A C++ FuzzyEngine library is implemented.
Go to the Fuzzy directory:
cd build
cmake ..
make -j
Build and run unit tests:
cd build
cmake ..
make -j
make test # or: ctest --output-on-failure
Or run test binaries directly:
./bin/testFuzzyEngine
./bin/testDySySim
This project contains some unit tests for Fuzzy and DySySim. Install the C++ UnitTest++ library.
DySySim uses C++ Boost Spirit X3.
Install logging lib spdlog.
For creating diagrams gnuplot is used.
Tipping problem fuzzy rules, using the FuzzyEngine library code:
IF service is poor OR food is rancid THEN tip is cheap
IF service is good THEN tip is average
IF service is excellent OR food is delicious THEN tip is generous with 0.5
IF service is excellent AND food is delicious THEN tip is generous with 1.0
- TippingFIS v0.4.1 --------------------------------------------------
Service = poor (0, 0, 2.5, 5) input = 0 mbs = 0
Service = good (2.5, 5, 7.5) input = 0 mbs = 0
Service = great (5, 7.5, 10, 10) input = 0 mbs = 0
Food = rancid (0, 0, 1.5, 4) input = 0 mbs = 0
Food = delicious (6, 8.5, 10, 10) input = 0 mbs = 0
Tip = cheap (0, 5, 10) mbs = 0
Tip = average (10, 15, 20) mbs = 0
Tip = generous (20, 25, 30) mbs = 0
not a = 1 - a
a and b = min(a, b)
a or b = max(a, b)
not a = 1 - a
a and b = max(a+b-1, 0)
a or b = min(a+b, 1)
Current dir is build:
../bin/Tipping
Defuzzification example 1:
-- Defuzzification: Weighted Average
service
10 11 11 12 16 25 25 25 25 25 25 25
9 11 11 12 16 25 25 25 25 25 25 25
8 11 11 12 16 25 25 25 25 25 25 25
7 11 11 12 15 21 21 21 21 23 23 23
6 10 10 11 13 17 17 17 19 19 19 19
5 10 10 10 12 15 15 15 16 17 18 18
4 8 8 9 11 11 11 11 13 15 15 15
3 6 6 7 7 7 7 7 10 12 13 13
2 5 5 5 5 5 5 5 8 10 11 11
1 5 5 5 5 5 5 5 8 10 11 11
0 5 5 5 5 5 5 5 8 10 11 11
food 0 1 2 3 4 5 6 7 8 9 10
Defuzzification example 2 and 3:
-- Defuzzification: Center of Sums, logic model: Zadeh
service
10 11 11 12 16 25 25 25 25 25 25 25
9 11 11 12 16 25 25 25 25 25 25 25
8 11 11 12 16 25 25 25 25 25 25 25
7 11 11 12 15 21 21 21 21 23 23 23
6 10 10 11 13 17 17 17 19 19 19 19
5 10 10 10 12 15 15 15 16 17 18 18
4 8 8 9 11 11 11 11 13 15 15 15
3 6 6 7 7 7 7 7 10 12 13 13
2 5 5 5 5 5 5 5 8 10 11 11
1 5 5 5 5 5 5 5 8 10 11 11
0 5 5 5 5 5 5 5 8 10 11 11
food 0 1 2 3 4 5 6 7 8 9 10
-- Defuzzification: Center of Sums, logic model: Lukasiewicz
service
10 11 11 12 16 25 25 25 25 25 25 25
9 11 11 12 16 25 25 25 25 25 25 25
8 11 11 12 16 25 25 25 25 25 25 25
7 11 11 12 15 21 21 21 22 22 23 23
6 10 10 11 13 17 17 17 19 19 19 19
5 10 10 10 12 15 15 15 16 17 18 18
4 8 8 8 9 11 11 11 13 15 15 15
3 6 6 6 6 7 7 7 10 12 13 13
2 5 5 5 5 5 5 5 8 10 11 11
1 5 5 5 5 5 5 5 8 10 11 11
0 5 5 5 5 5 5 5 8 10 11 11
food 0 1 2 3 4 5 6 7 8 9 10
DySySim is a very basic TUTSIM look-a-like. TUTSIM was developed in the early 70's for script based simulation of time continuous systems represented in simulation blocks. Only a subset of the TUTSIM simulation blocks are implemented, sometimes with slightly other names.
DySySim uses the .dss extension for the simulation text scripts.
Fuzzy logic and control functions are not (yet) implemented in DySySim simulation blocks. Some hardcoded examples are available.
Go to the scriptsDySySim/RCnetwork directory.
The following DySySim script is used for this example:
// Pulse response RC-network
delta_t = 0.005 t_end = 1.3 width_t = 6 precision_t = 3
1 PLS off=0.0 on=1.0 t_on=0.1 t_off=0.5
2 ATT 1,-3 RC=0.1 // RC time constant = 0.1 sec
3 INT 2 out_t0=0.0
4 LOG 1,3 w=4 p=3 w=4 p=3
Use the next command in the scriptsDySySim/RCnetwork directory:
./simplot.sh
All simulation values are calculated. The response diagram is plotted by gnuplot in a .png file in in the scriptsDySySim/RCnetwork directory:
Further development, testing and debugging are still in progress.