Intelligent renewable energy storage and management system for rural household
The battery in conventional standalone photovoltaic (PV) system frequently undergoes deep cycles and irregular charging patterns, which can significantly reduce the battery lifetime and increase the replacement cost of the system. Battery-Supercapacitor Hybrid Energy Storage System (HESS) is the mos...
| Main Author: | |
|---|---|
| Format: | Thesis (University of Nottingham only) |
| Language: | English |
| Published: |
2020
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/58999/ |
| _version_ | 1848799576180391936 |
|---|---|
| author | Chong, Lee Wai |
| author_facet | Chong, Lee Wai |
| author_sort | Chong, Lee Wai |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | The battery in conventional standalone photovoltaic (PV) system frequently undergoes deep cycles and irregular charging patterns, which can significantly reduce the battery lifetime and increase the replacement cost of the system. Battery-Supercapacitor Hybrid Energy Storage System (HESS) is the most promising solution to prolong the lifespan of the battery. The control strategy is implemented to distribute the resources of the HESS based on the real-time operating conditions. Despite the intelligent control strategy, such as Fuzzy Logic Controller (FLC), is more effective than the classical control strategies, there is limited studies on intelligent control strategy that is optimized based on predicted power demand. Also, it is challenging to integrate the intelligent control strategy with high computation complexity into an actual system with commercial non-programmable charge controller. Therefore, two intelligent control strategies, namely Particle Swarm Optimization (PSO) optimized FLC (PSO-optimized FLC) and PSO-optimized Power Distribution Algorithm (PSO-PDA), are proposed to prolong the battery lifespan in a standalone PV system with Battery-Supercapacitor HESS. The objectives of this research work are the development of prediction models for very short term prediction of power demand using basic features, the development of PSO-optimized FLC and PSO-PDA, as well as the implementation of PSO-PDA in actual standalone PV system with Battery-Supercapacitor HESS.
The PSO-optimized FLC is novel in terms of its structure as a moving average filter is used to first extract the high frequency power from the power demand then subsequently a FLC decides the sharing ratio of the supercapacitor and the battery. The membership functions of the FLC are optimized using Particle Swarm Optimization (PSO) based on predicted power demand to reduce the battery peak power. It outperforms conventional systems in terms of battery lifetime improvement (12.81 %) and peak power reduction (64.26 %). However, this system is impractical for real-life implementation as optimizing the membership functions requires high optimization complexity.
An alternative novel control strategy, PSO-PDA, is developed to achieve the research aim. The parameters of power distribution algorithm are optimized every minute based on predicted h-mins (h=30, h=60, and h=90) ahead power demand and state-of-charge of the supercapacitor to mitigate peak demand in battery power. With short optimization interval, it can compensate for the prediction error of power demand from basic prediction models and adapt to the varying power demand. The PSO-PDA outperforms the PSO-optimized FLC, in term of battery lifetime improvement (up to 22 %) and battery peak demand reduction (up to 57 %) and requires significantly shorter optimization time per optimization (not more than 20.647 s) than PSO-optimized FLC.
Moreover, the PSO-PDA requires only one day of pre-training to be fully implemented in an actual 2-kW rated standalone PV system with Battery-Supercapacitor HESS. Also, it can be tuned to work with the commercial non-programmable charge controller. The experimental results show that the PSO-PDA can improve the battery lifetime by 16.50 % and reduce the battery peak power by 58.58 %. Moreover, the experimental results are compared with simulation results where the findings highlight that the results of the Simulink model only overestimate the battery lifetime improvement by 7.66 %. |
| first_indexed | 2025-11-14T20:37:51Z |
| format | Thesis (University of Nottingham only) |
| id | nottingham-58999 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T20:37:51Z |
| publishDate | 2020 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-589992025-02-28T14:39:10Z https://eprints.nottingham.ac.uk/58999/ Intelligent renewable energy storage and management system for rural household Chong, Lee Wai The battery in conventional standalone photovoltaic (PV) system frequently undergoes deep cycles and irregular charging patterns, which can significantly reduce the battery lifetime and increase the replacement cost of the system. Battery-Supercapacitor Hybrid Energy Storage System (HESS) is the most promising solution to prolong the lifespan of the battery. The control strategy is implemented to distribute the resources of the HESS based on the real-time operating conditions. Despite the intelligent control strategy, such as Fuzzy Logic Controller (FLC), is more effective than the classical control strategies, there is limited studies on intelligent control strategy that is optimized based on predicted power demand. Also, it is challenging to integrate the intelligent control strategy with high computation complexity into an actual system with commercial non-programmable charge controller. Therefore, two intelligent control strategies, namely Particle Swarm Optimization (PSO) optimized FLC (PSO-optimized FLC) and PSO-optimized Power Distribution Algorithm (PSO-PDA), are proposed to prolong the battery lifespan in a standalone PV system with Battery-Supercapacitor HESS. The objectives of this research work are the development of prediction models for very short term prediction of power demand using basic features, the development of PSO-optimized FLC and PSO-PDA, as well as the implementation of PSO-PDA in actual standalone PV system with Battery-Supercapacitor HESS. The PSO-optimized FLC is novel in terms of its structure as a moving average filter is used to first extract the high frequency power from the power demand then subsequently a FLC decides the sharing ratio of the supercapacitor and the battery. The membership functions of the FLC are optimized using Particle Swarm Optimization (PSO) based on predicted power demand to reduce the battery peak power. It outperforms conventional systems in terms of battery lifetime improvement (12.81 %) and peak power reduction (64.26 %). However, this system is impractical for real-life implementation as optimizing the membership functions requires high optimization complexity. An alternative novel control strategy, PSO-PDA, is developed to achieve the research aim. The parameters of power distribution algorithm are optimized every minute based on predicted h-mins (h=30, h=60, and h=90) ahead power demand and state-of-charge of the supercapacitor to mitigate peak demand in battery power. With short optimization interval, it can compensate for the prediction error of power demand from basic prediction models and adapt to the varying power demand. The PSO-PDA outperforms the PSO-optimized FLC, in term of battery lifetime improvement (up to 22 %) and battery peak demand reduction (up to 57 %) and requires significantly shorter optimization time per optimization (not more than 20.647 s) than PSO-optimized FLC. Moreover, the PSO-PDA requires only one day of pre-training to be fully implemented in an actual 2-kW rated standalone PV system with Battery-Supercapacitor HESS. Also, it can be tuned to work with the commercial non-programmable charge controller. The experimental results show that the PSO-PDA can improve the battery lifetime by 16.50 % and reduce the battery peak power by 58.58 %. Moreover, the experimental results are compared with simulation results where the findings highlight that the results of the Simulink model only overestimate the battery lifetime improvement by 7.66 %. 2020-02-22 Thesis (University of Nottingham only) NonPeerReviewed application/pdf en arr https://eprints.nottingham.ac.uk/58999/1/Thesis%20-%20Chong%20Lee%20Wai.pdf Chong, Lee Wai (2020) Intelligent renewable energy storage and management system for rural household. PhD thesis, University of Nottingham. renewable energy storage battery power supercapacitor rural household |
| spellingShingle | renewable energy storage battery power supercapacitor rural household Chong, Lee Wai Intelligent renewable energy storage and management system for rural household |
| title | Intelligent renewable energy storage and management system for rural household |
| title_full | Intelligent renewable energy storage and management system for rural household |
| title_fullStr | Intelligent renewable energy storage and management system for rural household |
| title_full_unstemmed | Intelligent renewable energy storage and management system for rural household |
| title_short | Intelligent renewable energy storage and management system for rural household |
| title_sort | intelligent renewable energy storage and management system for rural household |
| topic | renewable energy storage battery power supercapacitor rural household |
| url | https://eprints.nottingham.ac.uk/58999/ |