Description:CLAIM 1 : SUNFLOWER HELIOTROPISM ALGORITHM(SHA)

The goal of the Patent work is to develop an algorithm that can learn from experience and improve their performance without being explicitly programmed for each task named as reinforcement learning. This Patent work will help us to better understand how machines think and how they can be applied in various fields. In this chapter pre-processing techniques, feature extraction and proposed Sunflower Heliotropism Algorithm (SHA) have been discussed in detail.
METHODOLOGY - PROPOSED SUNFLOWER HELIOTROPISM ALGORITHM (SHA)

Heliotropism is the phenomenon where a plant's leaves or flowers orient themselves towards the sun in order to maximize the amount of light they receive. Sunflowers are known for their strong heliotropism, as their flowers will track the sun as it moves across the sky during the day. This behaviour is controlled by the plant's response to changes in light intensity on different parts of the stem, which in turn triggers growth responses that cause the stem to bend towards the light source. The exact mechanism by which sunflowers exhibit heliotropism is not fully understood, but it is thought to involve a combination of phototropism (response to light intensity) and geotropism (response to gravity).
Patent suggests that the cells on the side of the stem facing the sun contain more of the hormone aux in, which causes the cells on that side to elongate more than the cells on the shaded side. This causes the stem to bend towards the sun. Additionally, it is also thought that blue light may inhibit the growth of cells on the shaded side and promote growth on the illuminated side, further contributing to the bending of the stem towards the sun. Clustering is a technique for grouping similar data points together. There are various algorithms for clustering data, such as K-Means, Hierarchical Clustering, and DBSCAN. These algorithms can be applied to LDR datasets to group similar light intensity values together.
CLAIM – II : SEE-SAW ALGORITHM (SSA)
Machine learning algorithms can be used in solar panel tilting to optimize the panel's orientation and increase its energy production. The main goal of this Patent work is to find the optimal tilt angle that maximizes the amount of sunlight that the panel receives. This can be achieved through the following steps:
STEP 1. Collect data on the solar panel's energy production and the environmental conditions.
STEP 2. Preprocess the data and prepare it for modeling.

STEP 3. Train a machine learning model, such as a regression or neural network model, on the preprocessed data to predict the energy production as a function of the tilt angle and environmental conditions.
STEP 4. Use the trained model to estimate the energy production for different tilt angles and find the optimal tilt angle that maximizes the energy production.
STEP 5. Implement the optimal tilt angle in the real-world system and monitor the performance to fine-tune the model if necessary.
This process can be repeated periodically to account for changes in the environmental conditions and to further optimize the panel's performance. By using machine learning algorithms, the solar panel tilting system can adapt to changing conditions and maximize its energy production over time. The goal of the Patent work is to tilt the solar panel without the heavy energy consuming rotatory motors and 24/7 live tracking sensors to reduce the cost and loss of energy. From CLAIM I obtained data of high light intensity and low light intensity with time stamp the proposed SeeSaw Algorithm (SSA) performs well for angle prediction with respect to wait balancing which helps to tilt the solar panel using solid weight and liquid holder helps to increase the weight using the IoT enabled water pump.
SEESAW ALGORITHM
The SeeSaw Algorithm (SSA) is a weight balancing technique used to balance a seesaw-like structure. A seesaw is a lever that is balanced by placing weight at either end, and the SeeSaw algorithm is used to find the optimal distribution of weights to balance the SeeSaw Algorithm. The algorithm works by iteratively adjusting the weights at either end of the seesaw based on the current imbalance until the seesaw reaches a state of balance. The specific steps involved in the SeeSaw algorithm will depend on the particular problem being solved and the structure of the seesaw, but a general outline of the algorithm is as follows:
STEP 1 : Initialize the weights at either end of the seesaw, w_1 and w_2.
STEP 2 : Measure the current imbalance of the seesaw, I.
STEP 3 : Based on the current imbalance, adjust the weights at either end of the
seesaw, w_1 = w_1 + d_1 and w_2 = w_2 + d_2, where d_1 and d_2 are
correction terms that depend on the current imbalance.
STEP 4 : Repeat steps 2 and 3 until the seesaw reaches a state of balance, where the
current imbalance is below a predefined threshold.

PANEL TILTING ANGLE PREDICTION
Angle prediction techniques are methods used to predict the angle of an object or system based on historical or current data. There are several techniques that can be used for angle prediction, including:
TIME SERIES ANALYSIS
Time series analysis is a set of statistical techniques used to model and analyze Time-Based data. It can be used to predict the angle of an object based on historical angle measurements and trends in the data. These are some of the most common angle prediction techniques used in various applications.
1. In CLAIM II the goal of the Patent work is to predict the angle and to tilt the solar panel facing the sun direction to observe maximum light exposure. From the CLAIM I result the panels are get mounted with the base angle of 45 degree.
2.
3. A is the angle where the solar panel need to facing.
4. Where, VW is the Varying Weight where the water tank weight increase and decrease to swing up and down SeeSaw structure.
5. SVD stands for the Standard Varying Degree.
6. The final dataset have been generated with the angle values and then that is considered as the output of CLAIM II to make this angle tilting an automated process CLAIM III will be taken into action.

The formula is derived with 90 degrees where the sun is the mid of afternoon which makes the panel to be flat and the facing angle will be 90 degrees. This mathematical formula is implemented in orange data mining tool and there will be a change in angle for each varying weight.
WEIGHT BALANCING MODEL
A seesaw-like structure is a balanced system where the weight on one side must be equal to the weight on the other side in order for the system to remain balance. In such a system, the weights can be balanced by adjusting the weights on each side of the seesaw until they are equal. The exact details of the "Seesaw-like structure weight balancing technique" will depend on the specific application, but the general idea is to use algorithms to measure the weights on each side of the seesaw and adjust them as necessary to ensure balance. The technique may also involve using machine learning algorithms, such as reinforcement learning, to learn the optimal weights for different conditions and to adapt to changes in the environment over time. The goal of the "SeeSaw-like structure and weight balancing technique" is to provide a reliable and efficient method for a weight balancing.
A PSI water pump is designed to pump water at a specific pressure, typically measured in pounds per square inch (PSI). The pressure generated by the pump is regulated by a pressure switch, which turns the pump on and off based on the desired pressure level. PSI water pumps come in a variety of sizes and capacities, from small, portable pumps to large, industrial-sized pumps. The design and construction have been optimized for efficient and reliable performance.

COST-EFFECTIVE
Water motors are often more cost-effective than other types of motors, as they are simple in design and can be mass-produced. The following steps are necessary for connecting the Arduino with the PSI motor.
STEP 1. Connect the mini PSI water pump to the Arduino Uno using a relay module.
STEP 2. Write a program in the Arduino IDE to control the relay module and turn the pump on and off at specific times.
STEP 3. Use the built-in clock of the Arduino Uno to keep track of time and trigger the pump accordingly.
STEP 4. Upload the program to the Arduino Uno.
STEP 5. Connect a power source to the Arduino Uno to power the pump and the microcontroller.
STEP 6. The program should use the millis() function to keep track of the elapsed time and determine when to turn the pump on and off.
CLAIM – III : ENERGY OPTIMIZATION AND BEHAVIOR PREDICTION USING MEOA AND TPC ALGORITHM
In CLAIM III the energy generation by the solar panels and on and off of power gadgets are monitored and controlled with the developed mobile android application. The generated dataset was iterated with the proposed algorithm to find the pattern of energy consumption per product cluster and periodic usage of the daily appliances was taken into consideration with timestamp. The proposed Monetrization Energy Optimization Algorithm (MEOA) helps to control the anonymous heavy energy-consuming products. The proposed Timestamp Prediction Consumption Algorithm (TPCA) helps to predict low energy generation in future and alerts the user to be aware of energy optimization and usage. Hence, machine learning algorithms and data mining techniques are used to design and development of a novel method to predict the future behavior of energy consumption.
ANDROID TOOL

MIT App Inventor is a user-friendly platform for creating Android applications. It also includes a live testing feature that allows users to test their app in real-time on an Android device connected to the personal computer. For developing the android application these steps to be followed:

Step 1: Created a new project in MIT App Inventor and named it Smart Home application.
Step 2: Then designed the User Interface (UI) for the app. Create buttons, sliders, and other user interface elements to control home appliances and display information about solar panel energy generation and battery backup time.
Step 3: Connect the app to the home appliances via Bluetooth. The Patent work uses Bluetooth modules namely HC-05 to connect the appliances.
Step 4: Using the built-in sensors on the Android device to monitor the energy generated by the solar panels. This can be done by using the light sensor LDR to detect the intensity of light, which can be used to estimate the solar panel energy generation and to increase the accuracy of the output of the solar energy generation dataset taken for comparison.
Step 5: The proposed Monetrization Energy Optimization Algorithm (MEOA) to predict the battery backup time. This algorithm can be based on the remaining battery capacity, solar panel energy generation, and the estimated power consumption of the home appliances.
Step 6: The energy generation and battery backup time has been done with the Firebase database.
Step 7: Test the application on an Android device to debug the lag issues and user friendly.
MONETIZATION ENERGY OPTIMIZATION ALGORITHM (MEOA)
Monetization Energy Optimization Algorithm (MEOA) that to minimize energy consumption and maximize profits in a building or facility. The algorithm works by identifying the optimal energy usage patterns based on occupancy patterns and building usage to achieve the desired level of comfort while minimizing energy consumption. The algorithm uses a combination of historical data and real-time data to optimize the energy usage of the building. MEOA is designed to take into account various parameters such as the building's occupancy, ambient temperature, and the energy consumption of various devices in the building. The algorithm works by predicting the energy demand and usage patterns of the houses based on these parameters and optimizes the energy consumption to minimize the cost of energy. MEOA also takes into account the energy pricing structure to maximize cost savings. The algorithm optimizes energy consumption by adjusting the operation of HVAC systems, lighting, and other energy-consuming devices in the building.
PROPOSED TIMESTAMP PREDICTION AND CONSUPTION ALGORITHM (TPCA)
Timestamp Prediction and Consumption algorithm (TPCA) is a data-driven algorithm used to predict the energy consumption based on time series data. The algorithm analyze the past energy usage patterns and predicting future energy usage based on the analysis the predictions can be used to optimize the energy consumption by adjusting the operation of HVAC systems, lighting, and other energy-consuming devices. TPCA utilizes a combination of statistical modeling, machine learning, and data analytics to predict energy consumption. The algorithm also uses real-time data to adjust the predictions and optimize energy consumption in real-time. The algorithm can be integrated with building automation systems to provide real-time monitoring and control of the building's energy consumption. The algorithm can also be used to predict energy demand and inform energy shortage decisions.

CONCLUSION

The main objective of the proposed Patent is to design and develop an efficient prototype for solar energy generation and optimization. There are many existing techniques like dual axis solar panel and sun tracking sensors mounted solar panels are still available, but the developed prototype is a cost efficient and increase the power generation duration of solar panels compared to the existing techniques. This Patent explores novel techniques for solar energy generation, optimization and as an efficient energy utilization with the help of monitorization model.A novel method have been proposed to increase the energy generation of solar panels by implementing the three main CLAIMs methodology.
The CLAIM-I accomplish with the collection of real-time LDR dataset and the pre- processing of LDR dataset to remove the noise data and to obtain the informative data to proceed with the proposed sunflower heliotropism algorithm.
The CLAIM-II is proposed to tilt the panel without any heavy energy consuming rotatory motors and sensors. The proposed SEE-SAW algorithm is to make the panel tilt according to the sun direction with the LDR dataset of CLAIM-I.
The CLAIM-III is proposed for the future prediction and the effective usage of energy - consumption and monitoring. The android tool have been developed in CLAIM-III. , Claims:CLAIM
PROBLEM STATEMENT

Due to the proliferation of electric vehicles and other appliances, the high demand for fossil fuels, the scarcity of coal, and efforts to reduce environmental hazards, electric usage is expected to become increasingly popular in the future decades. Sunlight is one of the best renewable resources, therefore solar panels are respectably producing power. However, the degree of energy produced fell short of expectations so some of the major challenges in solar energy optimization are listed below.

• Analyzing the intensity of light is a challenging problem.
• Radiant of solar panels is too difficult.
• Utilization of optimal electricity for operating the solar panel is combustible.
• Regulating energy optimization for household appliances is not feasible.
OBJECTIVES
The main objective of this research work is to design and develop a novel method for solar power generation and optimization for energy consumption using Machine learning techniques. The specific objectives of the research are,
CLAIM 1 : To eliminate the noisy data and identify the levels of light intensity using Sunflower Heliotropism Algorithm (SHA) with Light Dependent Resistor (LDR) sensor.
CLAIM 2 : To develop a novel prototype to gradient the solar panel and to produce solar energy generation with load-balancing edifice using Sea-Saw (SSA) algorithm.
CLAIM 3 : To develop an application tool for regulating household appliances remotely and make aware of energy optimization using Monetarization with Energy Optimization Algorithm (MEOA) and Timestamp Prediction and Consumption Behavior (TPCB).
CLAIM 1 : SUNFLOWER HELIOTROPISM ALGORITHM
In Phase-I, Sunflower Heliotropism Algorithm (SHA) With LDR Sensor has been implemented to detect the Light Intensity. Light Dependent Resistor (LDR) is a sensor that measures light intensity. It is made of a material that changes resistance when exposed to light. The resistance of an LDR decreases with increasing light intensity. Due to the disturbance like clouds and birds, the dataset contains noise or error data. Sunflower Heliotropism Algorithm (SHA) With LDR Sensor which aids to remove the noise in the data. The LDR dataset is also termed as a real-time dataset where information like highlight and lowlight intensity of lights with respective time stamps can be identified for further research progress. Instead of using continuous live tracking sensors this research work efficiently utilizes the dataset and reduced the energy conserved for live tracking sensors. Analysis can render a deep-seated perceptivity for appropriate understanding of different large-scale databases. When the noise removal was done with the dataset the exact information of light intensity with the respective time can be taken as output. This improves the quantity of energy generation duration based on deterministic and statistical properties. Studies related to acquaintance and developments of knowledge are also very proficient and are one of the first and foremost utilities in energy optimization and data mining techniques which help to know about the Energy-conscious scheduling heuristic method and Parallel bi-objective genetic algorithm have been compared with the proposed Sunflower Heliotropism Algorithm (SHA) is to cluster the light intensities at various levels.
The SHA algorithm produces better results with LDR sensors and achieves high-intensity of light clusters. Therefore, the proposed SHA is more efficient in accumulating the light intensity clusters than the existing algorithms such as the Multi-objectives Genetic Algorithm (MOGA), Parallel Bi-Objective Genetic Algorithm (PBGA), and regression analysis. The ultimate intention of this phase is to generate real time LDR dataset and take account of the proficient performance of the proposed algorithm (SHA) in the analysis of the high light intensity and low light intensity with the respective time from a large set of raw sensor data and also compared the proposed algorithm with traditional energy optimization algorithms of MOGA, PBGA. In accordance to measure the noise removal rate, classification accuracy performance measures were applied.
CLAIM 2 : SEE-SAW ALGORITHM
In phase II, the research work has incorporated a new design and algorithm to make the solar panels tilt towards the sun with a minimal amount of mechanical energy and coined the name See-Saw Algorithm (SSA) this algorithm has been proposed to gradient the solar panels to the respective angles which helps to increase the solar energy generation. A novel prototype has been developed to pose the solar panels in a row-wise approach which is connected with a load-balancing. Mathematical and mechanical equations have been applied for calculating the total mass of the solar panels with the help of a novel prototype. This prototype aids to quantify the level of weights at both ends. This iteration has to be performed periodically to increase the duration of solar energy generation. The results are measured with performance matrices and ascertain that the Sea-Saw (SSA) algorithm has achieved more accuracy in attaining the intensity of light to the fullest extent on the solar panels to increase the solar energy generation to a larger scope.
CLAIM 3 : MEOA & TPCA
In phase III, the outcomes of phase II have been analyzed using the Timestamp Prediction and Consumption Algorithm (TPCA) and Monetrization Energy Optimization Algorithm (MEOA) algorithm in order to analyze solar panels backup and future energy usage prediction. As a result, remote control of power devices and appliances is useful for household applications. The proposed method iterated on the generated dataset to determine the amount of energy consumed by each product cluster, and over time, periodic usage of the daily appliance will be taken into account. The proposed MEOA assists the user in managing anonymous, high-energy-consumption products. The efficiency and the energy generation level of the solar panel on cloudy and rainy days varies with the seasons and may go down in the production of energy. As a result, the Timestamp Prediction Consumption Behavior of Power Utilization TPCA algorithm helps to forecast future periods of low energy generation time stamps and alerts the user to be aware of energy optimization and utilization. Therefore, a unique way to forecast the future behavior of energy consumption is designed and proposed as MEOA and TPCA.