DEVELOPMENT OF A CONTROL SYSTEM FOR THE PROCESS OF OXIDATIVE CHLORINATION OF ETHYLENE USING A VIRTUAL ANALYZER AND A FUZZY CONTROLLER

HighlightsAiming at the difficult problem of controlling the electric heating furnace, combining the advantages of Auto-Encoder and fuzzy control, a composite control algorithm is proposed for dynamic modelling of the electric heating furnace, predicting the future temperature and adjusting the control parameters in real time in order to achieve accurate and stable control of the temperature, which provides a new way of thinking for solving the control problems of nonlinear, time-varying, and large-time-lag systems.What are the main findings?A discrete mathematical model of an electric heating furnace was established, and unsupervised dynamic modelling was achieved through an auto-encoder.A control structure combining predictive compensation and fuzzy regulation was designed to achieve stable low-overshoot control under complex interference.What is the implication of the main finding?Improved the stability, accuracy, and robustness of the electric heating furnace temperature control system.Provided a new modelling and control framework for solving control problems in nonlinear, time-varying, and large-time-delay industrial heating systems, with good application prospects.Electric heating furnaces are widely used in industrial production and scientific research, where the quality of temperature control directly affects product performance and operational safety. However, precise control remains challenging due to the system’s nonlinear behaviour, time-varying characteristics, and significant time delays. To overcome these issues, this paper proposes a composite control method that integrates an auto-encoder-based prediction model with fuzzy PI control. Specifically, a discrete-time temperature model is constructed, in which the auto-encoder learns the system dynamics and predicts future temperatures, while the fuzzy controller adaptively tunes the PI parameters in real time. This approach improves both modelling accuracy and the adaptability of the control system. The simulation results on the MATLAB/Simulink platform show that the proposed method maintains the temperature overshoot within 2% under various disturbances, including a maximum delay of 243 s, ±2 °C measurement noise, 10% voltage fluctuation, and abrupt 10% gain variation. These results demonstrate the method’s strong robustness and indicate its suitability for advanced control design in complex industrial environments.

The self-adaptive fuzzy PID control method is not affected by the parameters of the system and the changes of the environment greatly. It has faster convergence and stronger robustness. The most prominent is that the control method can improve the performance of the control system by continuously correcting the control rules and adjusting the system parameters in real time. In this paper, PID control, fuzzy control and self-adaptive fuzzy PID control methods are respectively used to control the linear two-stage inverted pendulum system, and the advantages of self-adaptive fuzzy PID control method are gained by comparison.

This paper presents a design methodology of a dual-input single-output fuzzy logic controller where synthesizing the classical three stages, fuzzification, inference engine, and defuzzification, are replaced by the outcome control surface obtained from these stages which is treated as a tow dimensional table called fuzzy control table FCT. With this proposed approach, (8x8), (16x16), (32x32), and (64x64) FCTs were investigated each having 64, 256, 1024, and 2048 values respectively. To make this system adaptable to different operating states a supervisor fuzzy controller is designed to continuously adjust, on line, the output factor of the basic fuzzy controller based on the error and change in error signals. The proposed architecture is implemented in XC3S200 FPGA, Spartan-3 starter kit to control the position of a D.C. servo motor with unknown parameters in real time. Keywords: control table, FPGA, fuzzy control surface, supervisor controller.

To improve the dynamic performance of the output voltage of the new interleaved shunt bi-directional DC-DC converter, a double closed-loop control method of voltage and current based on fuzzy PI control is proposed. Firstly, based on the study of the operating principle and steady-state performance of the new interleaved parallel DC-DC converter, a small-signal mathematical model of the interleaved parallel bidirectional DC-DC converter with different modes is established, and based on the model, the transfer functions from duty cycle to output voltage and inductor current are derived, and accordingly, the double-closed-loop controller of the bidirectional DC-DC converter is designed. Then the design method of the fuzzy PI controller is described in detail, and the fuzzy control is applied to the voltage outer loop of the double closed-loop control so that the system can adjust the PI parameters in real time. Simulation experiments are carried out by Matlab/Simulink to compare the simulation experimental waveforms using fuzzy PI control and conventional PI control, and it is verified that the fuzzy PI control improves the output voltage response speed of the new interleaved shunt bi-directional DC-DC converter, reduces the voltage peak of the system, and reduces the overshoot.

Aiming at eliminating and controlling the swing of load and realizing the precise positioning of the two-degree-of-freedom bridge crane, this paper uses the Lagrange equation to establish the nonlinear differential equation of the dynamic model of the bridge crane. Considering the nonlinearity, time variation and uncertainty of the crane system, the fuzzy self-adaptive PID controller is the fusion of PID controller and fuzzy controller. The function of fuzzy language is to regulate the PID parameters in real time. The model and dynamic simulation of the anti-sway system are studied by SIMULINK, to verify the fuzzy self-adaptive PID control method is feasible. Compared with the traditional PID control method, the fuzzy self-adaptive PID controller has better adaptability and robustness according to the analysis of the simulation results. The control process is more stable and the response speed is faster, which can improve the dynamic performance of the crane anti-swing system.

Power conditioning system (PCS) is an interface between distributed generation and utility grid. It regulates voltage, current and power transmitted from the hybrid direct fuel-cell/turbine (DFC/T) power plant to utility grid. This paper presents a self-adaptive fuzzy PI controller of three phase inverter in PCS for the DFC/T power plant. One of the main tasks of PCS for distributed generations is power flow control. Traditionally, the control scheme for grid connected inverters is PI controller. However, PI control scheme would lead to large overshoot and long response time when the load increases sharply. Thus, a hybrid fuzzy PI control scheme is proposed in this paper in order to improve the power flow control. The fuzzy controllers can adjust the PI parameters according to the voltage error and the derivative of voltage error. The overall self-adaptive tuning fuzzy PI controller can update the PI parameters in real time, which makes the power flow control much more accurate than conventional PI controllers. Analysis and design methodologies of hybrid fuzzy PI controllers are presented in this paper. The model was developed in Matlab/Simulink environment. Simulation results show that the proposed control scheme can effectively reduce the overshoot and response time compared to the conventional PI controllers.

The purpose of the study is to evaluate the possibility of using artificial neural networks for electrical signals processing.
 
 Methods. The application of a multilayer perceptron for these purposes as the simplest neural feed forward network is considered. Its difference as the basis of neural network algorithms is that when analyzing dynamic processes, signal processing should be carried out in a “sliding time window”.
 
 Results. It is shown that neural network processing makes it possible to approximate the shape of the signal with high accuracy and determine its parameters in real time. Using the example of periodic signals and transients in electrical circuits, accuracy assessments are made and the features of neural network processing are analyzed. The necessary sizes of the training sample of signals and the level of errors that occur when testing a neural network are discussed. Estimates of the required signal sampling frequency, the “sliding window” duration and the variation range of signal parameters when creating a training sample are given.
 
 Conclusions. It is shown that the proposed approach does not require “deep learning” of neural networks with complex architecture. It enables to create a signals training sample based on simple analytical formulas and to control the neural network algorithm quality at intermediate stages of calculations.

In order to improve the ability of rotating turntable to resist the external disturbance in the single-axis rotation modulation inertial navigation system (SRMINS) and ensure smooth running of the turntable, we designed a dual closed-loop controller. The conventional proportional-integral-derivative (PID) algorithm was employed in the position loop, and a tracking differentiator based fuzzy PID algorithm was proposed and used in the speed loop, to realize the steady speed control of continuous rotation in the SRMINS. The introduction of the tracking differentiator (TD) guarantees the reliability of the differential signal extraction. Furthermore, the combination of the TD and the fuzzy adaptive PID controller leads to the smooth motion of the turntable without overshoot. The robustness and better stability margin of the system can also be achieved while the static error is eliminated and the performance of fuzzy control is improved. Simulation results demonstrate that the system has many merits such as fast response, no overshoot, and better robustness. It is also verified that regulating the PID parameters in real time by the fuzzy control rules effectively improves the dynamic and static performance of the system, and the requirement on stable rotation speed of the SRMINS is satisfied.

Improved beluga whale optimization-based variable universe fuzzy controller for brushless direct current motors of electric tractors

Humidity is a key factor affecting proton exchange membrane fuel cell (PEMFC) efficiency and output performance. Different working conditions have different requirements for humidity. Too high or too low water content inside the PEMFC will damage the output performance and even affect the PEMFC life. Therefore, how to control humidity appropriately is a crucial subject. This paper establishes a PEMFC internal water management and inlet humidity model, and analyzes the influence of the anode inlet humidity change on the performance of the PEMFC and the water content of the membrane through computational fluid dynamics (CFD) simulation. The direct control of the inlet humidity, which is difficult to be accurately measured, is converted to the temperature control of the bubble humidifier according to the proposed model, and a back propagation neural network proportion integration differentiation (BPPID) controller is proposed, which combines artificial neural network and digital PID control to adjust PID parameters in real time. The controller is applied to the temperature control of the bubble humidifier and compared with the traditional PID controller and the fuzzy PID controller. It is found that the performance of the BPPID controller is better through the comparison with the experimental results.

Smart de-watering and production system through real-time water level surveillance for Coal-Bed Methane wells

In the past few decades, Coal-Bed Methane (CBM) has become an important source of energy especially in North America. The methane adsorbed within the coal is in a near-liquid state. The open fractures in the cleats are commonly saturated with water. To develop CBM reservoir, water in the fracture and coal seam must be continuously pumped off from coal seam to reduce pressure and desorb gas from matrix. Although operators desire to produce hydrocarbon quickly, a too fast dewatering rate can irreversibly damage matrix desorption process, which can lead to an unfavorable ultimate recovery. Further, the aggressive production rate can potentially release the coal fines and drive them into pump, which increases maintenance effort and cost. Even worse, the de-watering process fluctuates because of the rock porosity and permeability changes resulting from the brittle coal seam and pressure reduction. Therefore, it is critical to adjust the pump operating parameters in a timely manner to maintain a continuous/intermittent production. Ten CBM wells are located in a remote area, which makes the access to wellsites difficult. Previously engineers had to evaluate well performance and optimize the pump on-site, which is limited by a monthly basis. We firstly developed an automatic data processing system using the advanced Echosounders, which can measure the water level in real time. The reservoir pressure can be then monitored dynamically through interpreting the detected water level. With an automatic-wireless data transferring system installed on-site and a closed-loop control program to receive, process, and interpret data, the pump operating parameters can be changed in real time through remote control. This system not only identifies the downhole problems in real time, but also reduces the pump maintenance frequency from 40 days to 75 days statistically and numbers of trip to well site. Further, the gas production rate has been averagely improved by 30% for the 10 wells. The authors firstly developed an automation data processing and control system in the favor of advanced echosounders. Based on the interpreted reservoir pressure, we can avoid aggressive production by adjusting the pump operating parameters in real time, which eventually results in a better ultimate recovery. The developed workflow (automatic echosounder data acquisition, real filed data transferred to central office, data processing, interpretation, and simulation in computational system, adjustment commands to operating system) is especially valuable for the locations difficult to access.

In order to meet the needs of scientific research in space medicine and biology, a new multifunctional automated cell sample culture device for a Chinese space station has been designed. The temperature and carbon dioxide concentration are adjustable, making it convenient for cell culture in microgravity environments of the space station. A centrifuge is used to simulate the microgravity environment, allowing for synchronous gravity and microgravity comparison during cell culture. An automated focusing visible light microscope has been designed, capable of real-time photography of cultured cells, which can receive ground commands to complete automatic focusing and image transmission. The thermal design of the cell sample culture device uses an air heating method, and the rationality of the thermal control measures has been verified through thermal simulation analysis. The designed cell sample preparation device can monitor and display the cell growth environment parameters and device performance parameters in real time on orbit. It can also control the internal temperature within the temperature range required for cell culture. Thus, it can meet the urgent needs of various cell cultures, experiments, and scientific research on a Chinese space station.

The regulating method for adaptive ventilation system is proposed. The method is based on a predicted assessment of air parameters in real time. It takes into account the dynamics of air temperature and carbon dioxide concentration in the supply and exhaust air ducts of ventilation system. The predicted parameter estimate in the subsequent time period is calculated using the derivative of parameter. The decision making algorithm of the intelligent adaptive control system has been developed.

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper OTC 31680, “Successful Development and Deployment of a Global ROP-Optimization Machine-Learning Model,” by Timothy S. Robinson, SPE, Exebenus, Peter Batruny, SPE, Petronas, and Dalila Gomes, SPE, Exebenus, et al. The paper has not been peer reviewed. Copyright 2022 Offshore Technology Conference. Reproduced by permission. _ Drilling rate of penetration (ROP) is influenced by many factors, both controllable and uncontrollable, difficult to distinguish with the naked eye. Thus, machine-learning (ML) models such as neural networks have gained momentum in the drilling industry. Previous models were field-based or tool-based, which affected accuracy outside of the trained field. The authors of the complete paper aim to develop one generally applicable global ROP model, reducing the effort needed to redevelop models for every application. Introduction The authors have identified a need for an ROP model that can recommend parameters in real time, which ideally requires a general ROP model that can be applied with low prediction errors. Formation properties are available in real time from logging tools; however, incorporating logging data into a global ROP model is challenging. A distance usually exists between the bit and logging tools where the properties of the formation drilled in real time at the bit are not available. While strides have been made in ML interpretation of logs, human interpretation is still needed. Equipment and Processes A deep neural network was used for modeling ROP. This model was queried as a black box within the scope of a constrained optimization algorithm tasked with identifying combinations of controllable drilling parameters expected to increase ROP. An overview of the ROP optimization system powered by the ROP estimation model is shown in Fig. 1. The ROP model was fitted using training and validation data sets composed of data from independent wells. Four additional wells were held in reserve for final testing purposes, most importantly for assessing the model’s ability to generalize to completely unseen data. For optimization, a three-stage conditional parameter sweep was used where the value expected to optimize ROP was found sequentially for each of the controllable parameters while fixing the values of the other variables. The optimizing value for each parameter was used as a fixed value for the subsequent optimization steps. This approach maintains the simplicity of a grid search and converges to a local optimum quickly because of the relatively small number of search iterations required under this formulation. A variety of well types with global geographical diversity was present in the data, using different downhole drives and bottomhole assemblies. Twenty-nine wells were used for training the ROP model, with an additional five wells used for validation and a further four wells reserved for holdout testing. No overlap existed between the wells constituting the training, validation, and holdout data sets. The drilling data sets were preprocessed by methods detailed in the complete paper. To ensure consistency across all operations, an average ROP was calculated independently based on depth measurements and their associated timestamps. This ROP calculation was used as the target variable for the ROP estimation model and for assessing its predictive performance. Training and validating the neural network for ROP estimation using data from a variety of independent wells is very useful for assessing model generalization. This technique is superior to the common approach of randomly splitting data sets because it minimizes information leakage between training and validation data.