Application of regression in algorithm of nonlinear stochastic adaptation of unstable multidimensional objects

The results of a study of applicability of kernel estimation in the synergetic control systems for the objects unstable in an open-loop state (without a stabilizing control) have been presented. The effectiveness of kernel estimates has been shown for four nonlinear objects with unstable limiting states. The estimate the effectiveness of embedding the kernel predictive estimate of the state variables of a nonlinear object, subjected to disturbances of an unknown nature, into the system of synergetic control is demonstrated.

Study of the solar power plant power generation forecasting model.

The question of determining the possible capacity of a photovoltaic power plant is very acute due to the growing demand for renewable energy, coupled with the fact that during the day we have limited time to generate energy from such a source. Thus, based on the obtained analytical data, which allows to predict weather conditions, it is possible to regulate the amount of energy supplied to the network in a certain way due to more maneuverable power plants. In previous years, electrical engineering scientists and researchers from different countries have developed and implemented methods for determining weather conditions, such as clouds, air temperature, atmospheric dust and others, as well as their impact on the energy output of a solar power plant. A photovoltaic panel is a complex nonlinear object with many variables. In addition to the structural features of the module, the output is most affected by solar radiation and panel temperature. When researching the prediction of the amount of energy produced, it is important to find sufficiently reliable and consistent data. At the forefront of these issues are US universities and research centers. For example, the University of Nevada in Las Vegas, in 2006 put into operation a set of measurements of weather conditions: the level of sunlight, ambient temperature, wind speed, humidity and others. When calculating the power generated by the panels, it is assumed that the system operates at the point of maximum power. The scheme works as follows: we set the values of temperature (Temperature) and irradiation (Irradiance); we apply voltage to the output terminals of the array by changing its value from 0 to Voc. We take current readings at each point, we find the power for each point, we find the maximum among the obtained array of points. Repeat over the entire range of input values. Thus, we obtain a graph of the output power of Figs. 4 pre-considering the losses in the inverter.

About Using Two-Level Probing Signal in Nonlinear Radio Location

A probing signal represented by two successive radio impulses having the same amplitude, but different energies is suggested for application in pulsed nonlinear radars, while for the receiver it is suggested to perform either correlation processing or optimal filtering of each of the reflected signal components at the carrier frequency. Due to the fact that the response of the optimal filter is proportional to the energy of the pulsed signal, the response levels of the two impulses reflected from an object lacking nonlinear properties will be equal. Should an object have nonlinear properties the response levels at the optimal processing device output at certain moments of time will be different thus indicating that a nonlinear object has been detected. Since the energies of the probing signal components are equal and optimal filtration is performed when receiving the reflected signal, this ensures that the noise interference equally affects the error while comparing levels of the received signal components. Depending on the error magnitude it is necessary to select upper and lower limits of the amplitude uncertainty within which response levels can be considered different. Decision about the presence of the nonlinear object is made if the difference in response levels goes beyond these limits. Suggested below is a block diagram of a decision-making device based on a successive correlation processing of each of the received signal components where the response level of the correlator at the moment when impulse ends is stored until the time when the decision is made i.e. when the second impulse ends.

Structural transformations of a neural network controller with a recurrent network type

The complexity of the objects of regulation, as well as the increase in the requirements for the productivity of the applied regulators, leads to the complexity of the applied neural network regulators. One of the complications is the appearance of feedback loops in the regulator. That is, the transition from direct distribution networks to re-current ones. One of the problems when using them is setting up weight coefficients using methods based on gradient calculation (for example, the error propagation method, the Levenberg-Marquardt method, etc.). It manifests itself in a suddenly "dis-appearing" or "exploding" gradient, which means that the learning process of the net-work stops. The purpose of this article is to develop proposals for solving some problems of con-figuring the weight coefficients of a recurrent neural network. As methods for achieving this goal, structural transformations of the architecture of a recurrent neural network are used to bring it to the form of a direct distribution net-work. At the same time, there is a slight increase in the complexity of its architecture. For networks of direct distribution methods based on the computation of the inverse gradient can be used without modification. In the future, it is planned to increase the performance of regulating the system with the help of a converted neuro-regulator, namely, to reduce the over-regulation of the system and, after some complications of the structure, use it to regulate a nonlinear object.

Synthesis of a neural network control regulator of a nonlinear model of an inverted pendulum on a cart

In this article, we consider a method for selecting a structure of a neural network used to regulate an "inverted pendulum on a cart" object taking into account its additional features of a mathematical description, namely, nonlinear parameters. The algorithm is illustrated by the example of control synthesis which includes two neuroregulators. One of them is responsible for bringing the cart to the specified position, and the second is responsible for holding the pendulum in a vertical position. The structure transformations will be performed for the controller responsible for bringing the cart to the specified position. The architecture of a neural network controller is based on a discrete controller synthesized using polynomial matrix decomposition. For the original controller, we define the limits of its possible control of a nonlinear system. To increase the range of control of a nonlinear object, we perform transformations of the neural network structure of the original controller. We will make some complications in the structure of the neural network of the regulator, namely, increase the number of neurons and replace some activation functions with nonlinear ones (hyperbolic tangent). Next, we suggest one of the ways to select initial values of weight coefficients. Then we train the neural network and check the performance of the resulting controller on a nonlinear object. At the next stage, we compare the obtained performance of a controller having a complicated neural network structure with the performance of a classical controller. Thus, the purpose of this study is to formalize the synthesis procedure for a neural network controller for controlling a nonlinear object using a calculated classical controller for a linearized object model. The proposed method of generating the architecture of a neural network of controllers makes it possible to increase the range of control by a nonlinear object in comparison with the controller obtained by the method of polynomial matrix decomposition for a linear object. Compared to the typical ones, the proposed neural network structure is not redundant and therefore does not require additional computing resources to configure it.