Bridge construction safety risk assessment method and system based on convolutional neural network

Abstract

The invention discloses a bridge construction safety risk assessment method and system based on a convolutional neural network, and the method comprises the steps: obtaining various safety factor data affecting the safety of bridge construction, and the safety factor data comprise environment factor data, equipment factor data and material factor data; preprocessing the safety factor data, inputting the preprocessed safety factor data into a convolutional neural network model, and predicting an accident severity value; inputting the safety factor data into an SD-MOP model, and calculating the accident occurrence probability of bridge construction; and combining the accident severity value with the accident occurrence probability, and outputting a comprehensive risk level assessment result.

Description

Technical Field

[0001] The present invention belongs to the technical field of engineering construction safety assessment, and more specifically, relates to a convolutional neural network-based bridge construction safety risk assessment method and system. Background Art

[0002] With the rapid development of the construction industry, the complexity and uncertainty of construction sites have increased, and safety accidents have occurred frequently, especially in high-risk areas such as bridge construction. Traditional safety risk assessment methods usually rely on manual experience and historical accident data for analysis, which has problems such as strong subjectivity, poor real-time performance and low accuracy. The construction site environment changes rapidly and there are many influencing factors. It is difficult for traditional assessment methods to predict accident risks in a timely and accurate manner.

[0003] In recent years, with the widespread application of big data, artificial intelligence and optimization algorithms, accident prediction using intelligent systems has gradually become a research hotspot. For example, risk assessment methods based on historical data and statistical models have been able to improve the accuracy of predictions to a certain extent, but these methods often cannot dynamically respond to changes in real-time data on construction sites, and have weak prediction capabilities for the probability of accidents in complex environments with multiple factors and dimensions. In addition, the existing risk assessment systems are relatively backward in research on accident severity prediction, making it difficult to give specific accident risk levels and corresponding preventive measures.

[0004] Therefore, there is an urgent need for an intelligent accident risk assessment system based on multi-factor dynamic analysis, which can combine real-time construction site data and historical data to accurately predict the probability and severity of accidents, and provide more effective risk warning and control measures to reduce the occurrence of construction safety accidents. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a bridge construction safety risk assessment method based on convolutional neural network, comprising:

[0006] Acquire various safety factor data that affect bridge construction safety, the safety factor data including: environmental factor data, equipment factor data and material factor data;

[0007] Preprocessing the safety factor data, inputting the preprocessed safety factor data into a convolutional neural network model, and predicting an accident severity value;

[0008] Inputting the safety factor data into the SD-MOP model to calculate the probability of accidents occurring in bridge construction;

[0009] The accident severity value is combined with the accident occurrence probability to output a comprehensive risk level assessment result.

[0010] Furthermore, preprocessing the safety factor data includes: quantizing the safety factor data and converting it into one-hot encoding.

[0011] Further, combining the accident severity value with the accident occurrence probability includes: multiplying the accident severity value with the accident occurrence probability to obtain a comprehensive risk value;

[0012] A threshold range for dividing risk levels is set, the comprehensive risk value is classified according to the threshold range for dividing risk levels, and finally a risk level assessment result is obtained.

[0013] Furthermore, the convolutional neural network model includes: a one-dimensional convolution layer, a batch normalization layer, a maximum pooling layer, a fully connected layer and a soft maximum output layer.

[0014] Furthermore, it also includes: displaying the risk level assessment result to the user through a user interface.

[0015] The present invention also proposes a bridge construction safety risk assessment system based on a convolutional neural network, comprising:

[0016] An input module is used to obtain various safety factor data that affect the safety of bridge construction, wherein the safety factor data includes: environmental factor data, equipment factor data and material factor data;

[0017] A convolutional neural network module, used to pre-process the safety factor data, input the pre-processed safety factor data into a convolutional neural network model, and predict the accident severity value;

[0018] An accident probability prediction module is used to input the safety factor data into the SD-MOP model to calculate the accident probability of bridge construction;

[0019] The risk level calculation module is used to combine the accident severity value with the accident occurrence probability and output a comprehensive risk level assessment result.

[0020] Furthermore, preprocessing the safety factor data includes: quantizing the safety factor data and converting it into one-hot encoding.

[0021] Further, combining the accident severity value with the accident occurrence probability includes: multiplying the accident severity value with the accident occurrence probability to obtain a comprehensive risk value;

[0022] A threshold range for dividing risk levels is set, the comprehensive risk value is classified according to the threshold range for dividing risk levels, and finally a risk level assessment result is obtained.

[0023] Furthermore, the convolutional neural network model includes: a one-dimensional convolution layer, a batch normalization layer, a maximum pooling layer, a fully connected layer and a soft maximum output layer.

[0024] Furthermore, it also includes: displaying the risk level assessment result to the user through a user interface.

[0025] In general, the above technical solution conceived by the present invention has the following beneficial effects compared with the prior art:

[0026] Through the technical solution of the present invention, the safety risks of bridge construction can be evaluated, thereby improving construction safety and improving construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 is a flow chart of the method of embodiment 1 of the present invention;

[0028] Figure 2 is a structural diagram of a system in Embodiment 2 of the present invention;

[0029] Figure 3 This is an output result interface diagram of Example 1 of the present invention;

[0030] Figure 4 This is a risk level assessment effect diagram of Example 1 of the present invention. DETAILED DESCRIPTION

[0031] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0032] The method provided by the present invention can be implemented in the following terminal environment, and the terminal may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method described in the following embodiment.

[0033] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts in the entire terminal, and executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and calling data stored in the storage medium.

[0034] The storage medium may include a random access memory (RAM) or a read-only memory (ROM). The storage medium may be used to store instructions, programs, codes, code sets or instructions.

[0035] The display screen is used to display the user interface of each application.

[0036] In addition, those skilled in the art can understand that the structure of the above terminal does not constitute a limitation on the terminal, and the terminal may include more or fewer components, or combine certain components, or arrange the components differently. For example, the terminal also includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, and a power supply, which will not be described in detail here.

[0037] Example 1

[0038] like Figure 1 As shown, an embodiment of the present invention provides a bridge construction safety risk assessment method based on a convolutional neural network, comprising:

[0039] Step 101, obtaining a variety of safety factor data affecting bridge construction safety, the safety factor data including: environmental factor data, equipment factor data and material factor data;

[0040] Specifically, environmental factor data, equipment factor data, and material factor data are classified as follows:

[0041] Environmental factor data E: severe natural climate, complex geological conditions, poor (dangerous) working space, etc., for example, wind force level, temperature or visibility during construction.

[0042] Equipment factor data M: equipment not maintained (checked), equipment defects, etc. For example, the life of construction equipment (excavators, bulldozers, etc.), or the time interval when the construction equipment exceeds the prescribed maintenance time, etc.;

[0043] Material factor data Q: material strength, durability and other parameters

[0044] The user can input the value of each factor through the graphical user interface (GUI), and the input value range is 0 to 1. The input data will be transmitted to the subsequent step through the system interface for processing. This step supports data collection through manual entry or automatic sensors.

[0045] Step 102, preprocessing the safety factor data, inputting the preprocessed safety factor data into a convolutional neural network model, and predicting the accident severity value;

[0046] Specifically, after the data is input, each safety factor data will be automatically processed by unique hot encoding. The specific implementation steps are as follows:

[0047] For each input factor, if its value is greater than or equal to 0.5, it is quantized to 1; otherwise, it is quantized to 0.

[0048] The processed data forms a multi-dimensional one-hot encoded vector, which indicates the presence or absence of each factor. The one-hot encoded data will be passed to the accident severity prediction module for further analysis.

[0049] This embodiment predicts the accident severity value through a convolutional neural network model, specifically including:

[0050] Based on the convolutional neural network (CNN) model, the accident severity (C) is predicted for the input one-hot encoded data. The implementation steps are as follows:

[0051] Data preprocessing: The input one-hot encoded data is standardized to meet the input requirements of the neural network.

[0052] Model structure: The first layer: a one-dimensional convolutional layer to extract the features of the input data;

[0053] The second layer: pooling layer, which reduces the data dimension through the maximum pooling operation while retaining important features;

[0054] The third layer: convolutional layer, used to further extract features in depth;

[0055] The fourth layer: fully connected layer, which integrates all features and performs classification prediction;

[0056] Output layer: The output of the model is the probability of four types of accident severity, corresponding to 1 (not serious), 2 (more serious), 5 (serious) and 10 (especially serious). The category corresponding to the maximum probability of the output is taken as the severity of the accident.

[0057] Step 103, inputting the safety factor data into the SD-MOP model to calculate the probability of an accident occurring during bridge construction;

[0058] Specifically, in the step, the SD-MOP model is used to predict the probability of an accident (L). The implementation steps are as follows:

[0059] According to the multi-dimensional one-hot encoding vector, the average values ​​of the environmental factor data, the equipment factor data, and the material factor data are calculated respectively.

[0060] Using the predefined accident type coefficient (δ i ), for example, the coefficient of falling from a height is 0.15, the coefficient of object impact is 0.20, etc. The average value of each factor category is multiplied by the weight coefficient of the accident type, and then the weighted sum is performed to obtain the probability L of the accident.

[0061] Specifically, according to the weight value of the accident type (δ i ) and the average value of each factor, and use the weighted sum method to calculate the probability L of the accident.

[0062]

[0063] The process involves weighted accumulation of factors and multiplication by coefficients to obtain the final probability value.

[0064] δ i : Predefined accident type weight values, such as high-altitude fall, object impact, etc. Each accident type has a corresponding weight value.

[0065] s i : The calculated average values ​​of various factors, such as the average value of environmental factors, the average value of material factors, etc.

[0066] The final L value is normalized and output in the range of 0 to 10 with one decimal place retained.

[0067] Step 104, combining the accident severity value with the accident occurrence probability, and outputting a comprehensive risk level assessment result, such as Figure 3 and 4 shown.

[0068] Specifically, this step combines the severity of the accident (C) and the probability of the accident (L) to calculate the comprehensive risk value (D). The implementation steps are as follows:

[0069] Risk level calculation: Calculate the comprehensive risk value according to the formula D = L*C. The larger the D value, the higher the risk of the accident.

[0070] Risk level classification: Based on the calculated D value, the risk level assessment results are divided into four categories:

[0071] The D value is between 0-5, and the risk level assessment result is "small".

[0072] The D value is between 5-20, and the risk level assessment result is "average".

[0073] The D value is between 20-55, and the risk level assessment result is "high".

[0074] The D value is between 55-100, and the risk level assessment result is "significant".

[0075] Classification of accident probability and severity:

[0076] Classification of accident probability, based on the L value, the probability of an accident is divided into five categories:

[0077] 0-1: Very low, very unlikely to happen

[0078] 1-3: Low, unlikely to occur

[0079] 3-6: Moderate, may occur

[0080] 6-9: High, prone to occur

[0081] 9-10: Very high, very likely to occur

[0082] Accident severity classification, based on the C value, the severity of the accident is divided into four categories:

[0083] 1: Not serious

[0084] 2: More serious

[0085] 5: Severe

[0086] 10: Very severe

[0087] The prediction result interface displays the accident probability L, accident severity C, comprehensive risk value D and their corresponding level classifications to help managers quickly understand the risk status.

[0088] In order to make the bridge construction safety risk assessment more accurate, a construction risk assessment model is set up to calculate a new comprehensive risk value D in conjunction with the accident probability L and the accident severity C. The construction risk assessment model is specifically as follows:

[0089]

[0090] Among them, R is the bridge construction safety risk assessment index, φ(E) is the influence function of environmental factor data E on bridge construction safety, α4 is the weight of environmental factor data, β4 is the adjustment factor of environmental factor data, ψ(M) is the influence function of equipment factor data M on bridge construction safety, γ4 is the weight of equipment factor data, χ(Q) is the influence function of material factor data Q on bridge construction safety, and δ4 is the weight of material factor data.

[0091] The influence function φ(E) of environmental factor data E on bridge construction safety is:

[0092]

[0093] Among them, E0 is the environmental reference value during bridge construction, E c is the environmental critical value of safety risk during bridge construction, λ1 is the first adjustment factor of environmental impact, γ1 is the second adjustment factor of environmental impact, η1 is the third adjustment factor of environmental impact, λ2 is the fourth adjustment factor of environmental impact, and κ1 is the fifth adjustment factor of environmental impact.

[0094] The influence function ψ(M) of equipment factor data M on bridge construction safety is:

[0095]

[0096] Among them, μ1 is the first adjustment factor of the device impact, δ1 is the second adjustment factor of the device impact, θ1 is the third adjustment factor of the device impact, λ3 is the fourth adjustment factor of the device impact, and μ4 is the fifth adjustment factor of the device impact.

[0097] The influence function χ(Q) of material factor data Q on bridge construction safety is:

[0098]

[0099] Among them, k1 is the first adjustment factor of material influence, β1 is the second adjustment factor of material influence, k2 is the third adjustment factor of material influence, γ2 is the fourth adjustment factor of material influence, Q min is the material critical value of safety risk during bridge construction, δ2 is the fifth adjustment factor of material influence, and k3 is the sixth adjustment factor of material influence.

[0100] New D=L*C*R.

[0101] Example 2

[0102] like Figure 2 As shown, the embodiment of the present invention also provides a bridge construction safety risk assessment system based on a convolutional neural network, comprising:

[0103] An input module is used to obtain various safety factor data that affect the safety of bridge construction, wherein the safety factor data includes: environmental factor data, equipment factor data and material factor data;

[0104] A convolutional neural network module, used to pre-process the safety factor data, input the pre-processed safety factor data into a convolutional neural network model, and predict the accident severity value;

[0105] Specifically, preprocessing the safety factor data includes: quantizing the safety factor data and converting it into a one-hot encoding.

[0106] Specifically, the convolutional neural network model includes: a one-dimensional convolution layer, a batch normalization layer, a maximum pooling layer, a fully connected layer and a soft maximum output layer.

[0107] An accident probability prediction module is used to input the safety factor data into the SD-MOP model to calculate the accident probability of bridge construction;

[0108] The risk level calculation module is used to combine the accident severity value with the accident occurrence probability and output a comprehensive risk level assessment result.

[0109] Specifically, combining the accident severity value with the accident occurrence probability includes: multiplying the accident severity value with the accident occurrence probability to obtain a comprehensive risk value;

[0110] A threshold range for dividing risk levels is set, the comprehensive risk value is classified according to the threshold range for dividing risk levels, and finally a risk level assessment result is obtained.

[0111] Specifically, it also includes: displaying the risk level assessment result to the user through a user interface.

[0112] Example 3

[0113] An embodiment of the present invention also proposes a storage medium storing a plurality of instructions, wherein the instructions are used to implement the bridge construction safety risk assessment method based on a convolutional neural network.

[0114] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0115] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: Step 101, obtaining a plurality of safety factor data affecting the safety of bridge construction, the safety factor data including: environmental factor data, equipment factor data and material factor data;

[0116] Step 102, preprocessing the safety factor data, inputting the preprocessed safety factor data into a convolutional neural network model, and predicting the accident severity value;

[0117] Specifically, preprocessing the safety factor data includes: quantizing the safety factor data and converting it into a one-hot encoding.

[0118] Specifically, the convolutional neural network model includes: a one-dimensional convolution layer, a batch normalization layer, a maximum pooling layer, a fully connected layer and a soft maximum output layer.

[0119] Step 103, inputting the safety factor data into the SD-MOP model to calculate the probability of an accident occurring during bridge construction;

[0120] Step 104: Combine the accident severity value with the accident occurrence probability to output a comprehensive risk level assessment result.

[0121] Specifically, combining the accident severity value with the accident occurrence probability includes: multiplying the accident severity value with the accident occurrence probability to obtain a comprehensive risk value;

[0122] A threshold range for dividing risk levels is set, the comprehensive risk value is classified according to the threshold range for dividing risk levels, and finally a risk level assessment result is obtained.

[0123] Specifically, it also includes: displaying the risk level assessment result to the user through a user interface.

[0124] Example 4

[0125] An embodiment of the present invention also proposes an electronic device, including a processor and a storage medium connected to the processor, wherein the storage medium stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the bridge construction safety risk assessment method based on a convolutional neural network.

[0126] Specifically, the electronic device of this embodiment may be a computer terminal, and the computer terminal may include: one or more processors, and a storage medium.

[0127] Among them, the storage medium can be used to store software programs and modules, such as a bridge construction safety risk assessment method based on a convolutional neural network in an embodiment of the present invention, and the corresponding program instructions / modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, realizing the above-mentioned bridge construction safety risk assessment method based on a convolutional neural network. The storage medium may include a high-speed random storage medium, and may also include a non-volatile storage medium, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include a storage medium remotely arranged relative to the processor, and these remote storage media may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0128] The processor can call the information and application program stored in the storage medium through the transmission system to perform the following steps: Step 101, obtaining multiple safety factor data affecting the safety of bridge construction, the safety factor data including: environmental factor data, equipment factor data and material factor data;

[0129] Step 102, preprocessing the safety factor data, inputting the preprocessed safety factor data into a convolutional neural network model, and predicting the accident severity value;

[0130] Specifically, preprocessing the safety factor data includes: quantizing the safety factor data and converting it into a one-hot encoding.

[0131] Specifically, the convolutional neural network model includes: a one-dimensional convolution layer, a batch normalization layer, a maximum pooling layer, a fully connected layer and a soft maximum output layer.

[0132] Step 103, inputting the safety factor data into the SD-MOP model to calculate the probability of an accident occurring during bridge construction;

[0133] Step 104: Combine the accident severity value with the accident occurrence probability to output a comprehensive risk level assessment result.

[0134] Specifically, combining the accident severity value with the accident occurrence probability includes: multiplying the accident severity value with the accident occurrence probability to obtain a comprehensive risk value;

[0135] A threshold range for dividing risk levels is set, the comprehensive risk value is classified according to the threshold range for dividing risk levels, and finally a risk level assessment result is obtained.

[0136] Specifically, it also includes: displaying the risk level assessment result to the user through a user interface.

[0137] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0138] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0139] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0140] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0141] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0142] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0143] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A bridge construction safety risk assessment method based on convolutional neural network, characterized in that: include: Acquire various safety factor data that affect bridge construction safety, the safety factor data including: environmental factor data, equipment factor data and material factor data; Preprocessing the safety factor data, inputting the preprocessed safety factor data into a convolutional neural network model, and predicting an accident severity value; Inputting the safety factor data into the SD-MOP model to calculate the probability of accidents occurring in bridge construction; The accident severity value is combined with the accident occurrence probability to output a comprehensive risk level assessment result.

2. A bridge construction safety risk assessment method based on convolutional neural network as claimed in claim 1, characterized in that: Preprocessing the safety factor data includes: quantizing the safety factor data and converting the data into a one-hot encoding.

3. A bridge construction safety risk assessment method based on convolutional neural network as claimed in claim 1, characterized in that: Combining the accident severity value with the accident occurrence probability includes: multiplying the accident severity value with the accident occurrence probability to obtain a comprehensive risk value; A threshold range for dividing risk levels is set, the comprehensive risk value is classified according to the threshold range for dividing risk levels, and finally a risk level assessment result is obtained.

4. A bridge construction safety risk assessment method based on convolutional neural network as claimed in claim 1, characterized in that: The convolutional neural network model includes: a one-dimensional convolution layer, a batch normalization layer, a maximum pooling layer, a fully connected layer and a soft maximum output layer.

5. A bridge construction safety risk assessment method based on convolutional neural network as claimed in claim 1, characterized in that: Also includes: The risk level assessment result is displayed to the user through a user interface.

6. A bridge construction safety risk assessment system based on convolutional neural network, characterized in that: include: An input module is used to obtain various safety factor data that affect the safety of bridge construction, wherein the safety factor data includes: environmental factor data, equipment factor data and material factor data; A convolutional neural network module, used to pre-process the safety factor data, input the pre-processed safety factor data into a convolutional neural network model, and predict the accident severity value; An accident probability prediction module is used to input the safety factor data into the SD-MOP model to calculate the accident probability of bridge construction; The risk level calculation module is used to combine the accident severity value with the accident occurrence probability and output a comprehensive risk level assessment result.

7. A bridge construction safety risk assessment system based on convolutional neural network as claimed in claim 6, characterized in that: Preprocessing the safety factor data includes: quantizing the safety factor data and converting the data into a one-hot encoding.

8. A bridge construction safety risk assessment system based on convolutional neural network as claimed in claim 6, characterized in that: Combining the accident severity value with the accident occurrence probability includes: multiplying the accident severity value with the accident occurrence probability to obtain a comprehensive risk value; A threshold range for dividing risk levels is set, the comprehensive risk value is classified according to the threshold range for dividing risk levels, and finally a risk level assessment result is obtained.

9. A bridge construction safety risk assessment system based on convolutional neural network as claimed in claim 6, characterized in that: The convolutional neural network model includes: a one-dimensional convolution layer, a batch normalization layer, a maximum pooling layer, a fully connected layer and a soft maximum output layer.

10. A bridge construction safety risk assessment system based on convolutional neural network as claimed in claim 6, characterized in that: Also includes: The risk level assessment result is displayed to the user through a user interface.

Images