طراحی مدل نظارت هوشمند بر تجهیزات صنعتی با رویکرد بهینه سازی تاخیر انتقال و سرعت پردازش داده ها(مطالعه موردی: توربین های بادی) (مقاله علمی وزارت علوم)
درجه علمی: نشریه علمی (وزارت علوم)
آرشیو
چکیده
امروزه در واحدهای صنعتی نظارت بر عملکرد تجهیزات توسط اپراتور انجام می شود و به علت گستردگی محیط عملیاتی و حجم بالای تجهیزات، هماهنگی بین واحدها به سختی امکان پذیر بوده و صدمات جبران ناپذیری در پی دارد. باوجود پیشرفت های چشمگیر تکنولوژی در زمینه بازرسی و نظارت، می توان این وظیفه را به ابزارهای هوشمند و اینترنت اشیاء سپرد. همچنین، با ظهور فناوری محاسبات «رایانش لبه»، بسیاری از محققان به دلیل مزایای آن، از طراحی های مبتنی بر محاسبات لبه ای بهره برده اند. لذا در این پژوهش مدلی ترکیبی از اینترنت اشیاء و پهپادهای غیرنظامی جهت نظارت هوشمند بر عملکرد تجهیزات صنعتی با رویکرد رایانش لبه ارائه شد که به عنوان مطالعه موردی توربین های بادی موردبررسی قرار گرفت. در این مدل، عملکرد پهپاد جهت نظارت هوشمند بر توربین های بادی در سه مرحله موردبررسی قرار گرفت. 1) فرآیند تشخیص 2) فرآیند تخلیه محاسباتی پهپاد 3) فرآیند محاسبات محلی پهپاد. با توجه به دو هدفه بودن مدل نهایی که ترکیبی از سه مرحله فوق بود، مدل توسط روش های ژنتیک با مرتب سازی ناچیره و روش محدودیت اپسیلن تقویت شده با استفاده از اعداد تصادفی حل شد و روش ژنتیک با مرتب سازی ناچیره عملکرد بهتری داشت؛ زیرا در مسائل با ابعاد بزرگ، به دلیل پیچیدگی مسئله، روش محدودیت اپسیلن تقویت شده قادر به پاسخگویی در زمان مناسب نبود.امروزه در واحدهای صنعتی نظارت بر عملکرد تجهیزات توسط اپراتور انجام می شود و به علت گستردگی محیط عملیاتی و حجم بالای تجهیزات، هماهنگی بین واحدها به سختی امکان پذیر بوده و صدمات جبران ناپذیری در پی دارد. باوجود پیشرفت های چشمگیر تکنولوژی در زمینه بازرسی و نظارت، می توان این وظیفه را به ابزارهای هوشمند و اینترنت اشیاء سپرد. همچنین، با ظهور فناوری محاسبات «رایانش لبه»، بسیاری از محققان به دلیل مزایای آن، از طراحی های مبتنی بر محاسبات لبه ای بهره برده اند. لذا در این پژوهش مدلی ترکیبی از اینترنت اشیاء و پهپادهای غیرنظامی جهت نظارت هوشمند بر عملکرد تجهیزات صنعتی با رویکرد رایانش لبه ارائه شد که به عنوان مطالعه موردی توربین های بادی موردبررسی قرار گرفت. در این مدل، عملکرد پهپاد جهت نظارت هوشمند بر توربین های بادی در سه مرحله موردبررسی قرار گرفت. 1) فرآیند تشخیص 2) فرآیند تخلیه محاسباتی پهپاد 3) فرآیند محاسبات محلی پهپاد. با توجه به دو هدفه بودن مدل نهایی که ترکیبی از سه مرحله فوق بود، مدل توسط روش های ژنتیک با مرتب سازی ناچیره و روش محدودیت اپسیلن تقویت شده با استفاده از اعداد تصادفی حل شد و روش ژنتیک با مرتب سازی ناچیره عملکرد بهتری داشت؛ زیرا در مسائل با ابعاد بزرگ، به دلیل پیچیدگی مسئله، روش محدودیت اپسیلن تقویت شده قادر به پاسخگویی در زمان مناسب نبود.Designing an Intelligent Monitoring Model for Industrial Equipment with the Approach of Optimizing Transmission Delay and Data Processing Speed (Case Study: Wind Turbines)
In today's industrial units, operators monitor equipment performance, and the challenging coordination between units in vast operating environments with high volumes of equipment can lead to irreparable damage. Despite considerable technological advancements in inspection and surveillance, this responsibility can be effectively delegated to smart devices and the Internet of Things (IoT). Furthermore, the emergence of "edge computing" technology has prompted researchers to explore edge-based computing designs due to their numerous benefits. This study presents a combined model of IoT and civilian drones for intelligent monitoring of industrial equipment performance, employing an edge computing approach. The model is specifically investigated through a case study involving wind turbines. The model evaluates the performance of drones for intelligent monitoring of wind turbines in three stages: 1) Detection process, 2) UAV computational evacuation process, and 3) UAV local computation process. Given the dual purpose of the final model, which involves a combination of the aforementioned three steps, a genetic method was employed for problem-solving with negligible sorting. The amplified epsilon restriction method, utilizing random numbers, was also considered, but the combination of genetic and negligible sorting methods outperformed it, particularly in large problems where the enhanced epsilon restriction method struggled to provide timely responses due to the inherent complexity of the problem. IntroductionToday, in various industries, the productivity and efficiency of equipment contribute to the advancement of production and the profitability of production units. Beyond repair costs, equipment breakdowns also result in the expense of lost opportunities for the production unit. Without a solution to prevent these costs, bankruptcy for production units becomes a real possibility. Therefore, consideration should be given to a solution for the optimal monitoring of equipment. Clearly, swift action is crucial when any equipment is damaged, and such rapid response is unattainable through human effort alone. Despite significant technological advances in inspection and monitoring, this task can be delegated to smart tools and the Internet of Things (IoT). The IoT is regarded as one of the most crucial factors for the prosperity and progress of today's and future industrial businesses. Modernizing equipment is a priority for today's industries to quickly adapt to the evolving market changes and harness existing technologies. Businesses incorporating IoT into their infrastructure experience substantial growth in areas such as security, productivity, and profitability. As the use of industrial IoT increases, productivity levels in industries are naturally expected to rise. The IoT can accumulate massive amounts of information and data, enabling factories and companies to optimize their systems and equipment without being hindered by technological and economic limitations. However, a challenge arises from the substantial volume of data generated by the IoT, which is sent to cloud computing centers for processing. Centralized (cloud) processing results in high communication delays and lowers the data transfer rate between IoT devices and potential users, creating operational challenges in the network. To address this issue, the concept of edge computing has been proposed. Edge computing allows IoT services to process data near their own data sources and data sinks instead of relying on the cloud environment. This approach leads to reduced communication delays and more efficient utilization of computing, storage, and network resources. It also minimizes execution time and energy consumption, proving to be highly beneficial for IoT applications. Consequently, with the advent of "edge computing" technology, many researchers have embraced edge computing-based designs due to its numerous advantages.Materials and Methods In this research, a combined model of the Internet of Things and civilian drones was presented for the intelligent monitoring of industrial equipment, utilizing an edge computing approach. The model was investigated through a case study involving wind turbines. The performance of UAVs for intelligent monitoring of wind turbines was examined in three stages: 1) Detection process, 2) UAV computational evacuation process, and 3) UAV local computing process. Given the dual purpose of the final model, which involved a combination of the aforementioned three steps, the model was addressed using genetic methods with sparse sorting and the enhanced epsilon constraint method employing random numbers. The genetic method with sparse sorting outperformed the enhanced epsilon limit method, particularly in problems with large dimensions. The complexity of the problem made it challenging for the enhanced epsilon constraint method to provide timely responses in such cases.ResultsThe findings of this research offer valuable insights for the effective and accurate management and monitoring of industrial equipment across various industrial units, aiming to optimize costs, quality, and inspection time. Additionally, this research can provide guidance in considering regulatory restrictions in equipment placement before constructing an industrial unit. During the equipment arrangement phase, the model presented in this research can be utilized for optimal energy consumption and time management. As the combined model of the Internet of Things and civilian drones for intelligent monitoring of industrial equipment is a novel concept in the literature, there exist numerous opportunities for further development in this field. This may include the application of the model in additional case studies, such as enhancing the intelligent monitoring of power supply systems, fire services, etc. Moreover, there is potential for refining the mentioned model under conditions where drones operate simultaneously without a specific sequence.ConclusionFailure to monitor industrial equipment properly can result in substantial financial losses for factories and production units. The improper operation of equipment may lead to complete failure, necessitating the need for replacement. Additionally, increased equipment downtime, quality issues, reduced production speed, safety hazards, and environmental pollution can be consequences of equipment failure, ultimately diminishing the profitability of the production unit. Considering factors such as embargoes, emphasis on domestic production, and self-sufficiency, accurate supervision becomes economically crucial for factories.Effective management of the proper operation of industrial equipment is a fundamental requirement for every production unit, given that industrial equipment represents a significant investment for the unit. If device maintenance is limited to repairs only after breakdowns occur, production devices will consistently face unexpected halts, preventing production productivity from reaching its predetermined goals. Therefore, designing a framework for the "intelligent monitoring of the performance of all relevant industrial equipment" stands as one of the most crucial actions for any production unit. Depending on the type of equipment, monitoring the performance of industrial equipment may encompass periodic inspections, maintenance and repair planning, and scheduling the optimal operational time for the equipment
تبلیغات
