Recent Advances in Electrical & Electronic Engineering - Volume 17, Issue 1, 2024

The electro-hydraulic load simulator can simulate any force (torque) load required by the loaded object under test conditions, which greatly reduces the equipment development time and cost. However, the process of force loading inevitably creates a surplus force (torque). The existence of surplus force (torque) in the system will seriously interfere with the loading accuracy of the system and even damage the equipment in serious cases. In this paper, based on the research status of the suppression and compensation of the surplus force (torque) in the electro-hydraulic load simulator, the effective control strategies and methods to overcome the surplus force (torque) in the system are summarized. This paper analyzes the generation mechanism of the surplus force (torque), and summarizes the effective strategies to overcome the surplus force (torque) of the system in structural compensation and control compensation. Using the control method to compensate for the surplus force (torque) has the advantages of simple structure, convenient debugging, and strong universality. It is widely used in the electro-hydraulic load simulator to suppress the surplus force (torque). With the development of control theory, more and more control strategies and methods are applied to restrain and compensate for the surplus force (torque) in the electro-hydraulic load simulator, which not only restrains the surplus force (torque) in the system but also improves the loading accuracy and anti-interference.

Background: Complex structures such as a high proportion of power electronic equipment has brought new challenges to the safe and stable operation of new-type power system, increasing the possibility of the system being attacked, especially the more complex Advanced Persistent Threat (APT). This kind of attack has a long duration and strong concealment. Objective: Traditional detection methods target a relatively single attack mode, and the time span of APT processed is relatively short. None of them can effectively capture the long-term correlation in the attack, and the detection rate is low. These methods can’t meet the safety requirements of the new-type power system. In order to solve this problem, this paper proposes an improved transformer model called STSA-transformer algorithm, and applies it to the detection of APT in new-type power systems. Methods: In the STSA-transformer model, the network traffic collected from the power system is first converted into a sequence of feature vectors, and the location information and local feature of the sequence, is extracted by combining position encoding with convolutional embedding operations, and then global characteristics of attack sequences is captured using the multi-head selfattention mechanism of the transformer encoder, the higher-frequency features of the attention are extracted through the self-learning threshold operation, combined with the PowerNorm algorithm to standardize the samples, and finally classify the network traffic of the APT. Results: After multiple rounds of training on the model, the expected effect can be achieved and applied to the APT detection of a new-type power system. Conclusion: The experimental results show that the proposed STSA-transformer algorithm has better detection accuracy and lower detection false-alarm rate than traditional deep learning algorithms and machine learning algorithms.

Introduction: In addressing the issue of power transformer oil temperature prediction, traditional back propagation (BP) neural network algorithms have been found to suffer from local optimization and slow convergence. This study proposes an oil temperature prediction model based on an improved particle swarm optimization (PSO) neural network algorithm, which introduces an asymmetric adjustment learning factor and a mutation operator. The BP neural network, genetic algorithm (GA) optimization neural network, and the improved PSO neural network are compared by considering various factors, such as ambient temperature, load changes, and the number of cooler groups under different working conditions. Results show that the proposed algorithm improves the actual change trend of oil surface temperature and makes the transformer operation more stable to a certain extent. Background: The mathematical model for predicting transformer oil temperature is clear, but the parameters in the model are uncertain and vary with time. When subjected to different operating conditions, such as ambient temperature, load changes, and the number of cooler groups acting independently or in combination, the prediction results of the oil temperature model vary with different system parameters. Objective: This paper aims to enhance the accuracy of transformer temperature prediction. In order to optimize the oil temperature prediction model, asymmetric adjustment learning factors and mutant operators are added to meet diverse system parameter requirements. Methods: The paper utilizes an oil temperature prediction model based on an improved PSO neural network algorithm, which introduces an asymmetric adjustment learning factor and a mutation operator to address the limitations of the standard PSO algorithm. Results: This paper has employed a fusion algorithm of the genetic algorithm of the BP neural network and the PSO algorithm, and conducted simulation and experimental analysis. The simulation and experimental results demonstrate the accuracy and effectiveness of the fusion algorithm. Conclusion: This study demonstrates enhanced prediction accuracy of transformer oil surface temperature using the improved particle swarm optimization neural network algorithm. This algorithm has less prediction error under different working conditions compared to other algorithms. By increasing population diversity and combining inertia weights, the algorithm not only greatly improves its search performance but also avoids local optimization.

Background: Nowadays implementation of Demand Response (DR) programs in the distribution grid is a necessary planning criterion for distribution utility. Implemented DR programs should be automated, intelligent, well-educated, and more competent than the conventional augmentation techniques to resolve Distribution Network (DN) constraints. Peak demand causes DN to approach its maximum capacities. Peak demand also exceeds the sustainable limit of the DN resulting disruption in electric supply, failures of various assets like transformers, feeders, etc. Objective: In this paper, a Direct Load Control (DLC) scheme for Flexible Loads (FLs) is modeled & implemented under Automated Demand Response (ADR) program and tested on real 54-bus DN. Methods: This ADR program is implemented through Demand Response Aggregator (DRA) and ADR Technology Solution Enablers (ADRTSE) to curtail the peak demand on the DN ADR is a recent technology that may put off new generation (conventional- and non-conventional both). Results: It also enables the distribution utility to curtail the peak demand & its period ensuring reliability of supply without restructuring, augmentation of existing infrastructure, and development of new infrastructure. Conclusion: The result validates the effectiveness of ADR program for peak demand curtailment, asset management, distribution network losses minimization, and for sustainable development of environment.

Background: The advancements in robotic technology have completely revolutionized day-to-day life. In industrial applications, the implementation of robotics is quite advantageous as it may help in performing dangerous tasks like climbing high walls, working in a high-temperature environment, high radiation exposure conditions etc. Methods: This paper presents the design and development of a wall-climbing robot for dam wall inspection using an adaptive aerodynamic adhesion technique. The optimization of a robot design is done using a differential evolutionary algorithm. Results: In the proposed model, the principle of Bernoulli adhesion is used for designing the suction pad. The optimization of various variables is done using a differential evolutionary algorithm to improve the efficiency and effectiveness of the wall climbing robot adhesion. Conclusion: The results of the proposed system show that the approach can find an optimal holding force and can be effectively used for applications like dam wall climbing for inspection.

Background: Customers expect quality, uninterrupted power with cost-effective electricity in the latest trend. However, outages, severe storms, old infrastructure, and cost pressures can lead to ambiguity in power generation and transmission. To improve line power transmission capability, the right flexible AC transmission systems (FACTS) device may save millions of dollars. Methods: In this study, a FACTS controller named Optimal Generalized Interline Power Flow Controller (OGIPFC) was developed. Furthermore, for optimization, the Modified Marine Predator Algorithm (MMPA), which is a modification of the recently developed Marine Predator Algorithm (MPA). The optimum technique was used to evaluate a set of prioritized considered objective minimizations. A variety of factors must be maximized, such as generation cost, emissions, and power loss. Results: The performance of the proposed algorithm was analysed on benchmark test functions, and then single objective optimization problems of standard IEEE-30 bus system were solved and compared with the existing algorithms. The proposed algorithm was restricted to solving the single objective problem only, so it was further implemented with non-dominating sorting to solve the multiobjective optimization problem. The proposed multi-objective version is named as Non-dominating Sorting Modified Marine Predator Algorithm (NSMMPA), and it was validated on benchmark test functions and the IEEE-30 bus system. Conclusion: Finally, the OPF problem was solved with the incorporation of OGIPFC using the proposed methods, which resulted in better solutions and made the system more effective in operation.

Background: The new power system is more vulnerable to Advanced Persistent Threat (APT) than the traditional power system. Objective: This study aims to grasp the intent of the APT attack better; a new generation method of power system APT attack graph based on DQN is proposed. Methods: First, the network topology of the new power system was extracted by Nessus scanning as the model input. Secondly, the agent in DQN was trained for multiple rounds. Starting from the set initial state, the agent selected the action with the highest Q value to act on the system in each round, and then the system entered the next state. Then the Q network function value was updated according to the obtained system feedback value until the target state appeared. Results: After multiple rounds of training agents, multiple APT attack paths were finally obtained, thus an APT attack graph can be generated. Conclusion: The experimental results showed that the efficiency of generating an APT attack graph based on the DQN method is obviously superior to the existing methods for the large-scale industrial control system, such as the new power system.

Aim and Background: In recent periods, micro-array data analysis using soft computing and machine learning techniques gained more interest among researchers to detect prostate cancer. Due to the small sample size of micro-array data with a larger number of attributes, traditional machine learning techniques face difficulty detecting prostate cancer. Methodology: The selection of relevant genes exploits useful information about micro-array data, which enhances the accuracy of detection. In this research, the samples are acquired from the gene expression omnibus database, particularly related to the prostate cancer GEO IDs such as GSE 21034, GSE 15484 and GSE 3325/GSE 3998. In addition, ensemble feature optimization technique and Bidirectional Long Short Term Memory (Bi-LSTM) network are employed for detecting prostate cancer from the microarray data of gene expression. Results: The ensemble feature optimization technique includes 4 metaheuristic optimizers that select the top 2000 genes from each GEO IDs, which are relevant to prostate cancer. Next, the selected genes are given to the Bi-LSTM network for classifying the normal and prostate cancer subjects. Conclusion: The simulation analysis revealed that the ensemble based Bi-LSTM network obtained 99.13%, 98.97%, and 94.12% of accuracy on the GEO IDs like GSE 3325/GSE 3998, GSE 21034, and GSE 15484.