Dynamic Tariff for Synchronized Micro-Grids
In Pakistan, electric power is being generated from multiple resources i.e. thermal, hydell, solar, wind power plants etc. The power generation cost is different from all these sources depending upon the availability and load demand. Based on the diversity of the generation cost, National El
2025-06-28 16:32:14 - Adil Khan
Dynamic Tariff for Synchronized Micro-Grids
Project Area of Specialization Electrical/Electronic EngineeringProject SummaryIn Pakistan, electric power is being generated from multiple resources i.e. thermal, hydell, solar, wind power plants etc. The power generation cost is different from all these sources depending upon the availability and load demand. Based on the diversity of the generation cost, National Electric Power Regulatory Authority (NEPRA) defines a centralized fixed price per unit for all consumers and this price is implemented by Distribution Companies (DISCOs) in the form of Block Rate Tariff. But Block Rate Tariff is not a suitable choice for consumers as well as for DISCOs. Residential as well as Industrial consumers pay the same price no matter, they utilize power during peak hours or off-peak hours.
Our Final Year Project proposes a solution to this problem in the form of Dynamic Tariff. Dynamic tariff is actually dynamic with respect to time i.e. price per unit of electricity is changing on the basis of power generation, cost and demand. This type of tariff is much better as compared to fixed price tariff. Following this tariff, consumers will pay for the exact generation cost of electricity at that specific time. Dynamic tariff can be calculated after every hour, based on the details of generation capability and generation cost of multiple power plants and load demand at that time. Minimum Dynamic tariff can be achieved by applying machine learning techniques.
The FYP will be implemented on multiple Microgrids synchronized with conventional Utility Grid. One of the microgrid is of renewable source i.e. solar power plant in our case and other may be steam or diesel power plant. All these grids will contribute to a common bus bar and will feed a load. Load requirement will be met along with minimum cost of electricity with the Economic Dispatch Algorithm.
Moreover, our FYP includes Unit Price Forecasting i.e. consumer will be informed about the unit price calculated on hourly basis, for the upcoming hours or days through Web App and Android App. This forecasting will be done using some Machine Learning Algorithm on the basis of previously recorded annual data of generation and trend of load demand. This will help people to reduce their bill price by shifting their heavy load usage to time slots when unit price is low. There will be a change in life style of people if this is implemented. By this, some of the peak load will be shifted to off-peak time and peak losses will be lower.
Project ObjectivesOur objective is to:
- Introduce a fair type of tariff, suitable for both consumers as well as DISCOs.
- Provide cheapest possible cost of electricity to consumers.
- Predict unit price on hourly basis for the upcoming days and inform consumers through web and android app.
- Give consumers chance to reduce their bill prices by accommodating their loads to off-peak hours.
- Distribute the load curve evenly over the span of a day by providing low cost during off-peak hours.
- Improve health of the power distribution system by reducing maximum power demand on system during peak hours.
- Reduce power loss once the load curve is evenly distributed.
- Bring down peak load demand during summer and reduce the load shedding in the peak hour.
- Increase the lifetime of transformers with respect to their ratings by lessening peak load.
- Promote the concept of independent power plants and microgrids as these are trending research areas in Pakistan.
This project is primarily based on hardware part and a software part. Hardware part comprises of multiple microgrids with full power flow control tied to a single bus. Software part comprises of Machine Learning algorithm for price forecasting and power flow control.
As a first step in implementing this project multiple microgrids will be created and they will be synchronized with full power flow control. Solar panels as energy source will be used for microgrids. DC-DC chopper circuit will convert the output generated by the solar panels to variable DC voltage. A DC-DC chopper circuit is nothing but a high-speed switch which connects and disconnects the load from source at a high rate to get variable or chopped voltage at the output. Hence, a desired constant output voltage will be available at the output of chopper circuit. After necessary filtering process an inverter will produce AC voltage that will be tied to the slag bus via synchronizing unit.
STM microcontroller for each microgrid will be used that will not only control the output DC level of chopper circuit by generating PWM of appropriate duty cycle but also provide the full power control. The data of all the microgrids will be synchronized and an efficient machine learning algorithm will use this dataset for price forecasting and power flow management. Once ML model is trained on the dataset from the microgrids, it can provide the best possible working arrangement of the synchronized microgrids to provide the cheapest generated electricity having least power loss to the consumers consequently making an efficient power transmission and distribution system. The implementation of dynamic tariff with synchronized microgrids will not only reduce stress on the power transmission and distribution system but it will also reduce the electricity cost for the consumer and GENCOs will receive the true price of power generation. Hence, this system will be good for both GENCOs and consumers as well as for health of the distribution system.
A website and an android/iOS application will provide complete billing information as well as price forecasting details to the users. Automated notification system will be developed that will notify users about the billing and peak/off-peak hours to allow them to make full use of dynamic tariff system. Website will be developed in HTML, CSS, php and JavaScript while android studio will be used for creating android application. Oracle dataset will be purchased to store all the data and a server for website hosting. Firebase platform can be used for all the backend functionalities of web app and android application.
Benefits of the ProjectDynamic tariff is one of the emerging areas of research in the retail electricity industry. The idea of charging different prices at different times depending upon the load demand and generation cost can reduce peak load efficiently and can distribute load curve evenly consequently decreasing the power loss during peak-hours. Unit price forecasting plays an important role in scheduling load in dynamic tariff environment. Peaks in load profiles are the result of unregulated demand, and huge capacity addition is required to meet peak load. This peak load capacity stays idle during off-peak hours resulting in a loss of opportunity to use the generated energy efficiently. Hence, with the implementation of dynamic tariff people will be able to shift their high loads to off-peak hours which will lead to the reduction of electricity bill. In block rate tariff, prices remain unchanged irrespective of demand. However, the costs of generation to meet peak demands are high as compared to those for off-peak demand, since most peak time generating units have higher operating costs than base load units. Thus, the above-mentioned electricity prices do not reflect the true costs of generation and distribution. In addition to the reduction in peak demand, dynamic prices also provide each consumer with an opportunity to reduce his/her electricity bill at a constant consumption level, just by changing the consumption pattern by shifting the load.
This FYP will promote the culture of microgrids. A microgrid is a discrete energy system consisting of distributed energy sources with full power flow control and interconnected loads, within clearly defined electrical boundaries, capable of operating in parallel with, or independently from, the main power grid. The primary purpose of microgrids is to ensure local, reliable, and affordable energy security for urban and rural communities. One of the major benefits of microgrids is that use of renewable energy sources will lower the greenhouse gas emissions and also it will lower the stress on the transmission and distribution system. Microgrids differ from traditional electrical grids by providing a closer proximity between power generation and power use, resulting in efficiency increase and transmission power loss reduction.
Technical Details of Final DeliverableThe final working form of this FYP will be a complete compact and intelligent prototype that will be Hardware & Software Integrated Smart Scaled System. This system will be consisting of 3 major units.
- Hardware unit (MODEL of Microgrids)
- Smart Prediction Unit (Machine Learning Algorithm)
- Website & Android Application to remotely access the devices and real time information
- Hardware Unit:
It will comprise of 3 different models of microgrids feeding power to a load on the same power bus in a fully synchronized way. These grids will be either of same or different topology. One of them will be based upon renewable Energy source while others may be standalone DC or AC systems or Conventional grid. Each Microgrid will be having power flow control units that will be connected before the bus to control the power flow according the instruction sets provided by the prediction unit. There will be a synchronizing unit to synchronize all the three grids.
The single renewable source based microgrid will be having the following parts:
- Renewable Energy Collection Unit
- DC-DC (PUSH PULL CONVERTOR CIRCUIT)
- Filters
- Voltage Regulators / Controllers
- Invertors
- Power Flow Control Unit
- Synchronizer
- Smart Prediction unit:
This unit will be responsible for the prediction of future load demand that will be with the help of some machine learning algorithm, it will be estimating and providing the instruction sets to Microcontrollers about the quantity of power flow required by each of the microgrid. All of these estimations will be based upon the past information like, it will first forecast the future load by some regression techniques using the data of past and then it will fit out the generation curves according to the algorithm of most economic power. These generation curves will be defining that how much power from a specific microgrid/source is required to fulfil the load demands in the most economic manner.
- Website & Android application unit:
This unit will be behaving like a communication protocol between the consumer and the company, as soon as the new instructions will be implemented on the system these to mediums will be informing the consumer about the updated rates of electricity or the future rate of the electricity depending upon the nature of generation.
On the whole this system will be providing the most economic power to the consumer depending upon the specific conditions, with automated real-time updates of tariff, consumption and penalties.
Final Deliverable of the Project HW/SW integrated systemCore Industry Energy Other Industries IT Core Technology Clean TechOther Technologies Artificial Intelligence(AI)Sustainable Development Goals Affordable and Clean Energy, Decent Work and Economic Growth, Responsible Consumption and ProductionRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 80000 | |||
| FPGA Board Nexys A7 | Equipment | 1 | 24000 | 24000 |
| Raspberry Pi | Equipment | 1 | 6500 | 6500 |
| Stm 32f407 | Equipment | 1 | 4500 | 4500 |
| PCB boards and fabrications | Equipment | 1 | 4000 | 4000 |
| IGBTS | Equipment | 1 | 18000 | 18000 |
| Gate Drivers | Equipment | 1 | 5000 | 5000 |
| Online Server Fee | Equipment | 1 | 1500 | 1500 |
| Online Cloud Fee | Equipment | 1 | 2000 | 2000 |
| Basic electronic components and modules | Equipment | 1 | 4500 | 4500 |
| Miscellaneous | Miscellaneous | 1 | 10000 | 10000 |