Fuzzy logic control of a fed-batch fermentor with significant measurement delay

A rule based fuzzy logic controller is developed for control of product concentration in a fed-batch fermentor with a significant measurement delay. The performance of the delay compensated fuzzy logic controller is compared by simulation with that of a delay uncompensated fuzzy controller and with that of a conventional proportional and derivative (PD) controller.     

Fuzzy logic control of a fed-batch fermentor

A rule-based fuzzy logic control is developed for control of penicillin concentration in a fed-batch bioreactor. The membership functions, fuzzy ranges for the error and for the controller output are defined. A fuzzy rule base is constructed relating error to the control output based on operators' knowledge. The performance of the fuzzy-logic controller is evaluated by simulating a mathematical model of the fed-batch bioreactor.     

Fuzzy control of ethanol concentration its application to maximum glutathione production in yeast fed-batch culture

A fuzzy logic controller (FLC) for the control of ethanol concentration was developed and utilized to realize the maximum production of glutathione (GSH) in yeast fedbatch culture. A conventional fuzzy controller, which uses the control error and its rate of change in the premise part of the linguistic rules, worked well when the initial error of ethanol concentration was small. However, when the initial error was large, controller overreaction resulted in an overshoot.An improved fuzzy controller was obtained to avoid controller overreaction by diagnostic determination of "glucose emergency states" (i.e., glucose accumulation or deficiency), and then appropriate emergency control action was obtained by the use of weight coefficients and modification of linguistic rules to decrease the overreaction of the controller when the fermentation was in the emergency state. The improved fuzzy controller was able to control a constant ethanol concentration under conditions of large initial error.The improved fuzzy control system was used in the GSH production phase of the optimal operation to indirectly control the specific growth rate mu to its critical value mu(c). In the GSH production phase of the fed-batch culture, the optimal solution was to control mu to mu(c) in order to maintain a maximum specific GSH production rate. The value of mu(c) also coincided with the critical specific growth rate at which no ethanol formation occurs. Therefore, the control of mu to mu(c) could be done indirectly by maintaining a constant ethanol concentration, that is, zero net ethanol formation, through proper manipulation of the glucose feed rate. Maximum production of GSH was realized using the developed FLC; maximum production was a consequence of the substrate feeding strategy and cysteine addition, and the FLC was a simple way to realize the strategy. (c) 1993 John Wiley & Sons, Inc.     

Novel Hybrid Adaptive Controller for Manipulation in Complex Perturbation Environments

In this paper we present a hybrid control scheme, combining the advantages of task-space and joint-space control. The controller is based on a human-like adaptive design, which minimises both control effort and tracking error. Our novel hybrid adaptive controller has been tested in extensive simulations, in a scenario where a Baxter robot manipulator is affected by external disturbances in the form of interaction with the environment and tool-like end-effector perturbations. The results demonstrated improved performance in the hybrid controller over both of its component parts. In addition, we introduce a novel method for online adaptation of learning parameters, using the fuzzy control formalism to utilise expert knowledge from the experimenter. This mechanism of meta-learning induces further improvement in performance and avoids the need for tuning through trial testing.     

An intelligent control method based on fuzzy logic for a robotic testing system for the human spine

In previous biomechanical studies of the human spine, we implemented a hybrid controller to investigate load-displacement characteristics. We found that measurement errors in both position and force caused the controller to be less accurate than predicted. As an alternative to hybrid control, a fuzzy logic controller (FLC) has been developed and implemented in a robotic testing system for the human spine. An FLC is a real-time expert system that can emulate part of a human operator's knowledge by using a set of action rules. The FLC provides simple but robust solutions that cover a wide range of system parameters and can cope with significant disturbances. It can be viewed as a heuristic and modular way of defining a nonlinear, table-based control system. In this study, an FLC is developed which uses the force difference and the change in force difference as the input parameters, and the displacement as the output parameter. A rule-table based on these parameters is designed for the controller Experiments on a physical model composed of springs demonstrate the improved performance of the proposed method.     

Incorporation of negative rules and evolution of a fuzzy controller for yeast fermentation process

The control of bioprocesses can be very challenging due to the fact that these kinds of processes are highly affected by various sources of uncertainty like the intrinsic behavior of the used microorganisms. Due to the reason that these kinds of process uncertainties are not directly measureable in most cases, the overall control is either done manually because of the experience of the operator or intelligent expert systems are applied, e.g., on the basis of fuzzy logic theory. In the latter case, however, the control concept is mainly represented by using merely positive rules, e.g., “If A then do B”. As this is not straightforward with respect to the semantics of the human decision-making process that also includes negative experience in form of constraints or prohibitions, the incorporation of negative rules for process control based on fuzzy logic is emphasized. In this work, an approach of fuzzy logic control of the yeast propagation process based on a combination of positive and negative rules is presented. The process is guided along a reference trajectory for yeast cell concentration by alternating the process temperature. The incorporation of negative rules leads to a much more stable and accurate control of the process as the root mean squared error of reference trajectory and system response could be reduced by an average of 62.8 % compared to the controller using only positive rules.     

A fuzzy logic controller.

This paper describes a fuzzy sets method which is very useful for handling uncertainties and essential for knowledge acquisition of a human expert. Kinetics of a reactor is often complex and not trivial to describe by mathematical equations. Reactor control by traditional control technology is therefore difficult. A novel technology is presented. In the following a fuzzy inference (approximate reasoning) is used for decision making in analogy to human thinking, facilitating a more sophisticated control. Readers of this paper do not need any advanced mathematics beyond the four basic operations in arithmetic (+, -, x, divided by) and using the maximum and minimum values. This fuzzy inference is introduced to construct a fuzzy logic controller which is suitable for a nonlinear, multivariable and time variant system applied to a bioreactor.     

SIR型传染病的模糊控制

针对SIR型传染病数学模型,将疾病的发展程度这一影响传染病传播的主要因素模糊化,利用条件S0〈P和S0〉p对疫情的影响,建立了一种模糊控制模型,使之在疫情发展的不同阶段对应不同控制措施。     

软计算在生态模型中的应用

由于生态系统的高度复杂性和非线性以及空间数据采集技术的快速发展,近年来越来越多的软计算方法开始应用到生态模拟中来。软计算是个非常广泛的领域,在模式上主要包括元胞自动机、基于个体和盒式模式等;在方法上代表性的有人工神经网络、模糊数学、遗传算法、混沌理论等。重点介绍元胞自动机和规律方法在生态模型中的应用,具体实例包括种群动态模拟、水华预警和生境栖息地模拟。     

Construction of Predictive Models to Describe Apparent and Complex Viscosity Values of O/W Model System Meat Emulsions Using Adaptive Neuro – Fuzzy Inference System (ANFIS) and Artificial Neural Networks (ANN)

This paper introduces an adaptive neuro ?C fuzzy inference system (ANFIS) and artificial neural networks (ANN) models to predict the apparent and complex viscosity values of model system meat emulsions. Constructed models were compared with multiple linear regression (MLR) modeling based on their estimation performance. The root mean square error (RMSE), mean absolute error (MAE) and determination coefficient (R ²) statistics were performed to evaluate the accuracy of the models tested. Comparison of the models showed that the ANFIS model performed better than the ANN and MLR models to estimate the apparent and complex viscosity values of the model system meat emulsions. Coefficients of determination (R ²) calculated for estimation performance of ANFIS modeling to predict apparent and complex viscosity of the emulsions were 0.996 and 0.992, respectively. Similar R ² values (0.991 and 0.985) were obtained when estimating the performance of the ANN model. In the present study, use of the constructed ANFIS models can be suggested to effectively predict the apparent and complex viscosity values of model system meat emulsions.     

Statistical geometry based prediction of nonsynonymous SNP functional effects using random forest and neuro-fuzzy classifiers

There is substantial interest in methods designed to predict the effect of nonsynonymous single nucleotide polymorphisms (nsSNPs) on protein function, given their potential relationship to heritable diseases. Current state-of-the-art supervised machine learning algorithms, such as random forest (RF), train models that classify single amino acid mutations in proteins as either neutral or deleterious to function. However, it is frequently the case that the functional effect of a polymorphism on a protein resides between these two extremes. The utilization of classifiers that incorporate fuzzy logic provides a natural extension in order to account for the spectrum of possible functional consequences. We generated a dataset of single amino acid substitutions in human proteins having known three-dimensional structures. Each variant was uniquely represented as a feature vector that included computational geometry and knowledge-based statistical potential predictors obtained though application of Delaunay tessellation of protein structures. Additional attributes consisted of physicochemical properties of the native and replacement amino acids as well as topological location of the mutated residue position in the solved structure. Classification performance of the RF algorithm was evaluated on a training set consisting of the disease-associated and neutral nsSNPs taken from our dataset, and attributes were ranked according to their relative importance. Similarly, we evaluated the performance of adaptive neuro-fuzzy inference system (ANFIS). The utility of statistical geometry predictors was compared with that of traditional structural and evolutionary attributes employed by other researchers, revealing an equally effective yet complementary methodology. Among all attributes in our feature set, the statistical geometry predictors were found to be the most highly ranked. On the basis of the AUC (area under the ROC curve) measure of performance, the ANFIS and RF models were equally effective when only statistical geometry features were utilized. Tenfold cross-validation studies evaluating AUC, balanced error rate (BER), and Matthew's correlation coefficient (MCC) showed that our RF model was at least comparable with the well-established methods of SIFT and PolyPhen. The trained RF and ANFIS models were each subsequently used to predict the disease potential of human nsSNPs in our dataset that are currently unclassified (http://rna.gmu.edu/FuzzySnps/).     

Experiments and ANFIS modelling for the biodegradation of penicillin-G wastewater using anaerobic hybrid reactor

The performance of an anaerobic hybrid reactor (AHR) for treating penicillin-G wastewater was investigated at the ambient temperatures of 30-35 °C for 245 days in three phases. The experimental data were analysed by adopting an adaptive network-based fuzzy inference system (ANFIS) model, which combines the merits of both fuzzy systems and neural network technology. The statistical quality of the ANFIS model was significant due to its high correlation coefficient R² between experimental and simulated COD values. The R² was found to be 0.9718, 0.9268 and 0.9796 for the I, II and III phases, respectively. Furthermore, one to one correlation among the simulated and observed values was also observed. The results showed the proposed ANFIS model was well performed in predicting the performance of AHR.     

A modular control design method for a flexible manufacturing cell including error handling

This paper introduces a new modular control design method for a cell controller with integrated error handling. To make the complexity of the cell controller manageable, its control logic is separated into two parts: resource allocation control and operation control. To create the operation control not only quickly but correctly, modular operation blocks integrated with error handling are developed. An algorithm automatically generates the operation control. The operation control created by the proposed method is proved to have desired control behaviors. The method is applied to an example system.     

Assessment of human operator functional state using a novel differential evolution optimization based adaptive fuzzy model

With the development of human–machine systems, there has been a growing concern about the consequences of operator performance breakdown under excessive level of workload, especially in safety-critical situations. Assessment and detection of the operator functional state (OFS) enable us to predict the high operational risks of operator. This paper adopts the psychophysiological signals and task performance measures to evaluate OFS under different levels of mental workload. Four indices extracted from electrocardiogram and electroencephalogram, including heart rate (HR), ratio of the standard deviation to the average of HR segment, task load indices (TLI1 and TLI2), are chosen as the inputs of the proposed model. A technique of differential evolution with ant colony search (DEACS) is developed to optimize the parameters of Adaptive-Network-based Fuzzy Inference System (ANFIS). The optimized ANFIS model is employed to estimate the OFS under a series of process control tasks on a simulated software platform of AUTOmation-enhanced Cabin Air Management System. The results showed that the proposed adaptive fuzzy model based on ANFIS and DEACS algorithm is applicable for the operator functional state assessment.     

Backpropagation through time training of a neuro-fuzzy controller

The paper considers gradient training of fuzzy logic controller (FLC) presented in the form of neural network structure. The proposed neuro-fuzzy structure allows keeping linguistic meaning of fuzzy rule base. Its main adjustable parameters are shape determining parameters of the linguistic variables fuzzy values as well as that of the used as intersection operator parameterized T-norm. The backpropagation through time method was applied to train neuro-FLC for a highly non-linear plant (a biotechnological process). The obtained results are discussed with respect to adjustable parameters rationality. Conclusions are made with respect to the appropriate intersection operations too.     

An adaptive fuzzy logic controller using the respiratory quotient as an indicator of overdosage in the baker's yeast process

The baker's yeast process was optimised with a fuzzy logic controller, which is capable of detecting (with the respiratory quotient as indicator) and eliminating overdosage. The controller was developed to enable automatic modification of the set value for the respiratory quotient according to glucose concentration in the broth. With this controller, a cell yield of 55% (w/w) from glucose and a maximum specific growth rate of 0.16 h^–1 were obtained.     

On‐line yeast propagation process monitoring and control using an intelligent automatic control system

The monitoring and control of bioprocesses is a challenging task. This applies particularly if the actions to the process have to be carried out in real‐time. This work presents a system for on‐line monitoring and control of batch yeast propagation under limiting conditions based on a virtual plant operator, which uses the concept of intelligent control algorithms by means of fuzzy logic theory. Process information is provided on‐line using a sensor array comprising the measurement of OD, operating temperature, pressure, density, dissolved oxygen, and pH value. In this context practical problems arising through on‐line sensing and signal processing are addressed. The preprocessed sensor data are fed to a neural network for on‐line biomass estimation. The root mean squared error of prediction is 4 × 10⁶ cells/mL. The proposed system then triggers temperature and aeration by usage of a temperature dependent metabolic growth model and sensor data. The deviation of the predicted biomass from that of the reference trajectory as modeled by the metabolic growth model and its temporal derivative are used as inputs for the fuzzy temperature controller. The inputs used by the fuzzy aeration controller are the deviation of measured extract from that of the reference trajectory, the predicted cell count, and the dissolved oxygen concentration. The fuzzy‐based expert system allows to provide the desired yeast cell concentration of 100–120 × 10⁶ cells/mL at a minimum residual extract limit of 6.0 g/100 g at the required point of time. Thus, a dynamic adjustment of the propagation process to the overall production schedule is possible in order to produce the required amount of biomass at the right time.     

A fuzzy logic controller to control nutrient dosage in a fed-batch baker's yeast process

A fuzzy logic controller designed to control glucose feeding in a fed-batch baker's yeast process is presented. Feeding is carried out in portions and the controller determines the time at which glucose should be added and computes the size of the portion to provide the maximum glucose uptake rate. Moreover, the controller detects and prevents the occurrence of overdosage. The experimental results indicate that yield and specific growth rate obtained with the controller approached 55% and 0.13 h^–1, respectively.