Mood Based Building Automation Using Fuzzy Petrinets

This paper deals with MOOD based building automation system. It categorises various aspects of MOOD and its relationship with dynamic setting of the environment. It uses Fuzzy Petrinets to fine tune the environment setting if it exists. Error and error Dot membership function based on Rule Matrix. Workflow is defined to generate the algorithm.

ZigBee is an open source hardware device which can be programmed for specific need.
Secondly which is the most important issue is low power solution for automation it can receive signal wirelessly from a device and can respond to it.Thirdly it provides a good collaboration among its peers working as a Multiagent System.The novelty of this paper is Fuzzy mood based environmental dynamics (Prakash, S., & Darbari, M., 2012;Ahmad, F., Darbari, M., & Asthana, R., 2015;Darbari, M., Yagyasen, D., & Tiwari, A., 2015).

Ambient Dynamics using fuzzy mood based Detection system
There are various emotional models being applied to judge the person's mood using the basic descriptors like "Depressing", "Energetic", "Jovial", "Palm".
The main difficulty comes in when we change from one mood another than how to set the ambience during that situation.
In order to analyze the emotional model providing the continuous connectivity we first of all deal with Emotional Categorization and its relationship with fuzzy Petrinets.

Emotional Analysis
Emotional analysis can be classified into various types of mood which need to judged and analyzed.We can classify it into two parts: Thayer described various categories of mood under some quantitative dimension defined in In order for our system to judge efficiently between various combinations of stress and energy using facial expression we have used Fuzzy Petrinets.
The figure 2-2 describes how the ambient environment can be changed according to his MOOD based on Fuzzy Rule base relation.

Error and Error Dot Memberships
The degree of membership for an "error" of -1.0 projects up to middle of overlapping part of the "negative" and "zero" function so the results is "negative" membership = 0.5 and "zero" membership = 0.5.It can derived from figure 2.6.This particular input condition indicates that the feedback has exceeded the command and is still increasing.Thus, there is a unique membership function associated with each input parameter.The membership associate a weighting factor with values of each input the effective rules.By computing the logical product of the membership weights for each active rule, a set of fuzzy output response magnitudes are produced.

Formal verification of MOOD using Fuzzy Petrinets
Consider a situation of Mood Detection where stress (e1) occurs more than once within period T2, this situation concerns the case where multiple occurrences of one event within a certain time period causing another single event to occur.This situation introduces the motion of expiration time of events .If an event is not consumed by a rule, it may expire after a time interval.The Petrinet model pertaining to these events can be defined by placing tokens e' and e'', (Figure 2.7) the transition t2 and t3 are enabled, but they cannot be fired immediately.When there are two tokens arriving in place e', and e'' (figure 2.7), transition t1 fires immediately and produces the event e2 is "Notification to Control Centre".The condition can be expressed in the linguistic format (Saastamoinen, 1995;Ahmad, S. S., Purohit, H., Mohammed, F. N., & Darbari, M., 2013) as: R1 : IF e'1 is s'3 AND e'11 is s'4t3 THEN e'111 is s'7t3.
Where s'3, s'4 and s'7 set of rules under which an event can occur.
We can write an algorithm for the above condition where the decision of ambient setting can be initiated as : input: mood based ambient control: mbac output: set of output rules : sorl sorl =  ; for each output place e' of sort do // creates set of input variables on whose e depends.for each input transition e of sorl e' do // add all inputs of transition 's' to input set e inputs = inputs Us inputs ; end for each input transition e of sorl do // construction of rule corresponding to the transition 's' rule =  // rule belongs to mood based ambient settings for each element in from inputs do if rule  , then rule = rule + AND; if in  s, inputs then rules = rule + in.name is edge (s, it).value; else rule = rule + in.name is UNDEF; end rule = rule + THEN e'.name is edge (in , e').value; rb  rule database rb  rbUrule ; // add rule of ambient settings to Database for further reference.end.
The above algorithm rules can be quantified in the form of Rule Matrix Workflow of Fuzzy Petrinets (Sen, 2007;Sethi, 2000).There will be some combination of variables which is

Input Degree of Freedom
We can draw the membership function by considering the degree of membership from an "error" of -1.0 so the middle of the overlapping part is a negative and zero function.For an error-dot of +2.5, a "zero" and "positive" membership of 0.5 is indicated.

Figure 1 -
Figure 1-1 shows the connectivity between sensor camera interfaces to ZigBee board for processing which are finally connected to Control Centre.

Figure 0 - 2 Figure
Figure 0-2 Mood based ambient setting of the previous function can be defined by the help of petrinet as: The set of IF-THEN rules, which forms the linguistic description: R1 : = IF X1 is A11 AND ..... Xn is AIn THEN Y is B1 : : Rm : = IF X1 is Am, AND .......... Xn is Amn THEN Y is Bm.

Figure 0 - 4
Figure 0-4 Fuzzy Modeling when one of the edge is missing

Table 0 -
10 Rule Matrix of Ambient Settings