Search results for "NEURAL NETWORK"
showing 10 items of 1385 documents
Neural Sensorless Control of Linear Induction Motors by a Full-Order Luenberger Observer Considering the End Effects
2012
This paper proposes a neural based full-order Luenberger adaptive speed observer for sensorless linear induction motor (LIM) drives, where the linear speed is estimated with the total least squares (TLS) EXIN neuron. A novel state space-vector representation of the LIM has been deduced, taking into consideration its dynamic end effects. The state equations of the LIM have been rearranged into a matrix form to be solved, in terms of the LIM linear speed, by any least squares technique. The TLS EXIN neuron has been used to compute online, in recursive form, the machine linear speed. A new gain matrix choice of the Luenberger observer, specifically taking into consideration the LIM dynamic end…
Multi-step ahead electricity price forecasting using a hybrid model based on two-layer decomposition technique and BP neural network optimized by fir…
2017
In the deregulated competitive electricity market, the price which reflects the relationship between electricity supply and demand is one of the most important elements, making it crucial for all market participants to precisely forecast the electricity price. However, electricity price series usually has complex features such as non-linearity, non-stationarity and volatility, which makes the price forecasting turn out to be very difficult. In order to improve the accuracy of electricity price forecasting, this paper first proposes a two-layer decomposition technique and then develops a hybrid model based on fast ensemble empirical mode decomposition (FEEMD), variational mode decomposition …
MRAS speed observer for high performance linear induction motor drives based on linear neural networks
2011
This paper proposes a Neural Network (NN) MRAS (Model Reference Adaptive System) speed observer suited for linear induction motor (LIM) drives. The voltage and current models of the LIM in the stationary reference frame, taking into consideration the end effects, have been obtained. Then, equations of the induced part have been discretized and rearranged so as to be represented by a linear neural network the TLS EXIN neuron, which has been used to compute the machine linear speed on-line and in recursive form. The proposed NN MRAS observer has been tested experimentally on a suitably developed test setup. Its performance has been also compared to the classic MRAS speed observer.
A neural network-based optimizing control system for a seawater-desalination solar-powered membrane distillation unit
2013
Abstract Several schemes have been proposed so far for coupling desalination processes with the use of renewable energy. One of their main drawbacks, however, is the nature of the energy source that requires a discontinuous and non-stationary operation, with some control and optimization problems. In the present work, a solar powered membrane distillation system has been used for developing an optimizing control strategy. A neural network (NN) model of the system has been trained and tested using experimental data purposely collected. Afterwards, the NN model has been used for the analysis of the process performance under various operating conditions, namely distillate production versus fee…
Nonlinear control of an activated sludge aeration process: use of fuzzy techniques for tuning PID controllers
1999
In this paper, several tuning algorithms, specifically ITAE, IMC and Cohen and Coon, were applied in order to tune an activated sludge aeration PID controller. Performance results of these controllers were compared by simulation with those obtained by using a nonlinear fuzzy PID controller. In order to design this controller, a trial and error procedure was used to determine, as a function of error at current time and at a previous time, sets of parameters (including controller gain, integral time and derivative time) which achieve satisfactory response of a PID controller actuating over the aeration process. Once these sets of data were obtained, neural networks were used to obtain fuzzy m…
Electronic noses: a review of signal processing techniques
1999
The field of electronic noses, electronic instruments capable of mimicking the human olfactory system, has developed rapidly in the past ten years. There are now at least 25 research groups working in this area and more than ten companies have developed commercial instruments, which are mainly employed in the food and cosmetics industries. Most of the work published to date, and commercial applications, relate to the use of well established static pattern analysis techniques, such as principal components analysis, discriminant function analysis, cluster analysis and multilayer perceptron based neural networks. The authors first review static techniques that have been applied to the steady-s…
PMSM Drives Sensorless Position Control with Signal Injection and Neural Filtering
2009
Vector Field Oriented Control (FOC) is one of the best control methods for high-dynamic electrical drives. To avoid the adoption of the speed/position sensor (resolver/encoder), a sensorless technique should be used. Among the various sensorless methods in literature, those based on machine saliency detection by signal injection seem to be most useful for thier giving the possibility of closing the position control loop. This paper proposes a method for enhancing both rotating and pulsating voltage carrier injection methods by a neural adaptive band filter. Results show the goodness of the proposed solution.
Robust Control Allocation for Spacecraft Attitude Stabilization under Actuator Faults and Uncertainty
2014
A robust control allocation scheme is developed for rigid spacecraft attitude stabilization in the presence of actuator partial loss fault, actuator failure, and actuator misalignment. First, a neural network fault detection scheme is proposed, Second, an adaptive attitude tracking strategy is employed which can realize fault tolerance control under the actuator partial loss and actuator failure withinλmin=0.5. The attitude tracking and faults detection are always here during the procedure. Once the fault occurred which could not guaranteed the attitude stable for 30 s, the robust control allocation strategy is generated automatically. The robust control allocation compensates the control …
Two-stage procedure based on smoothed ensembles of neural networks applied to weed detection in orange groves
2014
The potential impacts of herbicide utilization compel producers to use new methods of weed control. The problem of how to reduce the amount of herbicide and yet maintain crop production has stimulated many researchers to study selective herbicide application. The key of selective herbicide application is how to discriminate the weed areas efficiently. We introduce a procedure for weed detection in orange groves which consists of two different stages. In the first stage, the main features in an image of the grove are determined (Trees, Trunks, Soil and Sky). In the second, the weeds are detected only in those areas which were determined as Soil in the first stage. Due to the characteristics …
Sensorless control of PMSM by a linear neural network: TLS EXIN neuron
2010
Sensorless vector control applied to the Permanent Magnet Synchronous Motors (PMSMs) is a very challenging subject. It permits obtaining high dinamical performance by exploiting increased reliability and also reduced cost. Among the different methodologies proposed in literature, a model based approach has been proposed here. In particular, the space vector equations of the PMSM have been re-elaborated to permit the use of a Least Squares technique. The problem has been then faced-up to with the so-called TLS EXIN neuron, which is a linear neural network able to solve the TLS problem on-line. Simulation tests have been done on both interior mounted and surface mounted machines.