CONDUCTING A SYNTHESIS OF A DIGITAL AUTOMATON FOR AN AUTOMATED FIREFIGHTING SYSTEM

Every year the industry in the world is gaining momentum: the number of industrial enterprises is growing, and with it the number of accidents at them. Oil today is the most common product for the synthesis and production of products. Increasing the level of fire protection systems at oil refine¬ries remains one of the most important components of protecting people from technogenic hazards. The speed of innovation allows the application of artificial intelligence in the creation of automated fire protection systems. Research objective. This study is aimed at building a model of an automated integrated fire protection system (AISPPO).Through the synthesis of digital automata and minimizing the control functions of the digital model is created a system of automated integrated fire protection system. Materials and methods. To solve the problems of research used methods of constructing graphical algorithms of automated integrated fire protection system. This system is a new approach to solving the issue of safety of industrial facilities in the oil refining industry. Results. The proposed new model of the software implementation of a digital automaton in an automated integrated system of fire detection and monitoring of an oil refinery has made it possible to create a bank of calculated and analytical data on all potential types of failures in the structure of the enterprise in order to train personnel and make changes to existing methodological documents and instructions for personnel actions in a particular situation. Conclusion. The developed technology allows you to process the incoming signal contained in cyclograms into an intermediate form for the synthesis of digital automata using innovative tools.


Introduction
The model of a digital automaton (CA) of an automated firefighting system with input and output signals induces a one-to-one mapping of the set of commands in the input signals (the input command mapping) into the set of commands in the output signals. In this article we consider the stages of solving the problem of synthesizing automata by the mappings induced by them.
According to the developed algorithm ( Fig. 1), when selecting the states, one should take into account such recommendations as: -the correspondence of the set and the initial set; -the choice of the next state is made according to the ascending order after each PROCESS block; -before each DECISION block, after each line adjacency point, which indicates the transition direction [1] .

Refinery accident and fire analysis
According to the developed functioning algorithm decided that the scheme of the CA model of the automated fire suppression system (AFS) will include 14 0 1 2 13 , , , , a a a a  states, where 0 a -the initial state [2].
All 14 states of digital automaton will be encoded by four-bit binary numbers (Tables 1-3). The memory block, in this case, will be a four-bit parallel register on D-triggers, because storage of each bit of the binary code will use one trigger [3,14].
Based on the developed algorithm of functioning of the digital automaton ASFT we build a graph [4]. The state of the device in the graph will depend proportionally on the values of the vertices (vertices of the graph). The vertices of the graph of the ASPT CA model are connected by arcs, which show the direction of transition. At the top of the arcs we write transition conditions and output signals [5,11,12].  One of the loops is broken 12 a The 2nd fire detector is triggered 13 a There is a fire in the premises Read the graph as follows: the automaton is in the initial state 0 a , then under the signal from the fire detector it changes its state to 1 a , with this transition the output signals must be formed 1 5 6 , , y y y . This is followed by a transition 2 a to the state with the formation of output signals 1 3 6 , , y y y . From the state 2 a to 3 a , then to 4 a . From the state 4 a the transition to the state 5 a , or 8 a [5] is possible. The 5 a automaton enters the state if the external condition (fire is detected) 3 x is 1 ( 3 x ) with issuing of 1 3 4 , , y y y control signals , and the automaton 8 a enters the state if the same signal is 0 ( 3 x ), etc. After constructing the graph, fill in the table of functions of the vertices of the graph. Using this table you can write functions for any number of variables (Fig. 2). After that it is necessary to analyze it carefully in order to simplify (minimize) it, because the tabular method does not give an opportunity to obtain in perfect disjunctive normal form (DNF) for outputs the minimal disjunctive normal form (MDNF) or the minimal conjunctive normal form (MCNF) [6]. In this case it will be enough to apply the gluing law to some expressions [7,8].
On the transition column of the digital automaton of the automated integrated firefighting system let's fill in the table 4. Example for the first line: The initial state, which is coded as "0000", changes to the state with the code "0001". This transition is unconditional. We see that 4  , , D D D "Trigger Control Signals", and to supply 0 to the others, at this transition the signals are formed 1 5 6 , , y y y . All the following lines are completed in the same way.

Fig. 2. Transition graph of the digital automaton of the automated integrated firefighting system
According to the table of functioning of the digital automata graph of AISPT we make analytical expressions in the SDNF for output signals 4 3 2 1 , , , D D D D . The perfect disjunctive normal form of the function is a disjunction of elementary co junctions [7,15].
The output signal 1 y will be generated if the automaton or in 7 a , or in 13 a , and the sign other output signals and trigger control signals are written [10,13]. "Trigger Control Signals", and to supply 0 to the others, at this transition the signals are . All the following lines are completed in the same way.

Transition graph of the digital automaton of the automated integrated firefighting system
According to the table of functioning of the digital automata graph of AISPT we make analytical expressions in the SDNF for output signals 1 2 3 , , y y y , and also signals of control of triggers . The perfect disjunctive normal form of the function is a disjunction of elementary co will be generated if the automaton is in state 0 a , or in 4 a 3 x = 1, or in state 8 a and the sign 4 x = 0. Similarly the functions for other output signals and trigger control signals are written [10,13]. "Trigger Control Signals", and to supply 0 to the others, at this transition the signals are

Transition graph of the digital automaton of the automated integrated firefighting system
According to the table of functioning of the digital automata graph of AISPT we make analytical , and also signals of control of triggers . The perfect disjunctive normal form of the function is a disjunction of elementary con-, or in 1 a , or in 3 a , or in 5 a , = 0. Similarly the functions for  D a a a a a a x a x a x a x a   a a a a a a x a a x a  D a a x a a a x a x a x a x   a a x a a a x a  a x  x a x a x a x a x a x a x   a x a a x a  a x Formulas (4), (9), (10), and (11) have been simplified using the gluing law. Using the laws of double negation and de Morgan formulas, the initial expressions from the basis of AND, OR, NOT are converted to the basis of AND, NOT.   5 .

Conclusion
Minimized logic functions for KS 1 and KS 2 will be used in the construction of the model of digital automata of automated integrated fire protection system.