S.S. Kundu

Designing of an interval type-2 fuzzy logic controller for Magnetic Levitation System with reduced rule base

Magnetic Levitation System (MLS) is a typical multi-variable, nonlinear and unstable system. The basic aim of our research work is for suspending the steel ball without any mechanical support in desired position with help of an efficient controller. The simulation of MLS based on its mathematical model is carried out in MATLAB and its real time implementation is also achieved. In this paper an interval type-2 single input fuzzy logic controller (IT2SIFLC) is designed for a Magnetic Levitation System. For this task we have proposed an interval type-2 single input fuzzy logic controller (IT2SIFLC) based on theory of type-2 fuzzy logic systems (T2FLS) and single input theory of fuzzy logic control, which has the advantage of the total number of rules are abruptly reduced compared to interval type-2 fuzzy logic controller (IT2FLC). The tracking performance and robustness are also checked for this system. For tracking, two type of reference trajectory are modelled. One is sine wave and other is a set of constant point varying at different levels. Lastly for robust performance, disturbance is applied in MLS. The simulation and experiment results are indicating that the proposed control method has good control ability.

Fractionation of carbohydrate and protein content of some forage feeds of ruminants for nutritive evaluation.

Aim: To evaluate some forage feeds of ruminants in terms of their carbohydrate (CHO) and protein fractions using Cornell Net Carbohydrate and Protein System (CNCPS). Materials and Methods: Eleven ruminant feeds (six green fodders - maize, oat, sorghum, bajra, cowpea, berseem and five range herbages - para grass, guinea grass, hedge lucerne, setaria grass and hybrid napier) were selected for this study. Each feed was chemically analyzed for proximate principles (dry matter, crude protein [CP], ether extract, organic matter and ash), fiber fractions (neutral detergent fiber, acid detergent fiber, acid detergent lignin, cellulose and hemicellulose), primary CHO fractions (CHO, non-structural CHO, structural CHO and starch) and primary protein fractions (neutral detergent insoluble CP, acid detergent insoluble CP, non-protein nitrogen and soluble protein). The results were fitted to the equations of CNCPS to arrive at various CHO (CA - fast degrading, CB1 - intermediate degrading, CB2 - slow degrading and CC - non-degrading or unavailable) and protein (PA - instantaneously degrading, PB1 - fast degrading, PB2 - intermediate degrading, PB3 - slow degrading and PC - non-degrading or unavailable) fractions of test feeds. Results: Among green fodders, cowpea and berseem had higher CA content while except hedge lucerne all range herbages had lower CA values. CB1 content of all feeds was low but similar. All feeds except cowpea, berseem, and hedge lucerne contained higher CB2 values. Oat among green fodders and hybrid napier among range herbages had lower CC fraction. Feeds such as bajra, cowpea, berseem and the setaria grass contained lower PA fraction. All green fodders had higher PB1 content except maize and cowpea while all range herbages had lower PB1 values except hedge lucerne. Para grass and hybrid napier contained exceptionally low PB2 fraction among all feeds. Low PC contents were reported in oat and berseem fodders. Conclusion: Based on our findings, it was concluded that feeds with similar CP and CHO content varied significantly with respect to their CHO and protein fractions. Due to lower CC fraction, oat and hybrid napier were superior feeds in terms of CHO supply to ruminants. Similarly, among all feeds oat and berseem had a lower PC fraction, thus were considered good sources of protein for ruminants.

An Intelligent and Robust Single Input Interval Type-2 Fuzzy Logic Controller for Ball and Beam System

The Ball and Beam system (BBS) is a nonlinear and unstable system which resembles with many real-time complicated systems. Providing an appropriate beam angle to give the stability to the ball on the beam in a specific position, is a challenging task for the control system researchers. In this paper a robust interval type-2 fuzzy logic controller (IT2FLC) is designed. The dimension of the rule base is reduced by using signed distance method. This signed distance method makes the IT2FLC to a Single input Interval Type-2 Fuzzy logic Controller (SIIT2FLC). The type-2 fuzzy sets resolve the problem of determining membership functions in type-1 fuzzy systems. The membership function of a type-2 fuzzy set is three dimensional, where third dimension is the value of membership function at each point on its two dimensional domain that is called its footprint of uncertainty (FOU). The ability of FOU to represent more uncertainties enables one to cover the input and output domains with less number of fuzzy sets. This SIIT2FLC gives the smooth 2-D control surface and robustness to the system. The simulation work is carried out in simulink environment of MATLAB (7.8.0) software. The simulation results are also validated in the real-time implementation of the BBS, designed by Googol Technology. Experimental results show that the performance of the proposed controller is better than the single input type-1 fuzzy logic controller in terms of transient and steady state response. The ability of handling uncertainty (robustness) of proposed controller has been checked by applying a disturbance signal to the position sensor’s output and also by parameter variation of the BBS.

Evaluation of various feedstuffs of ruminants in terms of chemical composition and metabolisable energy content

Aim: The aim was to determine the chemical composition and metabolisable energy (ME) content of feedstuffs used in ruminant animals using in vitro method. Materials and Methods: A total of 18 feedstuffs used for ruminant feeding including cultivated non-leguminous fodders like maize, sorghum, pearl millet, and oat; leguminous fodders like cowpea and berseem; agro-industrial by-products such as wheat bran, deoiled rice bran, rice polish, wheat straw, and concentrates such as mustard oil cake, groundnut cake, soybean meal, cotton seed cake, grains like maize, oat, wheat, and barley were taken for this study. Chemical compositions and cell wall constituents of test feeds were determined in triplicate. The crude protein (CP) content was calculated as nitrogen (N) × 6.25. True dry matter digestibility (TDMD), true organic matter digestibility (TOMD), ME, and partitioning factor (PF) values were determined by in vitro gas production technique (IVGPT). Results: The CP content of non-leguminous fodders varied from 7.29% (sorghum) to 9.51% (maize), but leguminous fodders had less variation in CP. Oilseed cakes/meals had high CP and ether extract (EE) content than other feedstuffs except rice polish, which had 12.80% EE. Wheat straw contained highest fiber fractions than the other ingredients. ME content was highest in grains (wheat-12.02 MJ/kg) and lowest in wheat straw (4.65 MJ/kg) and other roughages. TDMD of grains and oilseed cakes/meals were higher than the fodders and agro-industrial by-products. The same trend was observed for TOMD. Conclusions: It was concluded that the energy feeds showed a great variation in chemical composition and ME content. The results of this study demonstrated that the kinetics of gas production of energy feed sources differed among themselves. Evaluation of various feedstuffs is helpful in balanced ration formulation for field animals and under farm conditions for better utilization of these commonly available feed resources.

Metabolizable Protein Requirements of Lactating Murrah Buffaloes (Bubalus bubalis) Fed on Silage Based Diets

The aim of present study was to determine the metabolizable protein (MP) requirements of lactating buffaloes fed on silage based diet. Fifteen lactating Murrah buffaloes (Bubalus bubalis) were allocated into three treatments (n=5) on the basis of milk production (7.10±0.40) and days in milk (118±5.0 d). The nutrient requirements of buffaloes were met by feeding maize silage: concentrate mixture in proportion of 60: 40 for a period of 96 days. The MP level in second treatment (Control; 11.1% on DM basis) was provided as per ICAR (2013) standards whereas in treatment one and three, MP levels were 10% lower (10% LMP) and 10% higher (10% HMP), respectively than ICAR (2013) requirements. Feeding of diets containing varying levels of MP did not affect body weight, milk production and milk composition, urinary levels of purine derivatives and microbial N production. The N intake and N excretion showed positive correlation with dietary MP content and reported highest (P<0.05) in 10% HMP fed buffaloes. The estimated MP requirement for maintenance plus 6% fat corrected milk yield (6% FCM) in experimental buffaloes were 2.56 g/kg W0.75 and 66.78 g/kg, respectively. The estimated MP requirements for maintenance and 6% FCM were 3.11% lower and 1.13% higher than recommendation of ICAR (2013), respectively. In conclusion, performance of lactating buffaloes was not affected by feeding diets containing 10% higher or 10% lesser MP than ICAR (2013) recommendation. As per current study, ICAR (2013) recommendation can be used for feeding lactating buffaloes.

Residual Feed Intake as a Tool for Selecting More Efficient Animals: A Review

Feeds alone contributes about 60–70% of total cost of ruminants production. There is deficiency to the tune of 10, 37 and 35% for dry fodder, concentrates and green fodders, respectively in India. Therefore, selection of efficient animals could help in reducing the burden of already scarce feed resources. A significant improvement in profitability could be achieved through a reduction in production costs by implementation of selection strategies to improve feed efficiency. There are several measures of feed efficiency viz., feed conversion ratio, relative growth rate, partial efficiency of growth, gross efficiency and residual feed intake (RFI). RFI is difference between actual feed intake and the feed an animal is expected to consume based on its body size and growth rate. Cattle with low RFI eat less than expected for their weight and growth rate and are, therefore, more efficient than cattle with high RFI. Among the feed efficiency measures, RFI is considered to be the best as it is moderately heritable (h2=0.29–0.46) and phenotypically independent of growth rate and body weight in growing cattle and basically reflects the variation in animals maintenance. This makes RFI as selection tool for efficient animals. About one-third of the biological variation in RFI of these calves could be explained by differences in feed intake (2%), digestion (10%), heat increment (9%), composition of gain (5%) and activity (5%) while the remaining variation in RFI was related to differences in protein turnover, stress and tissue metabolism (37%) and differences in cellular energy expenditures such as ion pumping and mitochondrial proton leakage (27%). The low RFI animals consume less feed and produce less enteric methane. hence, there are economic and environmental benefits by selecting such efficient animals. Phenotypic selection for RFI poses challenges because it requires measurement of each individuals feed intake for a minimum of 70 days which is expensive, time and labour intensive. As a result, there is need to develop alternative approaches like biochemical and physiological markers which could differentiate between high and low RFI animals within short period of time. Recently, genomic selection is being used as an appealing alternative tool for selection of low RFI animals. However, such studies need to be conducted using large number of animals so that more efficient animals could be selected for future breeding programmes for economic and environmental benefits.