Underground pipeline construction area prediction device and method

Abstract

The invention discloses an underground pipeline construction area prediction device and a control method thereof, the device comprises an input module, a division module, an extraction module, a modeling module, a prediction module, a display module and a feedback module, the division module performs meshing division on the urban area, each grid serves as a basic unit for construction area prediction, the grids are equally divided according to geographic areas or underground pipelines, and meanwhile third-party construction data are divided into a training set and a test set; the extraction module is used for analyzing rules and characteristics of historical third-party construction data and extracting key factors influencing construction. According to the underground pipeline construction area prediction device and the control method thereof, the risk of urban area grid construction can be predicted according to urban area historical construction data in combination with other elements, and a pipeline ownership unit is guided to optimize an inspection route.

Description

Technical Field

[0001] The present application relates to the field of construction area prediction, and in particular to a device and method for predicting underground pipeline construction areas. Background Art

[0002] Urban underground pipelines provide residents and industrial and commercial users with a steady supply of safe, economical, convenient and environmentally friendly natural gas, electricity and water, and are important operating arteries of the city. With the continuous acceleration of urban construction and old urban area renovation, various municipal projects and community renovation construction projects are in full swing, and the risk of third-party construction damage to underground pipeline facilities has increased accordingly.

[0003] The construction units have a fluke mentality and a weak sense of safety. They fail to strictly follow the relevant rules and regulations and the requirements of the construction plan during the construction process. The operators of construction machinery carry out barbaric construction without a detailed investigation of the existing pipelines at the construction site, such as using large machinery to dig within the protection range of gas pipeline facilities. In order to meet the deadline, the construction units damage the underground pipelines of unknown location without informing the underground pipeline operation and maintenance units, or knowingly carry out construction and cause damage to the underground pipelines.

[0004] Currently, the monitoring of third-party construction mainly relies on daily inspections and some conventional monitoring equipment such as cameras and vibration monitoring. However, due to factors such as inspection density, equipment cost and equipment deployment density, a large amount of third-party construction is not discovered in time, resulting in accidental damage to underground pipelines and causing pipeline accidents.

[0005] Therefore, the present invention mainly provides an underground pipeline construction area prediction device and method, so as to predict the risks of construction occurring in urban areas, conduct targeted inspections, improve the detection rate of third-party construction, and reduce the possibility of pipeline accidents caused by third-party construction in underground pipelines. Summary of the invention

[0006] The present invention provides the following technical solutions:

[0007] In a first aspect, this specification provides an underground pipeline construction area prediction device, comprising

[0008] An input module, a partitioning module, an extraction module, a modeling module, a prediction module, a display module, and a feedback module, wherein:

[0009] The input module realizes the input of urban area maps and urban historical third-party construction data.

[0010] The division module divides the urban area into grids, where each grid is used as the basic unit for construction area prediction. The grid division is divided equally according to geographical area or underground pipelines. At the same time, the third-party construction data is divided into training sets and test sets.

[0011] The extraction module is used to analyze the patterns and characteristics of historical third-party construction data and extract key factors that affect construction.

[0012] The modeling module builds a construction area prediction model, uses a third-party construction data training set to train the model, and uses a third-party construction data test set to verify and correct the model effect.

[0013] The prediction module uses the construction area prediction model built by the modeling module to predict the construction risk of each grid in the urban area and obtain the construction incidence rate of each area.

[0014] The display module is used to directly display the risks of construction in each grid area. The display content of this module is superimposed with the administrative area map, underground pipeline map and construction risks, which is convenient for underground pipeline operation and maintenance personnel and units to grasp the possibility of construction in each area in real time and adjust the operation and maintenance strategy according to the display results. For example, it can intensify the operation in areas with higher construction risks and reduce the operation and maintenance frequency in areas with lower construction risks, so as to improve the detection rate of third-party construction.

[0015] The feedback module is used to receive feedback from operation and maintenance personnel. When the operation and maintenance personnel find a third-party construction entry system based on the inspection suggestions, the modeling module will continue to optimize and adjust the model according to the feedback results to improve the model prediction accuracy.

[0016] In a second aspect, the present invention provides a method for controlling an underground pipeline construction area prediction device, comprising:

[0017] Obtain the regional vector map and historical third-party construction data of the target city. The historical third-party construction data should include construction time, construction area, construction type, construction unit, construction tools, whether the construction caused damage to the pipeline, and whether it was nighttime construction;

[0018] Grid the urban area and divide the third-party construction data into training sets and test sets. The training set is used to train the model, and the test set is used to test and verify the model's effect.

[0019] Sort out historical construction data and analyze the key factors that affect construction data. Sort out and extract the key factors that affect construction from the dimensions of grid construction data, geographic data, pipeline data, weather data, and major event data;

[0020] The grid construction data includes the number of constructions that occurred in the grid within N days, and the number of constructions that occurred in similar areas (similar geographical features and pipeline data) within N days;

[0021] Geographic data includes the grid's address location, the main features of the area,

[0022] Pipeline data includes pipeline laying time, pipeline material, historical maintenance times, pipeline pressure, pipeline medium and pipeline ownership unit.

[0023] Weather data includes weather conditions of the previous N days, such as sunny, cloudy, rainy, and snowy.

[0024] Major event data refers to the major events in the previous N days. Major events are collective activities of the whole city and major holidays such as May Day, National Day, and Spring Festival.

[0025] N is generally set to 7, which means that the situation in the past week is considered as the joint vector of construction prediction elements in the previous 7 days of a certain grid, which is used to predict the risk of construction in the area at present.

[0026] The construction area prediction model is constructed by using a neural network-based machine learning method. The training set is used to train the model using a machine learning method to obtain a weight coefficient matrix. The generated prediction model is verified using a test set, and cross-validation is performed repeatedly until the model prediction performance meets the requirements.

[0027] By sorting out and analyzing the risk factors of construction, a data set was constructed to train a prediction model for high-incidence construction areas. 4 / 5 of the data were randomly selected as the model training set, and the remaining 1 / 5 of the data were used as the model test set. The joint vector of prediction factors was input to comprehensively consider the situation in the previous 7 days. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A schematic diagram of the input of the model in an embodiment of the present invention;

[0029] Figure 2 A schematic diagram of converting a weighted sum into a vector inner product in an embodiment of the present invention;

[0030] Figure 3 Schematic diagram of a long short-term memory neural network in an embodiment of the present invention;

[0031] Figure 4 Schematic diagram of the model prediction process in an embodiment of the present invention.

[0032] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of this specification more clear, the technical solutions of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this specification.

[0034] As described herein, the term “including” and various variations thereof may be understood as open-ended terms meaning “including but not limited to,” and the term “one embodiment” may be understood as “at least one embodiment.”

[0035] Embodiment 1

[0036] This embodiment provides an underground pipeline construction area prediction device and method, which predicts the risk of construction occurring in an urban area, and then guides underground pipeline operation and maintenance units to conduct effective inspections and improve the detection rate of third-party construction.

[0037] The underground pipeline construction area prediction device comprises an input module, a division module, an extraction module, a modeling module, a prediction module, a display module and a feedback module.

[0038] The input module realizes the input of urban area maps and urban historical third-party construction data.

[0039] The division module divides the urban area into grids, where each grid is used as the basic unit for construction area prediction. The grid division can be divided equally according to the geographical area or according to the underground pipelines. At the same time, the third-party construction data is divided into training sets and test sets.

[0040] The extraction module is used to analyze the patterns and characteristics of historical third-party construction data and extract key factors that affect the occurrence of construction.

[0041] The modeling module constructs a construction area prediction model, uses a third-party construction data training set to train the model, and uses a third-party construction data test set to verify and correct the model effect.

[0042] The prediction module uses the construction area prediction model built by the modeling module to predict the construction risk of each grid in the urban area and obtain the construction incidence rate of each area.

[0043] The display module is used to directly display the risks of construction occurring in each grid area. The module displays content superimposed with the administrative area map, underground pipeline map and construction risks, so that underground pipeline operation and maintenance personnel and units can grasp the possibility of construction occurring in each area in real time, and adjust the operation and maintenance strategy according to the displayed results. For example, intensify operations in areas with higher construction risks, and reduce the operation and maintenance frequency in areas with lower construction risks, so as to improve the detection rate of third-party construction.

[0044] The feedback module is used to receive feedback from operation and maintenance personnel. When the operation and maintenance personnel find a third-party construction entry system based on the inspection suggestions, the modeling module will continue to optimize and adjust the model according to the feedback results to improve the model prediction accuracy.

[0045] Embodiment 2

[0046] Based on the same technical concept, an embodiment of the present invention also provides a method for controlling an underground pipeline construction area prediction device. Since the principle of solving the problem by the above method is similar to that of the underground pipeline construction area prediction device, the implementation of the above method can refer to the implementation of the device, and the repeated parts will not be repeated.

[0047] An embodiment of the present invention provides a method for controlling an underground pipeline construction area prediction device, the method being executed by a processor and comprising the following steps:

[0048] 1. Obtain the regional vector map and historical third-party construction data of the target city. The historical third-party construction data should include construction time, construction area, construction type, construction unit, construction tools, whether the construction causes damage to the pipeline, whether it is nighttime construction, etc.

[0049] 2. Grid the urban area and divide the third-party construction data into training sets and test sets. The training set is used to train the model, and the test set is used to test and verify the model's effectiveness.

[0050] 3. Sort out historical construction data and analyze the key factors that affect construction data. Sort out and extract the key factors that affect construction from several dimensions, such as grid construction data, geographic data, pipeline data, weather data, and major event data.

[0051] The grid construction data includes the number of constructions that occurred in the grid within N days and the number of constructions that occurred in similar areas (with similar geographical features and pipeline data) within N days.

[0052] The geographic data includes the address location of the grid and the main features of the area (shopping malls, parks, green spaces, schools, hospitals, etc.).

[0053] Pipeline data includes pipeline laying time, pipeline material, historical maintenance times, pipeline pressure, pipeline medium and pipeline ownership unit, etc.

[0054] Weather data includes weather conditions of the previous N days, such as sunny, cloudy, rainy, snowy, etc.

[0055] The major event data refers to the major events in the previous N days. Major events are collective activities of a city-wide nature and major holidays such as May Day, National Day, and the Spring Festival.

[0056] N is generally set to 7, which means that the situation in the past week is taken into account as the joint vector of construction prediction elements for a certain grid in the previous 7 days, which is used to predict the risk of current construction in the area.

[0057] 4. Use neural network-based machine learning to build a construction area prediction model. Use the training set and machine learning methods to train the model to obtain the weight coefficient matrix. Use the test set to verify the generated prediction model, and repeatedly adjust and participate in cross-validation until the model prediction performance meets the requirements.

[0058] By combing and analyzing the risk factors of construction, a data set for training the prediction model of high-incidence construction areas was constructed, of which 4 / 5 of the data were randomly selected as the model training set, and the remaining 1 / 5 of the data were used as the model test set. The joint vector of prediction factors was input: (geographic data) + (pipeline data) + (whether there is construction on a specific date in the area) + (the number of construction on a specific date in similar areas) + (weather conditions), taking into account the situation in the previous 7 days.

[0059] like Figure 1 As shown, the input of the model is the various factors considered, x = [x1, x2, .... xn] T, each factor corresponds to a weight wi, i = 1, 2, ... n. First, the weighted sum of all factors is calculated, and then the threshold w0 inside the model is added. The sum is input into a nonlinear activation function f, and finally the model output y is generated.

[0060] The input x is represented by a vector, and the corresponding weight is also represented by a weight vector w. Then the weighted sum becomes the vector inner product, so we get Figure 2 The formula shown.

[0061] The activation function selects the Sigmoid function:

[0062]

[0063] The parameters of the model (weights and thresholds) are solved by stochastic gradient descent:

[0064]

[0065] like Figure 3 As shown, the neural network selected is a long short-term memory neural network.

[0066] like Figure 4 The model prediction process is shown, which combines the construction situation element vector of the selected area in the previous N days to predict the probability of construction occurring in the area at present.

[0067] The above technical solution of the present application has the following advantages:

[0068] The underground pipeline construction area prediction device and method provided in the present application can predict the risks of grid construction in urban areas based on historical construction data of urban areas in combination with other factors, guide pipeline right-holding units to optimize inspection routes, increase third-party construction inspection rates, and reduce the possibility of third-party construction damaging pipelines. The device can also receive result feedback, continuously perform self-correction and optimization, and improve prediction accuracy.

[0069] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0070] Those skilled in the art know that, in addition to implementing the device and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to implement the same functions of the device and its various devices, modules, and units provided by the present invention in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, etc. by performing logic programming on the method steps. Therefore, the device and its various devices, modules, and units provided by the present invention can be regarded as structures within hardware components or as software modules for implementing methods.

[0071] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A device for predicting underground pipeline construction areas, characterized in that: It includes an input module, a division module, an extraction module, a modeling module, a prediction module, a display module and a feedback module, wherein: The input module realizes the input of urban area maps and urban historical third-party construction data. The division module divides the urban area into grids, where each grid is used as the basic unit for construction area prediction. The grid division is divided equally according to geographical area or underground pipelines. At the same time, the third-party construction data is divided into training sets and test sets. The extraction module is used to analyze the patterns and characteristics of historical third-party construction data and extract key factors that affect construction. The modeling module builds a construction area prediction model, uses a third-party construction data training set to train the model, and uses a third-party construction data test set to verify and correct the model effect. The prediction module uses the construction area prediction model built by the modeling module to predict the construction risk of each grid in the urban area and obtain the construction incidence rate of each area. The display module is used to directly display the risks of construction in each grid area. The display content of this module is superimposed with the administrative area map, underground pipeline map and construction risks, which is convenient for underground pipeline operation and maintenance personnel and units to grasp the possibility of construction in each area in real time and adjust the operation and maintenance strategy according to the display results. For example, it can intensify the operation in areas with higher construction risks and reduce the operation and maintenance frequency in areas with lower construction risks, so as to improve the detection rate of third-party construction. The feedback module is used to receive feedback from operation and maintenance personnel. When the operation and maintenance personnel find a third-party construction entry system based on the inspection suggestions, the modeling module will continue to optimize and adjust the model according to the feedback results to improve the model prediction accuracy.

2. A method for controlling the underground pipeline construction area prediction device according to claim 1, characterized in that: include: Obtain regional vector diagrams and historical third-party construction data for the target city. The historical third-party construction data should include construction time, construction area, construction type, construction unit, construction tools, whether the construction caused damage to the pipeline, and whether it was carried out at night. Grid the urban area and divide the third-party construction data into training sets and test sets. The training set is used to train the model, and the test set is used to test and verify the model's effect. Sort out historical construction data and analyze the key factors that affect construction data. Sort out and extract the key factors that affect construction from the dimensions of grid construction data, geographic data, pipeline data, weather data, and major event data. The grid construction data includes the number of constructions that occurred in the grid within N days, and the number of constructions that occurred in similar areas (similar geographical features and pipeline data) within N days; Geographic data includes the grid's address location, the main features of the area, Pipeline data includes pipeline laying time, pipeline material, historical maintenance times, pipeline pressure, pipeline medium and pipeline ownership unit. Weather data includes weather conditions of the previous N days, such as sunny, cloudy, rainy, and snowy. Major event data refers to the major events in the previous N days. Major events are collective activities of the whole city and major holidays such as May Day, National Day, and Spring Festival. N is generally set to 7, which means that the situation in the past week is considered as the joint vector of construction prediction elements in the previous 7 days of a certain grid, which is used to predict the risk of construction in the area at present. The construction area prediction model is constructed by using a neural network-based machine learning method. The training set is used to train the model using a machine learning method to obtain a weight coefficient matrix. The generated prediction model is verified using a test set, and cross-validation is performed repeatedly until the model prediction performance meets the requirements. By sorting out and analyzing the risk factors of construction, a data set was constructed to train a prediction model for high-incidence construction areas. 4 / 5 of the data were randomly selected as the model training set, and the remaining 1 / 5 of the data were used as the model test set. The joint vector of prediction factors was input to comprehensively consider the situation in the previous 7 days.

Images