Forest Fire Evolution Prediction Using a Hybrid Intelligent System

Fire Ecology, 2015

Forest fires importantly influence our environment and lives. The ability of accurately predicting the area that may be involved in a forest fire event may help in optimizing fire management efforts. Given the complexity of the task, powerful computational tools are needed for predicting the amount of area that will be burned during a forest fire. The purpose of this study was to develop an intelligent system based on genetic programming for the prediction of burned areas, using only data related to the forest under analysis and meteorological data. We used geometric semantic genetic programming based on recently defined geometric semantic genetic operators for genetic programming. Experimental results, achieved using a database of 517 forest fire events between 2000 and 2003, showed the appropriateness of the proposed system for the prediction of the burned areas. In particular, results obtained with geometric semantic genetic programming were significantly better than those produced by standard genetic programming and other state of the art machine learning methods on both training and out-of-sample data. This

AIAA Infotech@Aerospace (I@A) Conference, 2013

Wildland fires have consumed acres of land and affected natural habitat in ways beyond a common man's intuition. More often than not, rescue operations including evacuation of surrounding urban areas have failed in saving the damage to life and property. Case studies of historical fires hold lack of situational awareness the biggest obstacle in forest fire-fighting. Eliminating the considerable rack of accurate information about fire-spread behavior can help fire-managers enhance safety of fire-personnel during on-field attacks. Fire-behavior study tools like FARSITE provide a good platform for study of historical fire and help in better understanding. Similar mathematical algorithms can be applied to real time environmental and spatial information to predict the spread of fire-perimeters and intensities. Systems like this can be used with surveillance-based unmanned aircrafts and enable firefighters plan on-field fire-attacks and air-drops. The following is a step towards building such a system. The work uses topographical data from the GAP project for the West Virginia Land Cover. A decision making tool is developed using fuzzy logic to designate a fuel model for forest-fires. This fuel model is then subjected to surface fire-spread techniques provided by Huygen's Principle and Rothermel's equation to develop a real time fire-predicting system.

IFIP Advances in Information and Communication Technology, 2016

Forest fires are one of the major natural disaster problems of the Mediterranean countries. Their prevention-effective fighting and especially the local prediction of the forest fire risk, requires the rational determination of the related factors and the development of a flexible system incorporating an intelligent inference mechanism. This is an enduring goal of the scientific community. This paper proposes an Intelligent Soft Computing Multivariable Analysis system (ISOCOMA) to determine effective wild fire risk indices. More specifically it involves a Takagi-Sugeno-Kang rule based fuzzy inference approach, that produces partial risk indices (PRI) per factor and per subject category. These PRI are unified by employing fuzzy conjunction T-Norms in order to develop pairs of risk indices (PARI). Through Chi Squared hypothesis testing, plus classification of the PARI and forest fire burned areas (in three classes) it was determined which PARI are closely related to the actual burned areas. Actually we have managed to determine which pairs of risk indices are able to determine the actual burned area for each case under study. Wild fire data related to specific features of each area in Greece were considered. The Soft computing approach proposed herein, was applied for the cases of Chania, and Ilia areas in Southern Greece and for Kefalonia island in the Ionian Sea, for the temporal period 1984-2004.

Suara Teknik: Jurnal Ilmiah

The tropical forest there are being one of natural resources that need to protect and explorate wise and safely also sustainabilty concept. But at several years later, there are some problems causes by improperly exploitation. One of disaster is forest fire. This disater not only eliminate forest natural resources but also produce other disaster like fogging and disturbing ecosystem. On the other side, research development at intelligence system especially neurofuzzy produce design system that able to make monitoring to real data, proceed with rule base form human thinking and produce proediction system and supporting information from decision making. Regarding to two point of view describe above, exist one idea to using development and achievment at neurofuzzy concept to make monitoring about ecological especially forest fire condition and proceed and make prediction to anticipate it.

Evolving Systems

through the years till 2100, based on the variance of average monthly temperature and average rain height (due to climate change) for the months May-October based on various climate models. Historical data for the period 1984-2004 were used to test the system for the areas of Chania and Ilia.

Expert Systems with Applications, 2011

In this study Forest Fire Decision Support System (FOFDESS) which is a multi-agent Decision Support System for Forest Fire has been presented. Depending on the existing meteorological state and environmental observations, FOFDESS does the fire danger rating by predicting the forest fire and it can also approximate fire spread speed and quickly detect a started fire. Some data fusion algorithms such as Artificial Neural Network (ANN), Naive Bayes Classifier (NBC), Fuzzy Switching (FS) and image processing have been used for these operations in FOFDESS. These algorithms have been brought together by a designed data fusion framework and a novel hybrid algorithm called NABNEF (Naive Bayes Aided Neural-Fuzzy Algorithm) has been improved for fire danger rating in FOFDESS. In this state, FOFDESS is an integrated system which includes the dimensions of prediction, detection and management. As a result of the experiments, it was found out that FOFDESS helped determining the most accurate strategy for fire fighting by producing effective results.

… Information Society (IS …, 2006

INDUSTRIAL, MECHANICAL AND ELECTRICAL ENGINEERING

Every year forest fires destroy a huge forest cover area, leaving large-scale destruction of flora and fauna in its wake. Forest fires play a major role in driving thousands of species of wildlife to extinction year. Machine Learning helps us predict the future. Using it in this domain can successfully help us predict forest fires and save wildlife. Any fire essentially depends upon three factors, which are oxygen, temperature, and humidity. This research aims at predicting the possibility of a forest fire taking place, given the oxygen, humidity, and temperature content of a given place. A concept website that can be created to take inputs from the user and predicts the forest fire probability in real-time is also shown. The paper aims to detect and alert the Forest fire occurrence using Temperature, Humidity, and oxygen values taken from the dataset and create a Web page to detect the forest fire.

Operational Research, 2005

Forest fires are considered natural phenomena but the magnitude of the problem can be attributed to poor forest management and to the extreme weather conditions. Furthermore, the abolishment of traditional activities related to forests has sent away people from the forests that they used to protect them. The aforementioned reasons result in the annual destruction of large forest areas, agricultural cultivations, industries, animals and even human lives. At the same time, catastrophic floods entail the gradual desolation of the affected areas and large amounts of carbon dioxide along with smoke particles aggravate air quality. In the present paper, a Decision Support System is proposed capable to support policy makers and services to counteract wildfires destruction danger in lowland Pine forests. The system is composed of two major components: (a) A Wildfire Destruction Danger Index useful for preventive and suppressive measures planning and decision policy making and (b) a Forest Management Planning Decision Support System useful for fire risk reduction through forest management plan. The open architecture of the system allows incorporation of data coming from external sources e.g. satellite systems, meteorological stations etc. All its subsystems can stand alone so as to satisfy the needs of responsible organisations (peripheral institutions of fire brigade, forest inspection and local authorities). The proposed system can be both used as an intergraded operational tool in forest fire management and as a training tool for personnel of various services involved.

In this paper, we present an application of artificial neural networks to the real-world problem of predicting forest fires. The neural network used for this application is a multilayer perceptron whose architectural parameters, i.e., the number of hidden layers and the number of neurons per layer were heuristically determined. The synaptic weights of this architecture were adjusted using the backpropagation learning algorithm and a large set of real data related to the studied problem. We also present and discuss some examples of illustrating results that show the performance and the usefulness of the resulting neural system.