Building energy modeling (BEM) is fundamental for achieving optimized energy control, resilient retrofit designs, and sustainable urbanization to mitigate climate change. However, traditional BEM requires detailed building information, expert knowledge, substantial modeling efforts, and customized case-by-case calibrations. This process must be repeated for every building, thereby limiting its scalability. To address these limitations, we developed a modularized neural network incorporating physical priors (ModNN), which is improved by its model structure incorporating heat balance equations, physically consistent model constraints, and data-driven modular design that can allow for multiple-building applications through model sharing and inheritance. We demonstrated its scalability in four cases: load prediction, indoor environment modeling, building retrofitting, and energy optimization. This approach provides guidance for future BEM by incorporating physical priors into data-driven models without extensive modeling efforts, paving the way for large-scale BEM, energy management, retrofit designs, and buildings-to-grid integration.

Industrial process optimization and control is crucial to increase economic and ecologic efficiency. However, data sovereignty, differing goals, or the required expert knowledge for implementation impede holistic implementation. Further, the increasing use of data-driven AI-methods in process models and industrial sensory often requires regular fine-tuning to accommodate distribution drifts. We propose the Artificial Neural Twin, which combines concepts from model predictive control, deep learning, and sensor networks to address these issues. Our approach introduces decentral, differentiable data fusion to estimate the state of distributed process steps and their dependence on input data. By treating the interconnected process steps as a quasi neural-network, we can backpropagate loss gradients for process optimization or model fine-tuning to process parameters or AI models respectively. The concept is demonstrated on a virtual machine park simulated in Unity, consisting of bulk material processes in plastic recycling.

In response to the fact that autonomous vehicles cannot avoid obstacles by emergency braking alone, this paper proposes an active collision avoidance method for autonomous vehicles based on model predictive control (MPC). The method includes trajectory tracking, adaptive cruise control (ACC), and active obstacle avoidance under high vehicle speed. Firstly, an MPC-based trajectory tracking controller is designed based on the vehicle dynamics model. Then, the MPC was combined with ACC to design the control strategies for vehicle braking to avoid collisions. Additionally, active steering for collision avoidance was developed based on the safety distance model. Finally, considering the distance between the vehicle and the obstacle and the relative speed, an obstacle avoidance function is constructed. A path planning controller based on nonlinear model predictive control (NMPC) is designed. In addition, the alternating direction multiplier method (ADMM) is used to accelerate the solution process and further ensure the safety of the obstacle avoidance process. The proposed algorithm is tested on the Simulink and CarSim co-simulation platform in both static and dynamic obstacle scenarios. Results show that the method effectively achieves collision avoidance through braking. It also demonstrates good stability and robustness in steering to avoid collisions at high speeds. The experiments confirm that the vehicle can return to the desired path after avoiding obstacles, verifying the effectiveness of the algorithm.

In this paper, a robust control method is introduced for autonomous vehicle control in different scenarios. Dual controllers have been used in this method to ensure high performance and low errors during the vehicle\'s trip. The new control system is called Model Predictive and Stanley based controller (MPS), which is an integration of a model predictive controller and a Stanley controller. Each of these two controllers has its drawbacks and weaknesses. The proposed method tries to overcome these points and come up with a high-performance control system. This hybrid way of combining two of the famous controllers has the benefit of using the best part of each one and trying to enhance the other part. The MPS is tested for both path-following and vehicle control in different scenarios and on both straight and curved roads. This controller has shown high performance and flexibility to deal with different scenarios of autonomous driving. The results are compared to previous types of controllers, and the proposed system outperformed these types.

OBJECTIVE: Precise control of neural systems is essential to experimental investigations of how the brain controls behavior and holds the potential for therapeutic manipulations to correct aberrant network states. Model predictive control, which employs a dynamical model of the system to find optimal control inputs, has promise for dealing with the nonlinear dynamics, high levels of exogenous noise, and limited information about unmeasured states and parameters that are common in a wide range of neural systems. However, the challenge still remains of selecting the right model, constraining its parameters, and synchronizing to the neural system.
METHODS: As a proof of principle, we used recent advances in data-driven forecasting to construct a nonlinear machine-learning model of a Hodgkin-Huxley type neuron when only the membrane voltage is observable and there are an unknown number of intrinsic currents.
RESULTS: We show that this approach is able to learn the dynamics of different neuron types and can be used with MPC to force the neuron to engage in arbitrary, researcher-defined spiking behaviors.
CONCLUSIONS: To the best of our knowledge, this is the first application of nonlinear MPC of a conductance-based model where there is only realistically limited information about unobservable states and parameters.

This research introduces a new technique to control constrained nonlinear systems, named Lyapunov-based neural network model predictive control using a metaheuristic optimization approach. This controller utilizes a feedforward neural network model as a prediction model and employs the driving training based optimization algorithm to resolve the related constrained optimization problem. The proposed controller relies on the simplicity and accuracy of the feedforward neural network model and the convergence speed of the driving training based optimization algorithm. The closed-loop stability of the developed controller is ensured by including the Lyapunov function as a constraint in the cost function. The efficiency of the suggested controller is illustrated by controlling the angular speed of three-phase squirrel cage induction motor. The reached results are contrasted to those of other methods, specifically the fuzzy logic controller optimized by teaching learning-based optimization algorithm, the optimized PID with particle swarm optimization algorithm, the neural network model predictive controller based on particle swarm optimization algorithm, and the neural network model predictive controller using driving training based optimization algorithm. This comparative study showcase that the suggested controller provides good accuracy, quickness and robustness due to the obtained values of the mean absolute error, mean square error root mean square error, enhancement percentage, and computing time in the different simulation cases, and it can be efficiently utilized to control constrained nonlinear systems with fast dynamics.

Nonholonomic constrained wheeled mobile robot (WMR) trajectory tracking requires the enhancement of the ground adaptation capability of the WMR while ensuring its attitude tracking accuracy, a novel dual closed-loop control structure is developed to implement this motion/force coordinated control objective in this paper. Firstly, the outer-loop motion controller is presented using Laguerre functions modified model predictive control (LMPC). Optimised solution condition is introduced to reduce the number of LMPC solutions. Secondly, an inner-loop force controller based on adaptive integral sliding mode control (AISMC) is constructed to ensure the desired velocity tracking and output driving torques by combining second-order nonlinear extended state observer (ESO) with the estimation of dynamic uncertainties and external disturbances during WMR travelling process. Then, Lyapunov stability theory is utilised to guarantee the consistent final boundedness of the designed controller. Finally, the system is numerically simulated and practically verified. The results show that the double-closed-loop control strategy devised in this paper has better control performance in terms of complex trajectory tracking accuracy, system resistance to strong interference and computational timeliness, and is able to realise effective coordinated control of WMR motion/force.

Linear induction machines (LIMs) find widespread adoption in various applications, owing to their inherent advantages such as low noise, compact turning radius, and excellent climbing capability. LIMs are extensively utilized in linear metro applications. However, in practical operations, the output thrust shrinks with the increase in speed, which is attributed to end effects. This phenomenon leads to a reduction in efficiency. In addition, fluctuations in the normal force impact system stability, posing disturbances to the suspension system. To address these challenges, this paper suggests a finite-set model predictive thrust control (FS-MPTC) for a drive system employing a linear induction motor (LIM). The FS-MPTC optimizes the duty cycle for the active voltage vector and allocates the remaining period to one of the zero voltage vectors. The selected zero voltage vector reduces the switching transition between it and the active voltage vectors. The duty cycle is calculated for the six voltage vectors by incorporating the deadbeat concept and the derivative value for the electromagnetic thrust. In the prediction stage of the FS-MPTC, the computed duty cycle and the corresponding voltage vector are used simultaneously and repeated for the six voltage vectors. The cost function comprises two terms: the error between the reference thrust and predicted thrust value as the first term and the error between the rated primary flux linkage and its predicted value as the second term. The reference thrust is generated from the outer speed control loop. To validate the effectiveness of the proposed control approach, Japanese 12000 linear induction machine parameters are employed for verification. Comparative analysis of the performance between the suggested control method with the optimal duty cycle and the same process with a fixed duty cycle demonstrates superior performance when utilizing the optimal duty cycle. Finally, the proposed FS-MPTC with the optimal duty cycle offers a promising solution to enhance the operational efficiency of the LIM-based drive systems.

The optimal control design of the boiler-turbine system is vital to ensure feasibility and high responsiveness over desired load variations. Using the traditional linear control techniques realization of this task is difficult, as the boiler-turbine mechanism has strong nonlinearities. Besides, environmental and economic concerns have replaced existing tracking control ones as the primary concerns of advanced power plants. Thus, this study proposes an optimal economic model predictive controller (EMPC) scheme for this unit on the basis of the input/output feedback linearization (IOFL) method. By employing the IOFL method, this unit is decoupled into a new linearized model that is utilized for developing the suggested optimal IOFL EMPC technique. The proposed control scheme is formulated in an economic quadratic programming form that considers the input-rate and input limits of the unit for optimal economic performance. In addition, an adaptive iterative algorithm is utilized for constraints mapping with guaranteeing a feasible solution in a finite number of steps without violation of original constraints over the entire predictive horizon. The outcomes of the simulation show that the suggested optimal IOFL EMPC scheme offers an improved dynamic and economic output performance over fuzzy hierarchical MPC, fuzzy EMPC, and nonlinear EMPC techniques during various load variations.

The effectiveness of an aquaponic system significantly relies on the habitat provided for both the fish and plants. As an integral component of aquaponics, hydroponic cultivation benefits greatly from the controlled environment of a greenhouse. Within this environment, factors such as temperature, carbon dioxide levels, humidity, and light can be carefully adjusted to maximize plant growth and development. This precise regulation ensures an ideal growing environment, fostering the flourishing of plants and contributing to the overall success of the aquaponic ecosystem. This study presented a control approach for an aquaponic greenhouse system. It aims to keep the greenhouse climate parameters (temperature, CO2 concentration, and humidity) at their ideal levels. The proposed control strategy is a two-layered mechanism in which the first layer presents an optimization framework using particle swarm optimization (PSO) algorithm to give the setpoints for the controller, and the second layer demonstrates a constrained discrete model predictive control (CDMPC) strategy to maintain the desired trajectories received from the optimization layer. To validate the results obtained using PSO, this study incorporates genetic algorithms (GA) and assesses their performance in comparison. Given similar computational efficiency and low computational time for both algorithms, the optimal values identified by particle swarm optimization (PSO) are adopted as the setpoints. Two performance criteria, relative average deviation (RAD) and mean relative deviation (MRD), are derived to evaluate the tracking performance of the proposed CDMPC controller under external disturbances. A comparison of the proposed CDMPC with the PI controller is also offered. According to the comparison results, our proposed CDMPC performs better than the PI controller with lower RAD values (temperature, 1.1315; CO2 concentration, 0.9225; humidity, 2.547) and MRD values (temperature, 0.315; CO2 concentration, 0.25; humidity: 1.013). The controller is validated to be efficient by its strong control performance, highlighted by robustness, efficient setpoint tracking, and adequate disturbance rejection. This novel approach might prove to be a useful technique for developing environmental control strategies that can be used for potentially boosting production rates of aquaponic greenhouse systems, maximizing profitability, and reducing labor needs. By maintaining optimal conditions, it can enhance ecosystem health, improve yields, and streamline operations, paving the way for greater system performance and sustainability.