**2.1. Artificial neural networks modelling**

The physiology of neurons present in biological neural system such as human nervous system was the fundamental idea behind developing the artificial neural networks ANNs. This computational model was trained to capture a nonlinear relationship between input and output variables with scientific and mathematical basis. In recent days, a commonly used model is layered feed-forward neural network with multilayer perceptions and back propa‐ gation learning algorithms [1-4].

The ANNs are computing systems composed of a number of highly interconnected layers of simple neuron-like processing elements, which process information by their dynamic response to external inputs. The information passes through the complete network by linear connection with linear or nonlinear transformations. The weights were determined by training the neural nets. Once the ANN was trained, it was used for predicting new sets of inputs. Multi-layer feed-forward neural network architecture (Figure 1) was used for predicting the tenacity, initial modulus, air permeability, initial thickness, percentage compression, thickness loss, and compression resilience properties of fabrics [5-8]. The circle in Figure 3.5 represents the neurons arranged in five layers as one input, one output and three hidden layers. Three neurons in the input layer, three hidden layers, each layer consisting of three neurons and one neuron in the output layer. HL-1, HL-2, and HL-3 are the first, second, and third hidden layers, respectively, whereas i and j are two different neurons in two different layers. The neuron (i) in one layer was connected with the neuron (j) in next layer with weights (Wij) as presented in Figure 1.

**Figure 1.** Neural architecture of the fabric property [8]

nonwoven products are developed. These jute fibres are comparatively coarse in nature and have wide variation in fineness apart from its mesh-like structure. Its high moisture regain also places its suitability in certain applications. These properties make it more popular in the development of needle-punched nonwoven rather than the other nonwoven like thermal bonded and adhesive bonded nonwoven structures. Apart from these properties, the jute fibre

Jute includes good insulating and antistatic properties, as well as having low thermal conduc‐ tivity and a moderate moisture regain. It includes acoustic insulating properties and are manufacture with no skin irritations. Jute has the ability to be blended with other fibres, both synthetic and natural and accepts cellulosic dye classes such as natural, basic, vat, sulphur, reactive, and pigment dyes. Jute can also be blended with wool. By treating jute with caustic soda, crimp, softness, pliability, and appearance are improved, aiding in its ability to be spun with wool. Liquid ammonia has a similar effect on jute, as well as the added characteristic of

Jute has a long history of use in the sackings, carpets, wrapping fabrics (cotton bale), and construction fabric manufacturing industry. However, the major breakthrough came when the automobile, pulp and paper, and the furniture and bedding industries started to use jute and its allied fibres with their nonwoven and composite technology to manufacture nonwovens, technical textiles, composite production of sheet moulding compound, resin transfer mould‐ ing, vacuum pressing techniques, and injection. In this chapter, the emphasis has been made on the design and development of jute-needle-punched nonwoven and their characterization in specific applications. Also, different important applications of jute and jute-based needle

**2. Design and development of jute-based needle-punched nonwoven**

Designing of needle-punched nonwoven from jute and its blend plays a very important role as far as the end product is concerned. The variation on fabric properties changes on various factors, like fabric weight, needling density, depth of needle penetration, needle gauge, blending of jute with other fibres, type of jute fibre used, pre-treatment of jute fibre, etc. Out of these various factors, the most influencing parameters are fabric weight, needling density, depth of needle penetration, and blending of jute with other fibres. Thus, the designing and development of jute needle-punched nonwoven neural network modelling approach holds good for designing, modelling, and the prediction of important properties. The following

The physiology of neurons present in biological neural system such as human nervous system was the fundamental idea behind developing the artificial neural networks ANNs. This computational model was trained to capture a nonlinear relationship between input and output variables with scientific and mathematical basis. In recent days, a commonly used

is one of the cheapest natural fibre available commercially in countable amount.

improving flame resistance when treated with flame proofing agents.

punched nonwovens have been covered.

278 Non-woven Fabrics

discussions describe these aspects.

**2.1. Artificial neural networks modelling**

The data were scaled down between 0 and 1 by normalizing them with their respective values. The ANN was trained with known sets of input-output data pairs.
