Multi-port converters are considered as exceeding earlier period decade owing to function in a combination of different energy sources in a single processing unit. Renewable energy sources are playing a significant role in the modern energy system with rapid development. In renewable sources like fuel combustion and solar energy, the generated voltages change due to their environmental changes. To develop energy resources, electric power generation involved huge awareness. The power and output voltages are plays important role in our work but it not considered in the existing system. For considering the power and voltage, Gaussian PI Controller-Maxpooling Deep Convolutional Neural Network Classifier (GPIC-MDCNNC) Model is introduced for the grid-connected renewable energy system. The input information is collected from two input sources. After that, input layer transfer information to hidden layer 1 fuzzy PI is employed for controlling voltage in GPIC-MDCNNC Model. Hidden layer 1 is transferred to hidden layer 2. Gaussian activation is employed for determining the output voltage with help of the controller. At last, the output layer offers the last value in GPIC-MDCNNC Model. The designed method was confirmed using one and multiple sources by stable and unpredictable input voltages. GPIC-MDCNNC Model increases the performance of grid-connected renewable energy systems by enhanced voltage value compared with state-of-the-art works. The control technique using GPIC-MDCNNC Model increases the dynamics of hybrid energy systems connected to the grid.
Renewable Energy (RE) is measured as significant with different energy. A grid-connected system is used considerably by higher diversity. RE is used in many applications namely Solar electric, or Photo Voltaic (PV), systems, power generation, wind turbines. With the improvement of modern and innovative inverter configurations, competence, size, load, consistency performance is enhanced. Dynamic modelling by Multiple Input Multiple Output (MIMO) controller was introduced in [
A robust blended Iterative Linear Quadratic Gaussian (ILQG) controller was designed in [
A three-phase multi-level multi-input power converter topology was introduced in [
The contribution of the paper is given by. The main aim of the GPIC-MDCNNC Model is to attain rated current and voltage for grid-connected RE systems. GPIC-MDCNNC Model, input information is collected from two input sources, namely ‘photovoltaic’ and ‘fuel cell’. The input layer transfers information to hidden layer 1 fuzzy PI regulates the speed for attaining the required voltage in GPIC-MDCNNC Model. The hidden layer 1 result is transferred to hidden layer 2. Gaussian activation function determines the output voltage with help of the controller. At last, the output layer offers the last value in GPIC-MDCNNC Model.
The paper is summarized by: Section 2 explains related works. Section 3 describes the proposed GPIC-DCNLC Model methodology by architecture diagram. Section 4 presents experimental settings as well as a discussion. Section 5 describes the conclusion of this paper.
The coordinated control strategy is designed in [
A real-time Hybrid Control System (HCS) was designed in [
Robust model predictive control method was designed in [
A switched Z-source converter was employed in [
STATic synchronous COMpensator (STATCOM) was used in [
In the present situation, there is a large power demand because of the increasing consumer load with the need for RE sources. The power supply is the main problem for hybrid systems. To maintain the efficiency of the hybrid system, power converters play a vital role in providing the regulated output voltage. Electric power is produced via fossil fuels. With the fast electricity and raise of energy disaster, it was essential for changing fossil fuels by RE. Between renewable resources, solar energy and wind power fascinate large interest because of simple attainment. In the power filter, Grid-connected voltage was employed by online uninterruptable power supplies and RE systems. Grid-connected PV is the electricity-producing solar PV power linked to the effectiveness of the grid. It is parallel to an electric utility grid. It is developed in capacities from hundred watts to tens of megawatts.
In power filter, Grid-connected voltage is employed, online uninterruptable power supply, and RE systems. Fuel Cells (FC), and wind turbines are employed to address the increasing electrical energy demand consistent with the significance of greenhouse gas reduction, RE sources like PV systems. The PV energy application has gained large attention because of the widespread global existence of solar energy. With new developments in solar energy technology, the cost of PV systems is reduced. DC–DC and DC–AC converters are essential for MPPT of renewable source and PV output conversion into AC power correspondingly. From controller design, a grid-connected system is a demanding task because of the cascade connection of multiple converters. Grid-connected inverters were employed for increasing the current quality and active power filtering in the distribution system.
Multi-Port Converters (MPCs) play an essential part to enhance MIMO utilization. MPC assumes a crucial job in interfacing and incorporating these vitality sources to supply the heaps. Every subclass is classified into three types, namely Multiple Input Single Output (MISO), Single Input Multiple Output (SIMO) and MIMO. The majority of specialists design the MISO converter to join different vitality sources at different voltage levels. MPCs provide financial procedures as well as enhance the performance of a system by multiple single converters in MIMO. MICs attained better performance with a combination of voltage sources used as EVs and Grid utilization. Multi-Output DC–DC converters received large awareness due to the minimum cost and compact size.
The circuit topology comprised two separate DC–DC boost converters for the highest power of input sources. Choppers provided the DC link of the grid-connected inverter. MIMO model of a system is used for the controller design. The Control system is the DC–DC boost converters and DC–AC inverter. System controls are power from PV
In GPIC-MDCNNC Model, Deep Learning is a subfield of machine learning inspired by the structure of artificial neural networks. Deep learning comprised multiple layers of neural networks to perform the data processing and computation task. Deep learning in Gaussian PI Controller Deep Convolutional Neural Learned Controller
Let us consider, the input source (i.e., a maximum output voltage from PV and FC) was measured as input as well as transferred to the input layer. GPIC-MDCNNC Model determines input value with help of weight vector and bias. It is represented by,
From
From
From
From
Fuzzy controllers are carried out with specifications under parameter variations and load disturbances. The controller parameters vary accordingly to maintain the desired performance in GPIC-MDCNNC Model. The system used inner loop current control and outer speed control. The outer speed control uses the Takagi-Sugeno-Kang Fuzzy PI controller and their parameters change with the operating conditions while loading.
From
From
From
Algorithm 1 explains the algorithmic procedure of PI Controller-based Maxpooling Deep Convolutional Neural Network Classifier for an energy source. Input layer transferred the data to hidden layer 1 fuzzy PI controller is employed for regulating voltage in GPIC-MDCNNC Model. Hidden layer 1 was sent to hidden layer 2. Gaussian activation function was employed for determining output voltage with help of the controller. At last, the output layer demonstrates the last value in GPIC-MDCNNC Model.
The experimental evaluation of the GPIC-MDCNNC method is implemented with MATLAB Simulink by 3.4 GHz Intel Core i3 processor, To enhance the performance of hybrid grid-connected RE system, 4 GB RAM, and Windows 7 platform was employed. The efficiency of the GPIC-MDCNNC Model is determined along with two different responses, namely steady-state response and transient response for metrics PV panel, Fuel cell stack, DC-link capacitor voltage, and grid current. The main parameters of the PV module and power circuit of values are given below in the
Parameter | Values |
---|---|
Number of cells (Ns) | 36 |
Standard light intensity (So) | 1000 W/m2 |
Reference temperature (Tref) | 25°C |
Series resistance (Rs) | 0.008 Ω |
Short-circuit current (ISC0) | 5 A |
Saturation current (IS0) | 3.8074 × 10−8 A |
Band energy (Eg) | 1.12 eV |
Ideality factor (A) | 1.2 |
Temperature coefficient (Ct) | 0.00065 A/°C |
Parameter | Values |
---|---|
Switching frequency of inverter and DC–DC converters (fs) | 20 kHz |
PV inductor (L1) | 1 mH |
FC inductor (L2) | 1 mH |
Grid inductor (Lg) | 1 mH |
DC-link capacitor (C) | 470 μF |
Voltage of PV source (V1) | 159.3 V |
Voltage of PV source (V2) | 150 V |
AC grid frequency (f) | 50 Hz |
AC grid voltage (Vg) | 110 V |
The simulation response of the proposed GPIC-MDCNNC Model for the grid-connected PV/FC hybrid energy system in the nominal condition is illustrated during steady-state operation. It is imagined that Ta = 25°C and S = 1000 W/m2. The reference current of the PV panel is equal to 5 A in MPP. The reference current of the FC stack is equal to 8 A. The designed MIMO controller is used to regulate the DC–DC converter. With reference value, DC link capacitor voltage is regulated. All active power obtained by the renewable source is injected into the grid.
In
In
In
In
In
In
Deep Learning is employed in the proposed GPIC-MDCNNC model for attaining higher output voltage and current. Input layer transfer information to hidden layer 1 through fuzzy PI and then information is transferred to hidden layer 2. It regulates the speed for attaining the required voltage in GPIC-MDCNNC Model. By using the Gaussian activation function output voltage is determined and achieves the required grid current. Finally GPIC-MDCNNC model achieves higher output voltage.
A new model termed GPIC-MDCNNC is introduced to attain higher output voltage for grid-connected RE systems. In GPIC-MDCNNC Model, the input layer transferred data information to hidden layer 1. The fuzzy PI controller regulates the output voltage of the GPIC-MDCNNC Model in hidden layer 1. Hidden layer 1 is transferred to hidden layer 2. Gaussian activation function determines the output voltage with help of the controller and attains the required grid current. Output layer demonstrates the last resultant value in GPIC-MDCNNC Model. The proposed controller succeeded to manage energy between microgrids under different scenarios. The proposed GPIC-MDCNNC Model is capable of MPPT for RE sources and injection of generated power. GPIC-MDCNNC method is evaluated with help of simulations in the MATLAB/Simulink toolbox. Based on the simulation results, the GPIC-MDCNNC Model is completely fast and stable at diverse operation points with zero steady-state error. The proposed GPIC-MDCNNC Model increases the output voltage and output current with minimum time. But, computational complexity was not reduced and Voltage was not regulated at the required level. In future work, Artificial intelligence and soft computing technique are used to attain higher output voltage for grid-connected RE systems with minimum computational complexity.