Journal of Applied Dynamic Systems and Control
Volume:6 Issue: 3, Summer 2023

Core vector regression (CVR) is an extension of the core vector machine algorithm for regression estimation by generalizing the minimum bounding ball (MEB) problem. As an estimator, both the kernel function and its parameters can significantly affect the prediction precision of CVR. In this paper, a method to improve CVR performance using pre-processing based on data feature extraction and Grid algorithm is proposed to obtain appropriate parameters values of the main formulation and its kernel function. The CVR estimated mean absolute error rate here is the evaluation criterion of the proposed method that should be minimized. In addition, some benchmark datasets out of different databases were used to evaluate the proposed parameter optimization approach. The obtained numerical results show that the proposed method can reduce the CVR error with an acceptable time and space complexity. Therefore, it is able to deal with very large data and real world regression problems.

Increasing the information and services available on the web, providing tools such as recommender systems to websites and applications for users to find information and services according to their interests, seems necessary.Therefore, providing appropriate guidance and suggestions to users in different choices, according to the user's priorities, has found a special position in different fields. Recommender systems proactively recommends items that usersmay prefer. We proposed, a multi-agent recommender system that can provide suitable recommendations as a shopping assistant in the purchasing process. To analyze the proposed model, the sales dataset of an online store has been used.According to the results, in this evaluation, the accuracy of the proposed model was better than common models such as Naïve Bayesian and artificial neural networks.By combining multi-agent systems, multi-agent recommender systems were proposed that can provide suitable recommendations as a purchasing assistant in the purchasing process. The results of applying the proposed model on the data related to the purchase history of the customers of an online shopping showed that the proposed model has a good efficiency in evaluating the parameters used in comparison with the common methods in this property field.

Due to the growing popularity of renewable energy sources, grid-connected inver ters are becoming more and more common in distributed microgrid and smart-grid system. The appropriate characteristics of Quasi-Z-source inverters (QZSI), including continuous inp ut current, common DC rail, and high voltage gain, have made these inverters widely used in the renewable energy system. A battery is necessary for renewable energy systems in o rder to store energy when the demand for power is low. IN this study a configuration inv olving a battery across one of the capacitors on the DC side is proposed, through which t he DC control loop is adjusted. Also, Interconnection-Damping-Assignment Passivity-Based Control (IDA-PBC)approach has been used to adjust the battery current/voltage and the out put voltage. Compared to other controllers, the proposed controller can provide faster respo nse and better stability for QZSI when the variation of input and load. In addition, the pr oposed controller is not sensitive to the system’s initial operating point and is global asym ptotic stability. The simulations and theoretical design show the effectiveness of the propos ed controller.

In this article, solving and optimizing the problem of forward and inverse kinematics of SCARA is studied. This robot belongs to series robots and it has four degrees of freedom. First, we specify the coordinate axes for each joint and use it to extract the Danavit-Hartenberg parameters. Next, we examineforward kinematics of the robot and obtain the rotation matrices and the homogeneous transformation matrix and calculate the forward kinematics of the robot. Next, the method of solving the inverse kinematics problem of the robot is studied using different algorithms, including Cultural Algorithm, Genetic-Hybrid Algorithm, Gray Wolf Optimization, Firefly Algorithm, Ant Colony Optimization and Particle Swarm Optimization.Them, we optimize the inverse kinematics of the robot using these algorithms in two ways: fixed point and circular path. In the end, the effectiveness of the proposed approaches for solving the inverse kinematics problem of the SCARA robot is evaluated with multiple simulations.

In recent years, the emergence of various social networks has led to the growth of s ocial network users. However, activity in such networks depends on the level of trust that users hav e in each other. Therefore, trust is essential and important issue in these networks, especially whe n users interact with each other. In this article, we examine this issue and provide a method to evaluate it.It is not easy to measure the accuracy of trust for users who interact with social networks. Here,interactions are virtual. In this article, we have used the adaptive neuro-fuzz inference system to evaluate trustworthiness by considering different personality attributes of users such as reli ability,availability, interest, patience and adaptability. Using these features as input and based on the adaptive neuro-fuzzy inference system,we evaluated the trustworthiness of users in socialnetwork.Thep roposed adaptive neuro-fuzzy inference system is expandable because in this system,trustcan be defin ed as a set of one or more personality attributes.Epinions social network dataset is also used to sim ulate and validate the proposed method. In the proposed method, the absolute mean value of error is less than 0.0095 and the value of F-score is more than 0.9884. Based on theobtainedresults and com pared to the previous methods,the proposed adaptive neuro-fuzzy inference systemshows an acceptable accuracy for evaluating the trustworthiness of users.

Based on the study of the theoretical foundations of the research, it is determined that so far there is no detailed study on heating and cooling energy related to zero-energy buildings in the recent research based on energy waste in buildings. Therefore, in this article, by simulating commercial buildings and simulating the correct materials and strategies in the heating and cooling system, as well as investigating the insulation of buildings, we will study the effect of zero-energy building materials on energy wastage to model the temperature variations in building and control to achieve desire value.This article, taking into account the effects of heat transfer through building walls, the energy consumption model, and by genetic algorithm model predictive control (MPC) methodoptimizes the indoor temperature of the building. For this purpose, the genetic algorithm is used to determine the best control input in the form of building heating. The simulation of this process has been done in MATLAB software and the method of modeling heat loss and temperature change outputs shows that the proposed method has a good performance. The maximum of overshoot of the temperature is %4 and the cost function of GA algorithm is 165 based of minimum control effort and temperature error.