Search

Search Results

• ALGORITHM FOR CLASSIFICATION OF FIRE HAZARDOUS SITUATIONS BASED ON KOLMOGOROV-ARNOLD NETWORK

  Sanni Singh, A.V. Pribylskiy

  6-15

  2025-01-06

  Abstract ▼

  The problem of timely and accurate detection of fire hazardous situations is critical to ensure the safety of people and property. Traditional monitoring methods based on simple threshold values for smoke and temperature sensors are often insufficiently effective, as they can lead to false alarms or miss real fire hazardous situations. Modern methods using neural networks can significantly improve the accuracy of classifying an emergency situation by analyzing complex patterns in sensor data, which are complex nonlinear functions with dynamically changing parameters. The development of such models requires attention to the collection, labeling and processing of data, to the choice of neural network architecture for a specific task, because high-quality data labeling and the choice of the desired neural network architecture directly affect the selection of the desired patterns, as well as the detection of hidden patterns that are impossible or difficult to determine by traditional methods. The article examines an algorithm for classifying fire hazardous situations based on the Kolmogorov-Arnold network (KAN). This algorithm is used to process data from a complex of interconnected fire sensors and is designed to detect and classify various types of fire hazardous situations. The key element of the development is the use of the Kolmogorov-Arnold network, which, due to its architecture, is capable of modeling complex functional dependencies between input data. Readings from a complex of interconnected fire sensors, such as temperature and smoke sensors, are used as input data. To improve the accuracy of classification, data is labeled using expert knowledge. The Python programming language was used to implement the algorithm, together with the Pytorch, pykan, and scikit-learn libraries. The article presents the results of testing the model on real data and discusses possible directions for further improvement of the algorithm. During the experiments, it was shown that the proposed model demonstrates high accuracy in classifying fire hazardous situations, which is not inferior to traditional methods of data classification.

• DEVELOPMENT AND RESEARCH OF ALGORITHMS FOR FORECASTING FIRE HAZARDOUS SITUATIONS

  Singh Sanni, А.V. Pribylskiy, Е.Y. Kosenko

  2025-01-30

  Abstract ▼

  Early detection of fire hazard situations is a critical aspect of ensuring safety, as it helps to minimize
  the risk of material and human losses. Early detection of threats helps to preserve material assets,
  reduce the time for their restoration and, more importantly, save human lives. In this regard, a new approach
  to predicting fire hazard situations is proposed: an algorithm for training a model for predicting
  fire hazard situations, as well as an algorithm for predicting fire hazard situations, which are developed
  on machine learning models such as recurrent neural networks, random forest, optimization trees, autoregressive
  neural networks, etc. The study proposes to consider algorithms for predicting fire hazard situations
  developed on the basis of an analysis of existing forecasting algorithms, including methods based
  on machine learning, statistical models and simulation approaches, taking into account their advantages
  and disadvantages, accuracy indicators. The results of the study of the developed algorithms show that
  they are capable of predicting the outside temperature value of the sensor with an accuracy of 93.33%
  based on the test data from a complex of interconnected fire sensors, with errors of MAE = 1.72,
  MSE = 2.95 in the abnormal mode on the test data, and with an accuracy of 92.85% for the temperature
  inside the sensor, errors MAE = 1.66, MSE = 2.75. The accuracy on the test data in the normal mode for
  the outside temperature was 96.27%, errors MAE = 1.22, MSE = 1.48, and the accuracy of predicting the
  inside temperature was 96.16%, errors MAE = 1.24, MSE = 1.53. For the test sample of 500,000 readings,
  the errors of the predicted outside temperature were: MAE = 1.82, and MSE = 3.31, and the accuracy
  was 91.78%. The errors of the predicted temperature inside (temp2_inside) were: MAE = 1.89, and
  MSE = 3.57, and the accuracy was 91.35%.

• ALGORITHM FOR CLASSIFICATION OF FIRE HAZARDOUS SITUATIONS BASED ON NEURAL NETWORK TECHNOLOGIES

  Sanni Singh, А.V. Pribylskiy

  2024-08-12

  Abstract ▼

  Modern technological requirements and developing urban infrastructure pose the task of developing
  methods for recognizing and classifying fire hazardous situations. Quickly and effectively recognizing the
  initial signs of a fire becomes a vital aspect of ensuring the safety of people as well as property. In this
  regard, systems are developed, implemented, tested and implemented that can automatically recognize
  and classify fire hazardous situations. Classification of fire hazardous situations allows you to determine
  the degree of danger of detected deviations, which contributes to making more effective decisions to prevent
  the consequences of fires and their signs, such as a one-time short-term increase in temperature and
  smoke level, which may indicate failure of electrical components located near the sensors. The algorithm
  for classifying fire hazardous situations is developed for a complex of interconnected sensors, which in
  turn, due to its structure, allows you to detect even the slightest sign of fire. Within the framework of this
  study, an algorithm for classifying fire hazardous situations based on neural network technologies is presented.
  A description of existing classes of fire hazardous situations is provided, as well as the criteria by
  which data for these classes were marked. The algorithm was modeled on training and test samples, presenting
  the accuracy parameters used, the formula for their calculations, and the results of classifying fire
  hazardous situations. A study was carried out of the influence of the sample step in the database sample
  on the accuracy parameters and training time of the neural network. The developed algorithm is implemented
  in the Python programming language in the PyCharm IDE. The dataset for training and testing
  was obtained from real sources containing information about detected fire hazardous situations in subways in which a complex of interconnected sensors is installed. The results of modeling the algorithm
  showed that the proposed algorithm has high accuracy for predictive classification of fire hazardous situations
  in real objects.

• SYNTHESIS OF A SYSTEM FOR ULTRA-FAST DETECTION OF FIREHAZARDOUS SITUATIONS BASED ON A COMPLEX OF INTERCONNECTED SENSORS

  Sanni Singh , А.V. Pribylskiy

  2024-05-28

  Abstract ▼

  Modern technologies and urban infrastructure require innovative approaches to detecting fire hazards.
  Effective and ultra-fast fire detection is becoming an integral part of safety. For this purpose, systems capable of
  detecting and informing about a fire hazard situation in a matter of seconds are synthesized and implemented;
  one of such systems is synthesized in the article. The research and synthesis of a mathematical model of a digital
  universal fire sensor, which in turn is a complex of interconnected sensors, is relevant due to the constant development
  of system infrastructure, the increasing complexity of electrical equipment and the need to reduce damage
  arising from the outbreak and spread of fires. Predictive diagnostics of electrical equipment performance
  allows timely identification and elimination of potential fire safety threats. Within the framework of this research,
  a theoretical mathematical model of a real digital universal fire sensor is presented, first in a simplified
  version, then in a more complicated version, taking into account the design and statistical approach to the problem
  of finding the sensor response thresholds, a description of the parameters of the mathematical model and the
  sequential principle of operation is given. This sensor is an innovative fire safety solution that provides a high
  level of control and efficiency in real time. Based on the theoretical models presented in the article, a mathematical
  model of the sensor has been developed, which is simulated using the Simulink software tool on real data
  obtained from the sensor manufacturer. The simulation results showed that the model correctly describes the
  behavior of a real sensor on all channels and can be used in further research, such as predicting and detecting
  fire situations using neural networks. The synthesis of the proposed system is necessary for further research in
  the field of forecasting and detection of fire hazardous situations based on the obtained mathematical model.