1.

Field of project – Smart city

2. Abstract
Tree fall detection system on roads
Now a days accident and other problems arrives in travel and transport due to tree fall on roads. These
tree falls creates blockage and disturbance in transport. A tree becomes a hazard when it has a
structural defect that may cause the tree, or a portion of the tree, to fall and injure people or damage
property.

As a landowner or occupier you need to regularly inspect and maintain any trees on your land which
are adjacent to, or within falling distance of, the highway. This includes trees in hedgerows and banks
on private land .In our project we are making a tree fall detection system which will provide a display
which will show where the tree is fallen. This can provide a chance to make a decision for necessary
action. These action can make a better maintenance.

3.Introduction
Faults on transmission lines need to be detected, classified, and cleared as fast as possible. In power
transmission line protection, fault detection and faulty phase identification (fault classification) are
the two most important items which need to be addressed in a reliable and accurate manner.
Distance relaying techniques based on the measurement of the impedance at the fundamental
frequency between the fault location and the relaying point have attracted wide spread attention.
The sampled voltage and current data at the relaying point are used to detect and classify the fault
involving the line with or without fault resistance present in the fault path. Fault detection
algorithms in a relay employ either of these signals to discriminate the faulty situation from the
normal state. The sample-to-sample comparison of the current signal is a straightforward approach
for fault detection

.
Layout of system:

Advantages:

 It will provide quick information of tree fall.

 The tree fall some times become life taking in night. So it can avoid
unnecessary delay to take action.

 Tree fall maintenance can be easy with this system.

4. Literature survey
The use of neural network for identification of parameters has been reported by Chu et al
(1990); Narendra et al (1990). Bernieri (1994) has used a dynamic multilayer neural network
for parameter estimation in a secondorder lead-lag bandpass filter. The network is trained
for normal and different fault conditions based on the parameters obtained from system
transfer function during the steady state conditions.

Sorsa et al (1993) have used a bank of neural network models (each for known class of
system behaviour) that is similar to observer based schemes. The neural network replaces
the analytical model that describes the process and the models are developed with
appropriate data for different system behaviour.

Han et al (1997) have proposed a parameter estimation technique using neural networks in
which the physical parameters are estimated by applying the neural network universal
approximation property with the help of the measured input/output data. The deviations of
the parameters from normal values are then used for fault diagnosis. It is assumed that the
fault in the mentioned process can be described as changes in the parameter vector and the
nominal parameter values are known in advance or can be estimated online. (e.g. via RLS
method).

Patton et al (1999) have proposed a model of fault detection without residual generation. A
single layer neural network used for fault detection accepts current and past inputs [u(k),
u(k-1), u(k-2) …] and current and past values of output [y(k), y(k-1), y(k-2) …]. The network
directly indicates the occurrence of fault at the output. As an alternative Bhama et al (1993)
have proposed a technique for identifying the parameters A and B of an unknown second-
order dynamic system. The author has used a Single Layer Neural Network (SLNN) that uses
a gradient descent learning algorithm (also known as instant back propagation) to train and
identify the parameters of a linear Single-Input Single-Output system (SISO), where the
weights of the network represent the system parameters. The modified gradient descent
learning algorithm proposed by Yadaiah et al (2000) works well in terms of faster
convergence irrespective of the initial conditions where online identification can be done.

Chi-Sing Leung et al (2001) have used RLS based algorithm for online training of the
multilayer feedforward neural network. It has been shown through simulation result that
the trained network has improved generalization capability. Yong et al (2006) have
proposed a generalised RLS model which includes a general decay term in the energy
function for the training of feedforward neural network for four different problems. In this
work a single layer neural network is used to obtain the system parameters for deaerator
which is a Multi-Input Multi-Output system, using the RLS algorithm
5.problem identification

6. Objective
The circuit consists of a power supply, 4 line display, arduino and resistance measurement circuit. To
induce faults manually in the kit, fault switches are used. About 12 fault switches are used which are
arranged in three rows with each row having 4 switches. The 3 rows represent the 3 phases namely
R,Y and B.The fault switches: have 2 positions-No fault position(NF) and fault position(F).Main
component of the underground Tree fault detection circuit is low value resistance measurement. It
is constructed using a constant current source of 100mAmps. It can measure very low value
resistance as the cables have around 0.01 Ohm/meter resistance. For 10meter cable resistance
becomes 0.1 Ohm. This circuit can measure resistance up 50 Ohm, Maximum cable length it can
check up to 4 kilometers. So starting from the reference point 4 sets of resistances are placed in
series.These 4 sets of resistances represent the three plases and the neutral l.Short circuit faults,
Symmetrical and unsymmetrical faults can be determined by this method. This project uses three set
of resistances in series (ie)R10- R11-R12-R12,R17-R16-R14R21,R20-R19-R18-R25 one for each phase.
Each series resistor represents the resistance of the underground cable for a particular distance and
so here four resistances in series represent 1-4kms.Value of each resistance is 10kΩ

6. conclusion
As this system provides the location where tree is fallen, this can help in managing it early
before traffic gets highly intense. This system makes it easy to manage road safety and
clearity.

7.Refrences
1. A. G. Phadke and J. S. Thorp, Computer Relaying for Power Systems”,New York: Wiley, 1988.

2. G. B. Gilcrest, G. D. Rockefeller, and E. A. Udren, “High speed distance relaying using a digital
computer—Part I: System description, IEEE Trans. Power App. Syst., vol. 91, no. 3, pp. 1235–1243,
May1972.

3. J. Barros and J. M. Drake, “Real-time fault detection and classification in power systems using
microprocessors”, Proc. Inst. Elect. Eng., Gen. Transm. Distrib., vol. 141, no. 4, pp. 315–322, 1994.

4. F. N. Chowdhury, J. P. Christensen, and J. L. Aravena, “Power system fault detection and state
estimation using Kalman filter with hypothesis testing,” IEEE Transactions Power Del., vol. 6, no. 3,
pp. 1025–1030, Jul. 1991

5. T. S. Sidhu, D. S. Ghotra, and M. S. Sachdev, “An adaptive distance relay and its performance
comparison with a fixed data window distance relay”, IEEE Tranactions Power Del., vol. 17, no. 3, pp.
691–697, Jul. 2002.

6. O.A.S Yaussef, “New Algorithms for phase selection based on wavelet transforms”, IEEE
transactions on Power Delivery, vol17, pp. 908–914, 2002.
7. O.A.S Yaussef, “Online applications of wavelet transform to power system relaying”, IEEE
transactions on Power Delivery, vol-18, No. 4, pp. 1158–1165, 2003.

8. O.A.S Yaussef, “Combined Fuzzy-Logic wavelet based fault classification for power system
relaying”, IEEE transaction on Power Delivery, vol-19, No.2, pp. 582–589, 2004.

9. A.H.Osman, and O.P.Malik, “Transmission line protection based on wavelet transform”, IEEE
transaction on Power Delivery, vol-19, No. 2, pp. 515–523, 2004.

10. F. Martin, and J.A. Aguado, “Wavelet based ANN approach for Transmission line protection”, IEEE
transaction on Power Delivery, vol-18, No. 4, pp. 1572–1574, 2003.