Browsing by Author "Li, Zhengmao"

Now showing 1 - 16 of 16

Attention-based multi-task deep learning for short-term multi-energy load forecasting
(2024-06-10) Zhu, Xinghan
Insinööritieteiden korkeakoulu | Master's thesis
Finland produces 38.1 TWh of district heat annually, making it the leading producer of district heat per capita among the Nordic countries. Additionally, it generates 82 TWh of electricity per year, with 41.8% coming from renewable sources, placing it fifth among the Nordic countries. The country's significant role in district heating and its potential to increase renewable energy integration underscores the importance of an energy transition for optimal energy use and sustainable development. Although the energy transition in Finland is complex and influenced by multiple factors, effective decision-making is essential for facilitating this transformation. A key aspect of such decision-making occurs in the domain of day-ahead optimal dispatch within Multi-Energy Systems (MES). This process involves making forward-looking decisions on aspects such as electricity pricing, renewable energy integration, and operations for the following day. Accurate multi-energy short-term load forecasting is crucial in this context, as it provides the necessary information for optimal decision-making, thereby improv-ing efficiency, economic outcomes, and sustainability within MES. Therefore, this thesis seeks to establish an accurate Multi-energy Load Forecasting (MELF) framework using a comprehensive, machine learning-based approach. It validates the effectiveness of multi-task learning (MTL) in predicting multi-energy loads. Through three chapters of step-by-step validation, the study examines 1) traditional machine learning methods, 2) single task deep learning methods, and 3) MTL methods. The results demonstrate that MTL methods can effectively predict multi-energy loads, with Bidirectional GRU (BiGRU) as the best-performing model. An attention mechanism was employed within the MTL framework to avoid overfitting, addressing the challenge of managing a large number of features in multi-task learning objectives. Finnish community load data was used to validate the proposed frame-work. The MTL-BiGRU model significantly outperforms other models, achieving the lowest Mean Absolute Percentage Error (MAPE) and Root Mean Squared Error (RMSE) for electricity (0.5882%, 4.8618) and heat load (2.4660%, 11.3059) predictions, along with the highest R² values (0.9980, 0.9977). These results demonstrate superior accuracy, substantial error reduction, and excellent model fit.
Blockchain-enabled Carbon and Energy Trading for Network-Constrained Coal Mines with Uncertainties
(2023-07-01) Huang, Hongxu; Li, Zhengmao; Sampath, L. P.M.I.; Yang, Jiawei; Nguyen, Hung D.; Gooi, Hoay Beng; Liang, Rui; Gong, Dunwei
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
In this paper, a blockchain-enabled distributed market framework is proposed for the bi-level carbon and energy trading between coal mine integrated energy systems (CMIESs) and a virtual power plant (VPP) with network constraints. To maximize the profits of these two entities and describe their complicated interactions in the market, the bi-level trading problem is formulated as a Stackelberg game considering integrating the energy market and the 'cap-And-Trade' carbon market mechanism. Meanwhile, in the CMIES, energy recovery units and belt conveyors can be flexibly scheduled and the pumped hydroelectric storage in the VPP is scheduled for energy management. To tackle uncertainties from PV outputs, the joint trading, and the energy management is solved by the distributionally robust optimization (DRO) method. In addition, for participants' privacy, the alternating direction method of multipliers (ADMM)-based DRO algorithm is applied to solve the trading problem in a distributed framework. Further, the Proof-of-Authority (PoA) blockchain is deployed to develop a safe and anonymous market platform. Finally, case studies along with numerous comparison cases are conducted to verify the effectiveness of the proposed method. Simulation results indicate that the proposed method can effectively reduce the system operation cost and regional carbon emission, reduce the conservativeness and protect the privacy of each participant.
A Carbon Emission Allowance Bargaining Model For Energy Transactions Among Prosumers
(2024-07) Xiang, Yue; Qing, Guiping; Fang, Mengqiu; Li, Zhengmao; Yao, Haotian; Liu, Junyong; Guo, Zekun; Liu, Jichun; Zeng, Pingliang
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
The carbon pricing is the main issue of the carbon trading market for enabling cost-effective decarbonization in the energy networks. A nodal carbon pricing model is firstly proposed based on the sharing and integration of the intra-regional carbon emission allowance. In this regard, the game theory is introduced to construct a multi-agent carbon emission allowance bargaining model in this letter. The alternating direction multiplier method is adopted to solve the model considering the competitional burden and privacy-preserving. Numerical results demonstrate that it could significantly reduce the carbon emissions of regional energy networks and improve the economic benefits of prosumers.
A CCP-Based Distributed Cooperative Operation Strategy for Multi-Agent Energy Systems Integrated with Wind, Solar, and Buildings
(2024-07) Ding, Bing; Li, Zening; Li, Zhengmao; Xue, Yixun; Su, Jia; Jin, Xiaolong; Sun, Hongbin
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
To explore the bidirectional interaction between renewable energy and buildings in multi-agent energy systems, this paper proposes a distributed cooperative operation strategy for multi-agent energy systems integrated with wind, solar, and buildings based on chance-constrained programming (CCP). First, the multi-agent energy system integrated with wind, solar, and buildings is comprehensively modeled with detailed electric and thermal characteristics for flexibility enhancement. Then for maximizing the profits of the cooperative energy system and each engaged agent, a Nash bargaining model is presented and is divided into two subproblems: the coalition income and the power payment. To preserve the privacy of agents, the adaptive alternating direction method of multipliers (ADMM) is exploited to solve both subproblems. Meanwhile, the CCP method is applied to address diverse uncertainties from wind and solar power generation as well as outdoor temperature. Finally, the effectiveness of the proposed strategy is validated. The simulation results show that, besides the privacy of information among all agents being well preserved, our strategy enhances the profits not only for the energy system but also for all engaged agents.
Cooperative Operation of Renewable-Integrated Multi-Energy Microgrids Under Dynamic Rolling Horizon Strategy
(2023) Li, Zhengmao; Kyyrä, Jorma; Xu, Yan; Zhao, Tianyang; Wang, Yunqi
A4 Artikkeli konferenssijulkaisussa
In this paper, a cooperative operation method is proposed in a multi-energy microgrid with high penetration levels of renewable energy sources. To handle uncertainties rising from wind turbines and photovoltaic cells, the rolling optimization approach is thus applied to achieve online multi-energy management with the constantly updated information. Through the effective coordination of energy markets, multi-energy networks, energy storage systems, and generators, a reliable and economic operation scheme is fulfilled. At last, to show the effectiveness of our proposed method, a case study with two compassion cases for the operation of multi-energy microgrids is done. The simulation results indicate that our method is more cost-effective for the cooperative operation of renewable-integrated multi-energy microgrids under uncertainty sources.
Electric Vehicle Charging Planning: A Complex Systems Perspective
(2024-08-21) Zhao, Alexis Pengfei; Li, Shuangqi; Li, Zhengmao; Wang, Zhaoyu; Fei, Xue; Hu, Zechun; Alhazmi, Mohannad; Yan, Xiaohe; Wu, Chenye; Lu, Shuai; Xiang, Yue; Xie, Da
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
In this paper, we introduce an innovative framework for the strategic planning of electric vehicle (EV) charging infrastructure within interconnected energy-transportation networks. By harnessing the small-world network model and the advanced optimization capabilities of the Non-dominated Sorting Genetic Algorithm III (NSGA-III), we address the complex challenges of station placement and network design. Our application of the small-world theory ensures that charging stations are optimally interconnected, fostering network resilience and ensuring consistent service availability. We approach the infrastructure planning as a multi-objective optimization task with NSGA-III, focusing on cost minimization and the enhancement of network resilience and connectivity. Through simulations and empirical case studies, we demonstrate the efficacy of our model, which markedly improves the reliability and operational efficiency of EV charging networks. The findings of this study significantly advance the integrated planning and operation of energy and transportation networks, offering insightful contributions to the domain of sustainable urban mobility.
Enhancing Resilience of Reconfigurable Power-Water Systems with Mobile Distributed Generators and High-Proportional Renewables
(2024) Yang, Yesen; Li, Zhengmao; Zhang, Guangxiao; Costa, Alberto; Lo, Edmond Y.
A4 Artikkeli konferenssijulkaisussa
The intertwined interdependencies existing in power-water systems (PWS) increase the risk of cascading failures during post-interruption scenarios and affect the overall resiliency. Towards a more resilient operation of damaged PWS, this paper presents a resilience enhancement methodology to improve serviceability with mobile distributed generators (MDGs) and high-proportional renewables (HPR). First, the PWS is comprehensively modeled with component mechanisms and flow constraints. The interdependencies, including the power needs for water components, are modeled at component level. Second, various resources, such as extensively installed solar HPR units and MDGs, are coordinated to supply damaged PWS and reduce unsupplied loads. Reconfigurability of power distribution networks is incorporated in the proposed method. It is to adjust PDN topology by coordinating the behavior of switches and leverage HPR and MDG for optimally allocating energy. Third, the model and enhancement measures are formulated into mixed-inter linear programming to facilitate efficient solving. The developed method is applied to a benchmark PWS with 33 power buses and 25 water nodes. The simulation results demonstrate the effectiveness of our method.
Generation of input spectrum for electrolysis stack degradation test applied to wind power PEM hydrogen production
(2024-08) Xu, Yanhui; Li, Guanlin; Gui, Yuyuan; Li, Zhengmao
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
Hydrogen production by proton exchange membrane electrolysis has good fluctuation adaptability, making it suitable for hydrogen production by electrolysis in fluctuating power sources such as wind power. However, current research on the durability of proton exchange membrane electrolyzers is insufficient. Studying the typical operating conditions of wind power electrolysis for hydrogen production can provide boundary conditions for performance and degradation tests of electrolysis stacks. In this study, the operating condition spectrum of an electrolysis stack degradation test cycle was proposed. Based on the rate of change of the wind farm output power and the time-averaged peak-valley difference, a fluctuation output power sample set was formed. The characteristic quantities that played an important role in the degradation of the electrolysis stack were selected. Dimensionality reduction of the operating data was performed using principal component analysis. Clustering analysis of the data segments was completed using an improved Gaussian mixture clustering algorithm. Taking the annual output power data of wind farms in Northwest China with a sampling rate of 1 min as an example, the cyclic operating condition spectrum of the proton-exchange membrane electrolysis stack degradation test was constructed. After preliminary simulation analysis, the typical operating condition proposed in this paper effectively reflects the impact of the original curve on the performance degradation of the electrolysis stack. This study provides a method for evaluating the degradation characteristics and system efficiency of an electrolysis stack due to fluctuations in renewable energy.
Household electricity bill reduction by utilizing a home energy management system
(2024-05-20) Sinkkonen, Jussi
Sähkötekniikan korkeakoulu | Master's thesis
The recent energy crisis has raised concerns among homeowners about the unpredictable electricity costs. Many seek ways to reduce electricity expenses during peak-price periods and explore different local electricity production technology options, such as solar panels. Electric vehicles can play a role in this by exchanging energy with households, thus helping to manage electricity usage efficiently. In Nordic countries, a big issue is that household electricity demand peaks during winter when solar panels output is low. To tackle this, smart coordination of household electricity flow is necessary. This work aimed to evaluate the potential of smart electricity coordination in Nordic households equipped with solar panels, electrical energy storage, and electric vehicles capable of bidirectional energy exchange. A case example for the evaluation was performed on a household located in southern Finland, where additional home electrical energy storage, a bidirectional charger, and a solar system were artificially added. An algorithm called a home energy management system was developed to optimize daily electricity flows. This algorithm excluded home appliance control. The daily costs were simulated by using measured detached house consumption, production, and electricity price data from 2022. This simulation found that the produced algorithm led to a 44.97 % reduction in yearly electricity cost compared to the non-optimized case. An expected system lifetime of 10 years could lead to revenue of 1 934.13 €. Although this work is directly suitable only for one Finnish household, it demonstrates the potential for reducing yearly household electricity costs in Nordic environments with the adoption of home energy management systems.
Joint Planning of Utility-Owned Distributed Energy Resources in an Unbalanced Active Distribution Network Considering Asset Health Degradation
(2024-07-01) Leng, Ruoxuan; Li, Zhengmao; Xu, Yan
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
Rapid integration of distributed energy resources (DERs) in active distribution networks (ADNs) necessitates advanced planning methods to optimally determine the size, site, and installation time of DERs. However, existing approaches often assume balanced networks and neglect health degradation of DER assets, limiting the accuracy and practicality of the planning results. This paper proposes a new planning method for utility-owned distributed generators (DGs) and energy storage systems (ESSs) in an unbalanced ADN considering asset health degradation. First, the three-phase branch flow is modeled for unbalanced characteristics of ADNs, and host DERs separately in different phases. Then, based on the Wiener degradation process, the aging path of each DG unit is modeled to estimate its available capacity along with service time; the ESS aging is modeled to reflect the degradation cost during charging and discharging. Finally, a copula-based stochastic programming method is presented considering the correlations between renewables and power demands. The inclusion of market volatility in electricity price uncertainty further enhances planning realism. Numerical case studies on an IEEE-34 bus three-phase ADN demonstrate the effectiveness and advantages of the proposed method.
Resilience enhancement of a multi-energy distribution system via joint network reconfiguration and mobile sources scheduling
(2024) Shi, Zhao; Xu, Yan; Li, Zhengmao; Xie, Dunjian; Ghias, Amer M. Y. M.
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
This paper proposes a new resilience enhancement strategy for a multi-energy distribution system (MDS) through coordinated reconfiguration of the coupled-power-and-heat network, scheduling of mobile power sources (MPSs) as well as the dispatch of stationary distributed energy resources (DERs) including renewable energy generation and energy storage systems (ESSs). Firstly, a resilient-oriented joint network reconfiguration model is proposed for a radial MDS after the disaster. The heat network is formulated as a quasi-linear flow model which is independent of the flow rate and temperature for alleviating computation burdens and reconfiguration. Then, MPSs are dynamically dispatched in the MDS for power exchange with MPS stations, to support the restoration of both the power and heat loads. Thirdly, the whole restoration model is linearized as a mixed-integer linear program (MILP) problem with heterogeneous temporally-spatially, operational, and topology constraints. Finally, numerical case studies are done to validate the effectiveness of the proposed strategy.
Resilience-oriented Operation of Power Distribution Networks with Line Hardening and Comprehensive Reconfiguration Measures
(2023-11-03) Li, Zhengmao; Shi, Zhao; Ruan, Guangchun; Lin, Yuzhang; Zhao, Jin
A4 Artikkeli konferenssijulkaisussa
With the frequent occurrence of extreme events like natural disasters and man-made attacks, the resilience concept is attracting worldwide research attention. Thus, this paper proposes a resilient operation model for the power distribution network (PDN) to recover load and limit economic loss to the greatest possible extent. First, the line hardening measure is applied to strengthen the PDN to avoid the serious breakdown of distribution lines. In addition, a comprehensive single commodity flow-based reconfiguration approach is used which not only sets the buses connecting generation sources but also those at the end of broken lines as the slack buses. Then, the radiality of PDNs can still be guaranteed after reconfiguration. The resilience-oriented operation problem can be originally formulated as a nonlinear programming one while efficient linearization methods are applied to reduce computational burdens. Hence, the operation problem is reformulated as a mixed-integer linear programming one which will enjoy favorable solution performances. Finally, case studies on an IEEE 33 PDN including comparison cases with traditional benchmarks are done to show the effectiveness of our method.
Robust Coordinated Planning of Multi-Region Integrated Energy Systems With Categorized Demand Response
(2024) Dong, Yingchao; Li, Zhengmao; Zhang, Hongli; Wang, Cong; Zhou, Xiaojun
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
In this paper, categorized demand response (DR) programs are proposed to address the coordinated planning problem in multi-region integrated energy systems (MRIESs). The categorized DR programs comprise a discrete manufacturing production model for industrial areas, a real-time pricing-based DR program for commercial areas, and diverse operational tasks for various electrical appliances in residential areas. Subsequently, the detailed DR model is leveraged to minimize the operation cost and gas emissions in a renewable-integrated MRIES considering the uncertainties from wind and solar power. Then, a flexible adjustable robust optimization (FARO) approach is presented to deal with all uncertainty sources. The FARO approach aims to ensure the safe operation of the MRIES against any uncertainty while meeting predefined performance objectives. Furthermore, a bi-level solution algorithm is designed by combining the stochastic dichotomy method and the column-and-constraint generation (C&CG) algorithm to solve our coordinated planning model. Finally, case studies are conducted on a practical MRIES in Changsha, China. Experimental results indicate the effectiveness of the categorized DR programs in adjusting allocable resources to maximize holistic system profits. Besides, compared to the commonly used information-gap decision theory (IGDT) method, our FARO approach can maintain the optimality of the solution while reducing conservatism.
Robust optimization for integrated production and energy scheduling in low-carbon factories with captive power plants under decision-dependent uncertainty
(2025-02) Lv, Quanpeng; Wang, Luhao; Li, Zhengmao; Song, Wen; Bu, Fanpeng; Wang, Linlin
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
Low-carbon factories with captive power plants represent a new industrial microgrid paradigm of energy conservation and emission reduction in many countries. However, one of the most common challenges of low-carbon management is the joint regulation of factory production and power plant operations under uncertainty. To meet this challenge, a robust optimization-based integrated production and energy (IPE) scheduling approach is proposed in this paper. Firstly, a two-stage adaptive robust optimization model is established to cover all possible realizations of decision-independent uncertainties (e.g. market demands and output power of renewable sources) and decision-dependent uncertainties (e.g. carbon emission densities depending on the choice of production lines). Secondly, a novel parametric column-and-constraint generation algorithm is utilized to derive robust scheduling schemes. The non-trivial scenarios of decision-dependent uncertainties identified in the subproblem are parametrically characterized based on Karush–Kuhn–Tucker conditions, which can be included in the master problem. Finally, simulations on different cases are conducted to test the rationality and validity of the proposed approach. Compared with the separate scheduling of production and energy, IPE scheduling may increase production and energy costs to ensure the robustness of the resulting schemes. Moreover, the proposed approach can mitigate the impacts of uncertainties on IPE scheduling without significantly increasing the computational complexity.
A Two-stage Multi-agent Deep Reinforcement Learning Method for Urban Distribution Network Reconfiguration Considering Switch Contribution
(2024) Gao, Hongjun; Jiang, Siyuan; Li, Zhengmao; Wang, Renjun; Liu, Youbo; Liu, Junyong
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
With the ever-escalating scale of urban distribution networks (UDNs), the traditional model-based reconfiguration methods are becoming inadequate for smart system control. On the contrary, the data-driven deep reinforcement learning method can facilitate the swift decision-making but the large action space would adversely affect the learning performance of its agents. Consequently, this paper presents a novel multi-agent deep reinforcement learning method for the reconfiguration of UDNs by introducing the concept of 'switch contribution'. First, a quantification method is proposed based on the mathematical UDN reconfiguration model. The contributions of controllable switches are effective quantified. By excluding the controllable switches with low contributions during network reconfiguration, the dimensionality of action space can be significantly reduced. Then, an improved QMIX algorithm is introduced to improve the policy of multiple agents by assigning the weights. Besides, a novel two-stage learning structure based on a reward-sharing mechanism is presented to further decompose tasks and enhance the learning efficiency of multiple agents. In the first stage, agents control the switches with higher contributions while switches with lower contributions will be controlled in the second stage. During the two-stage process, the proposed reward-sharing mechanism could guarantee a reliable UND reconfiguration and the convergence of our learning method. Finally, numerical results based on a practical 297-node system are performed to validate our method's effectiveness.
Two-stage Robust Operation of Electricity-Gas-Heat Integrated Multi-Energy Microgrids Considering Heterogeneous Uncertainties
(2024-10-01) Zhang, Rufeng; Chen, Yan; Li, Zhengmao; Jiang, Tao; Li, Xue
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
With the widespread adoption of combined heat and power and power-to-heat technologies, multi-energy microgrids (MEMGs) have been garnering significant research attention from both industry and academia. However, dealing with uncertainties from renewable energy and load and coordinating multiple energy carriers are the main challenges for MEMG operation. In this regard, a two-stage robust operation method of electricity-gas-heat integrated MEMGs considering heterogeneous uncertainties is proposed in this paper. First, network models for an electricity-gas-heat-based distribution-level MEMG are formulated considering the dynamic characteristics of gas and heat networks. Then, the power-to-hydrogen-and-heat unit and ladder-type carbon trading mechanism are introduced to reduce the curtailment of wind power and carbon emissions. Further, a two-stage robust optimization (TSRO) method is applied to tackle uncertainties of wind power and load under extreme scenarios in the MEMG operation by iteratively solving the operation problem with the column and constraint generation (C&CG) algorithm. Finally, case studies are conducted to verify our proposed method, demonstrating that it can reduce the multi-energy supply cost while the stepped carbon trading mechanism can also significantly reduce carbon emissions.