Plataforma IoT para la provisión de servicios en procesos industriales

Abstract

Today, society faces a number of environmental and sustainability concerns that are having a major impact on the way of life we have been living. One of the main concerns is climate change, which is causing changes in the global climate and environment. In addition, the scarcity of vital resources such as fresh water and food is becoming an increasingly important issue due to the growth of the world’s population and the ever-increasing demand for these supplies. These environmental and sustainability issues have a direct impact on people’s quality of life and the functioning of economic and social systems. It is therefore crucial to address these concerns and seek sustainable solutions to ensure a livable future for all. In this context, the scarcity of natural resources and climate change are issues of concern to authorities globally and have led to the implementation of various socioeconomic initiatives and research projects to try to address them. Technology, and in particular Internet of Things (IoT) technology, are tools that can play a key role in addressing these issues, as it can help improve efficiency and sustainability in various areas of society and particularly in the industrial sector. IoT is an ever-growing technology that enables the connection of devices to collect and transmit information from the environment in which they are located. These environments are called IoT ecosystems and are composed of all smart objects with Internet connectivity capabilities. It is expected that in the near future there will be a significant increase in the number of IoT devices, enabling smart environments. However, this technology presents a number of challenges related to Internet communication and the management of data generated by the devices. These challenges are associated with the heterogeneous nature of IoT, which refers to the wide variety of devices, manufacturers and technologies used in IoT implementation. This includes devices of different sizes, shapes and functions, as well as different connectivity technologies and hardware and software platforms. This heterogeneity makes it difficult to integrate and effectively use the IoT, as it can be difficult to make the different devices and technologies work together in a coherent way. It is important to take this heterogeneity into account when designing and deploying IoT-based solutions and ensure that the needs and requirements of the different devices and technologies involved are taken into account. To date, most of the research work has been focused on the development of management platforms for monitoring the industrial sector, focusing on the investigation of new lines of improvement, such as the use of low-cost sensors and software for decision support with the aim of improving the sector’s production efficiency. These works have provoked a great concern for researchers to try to obtain more efficient and sustainable systems that try to help the difficulties of this technology. Therefore, IoT technology has entered a new phase of research, in which, the works must be aimed at providing solutions to the heterogeneity of ecosystems and efficient data management. This includes implementing solutions that enable the application of IoT technology in multiple domains with scalable platforms and improving the way data is stored, processed and analyzed, homogenizing and standardizing it to obtain valuable information and aid decision-making. The main objective of this doctoral thesis is to develop an IoT platform for the provision of services in industrial processes. This platform must integrate different types of data and bring them together in a single place to facilitate user management and access. The IoT platform must ensure homogeneity and interoperability between platforms and devices in syntactic and semantic terms by using standards in data models for information exchange. In addition, it must include predictive models to support decision-making and be accessible online without relying on software. The IoT platform must be versatile to allow scalability to other industrial sectors and cover the entire production cycle, providing cloud services for device provisioning, monitoring and plant management. The first phase of development of the PhD thesis focuses on a review of the relevant literature on IoT, characteristics, objectives, reference architectures, levels of interoperability, complementary technologies and the use of a model for resource optimization, to propose a solution to interoperability and traditional decision support systems. Next, we proceed with phase two of the thesis project, the development of an IoT platform for greenhouse crop management that can be easily adapted to different industrial sectors, promoting the standardization of services and the optimization of resources. For this purpose, a modular architecture divided into layers formed by microservices on containers is developed, which provides scalability to the system. In addition, a standard data model is used at the core to address interoperability issues. The next step is the development and implementation of greenhouse models as a service accessible through the Internet, providing climatic, production, and irrigation models, covering two of the main objectives of the thesis. This second phase was validated in a pilot greenhouse and later scaled to real production greenhouses in the province of Almeria. This results in the iVeg application, which has an intellectual property registration no. 04/2022/956. The proposed IoT platform has been applied to different heterogeneous and differentiated industrial sectors, associated to three European projects funded by the European Union and one national project. The third phase of the development of the doctoral thesis seeks to improve the proposed IoT solution by applying it to the large-scale data management of an industrial-scale biomass production system through the cultivation of microalgae. This is done in response to the lack of digitization of the sector and the problems of data accessibility. The solution allows monitoring of the complete production cycle of the plant, including provisioning of new sensors, photobioreactors and data query. A real biomass production plant has been used to validate the correct functioning of the platform. In addition, this solution provides graphical and query services to technicians, researchers, and companies for data extraction through cloud services with standardized data, fulfilling another objective of this thesis. The fourth phase of the development of the thesis focuses on the theoretical development of an IoT-based water management architecture for application in agro-industrial districts, with the aim of facilitating data integration and the use of optimization and control algorithms in the cloud. Simulation tests have been performed in an agro-industrial environment consisting of three real facilities including a desalination plant, utility grid connection and several consuming agents. This architecture offers cloud services based on interoperable and standard data models, allowing real-time control and optimization operations. The comparative cost analysis performed with a manual operation shows how around 75% of the operating cost can be saved.