A multi-sensor data fusion method for forest fire prevention based on vulture search optimization

Abstract

The present invention discloses a multi-sensor data fusion method for forest fire prevention based on vulture search optimization, comprising the following steps: S10, using multiple sensors configured in fire point detection equipment to collect multi-dimensional data; S20, transmitting the collected multi-dimensional data to a forest fire prevention multi-sensor data fusion model based on vulture search optimization for data feature fusion, resulting in more effective decision-making data for forest fire point detection. This method optimizes the IVD characteristics of forest fire prevention data and designs an optimization algorithm based on the vulture search and foraging principle. By optimizing the search and fitting of individual sensor data features and optimizing the search and fusion of multiple sensor combinations, the method significantly improves fire point detection accuracy and shortens algorithm execution time.

Description

Technical Field

[0001] The invention belongs to the technical field of forest fire monitoring, and in particular relates to a forest fire prevention multi-sensor data fusion method based on vulture search optimization. Background Art

[0002] Forest fires are sudden, destructive, and difficult to manage and respond to natural disasters. They not only destroy trees but also damage the forest ecosystem, causing air pollution and property damage. Therefore, early warning of forest and grassland fires is crucial.

[0003] Currently, forest fire prevention primarily relies on drone technology, satellite and remote video surveillance, ground patrols, and manual observation, all of which are generally costly. The rapid development of the electronics industry has spurred the development of various sensors. Forest fire monitoring can utilize a variety of sensors, such as smoke and temperature sensors. Using a single sensor alone can result in false detections, so simultaneous monitoring using multiple sensors is often recommended. However, the diverse data generated by these sensors often cannot be effectively integrated to facilitate decision-making.

[0004] Common methods for multi-sensor data fusion include Kalman fusion and first-order low-pass filtering. However, these fusion methods are limited to a limited number of sensors and operate in a single manner. Multi-layer perceptrons (MLPs) are commonly used for data fusion using neural networks. However, as data volume increases, the number of parameters in the MLP model increases dramatically, leading to increased computational costs. Furthermore, the MLP model is a black box, making it difficult to explain how it performs data fusion and decision-making. Summary of the Invention

[0005] In order to solve the above problems, the present invention proposes a forest fire prevention multi-sensor data fusion method based on vulture search optimization. According to the IVD characteristics of forest fire prevention data, an optimization algorithm based on the vulture search and foraging principle is designed for optimization. Through the search optimization fitting of the data characteristics of a single sensor and the search optimization fusion of multiple sensor combinations, the fire point detection accuracy can be greatly improved and the algorithm time can be shortened.

[0006] To achieve the above object, the technical solution adopted by the present invention is: a forest defense multi-sensor data fusion method based on vulture search optimization, comprising the steps of:

[0007] S10, using multiple sensors configured in the fire point detection equipment to collect multi-dimensional data;

[0008] S20, transmitting the collected multi-dimensional data to a forest fire prevention multi-sensor data fusion model based on vulture search optimization to fuse the data features and fuse them into decision data for forest fire point detection;

[0009] The sensor data is input into the forest fire prevention multi-sensor data fusion model based on vulture search optimization. The vulture search algorithm is used to search and optimize the input data of multiple sensors, fit their interval value data characteristics, seek the best sensor data combination features, and fuse the combination features using machine learning methods. After multiple data refinement and fusion, the measurement data of multiple sensors are finally combined and fused into one data as decision support data for forest fire detection.

[0010] Furthermore, the data collected by the multiple sensors are multi-dimensional data, including temperature, humidity and air particle size, and these data have interval value IVD data characteristics.

[0011] Furthermore, a sample data set is constructed, including data from multiple sensors under different environmental conditions, as well as corresponding descriptions of forest fire conditions.

[0012] Furthermore, the constructed sample data set includes: data under normal environment, dry environment and high temperature environment, or data under normal conditions, suspected fire points, low concentration smoke and high concentration smoke conditions; and the corresponding fused data can effectively support fire point detection.

[0013] Furthermore, the forest defense multi-sensor data fusion model based on vulture search optimization includes:

[0014] Aiming at the IVD characteristics of forest fire prevention data, an optimization algorithm based on the vulture's search and foraging principle was established;

[0015] For the optimization algorithm based on the vulture foraging principle, the data change characteristics of each sensor are fitted by optimizing the search of each sensor data; then the combination of multiple sensor fitting data is optimized and searched to find the optimal combination of multiple sensor feature data for fire point detection.

[0016] Furthermore, a variety of machine learning methods are used to test the decision support data output by the optimization algorithm based on the vulture foraging principle, to test the accuracy of the fusion value output by the algorithm for generating data labels, and to calculate the corresponding error. The error is then used to adjust the loss calculation in the algorithm. After fitting the input multi-sensor data to the fused data, the loss calculation is continuously performed. After multiple adjustments, the final decision error gradually decreases, the entire algorithm will fit the input data of multiple sensors, and the output fused data will be able to effectively support the fire point detection task.

[0017] The beneficial effects of adopting this technical solution are:

[0018] For forest fire detection, using multiple sensors to collect data from the current environment can more accurately determine whether a fire exists at the current location. However, due to the inherent nature of the data as interval-valued data, the various data generated by multiple sensors often cannot be well integrated to effectively support the final decision. Therefore, this invention uses a multi-sensor data fusion algorithm based on the vulture search optimization algorithm to achieve improved accuracy in forest fire detection.

[0019] Using a vulture search optimization algorithm to fuse data from multiple sensors effectively fits the changing trends of data from each sensor and enables more efficient selection of combinations of data collected by multiple sensors, effectively taking into account the numerical variations of data from different sensors. While learning-based methods can effectively fuse data features, the large amount of data input from multiple sensors compromises detection efficiency. Methods based on the vulture search optimization algorithm, however, enable faster, more convenient, and more lightweight data fusion. Furthermore, the data collected by the various sensors used in monitoring equipment is inherently interval-valued, making it easy to fit the data features of each sensor and identify more effective combinations of sensor data. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic diagram of a forest defense multi-sensor data fusion algorithm based on vulture search optimization according to the present invention.

[0021] Figure 2 This is a schematic diagram of an example of the overall construction process of the present invention.

[0022] Figure 3 Schematic diagram of each stage of the vulture search algorithm in an embodiment of the present invention.

[0023] Figure 4 Schematic diagram of the algorithm calculation flow in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described below with reference to the accompanying drawings.

[0025] In this embodiment, see Figure 1 and Figure 2 As shown, the present invention proposes a forest defense multi-sensor data fusion method based on vulture search optimization, comprising the steps of:

[0026] S10, using multiple sensors configured in the fire point detection equipment to collect multi-dimensional data;

[0027] S20, transmitting the collected multi-dimensional data to a forest fire prevention multi-sensor data fusion model based on vulture search optimization to fuse the data features and fuse them into decision data for forest fire point detection;

[0028] The sensor data is input into the forest fire prevention multi-sensor data fusion model based on vulture search optimization. The vulture search algorithm is used to search and optimize the input data of multiple sensors, fit their interval value data characteristics, seek the best sensor data combination features, and fuse the combination features using machine learning methods. After multiple data refinement and fusion, the measurement data of multiple sensors are finally combined and fused into one data as decision support data for forest fire detection.

[0029] As an optimization solution to the above embodiment, the data collected by the multiple sensors is multidimensional, including temperature, humidity, and air particle size. This data has the characteristics of interval-valued IVD data. This means that it has large fluctuations and inherent variability errors. Directly using this data will affect the accuracy of overall fire point detection. Furthermore, the total amount of data from multiple sensors is huge, and directly using this data as input will seriously affect the prediction speed of the prediction algorithm and increase its runtime.

[0030] Construct a sample dataset, including data from multiple sensors in different environmental conditions, and corresponding descriptions of forest fire conditions.

[0031] Specifically, the constructed sample data set includes: data under normal environment, dry environment and high temperature environment, or data under normal conditions, suspected fire points, low-concentration smoke and high-concentration smoke conditions; and the corresponding fused data can effectively support fire point detection.

[0032] As an optimization solution of the above embodiment, a forest defense multi-sensor data fusion model based on vulture search optimization is proposed. Figure 3 Shown, including:

[0033] Aiming at the IVD characteristics of forest fire prevention data, an optimization algorithm based on the vulture's search and foraging principle was established;

[0034] Based on the vulture foraging principle, the optimization algorithm optimizes the data of each sensor and fits the data variation characteristics of each sensor. It then optimizes the combination of fitted data from multiple sensors to find the optimal combination of multiple sensor feature data for fire point detection. This algorithm-processed multi-dimensional data collected by multiple sensors can greatly improve fire point detection accuracy and shorten algorithm processing time.

[0035] like Figure 4 As shown in the figure, a variety of machine learning methods are used to test the decision support data output by the optimization algorithm based on the vulture foraging principle, test the accuracy of the fusion value output by the algorithm for generating data labels, calculate the corresponding error, and then use the error to adjust the loss calculation in the algorithm. After fitting the input multi-sensor data to the fused data, the loss calculation is continuously performed; after multiple adjustments, the final decision error gradually decreases, the entire algorithm will fit the input data of multiple sensors, and the output fused data will be able to effectively support the fire point detection task.

[0036] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A forest defense multi-sensor data fusion method based on vulture search optimization, characterized in that: The method comprises the steps of: S10, collecting multi-dimensional data using a plurality of sensors configured in the fire point detection equipment; S20, transmitting the collected multi-dimensional data to a forest fire prevention multi-sensor data fusion model based on vulture search optimization to fuse the data features and fuse them into decision data for forest fire point detection; The sensor data is fed into a forest fire prevention multi-sensor data fusion model based on vulture search optimization. Using a method based on the vulture search algorithm, the input data of multiple sensors is searched and optimized, fitting their interval value data characteristics to find the best sensor data combination features. The combined features are then fused using machine learning methods. After multiple rounds of data extraction and fusion, the measurement data of multiple sensors are finally combined and fused into a single data set as decision support data for forest fire detection. The forest defense multi-sensor data fusion model based on vulture search optimization includes: Aiming at the IVD characteristics of forest fire prevention data, an optimization algorithm based on the vulture's search and foraging principle was established; For the optimization algorithm based on the vulture foraging principle, the data change characteristics of each sensor are fitted by optimizing the search of each sensor data; then the combination of multiple sensor fitting data is optimized and searched to find the optimal combination of multiple sensor feature data for fire point detection.

2. The method for multi-sensor data fusion based on vulture search optimization for forest fire prevention according to claim 1 is characterized in that: The data collected by the multiple sensors are multi-dimensional data, including temperature, humidity and air particle size, and these data have interval value IVD data characteristics.

3. The method for forest fire prevention multi-sensor data fusion based on vulture search optimization according to claim 2 is characterized in that: Construct a sample dataset, including data from multiple sensors in different environmental conditions, and corresponding descriptions of forest fire conditions.

4. The method for forest fire prevention multi-sensor data fusion based on vulture search optimization according to claim 3 is characterized in that: The constructed sample data set includes: data under normal environment, dry environment and high temperature environment, or data under normal conditions, suspected fire points, low concentration smoke and high concentration smoke conditions; and the corresponding fused data can effectively support fire point detection.

5. The method for forest fire prevention multi-sensor data fusion based on vulture search optimization according to claim 1 is characterized in that: The decision-support data output by the optimization algorithm based on the vulture foraging principle is tested using various machine learning methods. The accuracy of the output fusion value for generating data labels is tested, and the corresponding error is calculated. The error is then used to adjust the loss calculation in the algorithm. After fitting the input multi-sensor data to the fused data, the loss calculation is continuously performed. After multiple adjustments, the final decision error gradually decreases.

Images