In this paper, we propose a filter structure whose output satisfies the velocity and acceleration constraints for any input signals. In the field of factory automation, step signals are sometimes converted into trapezoidal waves to satisfy the intended velocity limitations. This helps the operators avoid overload in industrial robots. In some cases, physical protection of the equipment, safety, and ride quality can be ensured by limiting the characteristics of input signals in actual plants. A signal-limitation filter is proposed for the input signal to satisfy the intended signal limit. In the previous study, a signal-limitation filter structure was provided as a simple unit-feedback control with a saturation function. The filter structure in the previous study had a delay between the input and output signals because it is difficult to design gains considering both the saturating and non-saturating cases. To solve this problem, we propose a novel filter structure that includes feedforward and feedback components. Applying this filter structure with feedforward terms including saturation enables us to fulfill the desired limitations for arbitrary input signals. We evaluated the proposed structure in a signal-limitation filter that simultaneously limits the velocity and acceleration. The simulation results demonstrate the effectiveness of our proposed filter.
We formulate and analyze a profit maximization problem for one participant (aggregator) in a multiperiod electricity market with the consideration of profit fluctuations caused by uncertain photovoltaics (PV) output. We first introduce an aggregator's prosumption model, whereby an energy demand/supply profile can be achieved by adjusting the dispatchable devices, e.g., energy storage management systems and fossil fuel generators. Then, the uncertainty issue stemming from the PV profiles is handled by a classical stochastic programming framework and a robust optimization framework. A dispatchable tuning cost function is discussed based on the two frameworks above, and the distribution of the particular cost is analyzed. The simulation results show that although both of the two frameworks are able to overcome the PV uncertainty, an average strategy gives a higher expected value of profits with a higher risk (a larger variance) while a worst-case strategy gives a lower expected value of profits with a smaller risk.
This paper presents algebraic connectivity invariance of a network-of-networks derived from a graph product. The network is modeled as an interconnected system of several kinds of similar subsystems via a certain structure, where several vertices of each subsystem are attached to the structure. The algebraic connectivity of this network can be characterized by using an appropriate network having homogeneous subsystems under a certain condition, which is investigated in this paper. In particular, a necessary and sufficient condition for adding links (edges) to subsystems which preserve the algebraic connectivity is clarified.