Intelligent evaluation method and evaluation system for digital supervision of electric power project construction

Abstract

The embodiment of the invention provides an intelligent evaluation method and evaluation system for electric power project construction digital supervision, and belongs to the technical field of electric power project digital supervision. The intelligent evaluation method comprises the following steps: obtaining construction image data and quality image data of a historical construction project in a power grid system; respectively constructing a progress training set and a quality training set according to the construction image data and the quality image data; according to the method, the construction image data and the quality image data of the historical construction project in the power grid system are acquired, the training interfaces are constructed according to the data so as to train the project construction progress model and the project construction quality model, and then the real-time image of the current construction project is acquired according to the real-time image including the overall image and the local image. The progress value and the quality value of the current construction project are obtained, and the construction state of the current construction project can be evaluated / evaluated according to the progress value and the quality value, so that digital supervision of electric power project construction is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of digital supervision of electric power projects, and in particular to an intelligent evaluation method and an evaluation system for digital supervision of electric power project construction. Background Art

[0002] Digital technology refers to a series of technical means that convert various forms of content such as information, data, sound, and images into digital signals, and process, transmit, store, and apply them through computers and networks. With the development of digital technology and the requirements for digitalization of power grid systems, the current project construction in power grid systems is also gradually transforming to digitalization. Specifically, the project construction in power grid systems includes pole tower construction, transmission line construction, power station construction, etc.

[0003] In the prior art, for project construction in the power grid system, the team is divided into teams according to the construction requirements. The quality of the project construction process, the completion of the construction period, etc. are all manually recorded and counted to achieve the supervision and evaluation of the current construction project. However, this supervision method requires a lot of manpower, and there are subjective supervision errors, and the supervision and evaluation efficiency is low; it is also difficult to implement standardized supervision, which in turn affects the reliability of the supervision and evaluation of power grid project construction.

[0004] The inventors of the present application discovered during the process of realizing the present invention that the above-mentioned scheme of the prior art has the defects of low efficiency and poor reliability of supervision and evaluation in the construction of power grid projects. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide an intelligent evaluation method and an evaluation system for digital supervision of power project construction, which have the functions of high supervision and evaluation efficiency and high supervision and evaluation reliability.

[0006] In order to achieve the above object, an embodiment of the present invention provides an intelligent evaluation method for digital supervision of power project construction, comprising:

[0007] Acquire construction image data and quality image data of historical construction projects in the power grid system;

[0008] Constructing a progress training set and a quality training set respectively according to the construction image data and the quality image data;

[0009] Construct project construction schedule model and project construction quality model;

[0010] Using the progress training set and the quality training set to train the project construction progress model and the project construction quality model respectively;

[0011] Get the overall and local images of the current construction project;

[0012] Obtaining a progress value and a quality value of the current construction project according to the overall image and the local image of the current construction project, the trained project construction progress model and the project construction quality model;

[0013] The construction status of the current construction project is evaluated based on the progress value and quality value of the current construction project.

[0014] Optionally, the construction image data includes multi-scale images of historical construction projects and construction project pixel labels, and the quality image data includes quality defect images and quality defect names of the historical construction projects.

[0015] Optionally, constructing a progress training set and a quality training set according to the construction image data and the quality image data respectively includes:

[0016] Construct a progress training set based on multi-scale images of historical construction projects and corresponding pixel labels of construction projects;

[0017] Construct a quality training set based on quality defect images and corresponding quality defect names;

[0018] The progress training set and the quality training set are preprocessed.

[0019] Optionally, using the progress training set and the quality training set to respectively train the project construction progress model and the project construction quality model comprises:

[0020] Inputting the progress training set into the project construction progress model;

[0021] According to formula (1), the loss function of the project construction progress model is obtained:

[0022]

[0023] Among them, LOSS P is the loss function of the project construction progress model, M is the total number of pixels in the image, y i ′ is the construction project pixel label of the i-th pixel, y i is the predicted probability of the i-th pixel of the project construction progress model, where i is an integer number;

[0024] The parameters of the project construction progress model are modified according to the loss function of the project construction progress model.

[0025] Optionally, using the progress training set and the quality training set to respectively train the project construction progress model and the project construction quality model further includes:

[0026] Inputting the quality training set into the project construction quality model;

[0027] According to formula (2), the loss function of the project construction quality model is obtained:

[0028] LOSS Q =min1-(· j Q j ′) / (||· j ||||· j ′||), (2)

[0029] Among them, LOSS Q The loss function for building the quality model for the project, Q j The prediction vector of the jth image of the project quality model, Q j ′ is the true vector of the jth image;

[0030] The parameters of the project construction quality model are modified according to the loss function of the project construction quality model.

[0031] Optionally, obtaining the progress value and the quality value of the current construction project according to the overall image and the local image of the current construction project, the trained project construction progress model and the project construction quality model includes:

[0032] Inputting the overall image of the current construction project into the project construction progress model to obtain a feature image;

[0033] The similarity between the characteristic image and the characteristic image of the historical construction project is obtained by using the Euclidean distance, and the highest similarity value is used as the progress prediction value of the current construction project;

[0034] Obtain the difficulty rating of the current construction project in question;

[0035] According to formula (3), the progress value of the current construction project is obtained.

[0036] P t =1-log -1 γ·(S t ′-S t ), (3)

[0037] Among them, P t is the progress value of the current construction project, γ is the difficulty rating, S t is the progress forecast value of the current construction project, S t ′ is the standard progress value of the construction project in stage t, and t is an integer number.

[0038] Optionally, obtaining the progress value and the quality value of the current construction project according to the overall image and the local image of the current construction project, the trained project construction progress model and the project construction quality model further includes:

[0039] Inputting a partial image of the current construction project into the project construction quality model to obtain a defect rate of the current construction project;

[0040] According to formula (4), the quality value of the current construction project is obtained.

[0041] Q t =1-log -1 γ·(q t -q t ′), (3)

[0042] Among them, Q t is the quality value of the current construction project, q t is the defect rate of the current construction project, q t ′ is the standard defect rate of the construction project at stage t.

[0043] Optionally, evaluating the construction status of the current construction project according to the progress value and quality value of the current construction project includes:

[0044] Determine whether the progress value of the current construction project is less than the standard progress threshold of the current construction stage;

[0045] In the case where it is determined that the progress value of the current construction project is less than the standard progress threshold of the current construction stage, determining whether the progress value of the current construction project is greater than or equal to the warning progress threshold;

[0046] When it is determined that the progress value of the current construction project is greater than or equal to the warning progress threshold, a progress warning is issued and the project is executed according to the pre-construction plan;

[0047] When it is determined that the progress value of the current construction project is less than the warning progress threshold, modify the pre-construction plan and execute it according to the modified construction plan;

[0048] When it is determined that the progress value of the current construction project is greater than or equal to the standard progress threshold of the current construction stage, it is executed according to the pre-construction plan.

[0049] Optionally, evaluating the construction status of the current construction project according to the progress value and quality value of the current construction project further includes:

[0050] Determine whether the quality value of the current construction project is less than the standard quality threshold of the current construction stage;

[0051] In the case where it is determined that the quality value of the current construction project is less than the standard quality threshold of the current construction stage, determining whether the quality value of the current construction project is greater than or equal to the warning quality threshold;

[0052] When it is judged that the quality value of the current construction project is greater than or equal to the warning quality threshold, a quality warning is issued and local defect rectification is arranged;

[0053] When it is determined that the quality value of the current construction project is less than the warning quality threshold, the project construction is stopped and a rectification plan is output;

[0054] When it is determined that the quality value of the current construction project is equal to the standard quality threshold of the current construction stage, it shall be executed according to the pre-construction plan.

[0055] On the other hand, the present invention also provides an intelligent evaluation system for digital supervision of power project construction, comprising:

[0056] A camera module is fixedly arranged around the construction project and is used to obtain multi-scale images of the construction project;

[0057] A digital supervision platform is connected to the camera module and is used to execute any of the intelligent assessment methods described above.

[0058] Through the above technical scheme, the intelligent evaluation method and system for digital supervision of power project construction provided by the present invention obtains the construction image data and quality image data of historical construction projects in the power grid system, and constructs training links according to the above data to train the project construction progress model and the project construction quality model, and then obtains the progress value and quality value of the current construction project according to the real-time image of the current construction project, including the overall image and the local image. According to the progress value and the quality value, the construction status of the current construction project can be evaluated / assessed to realize the digital supervision of the power project construction, effectively reduce manpower and material resources, improve the supervision efficiency, ensure the reliability of the power project construction, have a high degree of visualization, and be more intelligent.

[0059] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0061] Figure 1 is a flow chart of an intelligent evaluation method for digital supervision of power project construction according to an embodiment of the present invention;

[0062] Figure 2 It is a flowchart of constructing a training set in an intelligent evaluation method for digital supervision of power project construction according to an embodiment of the present invention;

[0063] Figure 3 It is a flowchart of training a model in an intelligent evaluation method for digital supervision of power project construction according to an embodiment of the present invention;

[0064] Figure 4 It is a flow chart of obtaining progress value and quality value of an intelligent evaluation method for digital supervision of power project construction according to an embodiment of the present invention;

[0065] Figure 5 The present invention is a flowchart of evaluating a construction project using an intelligent evaluation method for digital supervision of power project construction according to an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0067] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. In the embodiments of this application, some existing solutions in the industry such as certain software, components, and models may be mentioned, which should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use the solution.

[0068] Figure 1 The figure is a flow chart of an intelligent evaluation method for digital supervision of power project construction according to an embodiment of the present invention. Specifically, the intelligent evaluation method may include:

[0069] In step S10, the construction image data and quality image data of the historical construction projects in the power grid system are obtained. The construction images of the historical construction projects may include multi-scale standard images of different construction projects, that is, standard images of multiple points of standard construction projects. In order to reduce the interference of the background in the standard image on the construction project, the construction image data may also include pixel labels of the construction project in the image, so as to construct image denoising / background removal training data with the image. The quality image data may include all local image data of quality defects in the historical construction projects and the corresponding defect names / types, so as to construct defect recognition training data.

[0070] In step S11, a progress training set and a quality training set are constructed according to the construction image data and the quality image data. The construction of the training set may include: Figure 2 Specifically, Figure 2 In the construction step, the following steps may be included:

[0071] In step S110, a progress training set is constructed based on the multi-scale images of historical construction projects and the corresponding construction project pixel labels. The progress training set may be composed of multi-scale / multi-point standard images and the corresponding construction project pixel labels in the standard images.

[0072] In step S111, a quality training set is constructed according to the quality defect images and the corresponding quality defect names. The quality training set may be composed of quality defect images and the corresponding defect names.

[0073] In step S112, the progress training set and the quality training set are preprocessed. After the training set is constructed, the training set needs to be preprocessed, including but not limited to Gaussian filtering, interpolation processing, etc.

[0074] In steps S110 to S112, a progress training set is constructed based on the multi-scale images of historical construction projects and the corresponding pixel labels of the construction projects. Then, a quality training set is constructed based on the quality defect images and quality defect names of the construction projects. Finally, the constructed training set is preprocessed to ensure the integrity and validity of the training data.

[0075] In step S12, a project construction progress model and a project construction quality model are constructed, wherein the specific forms of the project construction progress model and the project construction quality model include but are not limited to neural network models such as CNN and RNN.

[0076] In step S13, the progress training set and the quality training set are used to train the project construction progress model and the project construction quality model respectively. After the model is constructed, it can be trained with the corresponding training set.

[0077] In step S14, the overall image and local image of the current construction project are obtained. The overall image of the current construction project is a multi-scale / multi-point real-time image. Specifically, the multi-point image can correspond to the multi-point image of the standard construction project. The real-time image is the overall construction image. The local image can be photographed according to the possible location of the defect, without limiting the shooting angle, and can also be photographed at multiple scales.

[0078] In step S15, the progress value and quality value of the current construction project are obtained according to the overall image and the local image of the current construction project, the trained project construction progress model and the project construction quality model. Wherein, the overall image is input into the project construction progress model to obtain the construction progress of the current construction project, and converted into a progress value. The local image is input into the project construction quality model to obtain the construction quality of the current construction project, and converted into a quality value. The progress value and the quality value reflect the construction progress and construction quality of the current construction project respectively.

[0079] In step S16, the construction status of the current construction project is evaluated according to the progress value and quality value of the current construction project. After obtaining the progress value and quality of the current construction project, the current construction project can be comprehensively evaluated to determine whether to execute according to the original construction plan or modify the original construction plan, so as to ensure the reliability of the construction project.

[0080] In step S10 to step S16, the construction image data and quality image data of the historical construction projects in the power grid system are first obtained, and then the progress training set and the quality training set are respectively constructed according to the construction image data and the quality image data, and the project construction progress model and the project construction quality model are respectively trained using the progress training set and the quality training set. The overall image and the local image of the current construction project are respectively input into the project construction progress model and the project construction quality model to obtain the progress value and the quality value of the current construction project. Finally, the current construction project is evaluated according to the progress value and the quality value to determine whether the current construction project is qualified and compliant, and whether the construction plan needs to be modified.

[0081] In traditional power grid systems, project construction is carried out through team division of labor according to construction requirements. The quality of the project construction process, the completion of the construction period, etc. are all manually recorded and counted to achieve the supervision and evaluation of the current construction project. However, this supervision method requires a lot of manpower, and there are subjective supervision errors, and the supervision and evaluation efficiency is low; and it is difficult to implement standardized supervision, which in turn affects the reliability of the supervision and evaluation of power grid project construction. In this embodiment of the present invention, the progress value and quality value of the current construction project are obtained, so that the construction status of the current construction project can be stably evaluated / assessed, so as to realize the digital supervision of the power project construction, effectively reduce manpower and material resources, improve supervision efficiency, ensure the reliability of power project construction, and have a high degree of visualization and greater intelligence.

[0082] In this embodiment of the present invention, after obtaining the progress training set and the quality training set, it is necessary to train the project construction progress model and the project construction quality model. The specific training steps can be as follows: Figure 3 Specifically, Figure 3 In the training step, the training step may include:

[0083] In step S130, the progress training set is input into the project construction progress model.

[0084] In step S131, the loss function of the project construction progress model is obtained according to formula (1):

[0085]

[0086] Among them, LOSS P is the loss function of the project construction progress model, M is the total number of pixels in the image, and y i ′ is the construction project pixel label of the i-th pixel, y i is the predicted probability of the i-th pixel in the project construction progress model, where i is an integer number.

[0087] In step S132, the parameters of the project construction progress model are corrected according to the loss function of the project construction progress model.

[0088] In step S133, the quality training set is input into the project construction quality model.

[0089] In step S134, the loss function of the project construction quality model is obtained according to formula (2):

[0090] LOSS Q =min1-(Q j Q j ′) / (||Q j ||||Q j ′||), (2)

[0091] Among them, LOSS Q is the loss function of the project quality model, Q j is the prediction vector of the jth image of the project construction quality model, Q j ′ is the true vector of the jth image. Specifically, the loss function can also be calculated by summing multiple feature vectors in the image.

[0092] In step S135 , the parameters of the project construction quality model are corrected according to the loss function of the project construction quality model.

[0093] In step S130 to step S135, the progress training set and the quality training set are respectively input into the project construction progress model and the project construction quality model, and their respective model parameters are modified according to the corresponding loss functions, so as to obtain the finally trained project construction progress model and project construction quality model.

[0094] In this embodiment of the present invention, after obtaining the overall image and the local image of the current construction project, they can be input into the corresponding model, and the obtained values ​​can be converted into progress values ​​and quality values. The specific steps can be as follows: Figure 4 Specifically, Figure 4 In the embodiment, the intelligent evaluation method may further include:

[0095] In step S150, the overall image of the current construction project is input into the project construction progress model to obtain a characteristic image. The overall image is input into the project construction progress model, and the construction project part in the overall image can be identified to form a characteristic image of the current construction project.

[0096] In step S151, the Euclidean distance is used to obtain the similarity between the feature image and the feature image of the historical construction project, and the highest similarity value is obtained. Among them, the feature image of the historical construction project is the feature image of the standard construction image of the historical construction project after background removal. The feature image of the current construction project is calculated for similarity with the feature images of the standard construction images of all historical construction projects, and the Euclidean distance method known to those skilled in the art is used to obtain multiple similarity values, and the value with the highest similarity is selected. The highest similarity is also the construction progress of the current construction project. Specifically, the feature images of multiple fixed points (shooting locations of standard images) of the current construction project can be selected for calculation respectively, and the average value is finally calculated as the final value.

[0097] In step S152, the difficulty rating of the current construction project is obtained. The difficulty rating of the current construction project can be scored from 2 to 10, which is determined according to the construction difficulty of the construction project.

[0098] In step S153, the progress value of the current construction project is obtained according to formula (3):

[0099] P t =1-log -1 γ·(S t ′-S t ), (3)

[0100] Among them, P t is the progress value of the current construction project, γ is the difficulty rating, S t is the progress forecast value of the current construction project, S t′ is the standard progress value of the construction project at stage t, and t is an integer number. Taking into account the construction difficulty of different construction projects, the more difficult the construction project is, the longer the construction period is required, and the project difficulty and project period are not in a linear relationship. Therefore, for construction projects with a higher difficulty rating, the conversion of their progress should be relatively relaxed, and the value calculated according to formula (3) can reflect the progress value of the current construction project. Specifically, for the progress calculation of the construction project, it can be divided according to the construction stage of the construction project. For example, stage t is the current construction stage, and the corresponding image is obtained once for each stage. The stage can include but is not limited to one day, half a month, etc.

[0101] In step S154, the partial image of the current construction project is input into the project construction quality model to obtain the defect rate of the current construction project. The multiple partial images of the current construction project can be taken according to the construction content of stage t or other points, or can be taken at multiple points according to the locations where construction defects are prone to occur. Specifically, the defect rate of the current construction project can be obtained by dividing the number of defects output by the project construction quality model by the number of partial images input.

[0102] In step S155, the quality value of the current construction project is obtained according to formula (4):

[0103] Q t =1-log -1 γ·(q t -q t ′), (3)

[0104] Among them, Q t is the quality value of the current construction project, q t is the defect rate of the current construction project, q t ′ is the standard defect rate of the construction project at stage t.

[0105] In step S150 to step S155, the overall image of the current construction project is first input into the project construction progress model to obtain the corresponding feature image. The feature image of the current construction project and the feature image of the historical standard image are respectively calculated to obtain the feature image of the historical standard image with the highest similarity value. And this similarity value also reflects the construction progress of the current construction project, so it can be combined with the difficulty rating of the current construction project to obtain the progress value of the current construction project. Similarly, multiple local images of the current construction project are input into the project construction quality model to obtain the defect rate of the current construction project, and the defect rate is combined with the difficulty rating of the construction project to obtain the quality value of the current construction project. The above method can be used to digitally manage the real-time information of the current construction project, which is more convenient and reliable.

[0106] In this embodiment of the present invention, after obtaining the progress value and quality value of the current construction project, the construction status of the current construction project can be evaluated to determine the next construction plan. The specific evaluation steps can be as follows: Figure 5 Specifically, Figure 5 In the example, the evaluation step may include:

[0107] In step S160, it is determined whether the progress value of the current construction project is less than the standard progress threshold of the current construction stage, wherein the standard progress threshold is 100%, that is, the construction tasks of the current construction stage need to be completed according to the plan.

[0108] In step S161, when it is determined that the progress value of the current construction project is less than the standard progress threshold of the current construction stage, it is determined whether the progress value of the current construction project is greater than or equal to the warning progress threshold. The warning progress threshold is also the minimum completion of the current construction stage, which can be determined according to the actual construction period requirements, such as 95%.

[0109] In step S162, if the progress value of the current construction project is greater than or equal to the warning progress threshold, a progress warning is issued and the construction plan is executed in accordance with the pre-construction plan. If the progress value of the current construction project is between the warning progress threshold and the standard progress threshold, it means that there is a controllable delay in the current construction progress, so a progress warning can be issued and the construction plan can be executed in accordance with the original plan, and the delay can be gradually made up in the subsequent construction.

[0110] In step S163, if the progress value of the current construction project is less than the warning progress threshold, the pre-construction plan is modified and executed according to the modified construction plan. If the progress value of the current construction project is less than the warning progress threshold, it means that the current construction progress is seriously delayed, which may cause the overall construction project to be delayed, thereby affecting the overall project construction, so it is necessary to re-formulate the construction plan and execute it according to the modified one.

[0111] In step S164, if it is determined that the progress value of the current construction project is greater than or equal to the standard progress threshold of the current construction stage, the pre-construction plan is followed. If the progress value of the current construction project is greater than the standard progress threshold, it means that it is completed ahead of schedule and can be followed according to the pre-construction plan.

[0112] In step S165, it is determined whether the quality value of the current construction project is less than the standard quality threshold of the current construction stage, wherein the standard quality threshold is 100%, that is, the current construction stage needs to complete the construction tasks of the current construction stage without defects.

[0113] In step S166, when it is determined that the quality value of the current construction project is less than the standard quality threshold of the current construction stage, it is determined whether the quality value of the current construction project is greater than or equal to the warning quality threshold. The warning quality threshold is also the minimum achievement of the current construction stage, which can be determined according to the actual quality requirements, such as 98%.

[0114] In step S167, if the quality value of the current construction project is greater than or equal to the warning quality threshold, a quality warning is issued and local defect rectification is arranged. If the quality value of the current construction project is between the standard quality threshold and the warning quality threshold, it means that the quality problem of the current construction project is controllable, a warning is issued, and personnel are arranged to carry out targeted defect rectification locally.

[0115] In step S168, if the quality value of the current construction project is less than the warning quality threshold, the project construction is stopped and a rectification plan is output. If the quality value of the current construction project is less than the warning quality threshold, it means that the current construction quality problem is serious and may affect the subsequent project construction. Therefore, it is necessary to stop the project construction, study the defect rectification plan, and train the current construction team.

[0116] In step S169, when it is determined that the quality value of the current construction project is equal to the standard quality threshold of the current construction stage, the pre-construction plan is followed. Since the maximum quality value of the current construction project is 1, that is, there is no defect, when the quality value of the current construction project is equal to the standard quality threshold of the current construction stage, the pre-construction plan is followed.

[0117] In step S160 to step S169, after obtaining the progress value of the current construction project, it is compared with the standard progress threshold and the warning progress threshold of the current construction stage in turn, so as to take corresponding measures. If the progress value is greater than or equal to the warning progress threshold, it can be executed according to the pre-construction plan, otherwise it is necessary to modify the construction plan. Similarly, after obtaining the quality value of the current construction project, it is compared with the standard quality threshold and the warning quality threshold of the current construction stage in turn, so as to take corresponding measures. If the quality value is greater than or equal to the warning quality threshold, the defects can be rectified in a targeted manner without stopping work; otherwise, it is necessary to stop work, study the rectification plan and train the construction team.

[0118] On the other hand, the present invention also provides an intelligent evaluation system for digital supervision of power project construction. Specifically, the intelligent evaluation system may include a camera module and a digital supervision platform. Specifically, the camera module may include a high-definition camera, a drone, etc.

[0119] The camera module is fixedly arranged around the construction project to obtain multi-scale images of the construction project. The digital supervision platform is connected to the camera module to execute any of the above intelligent assessment methods.

[0120] Through the above technical scheme, the intelligent evaluation method and system for digital supervision of power project construction provided by the present invention obtains the construction image data and quality image data of historical construction projects in the power grid system, and constructs training links according to the above data to train the project construction progress model and the project construction quality model, and then obtains the progress value and quality value of the current construction project according to the real-time image of the current construction project, including the overall image and the local image. According to the progress value and the quality value, the construction status of the current construction project can be evaluated / assessed to realize the digital supervision of the power project construction, effectively reduce manpower and material resources, improve the supervision efficiency, ensure the reliability of the power project construction, have a high degree of visualization, and be more intelligent.

[0121] 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 in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0122] 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.

[0123] 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.

[0124] 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.

[0125] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0126] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0127] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0128] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0129] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. An intelligent evaluation method for digital supervision of power project construction, characterized in that: include: Acquire construction image data and quality image data of historical construction projects in the power grid system; Constructing a progress training set and a quality training set respectively according to the construction image data and the quality image data; Construct project construction schedule model and project construction quality model; Using the progress training set and the quality training set to train the project construction progress model and the project construction quality model respectively; Get the overall and local images of the current construction project; Obtaining a progress value and a quality value of the current construction project according to the overall image and the local image of the current construction project, the trained project construction progress model and the project construction quality model; The construction status of the current construction project is evaluated based on the progress value and quality value of the current construction project.

2. The intelligent evaluation method according to claim 1, characterized in that: The construction image data includes multi-scale images of historical construction projects and construction project pixel labels, and the quality image data includes quality defect images and quality defect names of the historical construction projects.

3. The intelligent evaluation method according to claim 2, characterized in that: Constructing a progress training set and a quality training set according to the construction image data and the quality image data respectively comprises: Construct a progress training set based on multi-scale images of historical construction projects and corresponding pixel labels of construction projects; Construct a quality training set based on quality defect images and corresponding quality defect names; The progress training set and the quality training set are preprocessed.

4. The intelligent evaluation method according to claim 3, characterized in that: Using the progress training set and the quality training set to train the project construction progress model and the project construction quality model respectively includes: Inputting the progress training set into the project construction progress model; According to formula (1), the loss function of the project construction progress model is obtained: Among them, LOSS P is the loss function of the project construction progress model, M is the total number of pixels in the image, y i ′ is the construction project pixel label of the i-th pixel, y i is the predicted probability of the i-th pixel of the project construction progress model, where i is an integer number; The parameters of the project construction progress model are modified according to the loss function of the project construction progress model.

5. The intelligent evaluation method according to claim 4, characterized in that: Using the progress training set and the quality training set to train the project construction progress model and the project construction quality model respectively also includes: Inputting the quality training set into the project construction quality model; According to formula (2), the loss function of the project construction quality model is obtained: LOSS Q =min1-(Q j ·Q j ′) / (||Q j ||||Q j ′||), (2) Among them, LOSS Q The loss function for building the quality model for the project, Q j The prediction vector of the jth image of the project quality model, Q j ′ is the true vector of the jth image; The parameters of the project construction quality model are modified according to the loss function of the project construction quality model.

6. The intelligent evaluation method according to claim 5, characterized in that: Acquiring the progress value and the quality value of the current construction project according to the overall image and the local image of the current construction project, the trained project construction progress model and the project construction quality model includes: Inputting the overall image of the current construction project into the project construction progress model to obtain a feature image; The similarity between the characteristic image and the characteristic image of the historical construction project is obtained by using the Euclidean distance, and the highest similarity value is used as the progress prediction value of the current construction project; Obtain the difficulty rating of the current construction project in question; According to formula (3), the progress value of the current construction project is obtained. P t =1-log -1 c·(S t ′-S t ), (3) Among them, P t is the progress value of the current construction project, γ is the difficulty rating, S t is the progress forecast value of the current construction project, S t ′ is the standard progress value of the construction project in stage t, and t is an integer number.

7. The intelligent evaluation method according to claim 6, characterized in that: Acquiring the progress value and the quality value of the current construction project according to the overall image and the partial image of the current construction project, the trained project construction progress model and the project construction quality model also includes: Inputting a partial image of the current construction project into the project construction quality model to obtain a defect rate of the current construction project; According to formula (4), the quality value of the current construction project is obtained. Q t =1-log -1 γ·(q t -q t ′), (3) Among them, Q t is the quality value of the current construction project, q t is the defect rate of the current construction project, q t ′ is the standard defect rate of the construction project at stage t.

8. The intelligent evaluation method according to claim 7, characterized in that: Evaluating the construction status of the current construction project based on the progress value and quality value of the current construction project includes: Determine whether the progress value of the current construction project is less than the standard progress threshold of the current construction stage; In the case where it is determined that the progress value of the current construction project is less than the standard progress threshold of the current construction stage, determining whether the progress value of the current construction project is greater than or equal to the warning progress threshold; When it is determined that the progress value of the current construction project is greater than or equal to the warning progress threshold, a progress warning is issued and the project is executed according to the pre-construction plan; When it is determined that the progress value of the current construction project is less than the warning progress threshold, modify the pre-construction plan and execute it according to the modified construction plan; When it is determined that the progress value of the current construction project is greater than or equal to the standard progress threshold of the current construction stage, it is executed according to the pre-construction plan.

9. The intelligent evaluation method according to claim 8, characterized in that: Evaluating the construction status of the current construction project based on the progress value and quality value of the current construction project also includes: Determine whether the quality value of the current construction project is less than the standard quality threshold of the current construction stage; In the case where it is determined that the quality value of the current construction project is less than the standard quality threshold of the current construction stage, determining whether the quality value of the current construction project is greater than or equal to the warning quality threshold; When it is judged that the quality value of the current construction project is greater than or equal to the warning quality threshold, a quality warning is issued and local defect rectification is arranged; When it is determined that the quality value of the current construction project is less than the warning quality threshold, the project construction is stopped and a rectification plan is output; When it is determined that the quality value of the current construction project is equal to the standard quality threshold of the current construction stage, it shall be executed according to the pre-construction plan.

10. An intelligent evaluation system for digital supervision of power project construction, characterized in that: include: A camera module is fixedly arranged around the construction project and is used to obtain multi-scale images of the construction project; A digital supervision platform is connected to the camera module and is used to execute the intelligent assessment method as described in any one of claims 1-9.

Images