Convergence Analysis of Backpropagation Algorithm for Designing an Intelligent System for Sensing Manhole Gases

Abstract

Human fatalities are reported due to the excessive proportional presence of hazardous gas components in manhole, such as Hydrogen Sulfide, Ammonia, Methane, Carbon Dioxide, Nitrogen Oxide, Carbon Monoxide, etc. Hence, predetermination of these gases is imperative. A neural network (NN) based intelligent sensory system is proposed for the avoidance of such fatalities. Backpropagation (BP) was applied for the supervised training of the neural network. A Gas sensor array consists of many sensor elements was employed for the sensing manhole gases. Sensors in the sensor array are responsible for sensing their target gas components only. Therefore, the presence of multiple gases results in cross sensitivity. The cross sensitivity is a crucial issue to this problem and it is viewed as pattern recognition and noise reduction problem. Various performance parameters and complexity of the problem influences NN training. In present chapter the performance of BP algorithm on such a real life application problem was comprehensively studied, compared and contrasted with the several other hybrid intelligent approaches both, in theoretical and in statistical sense.

Study of Climate Change in the Mandalika International Circuit Area Using Neural Network Backpropagation

“…Architectural design used to find the optimal NN configuration is also known as NN structural training, which consists of three NN layers, namely one input layer, one hidden layer, and one output layer. The number of neurons in the input layer and output layer depends on the problem itself [25,26]. This is in line with Haviluddin and Alfred [27], who stated that the backpropagation network architecture generally consists of three layers, namely the input layer, the hidden layer, and the output layer.…”

Identifying hazardousness of sewer pipeline gas mixture using classification methods: a comparative study

“…-Cross-sensitivity during sensing is an important factor in gas detection system, which was least reported in literature as yet. However, Ojha et al [22,23,24,25,26,27] offered a few methods such as neuro-genetic, neuro-swarm, ant-colony-based, neuro-simulated annealing, etc., where cross-sensitivity factor has been addressed to some extent. However, these works were primarily related to regression modeling.…”

A Hybrid of Artificial Bee Colony, Genetic Algorithm, and Neural Network for Diabetic Mellitus Diagnosing

“…This computation is the core fundamental of the learning process in any BP-based ANN structure. Through this process, a BPNN calculates and adjusts the weights utilizing gradient descent method (Atakulreka and Sutivong, 2007;Dai and Liu, 2012;Ojha, et al, 2016;Ojha, et al, 2017). Through this weights adjustment, BPNN can minimize the error rate at the output layer.…”

“…These equations are the core of learning for any BP based NN structure. The learning is the process of updating the weight values that connecting layers through existing neurons (Atakulreka and Sutivong, 2007;Dai and Liu, 2012;Ojha, et al, 2016;Ojha, et al, 2017). The new value of any weight depends on the learning rate and the rate of errors computed in the neurons of the output layer.…”