Evolving rule-based models: A tool for intelligent adaptation
read more
Citations
Evolving Fuzzy Systems from Data Streams in Real-Time
Handling drifts and shifts in on-line data streams with evolving fuzzy systems
Evolving Takagi‐Sugeno Fuzzy Systems from Streaming Data (eTS+)
A fuzzy controller with evolving structure
A fast learning algorithm for evolving neo-fuzzy neuron
References
System Identification: Theory for the User
Fuzzy identification of systems and its applications to modeling and control
ANFIS: adaptive-network-based fuzzy inference system
Computer-Controlled Systems: Theory and Design
Fuzzy Model Identification Based on Cluster Estimation
Related Papers (5)
DENFIS: dynamic evolving neural-fuzzy inference system and its application for time-series prediction
Fuzzy identification of systems and its applications to modeling and control
Frequently Asked Questions (15)
Q2. What is the way to reduce the number of membership functions?
In order to maximise the transparency, which also minimises the memory cost, it is necessary to minimise the number of membership functions describing each input variable.
Q3. What is the principle load on the coil?
One of the principle loads on the coil is generated due to the supply of ambient air; required to maintain a minimum standard of indoor air quality.
Q4. What are the common types of fuzzy models?
Fuzzy rule-based (FRB) models, and especially Takagi-Sugeno (TS) models [4], are widely used to represent complex non-linear systems.
Q5. How many rules and 7 membership functions were generated?
Using a batch estimation approach on the training data, the subtractive clustering generated a model with 7 rules and 7 membership functions describing each input variable.
Q6. What is the way to test and validate the model simplification method?
Engineering applications to a time series prediction problem based on data from a real indoor climate control system has been considered to test and validate the model simplification method.
Q7. What are the advantages of using genetic algorithms?
Alternative techniques for identification of both model structure and parameters that are, in principle, nonlinear optimisation problems, include direct use of genetic algorithms [10]-[12] or gradient-based backpropagation [13].
Q8. What is the effect of similarity measures on the model?
Application of similarity measures additionally improves the transparency and simplicity of the model, without significant degradation in the model precision.
Q9. How is the output of the TS model calculated?
The model output is calculated by weighted averaging of the individual rule contributions using the centre of area de-fuzzification operator.
Q10. What is the advantage of the latter?
The advantage of the latter is the higher precision that is gained by solving the parameter and structure identification simultaneously.
Q11. How many parameters are used to demonstrate the reduction in the number of model parameters?
To demonstrate the reduction in the number of model parameters, through similarity, the approach was applied to a time series prediction problem.
Q12. What is the root mean squared prediction error of the initial and simplified models on the training data?
The root mean squared prediction error of the initial and simplified models on the training data was 0.030 and 0.031 respectively (no units for control signal).
Q13. What is the potential of a data point to be a centre of a cluster?
Pi = Nj ijD 1 , (2a)2ji zz ij eD , i=1,2,…N , (2b)where Pi is the potential of the data point zi=[xi;,yi] to be a cluster centre, Dij denotes the contribution of every single distance, N is the number of training data samples and = 4/r 2 ;r is the cluster radii.
Q14. What is the real issue in many industries and organisations?
At present, the real issue in many industries and organisations is how to effectively cope with the information in exponentially growing data-bases.
Q15. What is the simplest way to identify a TS model?
1. (Off-line) identification of TS modelsTakagi-Sugeno model [4] could be represented as:R i: IF (x1 is LV i1) … AND (xn is LV in) THEN (yi= pi1x1+…+pinxn+qi); i=1,…,NR , (1)where Ri denotes the i th fuzzy rule, NR is the number of rules, x is the input vector x=[x1,x2,…,xn]