-
Members: FreePES
IEEE Members: $25.00
Non-members: $40.00Pages/Slides: 34
Panel
14 Sep 2022
This panel session contains the following presentations:
1. Combination of PSO and TS-F for Electricity Load Management Optimization: A Case Study of Poultry Sector
Malaysia's energy consumption is quickly growing as the country advances along the path of the Industrial Revolu- tion 4.0. Peak periods necessitate greater energy generation, and as a result, the cost is higher than during off-peak periods. It is for this reason that the demand side management (DSM) approach, utilizing the demand response (DR) scheme, was created in order to adjust the demand profile via the implementation of various strategies of actions. The goal of this proposed work is to enhance the power consumption profile of the industrial sector, and perhaps to examine the relevance of energy cost savings using an optimization approach in this sector. In this study, Particle Swarm Optimization (PSO), a bio-inspired approach, was used to optimize the demand profile pattern of the load shifting technique under the Enhance Time of Use (ETOU) tariff scheme. The results of the tests have been endorsed by the use of six (6) cases, which are follows: (1) conventional method for establishing baseline the existing E1 flat tariff rate; (2) baseline of the existing E1 flat tariff rate with Time Series Forecasting (TS-F); (3) E1 ETOU tariff rate without any optimization technique; (4) E1 ETOU tariff rate with TS-F data without optimization technique; (5) E1 ETOU tariff rate with PSO optimization technique without TS-F data; and (6) Combination of PSO optimization technique and TS-F data, respectively. Following that, the statistically significant simulation result of operating profit increase through 24-hour power usage has been thoroughly examined. It was discovered that the proposed strategy resulted in a reduction in the cost of electrical energy consumption across all price zones. Manufacturers are expected to gain from the findings of this study, which will aid them in transitioning to the ETOU tariff and will also help the national Demand Side Management (DSM) initiative programme.
2. Analysis of Fast Method of Electricity Prices in Scenarios of High Penetration of Renewable Energy Resources
In order to realize a carbon-neutral society in 2050, it is necessary to ensure the large-scale introduction of renewable energy without fail. Furthermore, from the viewpoint of the investment in power plants in the future, electricity market prices are also important factors. Therefore, various future scenarios should be analyzed to understand whether it will be possible to achieve equilibrium between electricity supply and demand and secure flexibility. In order to investigate future scenarios, the Unit Commitment (UC) method has been used in many studies. However, the analysis of the UC problem using the Mixed Integer Linear Programming (MILP) method has a heavy computational burden. This paper proposes a method to reduce the calculation time significantly. In the proposed method, the binary variables in the MILP are replaced by real variables in the UC and calculated using the Linear Programming Relaxation method. In the numerical experiment, the calculation accuracy for market prices is validated for scenarios of high penetration of renewable energy.
3. Day Ahead Market Price Scenario Generation Using a Combined Quantile Regression Deep Neural Network and a Non-Parametric Bayesian Network
In this manuscript we propose a methodology to generate electricity price scenarios from probabilistic forecasts. Using a Combined Quantile Regression Deep Neural Network, we forecast hourly marginal price distribution quantiles for the DAM on which we fit parametric distributions. A Non-parametric Bayesian Network (BN) is applied to sample from these distributions while using the observed rank-correlation in the data to condition the samples. This results in a methodology that can create an unbounded amount of price-scenarios that obey both the forecast hourly marginal price distributions and the observed dependencies between the hourly prices in the data. The BN makes no assumptions on the marginal distribution, allowing us to flexibly change the marginal distributions of hourly forecasts while maintaining the dependency structure.
1. Combination of PSO and TS-F for Electricity Load Management Optimization: A Case Study of Poultry Sector
Malaysia's energy consumption is quickly growing as the country advances along the path of the Industrial Revolu- tion 4.0. Peak periods necessitate greater energy generation, and as a result, the cost is higher than during off-peak periods. It is for this reason that the demand side management (DSM) approach, utilizing the demand response (DR) scheme, was created in order to adjust the demand profile via the implementation of various strategies of actions. The goal of this proposed work is to enhance the power consumption profile of the industrial sector, and perhaps to examine the relevance of energy cost savings using an optimization approach in this sector. In this study, Particle Swarm Optimization (PSO), a bio-inspired approach, was used to optimize the demand profile pattern of the load shifting technique under the Enhance Time of Use (ETOU) tariff scheme. The results of the tests have been endorsed by the use of six (6) cases, which are follows: (1) conventional method for establishing baseline the existing E1 flat tariff rate; (2) baseline of the existing E1 flat tariff rate with Time Series Forecasting (TS-F); (3) E1 ETOU tariff rate without any optimization technique; (4) E1 ETOU tariff rate with TS-F data without optimization technique; (5) E1 ETOU tariff rate with PSO optimization technique without TS-F data; and (6) Combination of PSO optimization technique and TS-F data, respectively. Following that, the statistically significant simulation result of operating profit increase through 24-hour power usage has been thoroughly examined. It was discovered that the proposed strategy resulted in a reduction in the cost of electrical energy consumption across all price zones. Manufacturers are expected to gain from the findings of this study, which will aid them in transitioning to the ETOU tariff and will also help the national Demand Side Management (DSM) initiative programme.
2. Analysis of Fast Method of Electricity Prices in Scenarios of High Penetration of Renewable Energy Resources
In order to realize a carbon-neutral society in 2050, it is necessary to ensure the large-scale introduction of renewable energy without fail. Furthermore, from the viewpoint of the investment in power plants in the future, electricity market prices are also important factors. Therefore, various future scenarios should be analyzed to understand whether it will be possible to achieve equilibrium between electricity supply and demand and secure flexibility. In order to investigate future scenarios, the Unit Commitment (UC) method has been used in many studies. However, the analysis of the UC problem using the Mixed Integer Linear Programming (MILP) method has a heavy computational burden. This paper proposes a method to reduce the calculation time significantly. In the proposed method, the binary variables in the MILP are replaced by real variables in the UC and calculated using the Linear Programming Relaxation method. In the numerical experiment, the calculation accuracy for market prices is validated for scenarios of high penetration of renewable energy.
3. Day Ahead Market Price Scenario Generation Using a Combined Quantile Regression Deep Neural Network and a Non-Parametric Bayesian Network
In this manuscript we propose a methodology to generate electricity price scenarios from probabilistic forecasts. Using a Combined Quantile Regression Deep Neural Network, we forecast hourly marginal price distribution quantiles for the DAM on which we fit parametric distributions. A Non-parametric Bayesian Network (BN) is applied to sample from these distributions while using the observed rank-correlation in the data to condition the samples. This results in a methodology that can create an unbounded amount of price-scenarios that obey both the forecast hourly marginal price distributions and the observed dependencies between the hourly prices in the data. The BN makes no assumptions on the marginal distribution, allowing us to flexibly change the marginal distributions of hourly forecasts while maintaining the dependency structure.
Chairs:
Mohamed Shaaban
