Crop Planner and Firtilizer Schedular in Smart Agriculture

Internet of Things (IoT) technology has brought revolution to every field of common man’s life. By introducing everything intelligent and smart. IoT refers to a network of things which make a self-configuring network. The development of Intelligent Smart Farming based on IoT devices is increasing day by day which help in turning the face of agriculture and also increase the productivity of farmer but making it cost efficient and help to reduce wastage . The basic objective of this project is to predict suitable crop based on the values obtained from environmental sensors (Temperature, soil, humidity) and second objective consist of fertilizer scheduler which will suggest required amount of fertilizer for proper crop growth.

Introduction

Purpose—

The next era of Smart Computing will be totally based on Internet of Things (IoT). Internet of Things (IoT), these days is playing a crucial role of transforming “Traditional Technology” from homes to offices to “Next Generation Everywhere Computing”. Agricultural production requires lots of activities like soil and plant monitoring, environmental monitoring like moisture and temperature, transportation, supply chain management, infrastructure management, control systems management, animal monitoring, pest control etc.

IoT based agricultural convergence technology creates high value in terms of quality and increased production and also reduces burden on farmers in ample manner. The objective of this research paper is to propose IoT Based Smart agriculture which will enable farmers to have live data of soil moisture, environment temperature at very low cost so that live monitoring can be done.

Objective—

Objective of the project is to study about smart agriculture system and various existing application on smart agriculture.

Get quality help now

KarrieWrites

Verified

Proficient in: Agriculture

5 (339)

“ KarrieWrites did such a phenomenal job on this assignment! He completed it prior to its deadline and was thorough and informative. ”

+84 relevant experts are online

Hire writer

Study various technologies like IOT, sensors, cloud Computing and Machine learning algorithm.

Scope—

To develop multidisciplinary model for smart agriculture based on the key technology. The details of the farmer, periodic soil properties of farmland are stored on cloud storage. soil and environment properties and sensed and periodically send to cloud through IOT. Model is beneficial for increase in agriculture production and for cost control of agro-product. The project will also help in development of agriculture sector in India.

Proposed System

Our app contains 2 modes: – Mode 1 & Mode 2. In mode 1, it consists of crop suggestion, depending upon the present condition of soil and temperature of region. We are going to take the real time data from different seniors and compare data with given prior information. On the result of the given sensor data and information we will suggest crop to farmer which can help in the growth of crop yield. Mode 2, consists of fertilizer scheduler, we are going to take soil health card from as an input from farmer and then we will calculate the total amount of fertilizer required for the plant. We will check the soil nutrient and then compare the data with plant data. Depending upon the result we will calculate total amount of fertilizer required for proper growth of crop.

Proposed Outcome

The outcome of project is we will suggest suitable crop to the farmer depending upon the present condition of soil and temperature. Secondly we will give the amount of fertilizer required for crop for high yield and proper growth.

Literature Survey

Existing Applications surveyed—

The application was launched on 2016 by the PM Narendra Modi to work toward the development of the farmers and development of villages, the app offers user-friendly interface.

It provides the knowledge on fertilizers, seeds, machinery, it gives detail about the current weather and also forecast for next 5 days.

The application can be used in different languages which is easy to use.

IFFCO kisan application was launched in 2015 and managed by Indian Farmers Fertilizer Cooperative Ltd.

It provide customized information to the farmer which he needs for proper farming.

Agriculture related information is stored in the form of text, images, audio and video in different languages. The module consist of market details, weather reports and agriculture information.

They also provide call services from Kisan Call Centre Services.

Benefits of IoT in Agriculture

The following are the benefits of IoT in Agriculture:

1. IoT enables easy collection and management of tons of data collected from sensors and with integration of cloud computing services like Agriculture fields maps, cloud storage etc., data can be accessed live from anywhere and everywhere enabling live monitoring and end to end connectivity among all the parties concerned.
2. IoT is regarded as key component for Smart Farming as with accurate sensors and smart equipment’s, farmers can increase the food production by 70% till year 2050 as depicted by experts.
3. With IoT productions costs can be reduced to a remarkable level which will in turn increase profitability and sustainability.
4. With IoT, efficiency level would be increased in terms of usage of Soil, Water, Fertilizers, Pesticides etc.

Different types of sensors:

• Soil moisture sensor: The sensors is used to measure the soil moisture and materials in the soil . The sensor has two large exposed pads which functions as probes for the sensor, together acting as a variable resistor. The moisture level of the soil is detected by this sensor. When the water level is low in the soil, the analog voltage will be low and this analog voltage keeps increasing as the conductivity between the electrodes in the soil changes. This sensor can be used for watering a flower plant or any other plants require automation.
• Temperature sensor: The temperature and humidity sensor can measure the temperature and humidity. When the system working, the temperature sensor directly turns the temperature signal into a digital signal, and then is read out by MCU (Micro Control Unit). The humidity sensor can get the analog signal from the air or the soil which the MCU can’t read out, so AD convertors are need to turn the analog signals into digital signal.
• Arduino kit: Sensors are connected to the Arduino to perform an action. Arduino will send the data to the data base using Ethernet shield and if emergency it also send message to the user by using a device called GSM module. Sensor values from arduino are transmitted to the GSM modem

Introduction to Wireless Sensor Network

Demand for farming increased production of food in modern days to accommodate the large population of the world. To accomplish this intent, new technologies and solutions as well as new methods are used in agriculture domain to increases productivity.

Technologies and standards used in agriculture

There are various types of wireless equipment’s present in the wireless technology. We can use any type from these equipment’s. They are as follows: – Wireless communication, ZigBee, Wi-Fi, Bluetooth, GPRS/3G/4G, WiMAX, etc.

Wireless sensor networks and its potential for agricultural Applications

There are two types of wireless networks which are widely used in agriculture domain. They are as follows:

• Terrestrial wireless sensor networks
• Wireless underground sensor networks

In the wireless sensor networks the wireless sensors are placed on the top of land and each wireless sensor have small battery with it for provide power to the sensor. The wireless sensors communicate with each other using the radio frequency (rf).One Gateway Node is their through the Gateway node other sensor send information to the database.

The Wireless underground sensor network is another type of wireless sensor network. In this network type the wireless sensors are placed under the ground and also one Gateway node is there for communicating with the Surface sink node And through the sink node the sensors send the data to the server.

K-mean Algorithum for Crop Suggestion

This works demonstrates an evaluation of modified k-Means clustering algorithm in crop prediction. The results show comparison of modified k-Means over k-Means and k-Means++ clustering algorithm and modified k-Means has achieved the maximum number of high quality clusters, correct prediction of crop and maximum accuracy count.

The proposed solution introduces a better way for clustering by doing enhancement in partitions. In order to understand the working of algorithm, The architecture of crop prediction includes the crop knowledge-base, feature selection, three clustering approaches such as kMeans, k-Means++, proposed modified k-Means algorithm, pattern visualization and sample testing and prediction.

Architecture of Crop Prediction

The architecture of drop prediction is given in the fig with each and every block description. The architecture is a system which convert the components or element into equivalent functions. Block wise decryptions is as follow.

Crop knowledge base: It consists of farm knowledge such type of crops like kharif crop, soil type, soil moisture, mineral content of the soil, and temperature required for proper growth. The knowledge-base also consist environmental parameter such as maximum and minimum temperature value and average rainfall, and region wise data of crops.

Feature Selection: The process consists of feature selection from the given data for the purpose of calculation. It selects one record at a time from 396 x 10 records and performs calculation for partitioning.

Clustering Approaches: The three clustering approaches is used for the prediction of crop such as modified k-Means, kMeans++ and traditional k-Means. Because of the number of clusters (k value) is required at starting for traditional k-Means and kMeans ++, the same calculated value of number of clusters is provided and initial cluster centers are uniformly chosen. All three approaches perform there work and provide output.

Decision Making algorithm for fertilizer scheduler

The algorithm working is totally based on weight optimal solution to calculate proper amount of fertilizer required for farming.it also uses interpolation algorithm which improves the reliability of fertilization. It also reduces the computation load so it becomes easy to implement the algorithm with reduce the efforts.

At present, three fertilization models are widely used, and have good results, and they have their own advantages and disadvantages.

(a) The Soil Fertilizer nutrient balance model is based on soil nutrient and yield goals in decision making. However, crop yield, soil nutrients and other factors, make the model show highly nonlinear relationship and reduce the accuracy of algorithm.

(b) The Dissimilar Subtraction method of soil fertility is designed to carry out fertilization methods according to an equal relationship between the fertilizer production rate and difference of target yield and the blank yield. The advantages of the dissimilar subtraction method is that it does not need soil tests, so we can avoid the trouble of the quarterly measurements of soil nutrients, also it is easier to calculate. But the blank field yield is the decision factor of the combined result of various yield factors, and we cannot get the exact value of fertilization according to the production forecast, so the test results can be biased easily.

(c) The Soil Fertilizer Effect Function method including “3414” fertilizer experiments is widely used at home and abroad, and is a statistical model about fertilization and crop yield, which is able to calculate a crop’s optimal yields in theory and the corresponding fertilization amounts. The soil fertilizer effect function method has a variety of mathematical function models, applicable to different crops, but sometimes it is prone to saddle surface, and has big deviation when predicting fertilization.

At present, fertilization decision-making systems are largely based on one of these three kinds of models. Because of the single type of model, they cannot guarantee that the model is the optimal one, and hardly avoid the model errors, thus easily cause inaccurate results of fertilization.

References

1. [Ken Cai, “ Internet of Things Technology Applied in Field Information Monitoring”, Advances in information Sciences and Service Sciences (AISS) Volume 4, Number 12, July 2012.
2.  Xiaohui Wang, Nannan Liu, “The Application of Internet of Things in Agricultural means of production supply chain management”, Research Article, Journal of Chemical and Pharmaceutical Research, 2014,6(7):2304-2310.
3.  Li Minbo, Zhu Zhu, Chen Guangyu, “Information Service System of Agriculture IoT”, AUTOMATICA 54(2013) 4, 415-426.
4. Shitala Prasad, Sateesh K. Peddoju, Debashis Ghosh, “AgroMobile: A Cloud-Based Framework for Agriculturists on Mobile Platform”, International Journal of Advanced Science and Technology Vol.59,(2013), pp.41-52.
5. Snehal S.Dahikar, Sandeep V.Rode, “Agricultural Crop Yield Prediction Using Artificial Neural Network Approach”,INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN ELECTRICAL, ELECTRONICS, INSTRUMENTATION AND CONTROL ENGINEERING, Vol.2, Issue 1, January 2014.
6. Zhao Dong, Design and implementation of fertilizing expert system, ChangChun: ChangChun University of Science and Technology, 2008.
7. Pan Dafeng, Li QunAgricultural expert system fuzzy calculation scheme System Sciences and Comprehensive Studies in Agriculture, 19 (3) (2003), pp. 192-198
8.  A. RoyArtificial neural networks-a science in trouble ACM SIGKDD Explorations Newsletter, 1 (2) (2001), pp. 33-38