Optimization Time Research Articles

Application of improved particle swarm optimization in the energy saving strategy for a central air‐conditioning system

AbstractHere, a new energy‐saving optimization method, known as the Improved Particle Swarm Optimization (IPSO) algorithm, is proposed to solve the practical engineering problem of high energy consumption during the operation of central air‐conditioning water systems. The mathematical model of total energy consumption optimization and the calculation principle of the IPSO are established. It is shown via experiments that the proposed algorithm requires a short optimization time and can achieve the optimal operation conditions of water systems under various cooling loads, where these conditions include the optimal control strategy of equipment, and water temperature at the inlet and outlet of chillers. Furthermore, it ensures the lowest energy consumption of water system operation. The measured results show that the comprehensive energy saving rate of water system operation can reach more than 18% compared with that before the system transformation.

A novel machine learning-based model predictive control framework for improving the energy efficiency of air-conditioning systems

The dynamic optimization of key setpoints (e.g., supply water temperatures of chillers and cooling towers, or indoor temperature and humidity) can track the efficient performance point of the air-conditioning system (ACS), thus obtaining a great energy-saving effect. The model predictive control is an effective way to precisely adjust these setpoints. However, most of the existing studies focused on simple linear strategies, e.g., experience-based, rule-based, or component-based, which results in unsatisfactory energy efficiency. Aiming at this, the paper proposed a novel machine learning-based model predictive control (MLB-MPC) framework with predictive inputs of disturbances. The framework first employed the grid search and cross-validation to optimize the total energy consumption prediction models (TECPMs), which were built by data-driven models (e.g., multiple linear regression, artificial neural network, support vector regression (SVR), and random forest). And then the optimization performances of each TECPM to the controlled variables were obtained by using different optimization algorithms (e.g., genetic algorithm, particle swarm optimization (PSO), and simulated annealing). Finally, the optimal match of TECPMs and optimization algorithms was achieved by trade-off among prediction accuracies, optimization accuracies, and optimization time. The case studies demonstrated that the optimal match for the MLB-MPC is SVR and PSO as they had the highest prediction accuracy (mean absolute percentage error of 2.5%) and shortest optimization time (41 ms/per) and optimization accuracies with little difference. After adopting the optimal match, the MLB-MPC with predictive disturbing inputs achieved a great energy-saving ratio of 7.1% for the ACS, which was only less than the MLB-MPC with ideal disturbing inputs by 0.4%. This indicated that exact predictions of disturbances (e.g., cooling load and outdoor wet-bulb temperature) are important for the MLB-MPC. The proposed MLB-MPC framework would provide a method for improving the energy efficiency of ACSs.

Knowledge percolation threshold and optimization strategies of the combinatorial network for complex innovation in the digital economy

Digital economy expands the source of knowledge for innovation and accelerates the flow and combination of knowledge to form novel knowledge combinations, thereby generating the interdisciplinary knowledge production model. In this context, complex innovation which is characterized by the knowledge production consequence based on the combinations of multiple-field knowledge has become the new way for firms to seize new development opportunities and compete in the digital economy. Given that complex innovation emerged from a gradually forming large, multilayered, combinatorial network consists of collaboration networks in various knowledge fields that are initially separated, the challenge of facillatating the emergence of complex innovation is unveiling the minimum proportion of connected paths in the combinatorial network to trigger effective transmission of multi-fields knowledge and offering applicable optimization strategies to optimize that proportion. This study incorporated Ohm's law into the percolation theoretical framework and calculate the knowledge percolation threshold of the combinatorial network and its subnetworks with patent data of Chinese strategic emerging industries. We further examined the optimization results of six strategies in terms of their optimization effects and time costs. Accordingly, we revealed the probability of knowledge percolation occurring in a combinatorial network and its subnetworks, clarified knowledge transmission characteristics according to knowledge-based cluster dynamics, and determined strategies for optimizing the knowledge percolation threshold. This study is not only highly feasible and exercisable for academics to conduct future studies, but it also has vital implications for the practitioners to utilize and control the knowledge transmission of the combinatorial network to realize the complex innovation.

Optimisation of methanol distillation using GA and neural network hybrid

ABSTRACT Distillation is an energy-intensive non-stationary process represented using non-linear model equations and involves multiple objectives. For such processes, data-based multi-objective optimization methods are more suitable compared to conventional non-linear optimization methods. Therefore, a surrogate-assisted multi-objective optimization (SAMOO) approach is developed by hybridizing an artificial neural network (ANN) and genetic algorithm (GA) to simultaneously minimize the annualized capital expenditure cost (ACAPEX) and annualized operational expenditure cost (AOC) for the methanol separation process. The approach is then extended for operational optimization to maximize methanol purity and minimize heat duty. The Pareto optimal fronts obtained using the data-based SAMOO approach are found to be very close to the optimization results obtained using the actual physics-based Aspen Plus model. The coupling of the genetic algorithm and ANN modeling in SAMOO approach reduces the computing time of optimization by ∽ 50% with nearly the same results as that of the physics-based model.

Optimal Maneuvers Around Binary Asteroids Using Particle Swarm Optimization and Machine Learning

Designing optimal transfer trajectories and reference orbit tracking in binary asteroid systems is both challenging and computationally expensive. This paper proposes a method of bypassing the high computational overhead by leveraging a collection of known techniques. Indeed, the proposed framework is based on the combination of artificial intelligence techniques, such as the particle swarm optimization and neural networks, along with the inverse dynamics and the B-splines approximation of the trajectory. The real irregular shapes of the asteroids are considered in the free dynamics of the system, which are obtained via the mutual polyhedral model. The gravitational accelerations of the single asteroids acting on the spacecraft are approximated by using two single-layer neural networks trained via an extreme learning machine. By using a combination of these techniques, the computational time of the whole optimization is decreased from hours to minutes. The proposed approach is applied to the optimal trajectory design around the binary asteroid system, 1999 KW4, showing the feasibility of the proposed optimization approach, reducing the computational effort and time, and increasing the reliability of the obtained results. It is shown through a Monte Carlo analysis that our optimization strategy yields more accurate solutions than other optimization algorithms, such as the interior point and sequential quadratic programming methods, when a random initial guess is provided. Finally, the proposed optimization approach can be used in combination with other techniques to provide a feasible and reliable initial guess for a better solution refinement.t

Study of Optimization of Tourists' Travel Paths by Several Algorithms

The purpose of this paper is to optimize the tourism path to make the distance shorter. The article first constructed a model for tourism route planning and then used particle swarm optimization (PSO), genetic algorithm (GA), and ant colony algorithms to solve the model separately. Finally, a simulation experiment was conducted on tourist attractions in the suburbs of Taiyuan City to compare the path optimization performance of the three algorithms. The three path optimization algorithms all converged during the process of finding the optimal path. Among them, the ant colony algorithm exhibited the fastest and most stable convergence, resulting in the smallest model fitness value. The travel route obtained through the ant colony algorithm had the shortest distance, and this algorithm required minimal time for optimization. The novelty of this article lies in the enumeration and description of various algorithms used for optimizing travel paths, as well as the comparison of three different travel route optimization algorithms through simulation experiments. Doi: 10.28991/HIJ-2023-04-02-012 Full Text: PDF

Wind Energy Conversion System Using Advanced Speed Control and Model-Based Loss Minimization

This paper presents a new optimization strategy for a stand-alone wind energy conversion system (WECS). The WECS is comprised of a variable-speed wind turbine (WT) with a vector-controlled self-excited induction generator (SEIG), a three-phase full-bridge converter, and a DC-bus containing the excitation capacitor, batteries, and a load. The control strategy incorporates an advanced model-based SEIG loss minimization and fuzzy-logic WT optimization. The latter utilizes a hedge-algebra speed controller to ensure fast response with practically no overshoot in the whole WT operating range, which cannot be achieved with the conventional proportional-integral (PI) controller. Consequently, the WT optimization time step is shortened and its convergence accelerated. The proposed SEIG loss minimization is based on the corresponding mathematical model that accounts for magnetic saturation and variable stray load and iron losses. Simultaneous optimization of the WT and SEIG is enabled, which results in greater total energy output compared to the successive WT-SEIG optimization. The proposed control strategy is run in real-time using the DS1103 board (dSpace) with a 1.5 kW SEIG driven by an emulated WT. It is experimentally evaluated over a wide WT operating range and compared with several competing strategies involving successive optimization, PI speed control and/or less elaborate SEIG models.

Using artificial neural networks to accelerate flowsheet optimization for downstream process development.

An optimal purification process for biopharmaceutical products is important to meet strict safety regulations, and for economic benefits. To find the global optimum, it is desirable to screen the overall design space. Advanced model-based approaches enable to screen a broad range of the design-space, in contrast to traditional statistical or heuristic-based approaches. Though, chromatographic mechanistic modeling (MM), one of the advanced model-based approaches, can be speed-limiting for flowsheet optimization, which evaluates every purification possibility (e.g., type and order of purification techniques, and their operating conditions). Therefore, we propose to use artificial neural networks (ANNs) during global optimization to select the most optimal flowsheets. So, the number of flowsheets for final local optimization is reduced and consequently the overall optimization time. Employing ANNs during global optimization proved to reduce the number of flowsheets from 15 to only 3. From these three, one flowsheet was optimized locally and similar final results were found when using the global outcome of either the ANN or MM as starting condition. Moreover, the overall flowsheet optimization time was reduced by 50% when using ANNs during global optimization. This approach accelerates the early purification process design; moreover, it is generic, flexible, and regardless of sample material's type.

Deep optimization design of 2D repetitive control systems with saturating actuators: An adaptive multi-population PSO algorithm

This paper presents an optimization design method for a two-dimensional (2D) modified repetitive control system (MRCS) with an anti-windup compensator. Using lifting technology, a 2D hybrid model of the MRCS considering actuator saturation is established to describe the control and learning of the repetitive control. A linear-matrix-inequality (LMI)-based sufficient condition is derived to ensure the stability of the MRCS. Two tuning parameters, the selection of which is critical to the system design, are used in the LMI to adjust the control and learning, and hence the reference-tracking performance. A new cost function, developed through time domain analysis, directly evaluates the control performance of the system without calculating control errors, thus reducing the optimization time. Based on this cost function, an adaptive multi-population particle swarm optimization algorithm is presented to select an optimal pair of tuning parameters in which multiple populations cooperatively search in non-intersecting search intervals. An anti-windup term is added between the low-pass filter and the time delay in the modified repetitive controller to mitigate the undesirable effect of actuator saturation on system performance and stability. Simulations and experiments on the speed control of a rotation control system demonstrate the validity of the approach.

A simplified modeling approach for optimization of urban river systems

Effective strategies for operating gates and pumping stations can help urban river systems to better control flooding. However, the computing cost of optimizing urban river systems prevents their application to large-scale problems. We propose a new simplified approach to urban river system optimization that allows a portion of the river network to be simulated, while the remainder is represented by a simplified model consisting of multiple surrogate models. The surrogate model reflects changes at the boundaries of the region of interest at prediction points within the system. To verify the accuracy of the simplified model, a numerical model based on storm water management model (SWMM) is established. The accuracy of the numerical model, the surrogate model and the simplified model were tested by multiple precipitation events. The average simulation time has been reduced by 574 times, demonstrating that the simplified model has effectively replaced the urban river network. This method is used to optimize multi-objective flood control in coastal cities. By adjusting the opening and closing state of 28 gates and pumping stations, the water level of 4 observation sections is controlled. The results show that 60% of the flood volume and 84% of the flood duration of the observed section are reduced, indicating that this method can better control the flooding process of the section. Compared with the numerical model, the calculation time is reduced by 184 times, indicating that the proposed method can solve the problem of long calculation time for urban river system optimization. This approach will reduce the risk of flooding in cities.

4E analysis and machine learning optimization of a geothermal-based system integrated with ejector refrigeration cycle for efficient hydrogen production and liquefaction

This study presents a new geothermal-based system for power generation and liquid hydrogen production. Depending on the network requirements, part of the generated power is supplied to the network and the remaining part enters the PEM electrolyzer to produce hydrogen. The produced hydrogen is then pre-cooled by the ejector cooling cycle before entering the liquefaction cycle. The ejector refrigeration cycle not only uses geothermal waste heat but also facilitates the liquefaction process through precooling hydrogen; therefore, it significantly improves the overall performance of the integrated system. In addition, the modified liquefaction cycle is suggested, which is superior to common ones. After simulating, some key parameters are identified and a comprehensive parametric research is carried out to investigate the thermodynamic, economic and environmental performance in different conditions. The optimization procedure is implemented through a genetic algorithm approach and the objective functions are defined. The results reveal that the double flash power cycle has the greatest effect on exergy destruction rate and total cost rate. In the optimal case, the total cost rate, the liquefied hydrogen rate, the total exergy efficiency, net power generation, and CO2 reduction rate are determined as 181.71 $/h, 59.92 kg/h, 25.27%, 4.03 MW, and 1421.2 kg/h, respectively. Also, coupling an artificial neural network with genetic algorithm significantly lessens optimization time.

Application of the MPPT Control Algorithm Based on Hybrid Quantum Particle Swarm Optimization in a Photovoltaic Power Generation System

The Maximum Power Point Tracking method is a mainstream method for improving the operational efficiency of photovoltaic power generation, but it is difficult to adapt to the rapidly changing environment and lacks good steady-state and dynamic performance. To achieve fast and accurate tracking of the Maximum Power Point Tracking, the optimization of the contraction expansion coefficient of the Quantum Particle Swarm Optimization algorithm is studied, and then the Levy flight strategy is introduced to optimize the algorithm’s global convergence ability, thereby constructing the Hybrid Quantum Particle Swarm Optimization algorithm. Finally, the Hybrid Quantum Particle Swarm Optimization combined with the Maximum Power Point Tracking algorithm is obtained. The research results showed that the Hybrid Quantum Particle Swarm Optimization combined with the Maximum Power Point Tracking algorithm can always converge to the theoretical minimum value with a probability of more than 94% in the Roserock function and Rastigin function tests. The tracking error of the Hybrid Quantum Particle Swarm Optimization combined with the Maximum Power Point Tracking algorithm was less than 1% under lighting conditions. The convergence time of the Hybrid Quantum Particle Swarm Optimization combined with the Maximum Power Point Tracking algorithm in arbitrary shadow occlusion environments can reach a stable state within 0.1 s. In summary, the Hybrid Quantum Particle Swarm Optimization combined with the Maximum Power Point Tracking algorithm proposed in the study has excellent performance and very wide applicability. To a certain extent, it improves the total power generation capacity of the photovoltaic power generation system and the power generation efficiency of the photovoltaic array.

Parametric analysis and multi-objective optimization of the coupling beam pile structure foundation

To optimize the structure of a new coupling beam pile foundation, a finite element model that is verified by an indoor model test was established and compared with a pile group foundation to improve the uneven settlement phenomenon. The effects of different structural parameters on settlement and coupled beam axial forces were analysed. The corresponding mathematical model is created using the response surface method, and the NSGAII genetic algorithm is utilized to carry out multi-objective optimization while taking cost into account. The best structural parameters are then determined through the analysis of various weighting factors. The findings show that using the response surface method instead of finite element modelling can significantly shorten the optimization time while maintaining a high accuracy; that the cost weight factor should not be regarded too seriously; and that a value of 0.5 is recommended. When the optimum parameters were chosen, the coupling beam's axial force increased by 2.8%, the project cost of the structure was cut by 57.99%, and the settlement was lowered by 34.87%. The study offers a new concept for structural optimization and foundation treatment.

A Speculative Divide-and-Conquer Optimization Method for Large Analog/Mixed-Signal Circuits: A High-Speed FFE SST Transmitter Example

We propose a speculative divide-and-conquer (SDnC) method that enables optimization of a large analog/mixed-signal (AMS) circuit. Because modules of AMS circuits strongly interact with neighbor modules, they cannot be optimized individually. Therefore, design parameters of all modules must be co-optimized for the global optimization, and thus, the design space exponentially grows with the circuit size. Although metaheuristic algorithms can enhance optimization efficiency, they cannot handle very large circuits due to the exponentially increased size of design space to explore. The proposed method utilized the divide-and-conquer (DnC) strategy in the circuit optimization while taking into account modules’ interactions, allowing modules to be individually evaluated and optimized. Therefore, this DnC-based optimization method hierarchically and systematically reduces both the optimization complexity and the evaluation complexity, and thus, this method is very scalable with the circuit size. The proposed method can also be combined with other metaheuristic algorithms such as particle swarm optimization (PSO) or artificial intelligence for faster optimization. In experiment, the proposed method enabled optimization of a high-speed link transmitter that has 2400–15 000 transistors and 45–47 independent design parameters for the first time. The optimization time was improved by 43 and three orders of magnitudes compared to parameter sweep and PSO, respectively. The penalties of speed-up by DnC were only 3.4% and 9.3% estimation errors in power consumption and eye height, respectively. Because of the greatly improved speed, the proposed method also enables quantitative analysis on performance-power tradeoff of a large AMS circuit.

Online Learning of Stable Integer Variables in Unit Commitment Using Internal Information

In a set of similar unit commitment (UC) problems, a variable is called a stable integer variable (SIV) if its optimal value is 0 or 1 on a regular basis, which reflects an inherent pattern in UC solutions. The computational complexity can be significantly reduced by fixing SIVs with a minor chance of accuracy loss. To identify SIVs, this letter proposes a method to collect internal solution information from the initial branch-and-bound process of artificial mixed-integer programming problems generated from the given problem. A distinct advantage of this method is that because no prior offline training is needed, learning is instance-specific and has no dependency on the training set. With identified SIVs, machine learning-based acceleration is achieved. Based on 50 test cases from publicly available and practical data, the proposed method improves the percentage of solved problems within the clearing time window from 68% to 96% compared with commercial solvers, counting the total time of data collection, prediction, and optimization.

Multi-objective optimization of recompression S-CO2 cycle for gas turbine waste heat recovery

Recompression supercritical carbon dioxide Brayton cycle (RCSCBC) has important application prospects in the fields of gas turbine waste heat recovery. A RCSCBC model considering finite temperature difference heat transfer between heat reservoir and working fluid, irreversible compression, irreversible expansion and other irreversibility losses is established. Then the mass flow rate of the working fluid, cycle pressure ratio and the heat conductance distribution ratios of four heat exchangers are taken as optimization variables, the multi-objective coordinated optimization with net power output, exergy efficiency, thermal efficiency and ecological function as the goals is carried out, and an artificial neural network model is constructed to solve the problem of long optimization time. The results show that the decision-making result by Shannon Entropy method for the four-objective optimization has the lowest deviation index, and is the optimal. Compared with the initial design point, the thermal efficiency of the cycle for the optimization results by the Shannon Entropy method can reach 40.69%, which is increased by 29.8%. The net power output can reach 10.972 MW, which is increased by 9.5%. The ecological function can reach 7.265 MW, which is increased by 114.75%. The exergy efficiency can reach 74.72%, which is increased by 24.22%.

Vibration Suppression Trajectory Planning of Underwater Flexible Manipulators Based on Incremental Kriging-Assisted Optimization Algorithm

It is of great significance to expand the functions of submarines by carrying underwater manipulators with a large working space. To suppress the flexible vibration of underwater manipulators, an improved sparrow search algorithm (ISSA) combining an elite strategy and a sine algorithm is proposed for the trajectory planning of underwater flexible manipulators. In this method, the vibration evaluation function is established based on the precise dynamic model of the underwater flexible manipulator and considering complex motion and vibration constraints. Simulation results show that the ISSA algorithm requires only 1/3.68 of the time of PSO. Compared to PSO, SSA and the opposition-based learning sparrow search algorithm (OBLSSA), the optimization performance is improved by 17.3%, 13.1% and 9.7%, respectively. However, because the complex dynamics model of the underwater flexible manipulator leads to large computational effort and a long optimization time, ISSA is difficult to apply directly in practice. To obtain a large number of optimization results in a shorter time, an incremental Kriging-assisted ISSA (IKA-ISSA) is proposed in this paper. Simulation results show that IKA-ISSA has good nonlinear approximation ability and the optimization time is only 3% of that of the ISSA.

Algebraic surrogate‐based process optimization using Bayesian symbolic learning

AbstractHere, we propose a strategy for the global optimization of process flowsheets, a fundamental problem in process systems engineering, based on algebraic surrogates that are built from rigorous simulations via Bayesian symbolic regression. The applied method provides a closed‐form expression that can be optimized to global optimality using state‐of‐the‐art solvers, where BARON or ANTIGONE were the solvers of choice. When predicting unseen test data, the algebraic models show a similar accuracy level compared to traditional surrogates based on Gaussian processes. However, they can be more easily optimized to global optimality due to their analytical closed‐form structure, which allows the user to apply well‐established global deterministic solvers. We show the capabilities of our approach in several case studies, ranging from process units to full flowsheets. The performance of our approach is assessed by comparing the CPU time for model building, the prediction accuracy of the identified model, and the CPU time for the subsequent optimization with a proven benchmark.

Improving landscape ecological network connectivity in urbanizing areas from dual dimensions of structure and function

By building connectivity between ecological spaces, the landscape ecology network (LEN) can effectively prevent further habitat fragmentation and loss in urbanizing areas. The existing LEN construction method fails to consider the multiple optimization goals of improving network structural and functional connectivity. This paper aimed to provide a novel method to improve connectivity for existing LENs in urbanizing areas. We took an urbanizing area in China as an example and extract existing LEN based on ecological priorities. On this basis, from the dimension of structural connectivity optimization, we used the network link additional technique based on “integral index of probable connection” to identify 50 optimal locations of additional corridors that could significantly improve the network structural connectivity. From the dimension of functional connectivity enhancement, we applied the greedy algorithm based on hierarchical rules with different priorities to search for corridor paths with maximum ecological function and low ecological and construction resistance. The optimized LEN contained 43 ecological sources and 74 ecological corridors. The dual dimension method proposed in this paper can improve the functional and structural connectivity of existing LEN, and greatly save the optimization time and efficiency of the network, to provide targeted ecological space protection schemes for urbanizing areas.

An optimization method of adjoint assimilation based on the cumulative gradient: An application for correction of the bottom friction coefficient in the tidal wave model

In this paper, we explored the optimization process in adjoint method. An optimization method based on the cumulative gradient is proposed to minimize the cost function. On the basis of the gradient descent method, the cumulative gradient is added and the information of first-order gradient is further utilized to update the control variable. Instead of focusing on the gradient in the current iteration only, the method combines the gradient information in the previous iterations. By weighting and cumulating the previous gradients, it can provide constructive guidance for the update direction and update stride of the control variable. To test the optimization performance in the practical problem, the method is applied to the correction of the bottom friction coefficient in the tidal wave model with adjoint method and the gradient descent method is set as the control. The results show that the method based on the cumulative gradient can optimize the cost function to a lower level than the gradient descent method under the same number of iterations. With our method, the model had more accurate simulation results. The simulation effect in areas without assimilation data is also tested, and the results prove that the method has great stability and can work for the entire simulation area. Moreover, the optimization efficiency of the method is much higher than that of the gradient descent method. It can reach the same effect while the optimization time is greatly shortened. In further experiments, we compare some other optimization methods with the method in the paper, and the optimization results of the cost function show that our method is still competitive. Our method is effective and simply implemented. It requires few computation and storage resources, which is of great significance for large-scale optimization problems. The method has shown satisfactory performance for optimization of the cost function in adjoint method. It can enhance the assimilation effect and improve optimization efficiency to a certain extent. The method has certain engineering significance for solving the optimization problems in the actual assimilation system. As a general method, it may provide a reference for the selection and innovation of optimization methods in data assimilation process.