A decision tree-based array charge damage effectiveness evaluation method and system

Abstract

This invention discloses a method and system for evaluating the damage effectiveness of array charges based on decision trees, comprising: step S1: constructing a damage assessment dataset; step S2: constructing and training a decision tree model; step S3: collecting underwater obstacle information; step S4: collecting array charge information; step S5: calculating the damage level of underwater obstacles; step S6: evaluating the damage effectiveness of the array charge; the evaluation system includes: a damage database module, an underwater obstacle identification and localization module, an array charge detonation point identification and localization module, and a damage effectiveness evaluation module. This method constructs a damage assessment dataset of array charges and underwater obstacles, selects the proportional distance between the array charge detonation point and the underwater obstacle as a feature to divide the feature space, recursively constructs a decision tree and trains the model, and after generalization, can timely and accurately evaluate the damage effectiveness of the array charge based on the input distance matrix. The evaluation calculation is time-consuming, the evaluation model is highly accurate, and there is no danger.

Description

Technical Field

[0001] This invention relates to the field of array charge damage performance evaluation, and in particular to a decision tree-based method and system for array charge damage performance evaluation. Background Technology

[0002] When breaching underwater obstacles submerged in water, warheads mostly employ array charges, using a simultaneous detonation method to damage the target through the superposition of shock waves in the water. After the warhead explodes, its destructive effectiveness should be assessed promptly. Currently, there are three main technical approaches to assessing the destructive effectiveness of array charges.

[0003] One approach is the theoretical analytical method, which uses explosion mechanics to calculate the initial parameters, propagation process, and superposition laws of underwater shock waves. It also requires the use of materials mechanics and wave theory to calculate the impact of underwater shock waves on obstacles and targets, and their damage effects. While the theoretical analytical method has a rigorous theoretical foundation, its results are derived under numerous idealized assumptions. Due to the complexity of the underwater environment and the multitude of influencing factors, the theoretical calculations often deviate significantly from reality, resulting in an unsatisfactory assessment of damage effectiveness.

[0004] The second method is the finite element method, which uses mature finite element software to build a computational model and conduct numerical simulations. Because the explosion of an array-charged explosive in water involves a wide range of factors, the number of elements in the finite element model is large, requiring significant computational resources and resulting in lengthy calculations, making it difficult to meet the requirements for timely evaluation. Furthermore, the finite element method is highly dependent on the model; an inaccurate model can cause huge errors, or even completely wrong results.

[0005] Third is the basic test method, which uses live-fire tests to obtain damage data. This method is the most realistic and reliable. However, array charge testing is a huge project that requires a lot of manpower, material resources, and financial resources, and it is also highly dangerous. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for evaluating the damage effectiveness of arrayed explosive charges based on decision trees.

[0007] The innovative aspect of this invention lies in: constructing a damage assessment dataset for arrayed explosive charges and underwater obstacles; selecting the proportional distance between the explosive charge detonation point and the underwater obstacle as a feature to divide the feature space; recursively constructing a decision tree and training the model; and after generalization, being able to timely and accurately assess the damage effectiveness of the arrayed explosive charge based on the input distance matrix. The assessment calculation is time-consuming, the assessment model is highly accurate, and there is no danger.

[0008] To achieve the above-mentioned objectives, the technical solution of this invention is as follows:

[0009] The first aspect of this invention discloses a method for evaluating the damage effectiveness of arrayed explosive charges based on decision trees, comprising the following steps:

[0010] Step S1: Construct a damage assessment dataset; the damage assessment dataset includes underwater obstacle types, underwater obstacle structural parameters, underwater obstacle locations, array charge quantity, array charge location, obstacle damage level classification data, underwater obstacle-to-array charge ratio distance matrix, ammunition type, and environmental information;

[0011] Step S2: Construct and train the decision tree model; using the damage assessment dataset constructed in Step S1, after data cleaning and regularization, divide it into training and test sets, select the proportional distance between the array charge and underwater obstacles as the feature to divide the feature space, recursively construct and train the decision tree model using the training set; use the test set to test the accuracy of the decision tree model, and select the optimal damage assessment decision tree model accordingly.

[0012] Step S3: Collect underwater obstacle information; use an underwater obstacle detection device to collect information on the location distribution and shape structure of underwater obstacles, autonomously identify the types of underwater obstacles, and calculate the coordinate information of underwater obstacles;

[0013] Step S4: Collect array charge information; using the array charge detection device, collect information on the array charge detonation process, autonomously identify the detonation point location, and calculate the coordinate information of each detonation point of the array charge.

[0014] Step S5: Calculate the damage level of underwater obstacles; using the coordinate information of underwater obstacles and the coordinate information of array charge detonation points, generate a distance matrix between each underwater obstacle and all array charge detonation points, call the trained decision tree model, and calculate the damage level of each underwater obstacle;

[0015] Step S6: Assessment of the damage effectiveness of the array charge; assess the damage effectiveness of the array charge using the calculation results of the damage level of underwater obstacles; or predict the damage effectiveness of the array charge using the simulated coordinate information of underwater obstacles and the coordinate information of the array charge detonation point.

[0016] According to the method of the first aspect of the present invention, the ratio of the training set to the test set data in step S2 is 1:9 to 3:7.

[0017] According to the method of the first aspect of the present invention, the proportional distance in step S2 is defined as Where C represents the charge quantity and r represents the distance from the center of the charge to the target.

[0018] According to the method of the first aspect of the present invention, the underwater obstacle detection device in step S3 includes an underwater obstacle unmanned platform, an underwater obstacle image acquisition device, and an underwater obstacle unmanned platform controller; the underwater obstacle unmanned platform can acquire its own position in real time, the underwater obstacle image acquisition device is installed on the underwater obstacle unmanned platform and is used to capture images or videos of the underwater obstacle area; the underwater obstacle unmanned platform controller is used to control the underwater obstacle unmanned platform.

[0019] According to the method of the first aspect of the present invention, the array charge detection device in step S4 includes an array charge unmanned platform, an array charge image acquisition device, and an array charge unmanned platform controller; the array charge unmanned platform can acquire its own position in real time, the array charge image acquisition device is installed on the array charge unmanned platform and is used to capture images or videos of the moment the array charge explodes; the array charge unmanned platform controller is used to control the array charge unmanned platform.

[0020] A second aspect of this invention discloses a decision tree-based array charge damage performance evaluation system, comprising:

[0021] The damage database module is configured to build a damage assessment dataset and use the damage assessment dataset to build and train a decision tree model.

[0022] The underwater obstacle identification and positioning module is configured to use the underwater obstacle location information collected by the underwater obstacle detection device to achieve autonomous identification of underwater obstacles and calculation of underwater obstacle coordinate information;

[0023] The array charge detonation point identification and positioning module is configured to use the array charge detection device to collect information on the array charge detonation process to achieve autonomous identification of the array charge detonation point and calculation of the array charge detonation point coordinate information.

[0024] The damage effectiveness assessment module is configured to assess the damage effectiveness of the array charge using a damage assessment decision tree model, underwater obstacle coordinate information, and array charge detonation point coordinate information; or to predict the damage effectiveness of the array charge using simulated underwater obstacle coordinate information and array charge detonation point coordinate information.

[0025] A third aspect of the present invention provides an electronic device, the electronic device comprising a memory and a processor, the memory having a computer program, and the processor executing the computer program to implement the steps of the array charge damage performance evaluation method based on decision tree according to any one of the first aspects of the present invention.

[0026] The fourth aspect of the present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a decision tree-based array charge damage performance evaluation method according to any one of the first aspects of the present invention.

[0027] The beneficial effects of this invention are:

[0028] By training a decision tree model and generalizing it to the damage performance evaluation system, the damage level can be calculated using the proportional distance matrix between obstacles and the detonation points of array charges. This method is convenient, efficient, time-saving, reliable, and easy to iterate and optimize, greatly improving the effectiveness and accuracy of array charge damage performance evaluation. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the invention.

[0030] Figure 2 A structural diagram of an electronic device according to an embodiment of the present invention.

[0031] In the diagram: 10 is an underwater obstacle detection device, 11 is an unmanned underwater obstacle platform, 12 is an underwater obstacle image acquisition device, 13 is an unmanned underwater obstacle platform controller, 20 is an array charge detection device, 21 is an unmanned array charge platform, 22 is an array charge image acquisition device, 23 is an unmanned array charge platform controller, 30 is a damage database module, 40 is an underwater obstacle identification and positioning module, 50 is an array charge detonation point identification and positioning module, and 60 is a damage effectiveness assessment module. Detailed Implementation

[0032] The technical solutions in the embodiments of the present invention will now be clearly and completely described in conjunction with the accompanying drawings.

[0033] The first aspect of this invention discloses a method for evaluating the damage effectiveness of arrayed explosive charges based on decision trees, comprising the following steps:

[0034] Step S1: Construct a damage assessment dataset; the damage assessment dataset includes underwater obstacle types, underwater obstacle structural parameters, underwater obstacle locations, array charge quantity, array charge location, obstacle damage level, underwater obstacle to array charge ratio distance matrix, ammunition type, and environmental information;

[0035] Step S2: Construct and train a decision tree model; using the damage assessment dataset constructed in Step S1, after data cleaning and regularization, divide it into training and test sets, select the proportional distance between the array charge and underwater obstacles as the feature to divide the feature space, recursively construct and train the decision tree model using the training set; use the test set to test the accuracy of the decision tree model, and select the optimal damage assessment decision tree model accordingly; solve problems such as data duplication, missing data, and erroneous data through data cleaning, and use regularization to make the expected value of the data 0 and the variance 1, reducing the impact of outliers on the algorithm and improving the accuracy of the data;

[0036] Step S3: Collect underwater obstacle location information; use the underwater obstacle detection device 10 to collect underwater obstacle information, autonomously identify the type of underwater obstacle, and calculate the coordinate information of the underwater obstacle;

[0037] Step S4: Collect array charge position information; use array charge detection device 20 to collect array charge explosion process information, autonomously identify the explosion point, and calculate the array charge explosion point coordinate information;

[0038] Step S5: Calculate the damage level of underwater obstacles; using the coordinate information of underwater obstacles and the coordinate information of array charge detonation points, generate a distance matrix between each underwater obstacle and all array charge detonation points, call the trained decision tree model, and calculate the damage level of each underwater obstacle;

[0039] Step S6: Evaluation of the damage effectiveness of the array charge; using the calculation results of the damage level of underwater obstacles, evaluate the damage effectiveness of the array charge; at the same time, using the simulated underwater obstacle coordinate information and array charge detonation point coordinate information, predict the damage effectiveness of the array charge.

[0040] According to the method of the first aspect of the present invention, the ratio of the training set to the test set data in step S2 is 1:9 to 3:7.

[0041] According to the method of the first aspect of the present invention, the proportional distance in step S2 is defined as

[0042] Where C represents the charge quantity, and r represents the distance from the center of the charge to the target. Of course, based on the explosion similarity rate, parameters describing the explosive power, such as peak overpressure of the shock wave, duration of barotropic action, impulse, and energy density, can all be expressed as functions of the proportional distance.

[0043] According to the method of the first aspect of the present invention, in step S3, the underwater obstacle detection device 10 includes an underwater obstacle unmanned platform 11, an underwater obstacle image acquisition device 12, and an underwater obstacle unmanned platform controller 13; the underwater obstacle image acquisition device 12 is installed on the underwater obstacle unmanned platform 11 and is used to capture images or videos of the underwater obstacle area; the underwater obstacle unmanned platform controller 13 is used to control the underwater obstacle unmanned platform 11.

[0044] According to the method of the first aspect of the present invention, in step S4, the array charge detection device 20 includes an array charge unmanned platform 21, an array charge image acquisition device 22, and an array charge unmanned platform controller 23; the array charge image acquisition device 22 is installed on the array charge unmanned platform 21 and is used to capture images or videos of the moment the array charge explodes; the array charge unmanned platform controller 23 is used to control the array charge unmanned platform 21.

[0045] A second aspect of this invention discloses a decision tree-based array charge damage performance evaluation system, comprising:

[0046] Damage database module 30 is configured to construct a damage assessment dataset and use the damage assessment dataset to construct and train a decision tree model;

[0047] The underwater obstacle recognition and positioning module 40 is configured to use the underwater obstacle location information collected by the underwater obstacle detection device 10 to realize the autonomous recognition of underwater obstacles and calculate the coordinate information of underwater obstacles; the autonomous recognition of underwater obstacles is based on underwater obstacle image or video data and adopts mature recognition algorithms; at the same time, the obstacle recognition is transferred to train the underwater obstacle recognition model.

[0048] The array charge detonation point identification and positioning module 50 is configured to use the array charge position information collected by the array charge detection device 20 to achieve autonomous identification of the array charge detonation point and calculate the array charge detonation point coordinate information; the autonomous identification of the array charge detonation point is based on array charge detonation point image or video data, and a mature identification algorithm is adopted; at the same time, transfer learning is performed on the array charge detonation point identification to train the detonation point identification model;

[0049] The damage effectiveness assessment module 60 is configured to assess the damage effectiveness of the array charge using a damage assessment decision tree model, underwater obstacle coordinate information, and array charge detonation point coordinate information; or to predict the damage effectiveness of the array charge using simulated underwater obstacle information and array charge detonation point information.

[0050] Furthermore, the simulation generation method is generally either computer-generated randomly or manually input.

[0051] A third aspect of the present invention provides an electronic device, the electronic device comprising a memory and a processor, the memory having a computer program, and the processor executing the computer program to implement the steps of the array charge damage performance evaluation method based on decision tree according to any one of the first aspects of the present invention.

[0052] Figure 2 This is a structural diagram of an electronic device according to an embodiment of the present invention, such as... Figure 2 As shown, the electronic device includes a processor, memory, communication interface, display screen, and input device. The processor provides computing and control capabilities. The memory includes internal memory and a non-volatile storage medium. The non-volatile storage medium stores the operating system and computer programs. The computer programs include at least this decision tree-based array charge damage performance evaluation system and other software required to support the system's operation. The communication interface, display screen, and input device are existing technologies and will not be described further here.

[0053] The fourth aspect of the present invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a decision tree-based array charge damage performance evaluation method according to any one of the first aspects of the present invention.

[0054] The described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

Claims

1. A method for evaluating the damage effectiveness of arrayed explosive charges based on decision trees, characterized in that: Includes the following steps: Step S1: Construct a damage assessment dataset; the damage assessment dataset includes underwater obstacle types, underwater obstacle structural parameters, underwater obstacle location distribution, array charge quantity, array charge location distribution, obstacle damage level classification data, underwater obstacle to array charge ratio distance matrix, ammunition type and environmental information; Step S2: Construct and train the decision tree model; using the damage assessment dataset constructed in Step S1, after data cleaning and regularization, divide it into training and test sets, select the proportional distance between the array charge and underwater obstacles as the feature to divide the feature space, recursively construct and train the decision tree model using the training set; use the test set to test the accuracy of the decision tree model, and select the optimal damage assessment decision tree model accordingly. Step S3: Collect underwater obstacle information; The underwater obstacle detection device is used to collect information on the location distribution and shape structure of underwater obstacles, autonomously identify the types of underwater obstacles, and calculate the coordinate information of underwater obstacles. Step S4: Collect array charge information; using the array charge detection device, collect information on the array charge explosion process, autonomously identify the location of the explosion point, and calculate the coordinate information of each explosion point of the array charge. Step S5: Calculate the damage level of underwater obstacles; Using the coordinate information of underwater obstacles and the coordinate information of array charge detonation points, a distance matrix between each underwater obstacle and all array charge detonation points is generated. The trained decision tree model is then called to calculate the damage level of each underwater obstacle. Step S6: Evaluation of the damage effectiveness of the array charge; The damage effectiveness of the array charge is evaluated by using the damage level calculation results of underwater obstacles; at the same time, the damage effectiveness of the array charge is predicted by using the simulated underwater obstacle coordinate information and array charge detonation point coordinate information.

2. The method for evaluating the damage effectiveness of arrayed explosive charges based on decision trees according to claim 1, characterized in that: In step S2, the ratio of training set to test set data is 1:9 to 3:

7.

3. The method for evaluating the damage effectiveness of arrayed explosive charges based on decision trees according to claim 1, characterized in that: In step S2, the proportional distance is defined as ř=C. 1 / 3 / r, where C represents the charge amount and r represents the distance from the center of the charge to the target.

4. The method for evaluating the damage effectiveness of arrayed explosive charges based on decision trees according to claim 1, characterized in that: The underwater obstacle detection device in step S3 includes an underwater obstacle unmanned platform, an underwater obstacle image acquisition device, and an underwater obstacle unmanned platform controller; the underwater obstacle image acquisition device is installed on the underwater obstacle unmanned platform and is used to capture images or videos of the underwater obstacle area; the underwater obstacle unmanned platform controller is used to control the underwater obstacle unmanned platform.

5. The method for evaluating the damage effectiveness of arrayed explosive charges based on decision trees according to claim 1, characterized in that: In step S4, the array charge detection device includes an array charge unmanned platform, an array charge image acquisition device, and an array charge unmanned platform controller. The array charge image acquisition device is installed on the array charge unmanned platform and is used to capture images or videos of the array charge detonation process. The array charge unmanned platform controller is used to control the array charge unmanned platform.

6. The method for evaluating the damage effectiveness of arrayed explosive charges based on decision trees according to claim 1, characterized in that: The method is implemented based on a system, which includes: The damage database module is configured to build a damage assessment dataset and use the damage assessment dataset to build and train a decision tree model. The underwater obstacle identification and positioning module is configured to use the shape structure and location distribution information of underwater obstacles collected by the underwater obstacle detection device to achieve autonomous identification of underwater obstacle types and calculation of underwater obstacle coordinate information; The array charge detonation point identification and positioning module is configured to use the array charge detection device to collect information on the array charge detonation process to achieve autonomous identification of the array charge detonation point and calculation of the array charge detonation point coordinate information. The damage effectiveness assessment module is configured to calculate the damage level of obstacles and assess the damage effectiveness of array charges by using a damage assessment decision tree model, underwater obstacle coordinate information, and array charge detonation point coordinate information; or to predict the damage effectiveness of array charges by using simulated underwater obstacle coordinate information and array charge detonation point coordinate information.

7. An electronic device, characterized in that: The electronic device includes a memory and a processor. The memory contains a computer program. When the processor executes the computer program, it implements the steps in the method for evaluating the damage effectiveness of array charges based on a decision tree, as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the decision tree-based array charge damage performance evaluation method according to any one of claims 1-5.

Images