Multiproduct Pipeline Research Articles

Deeppipe: Theory-guided LSTM method for monitoring pressure after multi-product pipeline shutdown

The pressure changes dramatically during the shutdown process of the multi-product pipeline. When the pipeline pressure comes to decrease, it is often mistaken as pipeline leakage or other abnormal condition which increases the burden of the operator on-site. At present, the method of pipeline shutdown pressure analysis is mainly based on numerical simulation which can not monitor shutdown pressure in real-time. In this work, the time-series approximate ability of long short-term memory (LSTM) is taken advantage of to construct a shutdown pressure prediction model. To overcome the drawback of this deep learning algorithm that is trained only by ample data, the scientific principle and theory are integrated into LSTM. Subsequently, the theory-guided long short-term memory (TG-LSTM) is proposed for pipeline shutdown pressure prediction. The proposed model is trained with available data and simultaneously guided by the theory (physical principle and engineering theory) of the underlying problem. In the training process, the data mismatch, as well as monotonicity constraints, and boundary constraints are coupled into loss function. After acquiring the parameters of the neural network, a TG-LSTM model is established which not only fits the data, but also follows the physical principle and the engineering theory. The proposed model is verified by three real-world multi-product pipelines. The results indicate that TG-LSTM achieves better accuracy than other prediction models, with MAPE being 0.246%, 0.186%, and 0.143%, respectively. Finally, the sensitivity analysis of different hyper-parameter is conducted to illustrate the robustness of TG-LSTM in pipeline shutdown pressure prediction.

Support of Scheduling of Multiproduct Pipeline Systems Using Simulation in Witness

Support of Scheduling of Multiproduct Pipeline Systems Using Simulation in Witness

An effective decomposition algorithm for scheduling branched multiproduct pipelines

This article develops a novel two-stage algorithm for the detailed scheduling of branched multiproduct pipeline systems with a single refinery and multiple depots. It features continuous-time models allowing multiple batches to be injected/delivered over a slot. The first-stage model reduces the complexity by neglecting the lower flowrate limits, enforced by big-M constraints. The objective is to minimize the makespan. The second-stage model works with the first-stage assignments and with an extended time grid. Coupled with a backtracking mechanism, the proposed algorithm can significantly reduce computational time and improve the pipeline transportation capacity. We use four benchmark instances from the literature to test and validate the effectiveness and superiority of the method proposed in this paper. Compared to the literature, a new optimal solution is reported in one case, while the computational time is reduced by at least one order of magnitude in the other cases.

Multi-period optimal schedule of a multi-product pipeline: A case study in Algeria

The oil supply chain plays a vital role for both economic and social sectors in Algeria. Pipeline is the main oil transportation mode from production sites to distribution centers. However, finding the optimal schedule for a multi-product pipeline is among the most difficult optimization problems. In this article, we propose an operational schedule for a long-term horizon for a multi-product pipeline system. In our model, we link a refinery to multiple storage and distribution fuel centers. The proposed approach aims to satisfy the demands of different distribution centers in a timely manner by respecting quality and cost requirements. A mixed integer linear programming (MILP) model is proposed and solved by a branch and cut algorithm. Our model is tested on a real case of an Algerian oil company. The obtained schedule satisfies variable demand while respecting quality and cost requirements.

Deeppipe: A semi-supervised learning for operating condition recognition of multi-product pipelines

Intelligent operating monitoring of pipelines helps to detect anomalies in time to ensure pipeline safe, reducing potential risk. However, the operating conditions of the multi-product pipeline change frequently, and the recognition and monitoring by on-site personnel are easy to cause misjudgment, so the operating conditions of the pipeline cannot be accurately recognized. Noticeably, operating condition recognition is an important part of pipeline safety and risk management. Although ample operating data are stored in SCADA system, these data are lack of corresponding condition labels, making it hard to be mined. In this work, a semi-supervised learning for operating condition recognition is proposed to overcome aforementioned issues. Firstly, the operating parameters of each station are preprocessed and collected to construct into data matrices to overcome transient disturbance considering the pipeline space characteristics and time series of the operating data. Then stacked autoencoder (SAE) is used to pre-train the network parameters of multi-layer neural network (MLNN) based on a large amount of unlabeled operating data. After that, MLNN is fine-tuned based on a small amount of labeled data annotated by referring to the operation log. To verify the effectiveness of the semi-supervised learning, a real multi-product pipeline is taken as an example for operating condition recognition. The accuracy, precision, recall and F1 score is 95 %, 95 %, 80 % and 80 %, respectively. Results show that the condition recognition accuracy of the proposed model is better than other machine learning models. Finally, the sensitivity analysis is conducted to illustrate the importance of SAE in this classification model.

New Method Based on Model-Free Adaptive Control Theory and Kalman Filter for Multi-Product Unsteady Flow State Estimation

Abstract In this paper, a new methodology is proposed to realize real-time unsteady flow estimation for a multi-product pipeline system. Integrating transient flow model, adaptive control theory, and adaptive filter, this method is developed to solve the contradiction between the efficiency and accuracy in traditional model-based methods. In terms of improving computational efficiency, the linear flow model based on frequency response and difference transforming is established to replace the traditional nonlinear flow model for transient flow state estimation. To reduce the deviation between actual observations and linear model estimates, we first introduce a model-free adaptive control method as linear compensation of the reduced order unsteady flow state model. To overcome the interference of observation noise, the Kalman filter method is applied to the modified state space model to obtain the one-step-ahead transient flow estimation. The proposed method is applied to the transient flow state estimation of a multi-product pipeline system and compared with the model-based method and two data-driven methods. The proposed method can reduce the deviation of transient flow estimation between the reduced order linear model and the traditional nonlinear model to less than 0.5% under unforeseen conditions and shows strong robustness to noise interference and parameter drift.

Stress corrosion cracking in fuel-grade anhydrous ethanol from dedicated and multi-products pipelines

In this study, anhydrous ethanol fuel samples obtained from pipelines dedicated to the ethanol fuel transport and from multi-product pipelines were tested by standard slow strain stress tests and API 5L X65 notched specimens. The fractures of the specimens were observed by optical and scanning electron microscopes. The metallographic examination of a normal section of the fracture surface was carried out. It was observed that the ethanol samples from multi-product pipelines were more aggressive than the ethanol fuel from dedicated pipeline samples. On the other hand, ethanol fuel from dedicated pipeline samples presented quasi-cleavage fractures which have not been seen before in the literature. Nowadays, it has been known that Brazilian Midwest is producing fuel ethanol from second crop corn. This study sheds light on a possible change in the Brazilian fuel ethanol matrix that in the past was predominantly composed of sugar cane ethanol and is currently receiving the contribution of corn ethanol in order to avoid any transport and storage system failure.

(2004-5799) An interactive fuzzy programming approach for a new multi-objective multi-product oil pipeline scheduling problem

In this paper, a new fuzzy multi-objective multi-product pipeline scheduling problem is introduced. The system consists of a single refinery, a unique distribution center, and a multi-product pipeline. Restrictions such as batchsizing, discharging rate, forbidden sequences, batch volumetric, etc. are considered. Due to the uncertain nature of real-world problems, some parameters of the system are considered as fuzzy values. Tardiness and earliness penalties are considered with a time dependent non-linear function. The basic aim of this scheduling problem is to achieve the optimal sequence for pumping batches of oil products to maximize the financial benefit and simultaneously satisfies the customers with on-time delivery as a multi-objective problem. To tackle this problem, a two-stage methodology is proposed. In the first stage, the fuzzy formulation is converted to its equivalent crisp form by a credibility-based chance-constrained programming approach. In the second stage, the multi-objective crisp formulation is solved by some well-known approaches in the literature. Some test problems are generated and solved by the proposed approaches and the obtained Pareto-optimal solutions are analyzed and compared using some distance-based comparison metrics.

Detailed scheduling distribution of real multi-product pipeline

Detailed scheduling distribution of real multi-product pipeline

Detailed scheduling distribution of real multi-product pipeline

Detailed scheduling distribution of real multi-product pipeline

Efficient formulation for transportation scheduling of single refinery multiproduct pipelines

Multiproduct pipeline transportation scheduling is a complex operations research problem that is characterized by the movement of the cargo rather than the carrier. Hence, it cannot be solved using vehicle routing methods. While most formulations for short-term scheduling adopt a continuous-time representation, they often lead to suboptimal solutions because of the dependence on the number of time slots in the grid, which is difficult to predict. Furthermore, some of these formulations have poor linear relaxations due to the presence of inefficient big-M constraints. In this paper, we develop a discrete-time mixed integer linear programming (MILP) model for the detailed scheduling of a straight pipeline with a single refinery and multiple depots. The proposed formulation rigorously detects interface material generated between adjacent products and considers planned shutdowns in pipeline segments due to maintenance operations, as well as local market demands occurring at multiple intermediate due dates. The main novelty is that continuous tasks can span multiple time slots to enforce minimum batch sizes on injection and delivery nodes, which allows for the model to generate better schedules than those obtained with previously proposed formulations. To ensure an efficient model by design, we rely on generalized disjunctive programming (GDP) and on the convex hull reformulation of disjunctions, which results in stronger and often more computationally efficient formulations. We present numerical results for a set of benchmark instances and show that the proposed model applies to large-scale industrial cases.

A solution framework for the long-term scheduling and inventory management of straight pipeline systems with multiple-sources

This paper addresses the long-term scheduling of pumping and delivery operations on straight multiproduct pipeline systems connecting multiple-sources to multiple-destinations. A generic solution framework is proposed, combining heuristic procedures and mixed integer linear programming (MILP) models. The pipeline scheduling problem is decomposed into two sub-problems: the allocation and sequencing of pumping operations of the initial source, and the scheduling of delivery operations and also pumping operations of the intermediate sources. An innovative execution procedure using a rolling horizon strategy is developed to solve both problems. We focus on a new discrete-time MILP model formulation, covering several operational aspects of real-world pipeline systems, such as rigorous treatment of forbidden sequences and preventive tank maintenance and pipeline maintenance periods, while considering rigorous inventory management during the scheduling horizon. The approach is validated by solving an illustrative example and two real-world based pipeline problems. Results show that the proposed framework is able to obtain solutions in reasonable run time with detailed inventory management and respecting several operational constraints of the system.

An efficient decomposition approach for the low-energy scheduling of a straight multiproduct pipeline

This paper proposes a hybrid-time mixed integer linear programming model for low-energy scheduling of single-source multiproduct pipelines. The model divides the time horizon into slots by event points of batches reaching stations and fixed points of electricity price changing. Through this division, some practical factors are easily modelled as linear items, including time-varying electricity price and flowrate limits related to interfaces. The key technology for this model is the initialization of all time points’ sequence, so a decomposition approach is developed to efficiently seek satisfactory sequence of all time points. Its superiority is validated by two real-world cases.

An integrated methodology for the supply reliability analysis of multi-product pipeline systems under pumps failure

As the main way for the long-distance transportation of refined products, multi-products pipelines are of vital importance to the regional energy security. The supply reliability evaluation of multi-product pipeline systems can improve the effective response to unexpected disruptions and guarantee the reliable oil supply. Based on reliability theory and pipeline scheduling method, an integrated supply reliability evaluation methodology for multi-product pipeline systems is proposed in this paper and the pumps failure, of which influence is the most complex, is focused on. In the methodology, the discrete-time Markov process is adopted to describe the stochastic failure and the Monte Carlo method is used to simulate the system states transition. With the pipeline flowrate upper limits under various pumps failure scenarios optimized in advance, the maximum supply capacity to the downstream markets in each trial is calculated by the pipeline scheduling model. Three indicators are also developed to analyze the pipeline supply reliability from the holistic and individual perspectives. At last, the methodology application is performed on a real-world multi-product pipeline system in China and the supply reliability is analyzed in detail according to the simulation results. It is proved to provide a practical method for the emergency response decision-making and loss prevention.

Modelos de optimización matemática aplicables al sector downstream y midstream del petróleo. Revisión de la literatura y dirección de investigaciones futuras

Este artículo presenta una revisión de la literatura sobre los modelos de Programación Lineal y No lineal de Enteros Mixtos (MILP y MINLP, respectivamente por sus siglas en inglés) aplicados a la cadena de suministros del petróleo crudo, especialmente en el sector downstream y midstream. Los cuales se encargan de la producción, preparación y distribución de hidrocarburos de uso final como gasolina, diésel, jet y gas licuado de petróleo. El objetivo de esta investigación es comparar varios trabajos de investigación de operaciones, especialmente los que consideran el transporte de multi productos en una misma línea.
 El articulo tiene el enfoque principal de realizar un planteamiento de la dirección futura de la investigación para la cadena de suministro del petróleo en los sectores downstream y midstream. De todos los artículos analizados los 32 generan un modelo MILP, de los cuales 26 modelan la estructura de distribución de combustibles en el sector downstream, no existe evidencia de trabajos de investigación de modelos MILP o MINLP desarrollados en el Ecuador. Finalmente, se discute futuras líneas de investigación referente a esta temática como la aplicación de MILP para la programación operativa de la red de la cadena de suministros del petróleo en el Ecuador. Herramienta que a futuro permitirá la toma de decisiones de nivel táctico y operativo de manera efectiva, considerando todas las variables de la red de distribución de combustibles para satisfacer una demanda prevista.

Multiproduct Pipeline Leak Detection and Localization System Using Artificial Intelligence

The modeling and simulation of a leak detection system with incidence localization for a multiproduct unidirectional flow pipeline is presented in this paper. The research work employs the pressure profile of the pipeline using artificial intelligence (AI) with pressure sensors situated at regular intervals (100 m) along the pipeline. A pipeline of total length 1500 m was modeled with pressure sensors placed along the pipeline. The pressure along the line was measured as a time series data which was then used to train an artificial neural network (ANN) in order to detect leaks. To localize leaks, disturbance (leak) of different sizes was created at intervals along the pipe, which yielded different pressure profiles from the normal operation. This provides the required data for the learning algorithm. In this work, the Darcy–Weisbach equation was used to model the leak detection and localization while the Bernoulli and Colebrook equations were modeled for laminar and turbulent flow, respectively. The model was developed and simulated in Simulink/MATLAB 2017a, and real-time pressure was then adopted to estimate the functionality of the developed (simulated) system. The result shows that different products (fluids) produce different pressure profiles. The developed algorithm is suitable for a multiproduct pipeline. The evaluation of the model shows that leaks can be accurately detected with an accuracy of 98.56%.

Future scenario of China's downstream oil reform: Improving the energy-environmental efficiency of the pipeline networks through interconnectivity

Improving energy supply efficiency and quality is regarded as a key pathway to shifting towards a fully sustainable energy system. To address the low efficiency and high emissions in the downstream oil industry caused by the problem of vertically integrated monopoly, the Chinese government is making an effort to promote a multiproduct pipeline network reform. The fundamental and indispensable step for this goal is the pipeline network interconnectivity. This paper quantifies the energy-environmental impacts of the pipeline network interconnectivity reform on China's downstream oil supply chain to 2030. An integrated framework is developed to obtain the detailed design-scale information required for assessment, introducing demand forecasting and demand reallocation into a pipeline network optimal planning model. The model is formulated as a fuzzy mixed integer linear programming that optimizes the infrastructure development scheme and supply chain operation plan simultaneously while taking into account demand uncertainty. The results show that, compared with the baseline, the pipeline interconnectivity reform could reduce yearly energy consumption and CO2 emissions by 9.7–19.8% and 12.5–17.9%, respectively. It is shown that this reform policy could overcome infrastructure constraints, increase pipeline utilization, and improve both energy and environmental performance. The proposed framework can be a theoretical guideline for policymakers within and beyond China.

Simultaneous scheduling of multi-product pipeline distribution and depot inventory management for petroleum refineries

Multiple oil-product distribution through a long-distance pipeline to downstream multiple depots is critically important to petroleum refineries and has been extensively studied. Unfortunately, the issues of appropriate handling the trans-mix (TM) between oil-product slugs along the pipeline, as well as the depot inventory management in terms of tank receiving, storage, blending, and discharge operations have not been considered by the previous studies, which could potentially jeopardize their application profitability and operability. Facing these challenges, a general methodology together with a new continuous-time scheduling model have been developed in this paper for simultaneous multi-product pipeline distribution and depot inventory management (MPD-DIM) of a straight pipeline with an input node and multiple output nodes. The scheduling model consists of three sub-models: (i) a preprocessing sub-model for regulating new feeding pipeline slugs; (ii) a TM handling sub-model, and (iii) a product pipeline distribution sub-model with consideration of depot inventory management. Comparably, the development of this study could handle the mixed-interface of oil products appropriately and economically through blending and distillation operations. Meanwhile, the optimization of multi-product pipeline transportation and depot inventory management could be simultaneously accomplished. The efficacy of the developed methodology and MPD-DIM model has been demonstrated by six case studies.

Hybrid method based on particle filter and NARX for real-time flow rate estimation in multi-product pipelines

Real-time flow estimation plays a vital role in multi-product pipeline operations, and the accuracy of real-time flow estimation is affected by noise interference and instrument accuracy and cannot be performed by direct observation of flow meter. Pipeline flow models based on the first principle method are established and employed as soft sensors of pipeline real-time flow rate. However, these models are validated by the controlled experimental pipeline, which may be ineffective regarding actual pipelines with uncertain physical parameters. In this paper, a novel approach integrating data-driven and model-driven method is proposed to estimate the flow rate of petroleum products on-line. The difference between the theoretical model and actual state of a pipeline is accounted for by the friction coefficient, and on-line calibration is achieved by solving multi-objective optimisation problems with asynchronous operation data. The flow state of the pipeline is obtained in real time by the particle filter when new pressure observations with noise become available. The estimation performance of local pressure mutation points is improved by adopting the recurrent nonlinear autoregressive neural network modelling blue of the data-driven method. The effectiveness of the proposed method is evaluated blue by examining actual data of the pipeline over a period of time. The prediction results of some other model-driven and data-driven methods are also compared to blue that of the proposed method. The results blue indicate that the proposed method improves the accuracy and reliability of the product flow rate estimations even under unforeseen operation conditions.

Scheduling of a single-source multiproduct pipeline system by a matheuristic approach: Combining simulated annealing and MILP

Pipeline systems are used for distribution of oil derivatives, where the optimization of transport activities impact considerably in the oil company’s economy. We focused on the scheduling of pumping and delivery operations on straight multiproduct pipeline systems connecting a single-source to multiple-destinations. This paper develops new preprocessing heuristics and a discrete-time mixed integer linear programming (MILP) formulation for the allocation and sequencing activity (ASM model). The new MILP model includes constraints concerning the scheduling of pumping interruptions considering stoppages in order to avoid inventory problems. A preprocessing block is also proposed to better estimate the delivery operations for the initialization batches using a simulated annealing metaheuristic to provide a more robust input data for the ASM model. The results for two case studies of a real-world pipeline system were studied. Better inventory management is observed when facing low demand instances due to its capacity to effectively propose pipeline interruptions.