pole erector and determine the offset angle of the pole; Performing coordinate system conversion according to the relative rectangular coordinate system to convert the relative angle into an absolute angle of the geodetic coordinate system; Calculate the lateral offset, longitudinal offset and pile spacing of each pole relative to the baseline; Output multi-rod construction quality type.

4. The method for judging the effectiveness of digital construction of a pole erecting machine as claimed in claim 3, characterized in that: Using active learning models to select samples to construct training sets includes: S100: constructing an initial training set, and constructing and training a learning model based on the training set; S101: using the learning model to predict samples in the sample pool according to the most uncertain criterion, and selecting the most valuable samples; S102: labeling the most valuable samples, and adding and updating the training set; S103: training and optimizing the learning model according to the updated training set; S104: Repeat step S101 according to the optimized learning model until the learning model no longer selects samples; S105: Output the training set.

5. The method for judging the effectiveness of digital construction of a pole erecting machine as claimed in claim 4, characterized in that: The constructing of the initial training set includes: extracting some samples from the sample pool, and performing single-rod construction quality judgment and multi-rod construction quality judgment on each of the extracted samples to construct the initial training set.

6. The method for judging the effectiveness of digital construction of a pole erecting machine according to any one of claims 1 to 5, characterized in that: Initializing the judgment model and performing model training in combination with the training set includes: The judgment model is trained based on the training set, and the accuracy of the judgment model is corrected using the proportional verification method until the proportion of qualified samples in the training set exceeds ρ (ρ is a positive number greater than zero).

7. The method for judging the effectiveness of digital construction of a pole erecting machine as claimed in claim 6, characterized in that: The judgment model uses a random forest model.

8. The method for judging the effectiveness of digital construction of a pole erecting machine according to any one of claims 1 to 5, characterized in that: Obtaining construction data and building a sample pool includes: Monitor the location information of the pole erector and poles, and obtain the front and rear, left and right offsets and height of the poles based on computer vision automatic recognition technology; Construct a feature vector P representing the construction quality, where: P = [P1, P2, P3, P4, P5] In the formula, P1 represents the longitude and latitude coordinates of the pole, P2 represents the longitude and latitude coordinates of the pole erector, P3 represents the front and rear offset angle of the pole, P4 represents the left and right offset angle of the pole, and P5 represents the buried depth; Repeat the above steps to complete the sample pool construction.

9. A digital construction effectiveness judgment system for a pole erecting machine, characterized in that: It includes a processor, a user interface, a network interface, a communication bus and a memory, wherein the processor, the user interface, the network interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; A processor is used to execute the method for judging the effectiveness of digital construction of a pole erector according to any one of claims 1 to 8 by running a program stored in a memory.

10. A device for judging the effectiveness of digital construction of a pole erecting machine, characterized in that: include: Data acquisition module, used to obtain construction data; A calculation and analysis module, used to construct a sample pool according to the construction data, extract samples from the sample pool and classify them, actively learn to construct and optimize a training set, and use the training set to optimize and output a judgment model; The identification module is used to determine the construction quality category according to the judgment model.

Images