In-memory computing (IMC) quantized neural network (QNN) accelerators are extensively used to improve energy-efficiency. However, ternary (TNN) with bitwise operations in nonvolatile memory lacked. In addition, specific generally for a single algorithm limited applications. this report, multiply-and-accumulate (MAC) circuit based on spin-torque transfer magnetic random access (STT-MRAM) is prop...

This paper addresses the synchronization of chaotic gyros with unknown parameters and external disturbance via an adaptive dynamic neural network control ADNNC system. The proposed ADNNC system is composed of a neural controller and a smooth compensator. The neural controller uses a dynamic RBF DRBF network to online approximate an ideal controller. The DRBF network can create new hidden neuron...

Abstract: This project report emphasis enhancement of power quality by using UPQC with fuzzy logic controller (FLC), Artificial Neural Network (ANN) controller and with the conventional proportionalintegral (PI) controller. The unified power quality conditioner (UPQC) is being used as a universal active power conditioning device to mitigate both current and voltage harmonics at a distribution s...

This paper presents the design of neural networks compared with the conventional technique, a hysteresis controller for active power filter for three-phase four-wire electric system. A particular three-layer neural network structure is studied in some detail. Simulation and experimental results of the active power filter with both controllers are also presented to verify the feasibility of such...

This paper addresses the issue of trajectory tracking control based on a neural network controller for industrial manipulators. A new control scheme is proposed based on neural network technology and linear feedback approach for tracking a planned trajectory. In detail, the control system is designed with two parallel subsystems designed separately. One is a linear controller, and another one i...

We introduce a method to train Quantized Neural Networks (QNNs) — neural networks with extremely low precision (e.g., 1-bit) weights and activations, at run-time. At traintime the quantized weights and activations are used for computing the parameter gradients. During the forward pass, QNNs drastically reduce memory size and accesses, and replace most arithmetic operations with bit-wise operati...

In this paper Hybrid Direct Neural Controller (HDNC) with Linear Feedback Compensator (LFBC) has been developed. Proper initialization of neural network weights is a critical problem. This paper presents two different neural network configurations with unity and random weight initialization while using it as a direct controller and linear feedback compensator. The performances of these controll...

Abstract The traditional PID controller is widely used to improve the stability of system because its simplicity and convenience. However, disadvantage control method that parameters are difficult adjust in real-time, it can not adapt needs variable targets different states, affecting controller’s practical application effect. Therefore, using idea hierarchical learning self-adjusting control, ...