Fault monitoring and diagnosis of high-pressure heater system based on improved particle swarm optimization and probabilistic neural network

Abstract

The high-pressure heater system is an important part of the return heat system of thermal power units, which can significantly reduce the boiler fuel consumption and is of great significance to the safe and economic operation of the units. Aiming at the problem that the high-pressure heater system data has strong non-linear characteristics and the fault diagnosis accuracy is low, this paper proposes a hybrid model-based fault monitoring and diagnosis method for high-pressure heater system. Firstly, an improved particle swarm optimisation algorithm (IEDPSO) is proposed. A differential evolution operation is introduced to enhance particle diversity, and the inertia weight coefficients and learning factor parameters are improved to optimise the particle position and velocity update process. The problem that particle swarm optimisation (PSO) tends to fall into local optimum at the late stage of iterative optimisation search is solved. Numerical simulation experiments demonstrate that IDEPSO has high convergence speed and accuracy in the optimisation process of the test function. Secondly, a fault monitoring and diagnosis method based on a hybrid KPCA-IDEPSO-PNN model is proposed. Non-linear features are extracted using kernel principal component analysis (KPCA) for fault monitoring. The IDEPSO algorithm is used to iteratively find the best probabilistic neural network (PNN) to improve the fault diagnosis accuracy. Simulation experiments prove that compared with the traditional PNN model, the fault diagnosis accuracy of the KPCA-IDEPSO-PNN model is improved by 4.9% and the number of fault misclassifications is reduced by 34, effectively improving the fault diagnosis accuracy of the high-pressure heater system and ensuring the safe and stable operation of thermal power units.

Acoustic tunnel lining cavity detection using cepstral coefficients with optimized filter bank

“…Particle swarm optimization (PSO) is a search algorithm that simulates the foraging behavior of birds to achieve group collaboration [18]. In this algorithm, a particle swarm consisting of m particles is formed in an N-dimensional search space.…”

Research on neural network modeling of thermal power unit load system

“…In the model established in this paper, the initial weight values from the input layer to the hidden layer and from the hidden layer to the output layer are all within the range of [-1,1]. The classical BP algorithm with a fixed learning rate is used to train the network model [8]. The learning rate of the network training is 0.2, the training target is 0.001, and the number of training steps is 1000.…”

Machine learning-based rail corrugation recognition: a metro vehicle response and noise perspective

“…In recent years, machine learning and deep learning algorithms have also been widely used in the field of fault diagnosis. Wu et al [22] proposed an improved particle swarm optimization algorithm. The algorithm was used to iteratively find the smoothing factor of the probabilistic neural network (PNN).…”