Haidong Yang

An Ant Optimization Model for Unrelated Parallel Machine Scheduling with Energy Consumption and Total Tardiness

This research considers an unrelated parallel machine scheduling problem with energy consumption and total tardiness. This problem is compounded by two challenges: differences of unrelated parallel machines energy consumption and interaction between job assignments and machine state operations. To begin with, we establish a mathematical model for this problem. Then an ant optimization algorithm based on ATC heuristic rule (ATC-ACO) is presented. Furthermore, optimal parameters of proposed algorithm are defined via Taguchi methods for generating test data. Finally, comparative experiments indicate the proposed ATC-ACO algorithm has better performance on minimizing energy consumption as well as total tardiness and the modified ATC heuristic rule is more effectively on reducing energy consumption.

IoT-enabled real-time energy efficiency optimisation method for energy-intensive manufacturing enterprises

Rising energy prices, increasing fierce competition, new environmental legislation and concerns over climate change are forcing energy-intensive manufacturing enterprises to increase production energy efficiency and reduce their associated environmental impacts. Thanks to the rapid developments of technologies in Internet of Things (IoT), the real-time status of resources and the data of energy consumption from manufacturing processes can be collected easily. These manufacturing information can provide an opportunity to enhance the energy efficiency in real-time production management. To achieve this target, this work presents a real-time energy efficiency optimisation method (REEOM) for energy-intensive manufacturing enterprises. By this method, IoT technologies are applied to sense the real-time primitive production data, including the energy consumption data and the resources status data. Multilevel event model and complex event processing are used to obtain real-time energy-related key performance indicators (e-KPIs) which extend production performance indicators to the energy efficiency area. Then, the non-dominant sorting genetic algorithm II is used to schedule or reschedule the production plan in an energy-efficient way based on real-time e-KPIs. Finally, a case is used to demonstrate the presented REEOM.

Modelling and simulation of energy consumption of ceramic production chains with mixed flows using hybrid Petri nets

Ceramic production chain consisting of discrete flow and continuous flow energy-intensive processes consumes substantial amounts of energy. This study aims to evaluate energy consumption performance and energy-saving potentials of the ceramic production chain. According to the energy consumption characteristics of manufacturing processes and process interaction constraints in a ceramic production chain, an approach integrating the first-order hybrid Petri net (FOHPN) model, an objective linear programming model and a sensitivity analysis is proposed. The FOHPN model will simulate the energy consumption patterns of the ceramic production chain. Meanwhile, multi-objective linear programming model and sensitivity analysis will suggest the optimal specific energy consumption (SEC) of the production chain and identify the influences of input parameters (i.e. production rate of a process) on the SEC in the optimal production scheme. Finally, a real case study from bathroom ceramic plant validates the approach. It provides a tool for modelling and simulation of energy consumption of ceramic production chains with mixed flows and helps operators to perform energy-saving actions in the ceramic enterprise.

Transfer learning for aluminium extrusion electricity consumption anomaly detection via deep neural networks

Effective anomaly detection can reduce the electricity consumption and carbon emissions in aluminium extrusion processes. The following two steps identify anomalies: electricity consumption forecasting and anomaly detection. Data-driven modelling is typical paradigm for building an accurate forecasting model. For a new extruding machine, there is insufficient extruded data for model training. The research objective of this work is to determine whether a forecasting model can be trained by transferring knowledge from a data-sufficient domain to a data-insufficient domain. A shared connected deep neural network is proposed for electricity consumption time-series anomaly forecasting. Anomalies are detected by the difference of predicted and measured values at a confidence interval. The experimental results show that the proposed approach can identify electricity anomaly events in real time. Furthermore, it is shown that transferring learning knowledge between domains significantly improves the forecasting results.

Simulation-based approach to modeling the chip formation energy of polishing process

ABSTRACT Polishing is one typical material removal process, which is widely used for surface processing of porcelain tiles. Due to complex polishing head structure and kinematics features of polishing machine, polishing for porcelain tile is a high energy intensity process. To improve the energy efficiency by optimizing operation, it is essential to establish an energy consumption model for polishing process. This article divides the total energy of polishing process into constant energy and chip formation energy. Furthermore, this article focuses on modeling the chip formation energy for optimizing operation. Based on the energy conversion mechanism and energy flow characteristics, the chip formation energy of polishing process is further divided into three motion energies that govern the abrasive trajectory over the tile surface. A conceptual framework of simulation-based approach is then proposed for modeling chip formation energy of polishing process by integrating the above calculation algorithm of motion energy. Finally, a case study is implemented to illustrate the validation of the proposed approach, and the results show that it is a feasible tool to model the chip formation energy of polishing process and reveal the influence of different process operational parameters.

State space modelling carbon emission dynamics of machining workshop based on carbon efficiency

With increasing pressure to deal with climate change, to reduce carbon emissions has been one major challenge for manufacturing enterprises even though they face unprecedented levels of business competition. This paper proposes a state space model for evaluating carbon emission dynamics of a machining workshop based on carbon efficiency. A general hierarchical analysis framework for carbon emission dynamics for a three-level structure machining workshop is initially established from three perspectives: organisation, control loop and resource coupling input and output flow. Within this framework, carbon efficiency measuring emission dynamics is defined systematically, which reflects the balancing trends between benefit output (desired output) and emission output (undesired output) of any level system in a machining workshop. In order to evaluate the balancing trends, a state space based conceptual model is proposed. A state space model module for individual machine which consists of two sub-models, namely production process state space and emission process state space, forms the basic element of this conceptual model. A carbon emission dynamics profile based on carbon efficiency from machine tool level to workshop level is then determined through module integration. Finally, an experimental study is carried out to illustrate its feasibility and applicability of evaluating carbon emission dynamics through aligning economic and environmental dimension.