Data-Driven Multiobjective Predictive Control for Wastewater Treatment Process

Data-based robust model predictive control for wastewater treatment process

Social impact as growing concern is becoming an important aspect of the design and operation of wastewater treatment processes. A need has arisen for the development of quantitative indicators of social sustainability. Design and operation of wastewater treatment processes additionally require simple and effective methods to represent and understand the interconnections between the indicators of social sustainability. This paper presents an approach for the development of quantitative social sustainability indicators, and introduces a novel method for defining and visualizing indicator interdependence. It outlines equations for quantitative evaluation of health, safety and security, and comfort. Weighting method of the bipartite network of the relations between the indicators and stakeholders enables clear visualization of the interdependencies of the indicators and facilitates simplification of the set of social sustainability criteria. It creates a basis for reduction of amount of data needed for performing the analysis and reducing the social sustainability assessment’s costs.

Fault detection of sludge bulking using a self-organizing type-2 fuzzy-neural-network

This paper proposes a multi-objective predictive control based on fuzzy satisfactory goal programming. For the different control goals, the multi-objective predictive control problem with priority is formulated. In order to acquire the preferred result, the fuzzy satisfactory goal programming for three types of fuzzy relations is presented. The released order of the satisfactory degrees is modelled to replace priority according to that the higher priority achieves the higher satisfactory degree. The balance between optimization and priority is realized by regulating the parameter. The example shows the proposed method's effectiveness.

Strict effluent quality standards and physical limitations of the equipment make accurately controlling dissolved oxygen (DO) concentration in wastewater treatment processes (WWTPs) a challenging task. In this paper, a data-driven model predictive control strategy with performance constraint (PC-MPC) is proposed to ensure accurate tracking of DO concentration set-point. First, a data-driven model is established using the fuzzy neural network, effectively addressing the problem of complex nonlinear modeling of WWTPs and providing accurate predictive output for the control process. Second, a model predictive control (MPC) strategy with performance constraint is designed, in which tracking error is considered as a performance constraint index within the MPC framework, enabling accurate tracking of the DO concentration to the desired set-point. Third, an adaptive constraint optimization algorithm based on the Levenberg-Marquardt (L-M) method is developed. By utilizing the exponential penalty function to reformulate the system constraints, the finite horizon optimization control problem can be directly solved to determine the optimal control sequence. Finally, the stability of the proposed PC-MPC strategy is proven through theoretical analysis and its effectiveness is validated on the benchmark simulation model No. 1.

Since weather has a huge impact on the wastewater treatment process (WWTP), the prediction accuracy for the Biochemical Oxygen Demand (BOD) concentration in WWTP would degenerate if using only one single artificial neural network as the model for soft measurement method. Aiming to solve this problem, the present study proposes a novel hybrid scheme using a modular neural network (MNN) combining with the factor of weather condition. First, discriminative features among different weather groups are selected to ensure a high accuracy for sample clustering based on weather conditions. Second, the samples are clustered based on a density-based clustering algorithm using the discriminative features. Third, the clustered samples are input to each module in MNN, with the auxiliary variables correlated with BOD prediction input to the corresponding model. Finally, a constructive radial basis function neural network with the error-correction algorithm is used as the model for each subnetwork to predict BOD concentration. The proposed scheme is evaluated on a standard wastewater treatment platform—Benchmark Simulation Model 1 (BSM1). Experimental results demonstrate the performance improvement of the proposed scheme on the prediction accuracy for BOD concentration in WWTP. Besides, the training time is shortened and the network structure is compact.

Multi-objective nonlinear predictive control of process systems: A dual-mode tracking control approach

Adaptive type-2 fuzzy-neural switching control for wastewater treatment process under several operating conditions

This paper describes a real-time multi-objective predictive control strategy for the wheelchair ergometer platform, which allows simulating the manual wheelchair (MWC) propulsion in virtual reality (VR). A nonlinear least-square method is used to build the predictive wheelchair ergometer model, where the trust-region-reflective algorithm is adopted. A detailed study of the wheelchair dynamics, and its performance on the rollers, and the frictions are presented. The main contribution is the implementation of a wheelchair ergometer model into a proposed controller in order to provide realistic navigation within the VR world by accurately detecting and tracking the rotation of the driving wheels. In particular, the proposed strategy has two objectives: 1) to generate force feedback (haptics) during the push phase (hand-to-rim contact); 2) assistance control to make the wheelchair movement more realistic in the immersive virtual environment during the recovery phase (hand-to-rim no contact). Various numerical examples emphasize the flexibility of the approach and experimental results confirm that the proposed controller successfully tracks the speed reference. It is also shown that the real-time tracking ability of the explicit MPC (eMPC) controller is superior to that of the PI controller.

A PID-type controller incorporating neural network elimination scheme and sliding-mode control action for different objectives including H/sub 2/ tracking performance, H/sub /spl infin// tracking performance, and regional pole constraints is developed in robotic systems under plant uncertainties and external disturbances. The adaptive neural networks are used to compensate the plant uncertainties. The sliding-mode control action is included to eliminate the effect of approximation error via neural network approximation. The sufficient conditions are developed for different objectives in terms of linear matrix inequality (LMI) formulations. The interesting combinations of different objectives are considered in this paper, which include H/sub /spl infin// PID tracking control design with regional pole constraints and mixed H/sub 2//H/sub /spl infin// PID tracking control design with regional pole constraints. These multiobjective PID control problems are characterized in terms of eigenvalue problem (EVP). The EVP can be efficiently solved by the LMI toolbox in Matlab. The proposed methods are simple and the PID control gain for different objectives can be obtained systematically.

In this paper, we extend the notions of $$(\Phi ,\rho )$$ ( ? , ? ) -invexity and generalized $$(\Phi ,\rho )$$ ( ? , ? ) -invexity to the continuous case and we use these concepts to establish sufficient optimality conditions for the considered class of nonconvex multiobjective variational control problems. Further, multiobjective variational control mixed dual problem is given for the considered multiobjective variational control problem and several mixed duality results are established under $$(\Phi ,\rho )$$ ( ? , ? ) -invexity.

Because of highly complexity and large-scale operation, wastewater treatment process (WWTP) is considered as a robust adaptive control problem. This paper proposes a data-driven robust adaptive control with deep learning (DRAC-DL) for WWTP to improve the operational performance. In ths DRACDL framework, a robust controller is firstly designed to construct the closed-loop control scheme. Meanwhile, an adaptive deep belief network (ADBN) is developed based on the self-incremental learning strategy to approximate the ideal control law. The main advantage of DRAC-DL lies in its improved robustness and low computational burden, which benefit from Lyapunov-based closed-loop controller and efficient ADBN approximator. Finally, the effectiveness of DRAC-DL is demonstrated on the benchmark simulation model No.1 (BSMI) of WWTP.

Aiming at the frequency instability and the increase of control cost caused by the uncertainty of the island microgrid, this paper proposes a multi-objective control strategy for the island microgrid based on the Soft Actor Critic (SAC) algorithm. Firstly, considering the influence of the output distribution of micro gas turbines on the control cost, a multi-objective control model that coordinates control cost and frequency stability is constructed. Secondly, in order to cope with the sparse reward task of multi-objective control, this paper takes the frequency deviation and control cost as the state set, takes the unit output as the action set, and designs the corresponding reward function, thus completing the process design of the multi-objective control strategy. Finally, the calculation example shows that the SAC controller can effectively reduce the control cost on the premise of ensuring the frequency regulation requirements of the microgrid, and the comprehensive control effect is far superior to traditional control methods such as PI control and fuzzy control.

A fuzzy neural network-based soft sensor for modeling nutrient removal mechanism in a full-scale wastewater treatment system

The wastewater treatment process (WWTP) is of great significance to environmental protection. To improve the efficiency of the WWTP, it is crucial to ensure that the dissolved oxygen (DO) concentration tracks the set value efficiently. Due to the nonlinear and time-varying dynamics of the WWTP, traditional control methods cannot accurately control the DO concentration. To overcome these challenges, this paper proposes an online transferred heuristic dynamic programming (TrHDP) control design by combining transfer learning with adaptive critic design. First, we use the historical sample data to construct a mathematical model of the WWTP and learn the prior knowledge from the model. Then, the online control process of the DO concentration is guided by utilizing the prior knowledge. In order to avoid negative transfer and save computing resources, we design a novel decay function with the truncation mechanism. In addition, we prove the stability of the TrHDP control scheme by constructing a Lyapunov function. Finally, the performance of the TrHDP scheme is verified by the Benchmark Simulation Model No. 1. Compared with other methods, the TrHDP method possesses higher control accuracy for the DO concentration and overcomes the disadvantage of low learning efficiency of general online methods.