Plant diseases have a great influence on the productivity and economics of agriculture. In the traditional method where manual labour and experiences play important roles, early plant disease detection and prevention for field crop are inefficient. This can bring negative effects in stopping the spread of diseases. Applications of image processing techniques and computer vision may provide a solution to these problems. Deep learning is known as one of the most powerful techniques which is able to address more complicated tasks compared to traditional machine learning thanks to the embedded complex layers. This paper presents an image-based diseases identification method using a deep neural network with contrast enhancement and transfer learning from the PlantVillage dataset. Since there exist various sizes of leaf, rescaling and centre alignment are performed to standardize images. Besides, contrast enhancement is used to improve quality of visualization which gives a well-prepared step for further processing. By using a collected mango disease dataset, the proposed model is trained to identify 3 common diseases from the healthy one. The proposed model is then compared with other pre-trained models. An accuracy of 88.46% is achieved for the proposed model, which is higher than the results obtained from other pre-trained models.
This paper presents a method to design and control a two-wheeled self-balancing robot and it focus on hardware description, signal processing, discrete Kalman filter algorithm, system modelling and PID backstepping controller design. In the system, signals from angle sensors are filtered by a discrete Kalman filter before being fed to the PID backstepping controller. The objectives of the proposed controller are to stabilize the robot while try to keep the motion of robot to track a reference signal. The proposed PID backstepping controller has three control loops, in which the first loop uses a backstepping controller to maintain the robot at equilibrium, the second loop uses a PD controller to control the position of robot and the last uses a PI controller to control the motion direction. Simulations and experimental results show that the proposed control system has good performances in terms of quick response, good balance, stability .
27 March 2018 in IEEE Transactions on Control Systems Technology
This brief proposes a novel scheme based on fuzzy logic for automatically and appropriately adjusting the main control coefficient of a shunt active power filter (SAPF), which is used to efficiently regulate the voltage and current harmonic distortions in the grid with a large offshore wind farm connected. With the scheme, the SAPF can be operated in two different control modes based on the voltage detection method. In the first mode, the SAPF has an objective to control the total harmonic distortion (THD) of grid voltage at the point of common coupling (PCC). In the second mode, the SAPF can reduce the THD of both grid voltage and current at the PCC. In addition, a unique setting module is designed to suitably choose the output scaling factor of the proposed fuzzy logic controller in the scheme; this is to considerably enhance adaptability of the scheme. Limits for the actual current and the control coefficient of the SAPF are also utilized to ensure the safety in operation. Simulations demonstrate that the suggested scheme has good performance in automatically tuning the control coefficient of the SAPF to regulate the THD of grid voltage and current to acceptable values even under unexpected changes in configuration of the grid.
24 October 2023 in IEEE Transactions on Circuits and Systems I: Regular Papers
A design method based on event-triggered functional interval observers for estimating state vectors of a battery and motor circuit model (BMCM) is proposed in this paper. The main outputs of this paper are that, firstly, a novel battery and motor circuit model for electric vehicles is established. Then, the design of event-triggered functional interval observers (ETFIOs) is considered to estimate any linear functions of the state vector of this model. The distinctiveness of the ETFIOs compared with existing functional interval observers in the literature is that the information of the measurement vectors only updates when some event-triggered conditions hold, which may be effective in saving communication resources. Conditions for the existence of the ETFIOs and a design algorithm are provided.