An interval type-2 fuzzy set and Petri net combined risk assessment model for chemical production was proposed to solve the problems of disorganized hierarchy poorly targeted measures, as well requirement complex equipment associated with assessment. First, four different types databases were established according process cyclohexane. Considering intrinsic relationship between factors in fault database, used improve semantic transformation accuracy calculate confidence factors. The simulate...

Multivariable coupling, nonlinear and large time delays exist in the coil outlet temperature (COT) control system of ethylene cracking furnace, which make it hard to achieve accurate over COT furnace actual production. To solve these problems, an inverse controller based on interval type-2 fuzzy model strategy is introduced. In this paper, proposed scheme divided into two parts: one approach structure part (IT2-FM), utilized process output. The other (IT2-FMIC) part, output target value....

Abstract The change of PM2.5 concentration in air quality is nonlinear and difficult to predict. Therefore, a self-learning interval type-2 fuzzy neural network (SLIT2FNN) proposed. SLIT2FNN has two parts: online structure learning parameter learning. In learning, improve the training accuracy speed model, Possibilistic Fuzzy C-Means (PFCM) algorithm used process input data obtain number initial rules before model training. PFCM introduces concept possibility P (FCM), allowing overcome...

To address the issue of low accuracy in soft sensor modeling key variables caused by multi-variable coupling and parameter sensitivity complex processes, this paper introduces a TSK-type-based self-evolving compensatory interval type-2 fuzzy Long short-term memory (LSTM) neural network (TSECIT2FNN-LSTM) model. The proposed TSECIT2FNN-LSTM integrates LSTM with inference system to long-term dependencies sequence data utilizing gate mechanism network. structure learning algorithm uses firing...

In this paper, an improved interval type-2 fuzzy logic controller based on genetic algorithm is presented to deal with the problem of losing uncertain information in type reduction process controller. The four boundary values output are obtained by using reasoning and Wu-Mendel uncertainty bound algorithm. Then outputs re-optimized. quantization factor, scaling factor membership functions evolved By constructing fitness function as performance index, which can be directly related system...

Abstract In this paper, an interval type-2 fuzzy fractional order PID controller (IT2F-FOPIDC) based on particle swarm optimization is proposed. The proposed has two adjustable degree of freedom parameters, so the control range parameters becomes larger and can controlled object more flexibly. Aiming at difficult problem many tuning, paper introduces algorithm to optimize parameters. avoid problems long iteration time high computational cost caused by Karnik-Mendel reduction algorithm,...

The safety problems of petrochemical enterprises can not be collected and reflected in a short time, there is lack effective risk evaluation system. Firstly, the weight calculation method improved by interval type-2 fuzzy method, internal relationship between influencing factors considered, comprehensive each factor determined combining subjective objective weight; Then, model VIKOR based on improvement established, maximum benefit value regret degree scheme are analyzed, compromise...

Recently, Ballantine and Merca proved that if $ (a,b) \in \{(6,8),\ (8,12),\ (12,24),\ (15,40),\\ (16,48),\ (20,120),\ (21,168)\}$, then $\sum_{ak+1 \ {\rm square}}p(n-k)\equiv 1\ ({\rm mod}\ 2)$ only $bn+1$ is a square. In this paper, we investigate septuple $(a_1,a_2,a_3,a_4,a_5,a_6,a_7)\in \mathbb{N}^5\times \mathbb{Q}^2$ for which $\sum_{a_1k+a_2 square}}p(a_3a_4^\alpha n+a_6 a_4^\alpha+a_7-k) \equiv $a_5n+1$ We prove some new parity results of sums partitions squares in arithmetic...

In this paper, a self-organizing interval type-2 fuzzy neural network based on singular value decomposition and QR with column pivoting (SVD-QR)method is proposed to solve the identification problem of nonlinear systems. The model can realize both structure learning parameter learning. Firstly, in aspect learning, error criterion completeness rules are used judge whether growing or not, at same time, SVD-QR method find less active delete. Secondly, sliding-window secondorder algorithm...

Recently, Ballantine and Merca proved that if $ (a,b) \in \{(6,8), (8,12), (12,24), (15,40), (16,48), (20,120), (21,168)\}$, then $\sum_{ak+1 {\rm square}}p(n-k)\equiv 1 ({\rm mod} 2)$ only $bn+1$ is a square. In this paper, we investigate septuple $(a_1,a_2,a_3,a_4,a_5,a_6,a_7)\in \mathbb{N}^5\times \mathbb{Q}^2$ for which $\sum_{a_1k+a_2 square}}p(a_3a_4^\alpha n+a_6 a_4^\alpha+a_7-k) \equiv $a_5n+1$ We prove some new parity results of sums partitions squares in arithmetic progressions...

To solve the problem of low accuracy soft sensor monitoring variables due to strong nonlinear, multivariable coupling, parameters time-varying and large time delays in complex chemical process, a modeling method based on self-organizing interval type-2 fuzzy neural network (SOIT2FNN) is proposed. Firstly, that structure type 2 (IT2FNN) difficult determine, an algorithm for generation rules uses firing strength rule similarity define growth deletion indicators The determine whether generate...