Optimal Dynamic Strategy Research Articles

Optimal dynamic control of predator\u2013prey models

This paper combines work to use a decision support tool for sustainable economic development, while acknowledging interdependent dynamics of population density, and interferences from outside. We get new insights derived from experimental approaches: analytical models (optimal dynamic control of predator–prey models) provide optimal dynamic strategies and interventions, depending on different objective functions. Our economic experiments are able to test the applicability of these strategies, and in how far decision-makers can learn to improve decision-making by repeated applications. We aim to analyse a sustainable environment with diametrical goals to harvest as much as possible while allowing optimal population growth. We find interesting insights from those who manage the dynamic system. With the methodology of experimental economics, the experiment at hand is developed to analyse the capability of individual persons to handle a complex system, and to find an economic, stable equilibrium in a neutral setting. We have developed a most interesting simulation model, where it will turn out that prices play a less important role than availability of the goods. This aspect could become a new important aspect in economics in general and in sustainable environments especially.

Bidding strategy for trading wind energy and purchasing reserve of wind power producer – A DRL based approach

Wind power producers (WPP) are punished when take part in the short-term electricity market due to the inaccuracy of wind power prediction. The profit loss can be partially offset by strategic reserve purchasing. Due to the uncertainty of real-time wind power production, the reserve capacity price and the highly dynamic price regulation, it is difficult to determine the quantity of reserve to purchase, which might have a great impact on the profit of WPP. This paper investigates the possible influence on the revenue of WPP if they take part in both the energy and reserve market. This problem is first formulated as a Markov decision process (MDP). After that, the asynchronous advantage actor-critic (A3C) method is used to take care of this problem. Several agents are applied to explore the action space simultaneously. Neural networks are applied to fit the policy function and value function, which are trained against each other to learn the optimal dynamic bidding strategy from historical data by constantly trial and error. Simulation results of a wind farm in Denmark validate that the profit loss of the WPP can be significantly reduced when the A3C algorithm is used for bidding strategy formulation when WPP participate in both the energy and reserve market.

On Infinite Horizon Optimal Control Problems with a Feed Forward Neural Network Scheme

In this paper, a class of infinite-horizon nonlinear optimal control problems is considered. The main idea is to convert the infinite horizon problem to an equivalent finite-horizon optimal control problem. According to the Pontryagin minimum principle for optimal control problems and by constructing an error function, we define an unconstrained minimization problem. In the optimization problem, we use trial solutions for the state, costate and control functions where these trial solutions are constructed by using two-layer perceptron. We then minimize the error function where weights and biases associated with all neurons are unknown. Substituting the optimal values of the weights and biases into the trial solutions, we obtain the optimal solution of the original problem. We also use a dynamic optimization scheme to learning process and discuss the stability and convergence properties of it. Some examples are given to show the efficiency of the method.

Joint Quality Investment and Inventory Decisions for Perishable Items with Reference Quality Effect under the O2O Environment

Reference quality, as a benchmark against which the purchase quality of a product is judged, has an important impact on consumers’ purchase decision, especially perishable items. Thus, how to determine an appropriate quality investment and ordering strategies for perishable items to maximize retailers’ profit is an essential task. A joint dynamic quality investment and inventory replenishment problem for perishable items with consideration of reference quality effect is investigated, where the retailer sells its products using a combination of the offline and online stores, and the sales function is sensitive to products’ quality and consumers’ reference quality. The optimal control model is established to maximize the retailer’s total profit by making a joint quality and inventory strategy. The continuous time dynamic optimal quality investment and inventory replenishment strategies with reference quality effect are obtained by Pontryagin’s maximum principle. Numerical experiments are accounted for the key system parameters on the optimal strategies.

A real-time decentralized algorithm for task scheduling in multi-agent system with continuous damage

In this paper, a common model of task scheduling problems in agent rescue scenario is proposed, in which tasks with continuous dynamic damage are introduced to capture the emerging applications of using rescue robots and other resources to enhance human disaster rescue capability. Beyond this, we mainly focus on finding the optimal task scheduling strategy. We design a heuristic algorithm based on greedy strategy to obtain the optimal dynamic scheduling strategy of agents. Compared with solving global integer programming directly, the computational time is greatly reduced. The proof of the greedy strategy’s validity is also demonstrated under some specific damage functions. By comparing with the two strategies commonly used in real life, it is proved that our strategy is optimal. For practical application, we design an automatic negotiation framework, which realizes the real-time decentralized automated negotiation of agents. Then, using Game Description Language (GDL) as a tool, an automated negotiation algorithm is implemented, which enables agents to adjust the plan dispersedly. Experiments show that the algorithm is more efficient than the centralized algorithm in the case of limited communication.

RGV Dynamic Scheduling Strategy Based on Network Cellular Automaton and Marko Model

In an intelligent processing system, if a Rail Guide Vehicle (RGV), also known as automated guide vehicle, is provided with the optimal dynamic scheduling strategy, the work efficiency and economic profit of the workshop can be greatly improved. This paper considers the variable number as well as the types of Computer Number Controller (CNC) and regards the real problem as the Job Shop Scheduling Problem (JSSP). To solve JSSP, this paper use Network Cellular Automaton (NCA) to abstract the flexible workshop system and apply the rule of one-step trial-and-error for the core link between the network cellular automaton and the Markov Transition Probability Matrix. Further this system is simulated using Markov chain in probabilities to expand the solution space with the optimal solution among according to the Monte Carlo experimental principle. Finally, the result can jump out of the local optimal solution and approximate the global optimal dynamic scheduling strategy by trial-and-error rule and Transition Probability Matrix. To verify the NCA-Markov model, this paper use an intelligent flexible workshop which produces two-process materials. By simulating this system with the model, the efficiency of the intelligent processing system can be greatly improved.

A differential game approach to dynamic optimal control strategies for watershed pollution across regional boundaries under eco-compensation criterion

In recent years, transboundary pollution reflected by watershed pollution has emerged in an endless stream, manifesting as the overflow phenomenon of one-way or crossing between river basins. Such pollution significantly seriously affects the inherent requirements for achieving regional sustainable and coordinated development in China. By introducing the eco-compensation criterion, a cross-regional boundaries pollution abatement differential game model covering upstream and downstream regions in a continuous time is established. Subsequently, the optimal control theory is adopted by maximizing the net present value of river basin welfare to explore relevant optimal feedback equilibriums on the overall watershed environmental quality under the Stackelberg and cooperative game contracts. Eventually, a case study of Xiang Jiang River Basin (XJRB) in Hunan Province of China serves to verify the practicable effectiveness of the foregoing model combined with numerical simulation and support key insights as below: (i) the watershed environmental degradation cost factor as well as the payoff factor possess significant impact on emission capacities of upstream and downstream; (ii) the optimal trajectory of pollutant stock in the downstream affected by the upstream pollutant stock and pollution transfer coefficient is higher than that in the upstream; (iii) the setting of eco-compensation criterion can effectively stimulate upstream governments to invest more in environmentally friendly technologies; and (iv) the joint decisions of the two game parties improve the total welfare level of both river basin regions. Findings present an efficient approach for the design of environmental policies addressing conflicts over water pollution across regional boundaries.

Research on the novel discretized ratio control strategies and hardware-in-the-loop test for continuous variable transmission system

Based on traditional continuous control strategy for Continuously Variable Transmission (CVT) ratio, according to the principles of shift control strategy for stepped automatic transmission, the influences of throttle opening and external resistance or vehicle speed on CVT ratio control are analyzed on bumpy road. Under the same variation of external resistance condition, the differences between optimal economic control strategy and optimal dynamic control strategy are discussed. Then, the traditional continuous optimal dynamic and economic control lines are divided into multi-step upshift points. Meanwhile, the corresponding downshift points are set to avoid the interference near shift points. After that, the novel discretized ratio control methods for CVT system are proposed. By respectively discretizing throttle opening and vehicle speed, the discretized ratio control strategy for throttle opening, and the integrated discretized ratio control strategy for throttle opening and vehicle speed are further proposed and simulated. Furthermore, the hardware-in-the-loop (HIL) test system is built to further verify the feasibility and accuracy of discretized ratio control strategies. Both simulation and HIL test results show that the sensitivity of throttle opening and vehicle speed to ratio control is reduced dramatically, the fluctuation of ratio is decreased considerably, the transmission efficiency is increased significantly, and the jerk is declined moderately.

HEDGING EFFECTIVENESS OF CROSS-LISTED NIFTY INDEX FUTURES

This paper investigates the hedging effectiveness of cross-listed Nifty Index futures and compares the performance of constant and dynamic optimal hedging strategies. We use daily data of Nifty index traded on the National Stock Exchange (NSE), India and cross-listed Nifty futures traded on the Singapore Stock Exchange (SGX) for a period of six years from July 15, 2010 to July 15, 2016. Various competing forms of Multivariate Generalised Autoregressive Conditional Heteroscedasticity (MGARCH) models, such as Constant Conditional Correlation (CCC) and Dynamic Conditional Correlation (DCC), have been employed to capture the time-varying volatility. The results clearly depict that dynamic hedge ratios outperform traditional constant hedge ratios with the DCC–GARCH model being the most efficient with maximum variance reduction from the unhedged portfolio.

Life Cycle Goal Achievement or Portfolio Volatility Reduction?

This paper is concerned with the use of currently available technology to provide individuals, financial advisors and pension fund financial planners with detailed prospective financial plans tailored to an individual's financial goals and obligations. By taking account of all prospective cash flows of an individual, including servicing current liabilities, and simultaneously optimizing prospective spending, saving, asset allocation, tax, insurance, etc. using dynamic stochastic optimization, the paper addresses the question of the title by comparing the results of such a goal-based fully dynamic strategy with representative current best practices of the financial advisory industry. These include piecemeal fixed allocation portfolios for specific goals, target-date retirement funds and fixed real income post-retirement financial products, all using Markowitz mean variance optimization based on the very general goal of minimizing portfolio volatility for a specific portfolio expected return over a finite horizon. Making use of the same data and market calibrated Monte Carlo stochastic simulation for all the alternative portfolio strategies, we find that flexibility turns out to be of key importance to individuals for both portfolio and spending decisions. The performance of the adaptive dynamic goal-based portfolio strategy is found to be far superior to all the industry’s Markowitz-based approaches. Superiority is measured here by the certainty equivalent increase in expected utility of individual lifetime consumption (gamma) and the extra initial capital required by an individual to put the dominated strategy on the same expected utility footing as the optimal dynamic strategy (initial capital gap). These empirical results should put paid to the commonly held view amongst finance professionals that the extra complexity of holistic dynamic stochastic models is not worth the marginal extra value obtained from their employment. We hope that such approaches implemented in currently available technologies will rapidly find acceptance by individuals, financial advisors and pension funds to the genuine benefit of individual investors.

Fuel consumption reduction by introducing best-mode controller for hybrid electric vehicles

In this paper, a power management system for hybrid electric vehicles is developed and shown to improve the vehicle fuel consumption in various working conditions. A best-mode concept is defined based on the results of the dynamic programming global optimization strategy. It is shown that the use of exclusive control relations for each working mode improves fuel saving. The working state of the engine and one electric motor is used to determine the best-mode of powertrain operation. The best-mode classification also considers the battery state of charge. This enables the controller to specify near optimal working points in a wide range of state of charge. The control rule for each different work-mode is developed based on the dynamic programming results and applying the particle swarm optimization algorithm. The results show that the best-mode controller is capable of achieving fuel consumptions around 97% of that of the offline dynamic programming.

Stabilization of a class of nonlinear control systems via a neural network scheme with convergence analysis

In this paper, the stability of a class of nonlinear control systems is analyzed. We first construct an optimal control problem by inserting a suitable performance index; this problem is referred to as an infinite horizon problem. By a suitable change of variable, the infinite horizon problem is reduced to a finite horizon problem. We then present a feedback controller designing approach for the obtained finite horizon control problem. This approach involves a neural network scheme for solving the nonlinear Hamilton Jacobi Bellman equation. By using the neural network method, an analytic approximate solution for value function and a suboptimal feedback control law are achieved. A learning algorithm based on a dynamic optimization scheme with stability and convergence properties is also provided. Some illustrative examples are employed to demonstrate the accuracy and efficiency of the proposed plan. As a real-life application in engineering, the stabilization of a micro-electromechanical system is studied.

A comprehensive dynamic efficiency-enhanced energy management strategy for plug-in hybrid electric vehicles

This paper presents a comprehensive dynamic efficiency model for the overall powertrain efficiency that considers the internal combustion engine and electric motor generators in depth as well as a planetary coupling transmission system of PHEVs example. In particular, the dynamic forces of planetary coupling transmission system with an effect on the dynamic transmission efficiency are taken into consideration. Then a dynamic efficiency optimization strategy is proposed to improve the fuel economy, and an improved dynamic efficiency optimization strategy, consisting in adding to the dynamic efficiency optimization strategy a penalty coefficient, is applied to reach the trade-off between energy consumption and fatigue life of planetary coupling transmission system. Eventually, the dynamic efficiency optimization strategy and improved dynamic efficiency optimization strategy are validated by hardware-in-the-loop experiments. The main contribution of this study is to explore a novel way to optimize the comprehensive dynamic efficiency of whole powertrain, and also improve ride comfort by avoiding planetary coupling transmission system working in the inefficient resonance region.

A model-based dynamic optimization strategy for control of indoor air pollutants

The American Society for Heating, Refrigeration, and Air Conditioning Engineers (ASHARE) prescribes fixed ventilation rates for the operation of heating, ventilation, and air conditioning (HVAC) systems, which is sub-optimal in terms of controlling indoor air pollutants and energy consumption. This paper relies on a physics-based model of a residential house to predict the concentrations of indoor pollutants such as ozone, formaldehyde (HCHO), and particulate matter (PM) as function of time, outdoor conditions, and indoor emissions. The model captures real-world scenarios such as pollutants entering the house from ambient air through an air handling unit (AHU) and pollutants emitted in the house. A dynamic optimization problem is then formulated and solved to calculate the optimal ventilation rate at each time instant, that minimizes the total energy consumption, while maintaining the indoor pollutant concentrations below or as close as possible to their respective acceptable thresholds. For the considered example day, operation under optimized conditions reduces the peak pollutant concentration by 31%, time of exposure to undesirable concentrations by 48%, and energy consumption of the AHU by 17.7%.

Process integration for simulated moving bed reactor for the production of glycol ether acetate

This work designs and optimizes a simulated moving bed reactor (SMBR) within an overall process that includes two downstream units. The extract and raffinate streams are processed to improve product purity, and recycle streams leaving the downstream units are returned to the SMBR to reduce solvent consumption. Three different operating schemes are evaluated–the standard, limited superstructure, and full superstructure configuration.The proposed process flowsheet is optimized in a deterministic, nonlinear dynamic optimization scheme. The productivity of the target product, propylene glycol methyl ether acetate (PMA), is maximized while the minimum conversion is maintained between 70% and 99%, and the ratio of excess (solvent) reactant to the product is from 1.2 to 8. In addition to the standard configuration of SMBR, we found novel operating schemes through a superstructure formulation which allows feeding the recycle stream from downstream separation units. This work offers a realistic understanding of SMBR for implementation in an overall process flowsheet for industrial applications.

A stochastic control approach to managed futures portfolios

We study a stochastic control approach to managed futures portfolios. Building on the (Schwartz, 1997) stochastic convenience yield model for commodity prices, we formulate a utility maximization problem for dynamically trading a single-maturity futures or multiple futures contracts over a finite horizon. By analyzing the associated Hamilton–Jacobi–Bellman (HJB) equation, we solve the investor’s utility maximization problem explicitly and derive the optimal dynamic trading strategies in closed form. We provide numerical examples and illustrate the optimal trading strategies using WTI crude oil futures data.

Research on Dynamic Optimization Operation of Optical Storage Based on LSTM Load Forecasting

Affected by weather, geography and other factors, PV output has fluctuations and intermittentness. It is not suitable to supply power to the load independently. It needs to be used together with the energy storage device. At the same time, the power load has a great influence on the energy storage configuration. Therefore, accurate load forecasting has Helps improve the stability and reliability of the power system, saves electricity costs, and facilitates dynamic optimization of optical storage. In this regard, this paper proposes a dynamic optimization strategy for optical storage based on long-term and short-term memory networks. Firstly, the long-short-term memory neural network method is used to predict the power load. Secondly, according to the power difference adjustment and the photovoltaic output result, the daily-time load is divided into four sections, and the different sections respectively correspond to specific control strategies. Finally, according to each, the interval control strategy optimizes the energy storage and discharge power in real time. The method stores excess photovoltaic energy and releases it in the form of electric energy when the user's power grid needs it. This can effectively suppress the power fluctuation of photovoltaic power generation, improve the utilization rate of photovoltaics, and perform peak clipping for the user grid Valley regulation.

Dynamic defense strategy against advanced persistent threat under heterogeneous networks

Advanced persistent threats (APTs) pose a grave threat in cyberspace because of their long latency and concealment. In this paper, we propose a hybrid strategy game-based dynamic defense model to optimally allocate constrained secure resources for the target network. In addition, values of profits of players in this game are computed by a novel data-fusion method called NetF. Based on network protocols and log documents, the NetF deciphers data packets collected from different networks to natural language to make them comparable. Using this algorithm, data observed from the Internet and wireless sensor networks (WSNs) can be fused to calculate the comprehensive payoff of every node precisely. The Nash equilibrium can be computed using the value to detect the possibility of a node being a malicious node. Using this method, the dynamic optimal defense strategy can be allocated to every node at different times, which enhances the security of the target network obviously. In experiments, we illustrate the obtained results via case studies of a cluster of heterogeneous networks. The results guide planning of optimal defense strategies for different kinds of nodes at different times.

Dynamic portfolio allocation with time-varying jump risk

Abstract This paper solves the dynamic portfolio allocation problem with account of time-varying jump risk. We find that both the initial jump intensity as a state variable and the jump dynamics including the average jump intensity and jump persistence are important for the investor’s optimal portfolio decision. The risk-averse investor can benefit from the optimal dynamic strategy instead of the myopic strategy. The out-of-sample results show that compared with the no-jump model, constant-jump model or the equal weighted portfolio, the dynamic portfolio with account of time-varying jump risk can produce better performance, and is more preferred by the risk-averse investor.

A Special Points-Based Hybrid Prediction Strategy for Dynamic Multi-Objective Optimization

Dynamic multi-objective optimization problem (DMOP) is such a type of optimization problems that multiple contradictory objectives change over time. This paper designs a special point-based hybrid prediction strategy (SHPS) integrated into the decomposition-based multi-objective optimization algorithm with differential evolution (MOEA/D-DE) to handle DMOPs, which is denoted as MOEA/D-DE-SHPS. In the SHPS, when historical information is insufficient to establish prediction model of population prediction strategy (PPS), the prediction (PRE) and variation (VAR) method are adapted to generate the initial population of the new environment. Meanwhile, the PPS predicts the whole population of new environment according to the history information collected from past environments; therefore, once collected historical information is inaccurate, the predicted population may be located in the wrong search region. To overcome the shortcoming, we propose the special point-based strategy in which the initial population of the new environment consists of two parts of individuals: the predicted special points and the predicted population by PPS (except the special points). The empirical results show that MOEA/D-DE-SHPS is promising for handling DMOPs.