Mengshuo Jia

Last Name

Jia

First Name

Mengshuo

ORCID

0000-0002-2027-5314

Show all metadata

Search Results

Publications 1 - 10 of 30

A Fault Detection Scheme for PV Modules in Large Scale PV Stations With Complex Installation Conditions
Cao, Qianni; Jia, Mengshuo; Shen, Chen (2022)
Proceedings of the CSEE
Aiming at photovoltaic power plants with complex installation conditions, we proposed a method based on the comparison of working conditions between photovoltaic modules to achieve fault detection. First, to consider accuracy and simplicity, an analytical and linear model of module output was introduced. Moreover, a characteristic value is proposed, together with its extracting, calculation and probability analysis method. On this basis, a fault detection method for modules in PV stations, especially with complex installation conditions, is designed. Finally, the effectiveness of this method is verified by simulation. Specially, the influence of solving characteristic value based on linear model on the accuracy of fault detection is discussed, and a method to improve the accuracy of the fault judgment of this method is proposed.
User Manual for DALINE 1.1.5
Jia, Mengshuo; Chan, Wen Yi; Hug, Gabriela (2024)
Power flow linearization is a fundamental technique in power system operations, crucial for both academic research and industry applications. Despite its importance, many advanced linearization methods, particularly those driven by data, remain largely inaccessible, neither open-sourced nor integrated into widely used software platforms. To address this gap, we introduce Daline, an open-source MATLAB toolbox specifically designed for the power systems community. Daline provides a comprehensive suite of 57 linearization methods, including 53 data-driven techniques and 4 physics-driven approaches. Daline's robust capabilities cover a spectrum of functionalities: (1) Data Generation, (2) Data Pollution, (3) Data Cleaning, (4) Data Normalization, (5) Method Selection, (6) Method Customization, (7) Model Linearization, (8) Model Evaluation, and (9) Result Visualization. These functionalities enable users to execute complex tasks with minimal coding effort. This manual serves as an in-depth guide, providing a gentle introduction for new users and a detailed reference for advanced users, explaining Daline's features, functions, parameters, syntax, and practical applications, and all the other related contents. Our vision is to empower users by saving their valuable time and fostering innovation in power systems analysis through accessible, state-of-the-art linearization tools.
Daline: A Data-driven Power Flow Linearization Toolbox for Power Systems Research and Education
Jia, Mengshuo; Chan, Wen Yi; Hug, Gabriela (2024)
Power flow linearization has long been a fundamental tool in both academia and industry. While physics-driven power flow linearization (P-PFL) methods are relatively accessible, data-driven power flow linearization (D-PFL) approaches are significantly less so. As a promising field, D-PFL has demonstrated its potential to enhance linearization accuracy and address scenarios where traditional P-PFL methods are limited, particularly when physical parameters are unavailable. Despite their significance, over 95% of existing D-PFL algorithms lack publicly available codes, thereby limiting the utilization and comparison of the latest advancements in power flow linearization. To establish D-PFL methods into easily accessible fundamental tools, we developed Daline, a MATLAB-based, open-source toolbox that includes 57 linearization methods (42 D-PFL methods reproduced from existing literature, 11 D-PFL methods not reported before, and four widely used P-PFL methods), offering extensive flexibility in method selection and customization. Beyond linearization, Daline offers comprehensive functionality, ranging from data generation and processing, to model training, testing, and visualizable evaluation of accuracy or computational efficiency. These modules are grounded on Daline's comprehensive yet user-friendly architecture: with over 300 options, Daline caters to diverse needs with minimal coding, making complex tasks achievable with just one or two lines of code.
A Distributed Probabilistic Modeling Algorithm for the Aggregated Power Forecast Error of Multiple Newly Built Wind Farms
Jia, Mengshuo; Shen, Chen; Wang, Zhiwen (2019)
IEEE Transactions on Sustainable Energy
The extensive penetration of wind farms (WFs) presents challenges to the operation of distribution networks (DNs). Building a probability distribution of the aggregated wind power forecast error is of great value for decision making. However, as a result of recent government incentives, many WFs are being newly built with little historical data for training distribution models. Moreover, WFs with different stakeholders may refuse to submit the raw data to a data center for model training. To address these problems, a Gaussian mixture model (GMM) is applied to build the distribution of the aggregated wind power forecast error; then, the maximum a posteriori (MAP) estimation method is adopted to overcome the limited training data problem in GMM parameter estimation. Next, a distributed MAP estimation method is developed based on the average consensus filter algorithm to address the data privacy issue. The distribution control center is introduced into the distributed estimation process to acquire more precise estimation results and better adapt to the DN control architecture. The effectiveness of the proposed algorithm is empirically verified using historical data.
Data-driven Power Flow Linearization: Theory
Jia, Mengshuo; Hug-Glanzmann, Gabriela; Zhang, Ning; et al. (2024)
This two-part tutorial dives into the field of data-driven power flow linearization (DPFL), a domain gaining increased attention with the rise of data-centric methodologies. DPFL stands out for its higher approximation accuracy, wide adaptability, and better ability to implicitly incorporate power losses and the latest system attributes. This renders DPFL a potentially superior option for managing the significant fluctuations from renewable energy sources, a step towards realizing a more sustainable energy future, by translating the higher model accuracy into increased economic efficiency and less energy losses. To conduct a complete, rigorous, and deep reexamination, this tutorial first classifies all existing DPFL methods into DPFL training algorithms and supportive techniques. Their mathematical models, analytical solutions, capabilities, limitations, and generalizability are systematically examined, discussed, and summarized. In addition, this tutorial reviews existing DPFL experiments, examining the settings of test systems, the fidelity of datasets, and the comparison made among a limited number of DPFL methods. Further, this tutorial implements extensive numerical comparisons of all existing DPFL methods (40 methods in total) and four classic physics-driven approaches, focusing on their generalizability, applicability, accuracy, and computational efficiency. Through these experiments, this tutorial aims to reveal the actual performance of all the methods (including the performances exposed to data noise or outliers), therefore guiding the selection of appropriate linearization methods for researchers. Furthermore, this tutorial discusses open research questions and future directions based on the theoretical and numerical insights gained, contributing to the progression of the field of DPFL. As the first part of the tutorial, this paper mainly reexamines DPFL theories, covering all the training algorithms and supportive techniques in depth. Numerous capabilities, limitations, and aspects of generalizability, which were previously unmentioned in the literature, have been identified.
Tutorial on Data-driven Power Flow Linearization - Part I: Challenges and Training Algorithms
Jia, Mengshuo; Hug, Gabriela; Zhang, Ning; et al. (2023)
The widespread deployment of advanced metering infrastructure has led to an increased interest in data-driven power flow linearization (DPFL) methods. To date, substantial studies have already been carried out in the field of DPFL, which have shown that DPFL models are often more accurate than physics-based power flow linearization models. In this paper, a tutorial on this topic is provided. This tutorial reviews the motivations, challenges, training algorithms, supportive techniques, and numerical experiments with respect to DPFL. The derivations of the algorithms/techniques and their (closed-form) solutions are both mathematically revisited and discussed. Numerous future research directions for DPFL are identified and summarized. Additionally, an intuitive, unbiased, easily calculateable indicator for measuring the linearity level of any given system is proposed and employed on various power systems. This tutorial is divided into two parts, and this paper, as the first part, covers the surveys on the motivations, challenges, and training algorithms for DPFL.
Federated Clustering for Electricity Consumption Pattern Extraction
Wang, Yi; Jia, Mengshuo; Gao, Ning; et al. (2022)
IEEE Transactions on Smart Grid
The wide popularity of smart meters enables the collection of massive amounts of fine-grained electricity consumption data. Extracting typical electricity consumption patterns from these data supports the retailers in their understanding of consumer behaviors. In this way, diversified services such as personalized price design and demand response targeting can be provided. Various clustering algorithms have been studied for electricity consumption pattern extraction. These methods have to be implemented in a centralized way, assuming that all smart meter data can be accessed. However, smart meter data may belong to different retailers or even consumers themselves who are not willing to share their data. In order to better protect the privacy of the smart meter data owners, this paper proposes two federated learning approaches for electricity consumption pattern extraction, where the k-means clustering algorithm can be trained in a distributed way based on two frequently used strategies, namely model-averaging and gradient-sharing. Numerical experiments on two real-world smart meter datasets are conducted to verify the effectiveness of the proposed method.
Digital Twin of the Energy Internet and Its Application
Shen, Chen; Jia, Mengshuo; Chen, Ying; et al. (2020)
Global Energy Interconnection
Based on electrical power systems, leveraging renewable energy generation technology, and information technology, the energy Internet fuses power grids, natural gas networks, heat/cold supply networks, electric transportation networks, etc. into an interconnected energy sharing network. The energy Internet is an important technology for promoting renewable energy integration and improving energy efficiency. However, due to the complexity of multiple energy networks and the significant differences between them, the planning, operation, and control of the energy Internet presents several technical difficulties. Digital twins is an advanced simulation technology based on the Internet of things, communications, big data, and high performance computing that may provide effective solutions to the problems that the energy Internet is currently facing. In this paper, the concept of digital twins is firstly introduced, the construction and possible applications of digital twins to energy Internets are discussed, the problems that the energy Internet digital twin can solve are illustrated by taking the planning of energy Internets as an example, and a digital twin technology-based energy Internet planning platform — CloudIEPS—is introduced to further exemplify the role of the energy Internet digital twin through a concrete planning case.
Large Language Models RePower Autonomous Research of Data-Driven Tasks in Power Systems
Liu, Yu-Xiao; Jia, Mengshuo; Zhang, Yong-Xin; et al. (2024)
Large language models (LLMs) have demonstrated exceptional performance across various fields, including chemistry, mathematics, medical science, and materials science. However, their adoption in power systems research remains limited, and when utilized, LLMs are often confined to specific problem-solving tasks, with human researchers maintaining control over design and innovation. This perspective may restrict the full potential of LLMs in advancing the field. This paper presents a novel framework that repositions LLMs from mere problem solvers to active research designers, aimed at accelerating scientific progress in power systems. We propose RePower, an automated research system powered by LLM that employs a reflection-evolution strategy to autonomously handle complex research tasks. RePower supports human researchers by acquiring data, designing methods, and evolving algorithms to tackle problems that are challenging to solve but straightforward to evaluate. We validated RePower on three critical data-driven tasks in power systems: parameter prediction, power optimization, and state estimation. Our results show that RePower-designed and evolved methods outperform existing approaches, achieving, for example,an average 15.23% reduction in prediction error for the parameter prediction task. Overall, RePower offers a comprehensive framework for integrating LLMs into power systems research, enabling autonomous scientific discoveries and fostering innovation across various subfields.
Insulating materials for realising carbon neutrality: Opportunities, remaining issues and challenges
Li, Chuanyang; Yang, Yang; Xu, Guoqiang; et al. (2022)
High Voltage
The 2050 carbon-neutral vision spawns a novel energy structure revolution, and the construction of the future energy structure is based on equipment innovation. Insulating material, as the core of electrical power equipment and electrified transportation asset, faces unprecedented challenges and opportunities. The goal of carbon neutral and the urgent need for innovation in electric power equipment and electrification assets are first discussed. The engineering challenges constrained by the insulation system in future electric power equipment/devices and electrified transportation assets are investigated. Insulating materials, including intelligent insulating material, high thermal conductivity insulating material, high energy storage density insulating material, extreme environment resistant insulating material, and environmental-friendly insulating material, are categorised with their scientific issues, opportunities and challenges under the goal of carbon neutrality being discussed. In the context of carbon neutrality, not only improves the understanding of the insulation problems from a macro level, that is, electrical power equipment and electrified transportation asset, but also offers opportunities, remaining issues and challenges from the insulating material level. It is hoped that this paper envisions the challenges regarding design and reliability of insulations in electrical equipment and electric vehicles in the context of policies towards carbon neutrality rules. The authors also hope that this paper can be helpful in future development and research of novel insulating materials, which promote the realisation of the carbon-neutral vision.

Publications 1 - 10 of 30

Mengshuo Jia

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results