Forecasting methods of intelligent systems technical condition analysis

Modern information processing and control systems reached the level that makes it possible to use detailed information on device and human nervous system functioning. Using a cognitive psychological approach in intelligent systems development, allowed them to imitate human nervous system functions. The development of increasingly powerful intelligent information and control systems with learning and self-learning, including information and measuring systems, is expected as a result. They will have enhanced cognitive capabilities due to the stimulation of cognitive functions and processes responsible for perception, learning, thinking, and consciousness in the human nervous system. Development of this field allows to develop intelligent systems with thinking and behaviour analysis elements. Adding some creative possibilities, for example, related to automatic hypotheses and models creation and self-learning for new task solving, allows to improve the efficiency of intelligent systems. Due to this new approaches to the artificial brain and the artificial nervous system of robots development, which relate not only to artificial intelligence but also to its development in the form of an artificial mind. Thus, the main scientific and development task for models, methods, methodics, and algorithms for intelligent systems technical condition forecasting based on soft computing is important and relevant for science and practice. Intelligent technologies allowing to develop useful intelligent systems are continuously being improved. Currently, quite powerful tools for implementing the technology of expert systems, fuzzy logic, neural network systems, and multi-agent systems technology are used. They are rapidly improved by adding software packages and hardware tools. New technologies are being developed in the so-called neuromorphic systems field that models some brain structures and in the parallel computing and quantum computers field. These technologies aim to raise significantly the intelligence level of systems in the future.


Problem statement in general and its relation to important scientific or practical tasks
Nowadays, we have no doubt that artificial intelligence passed the following development stages of theoretical development and practical application in the expert and problem-oriented intellectual systems fields. Artificial cognitive systems and systems with creative abilities currently expand artificial intelligence.
Modern intelligent systems for data processing and control gained a level that allows use of indepth knowledge about the structure and operation of the human nervous system. Using a cognitive psychological approach in intelligent systems development allowed them to imitate human nervous system functions. The development of increasingly powerful intelligent information and control systems with learning and self-learning, including information and measuring systems, is expected as a result.
They will have enhanced cognitive capabilities due to the stimulation of cognitive functions and processes that are responsible for perception, learning, thinking, and consciousness in the human nervous system. Development of this field allows to develop the intelligent systems with thinking and behaviour analysis elements. Adding some creative possibilities, for example, related to automatic hypotheses and models creation, and self-learning for new task solving, allows to improve the efficiency of intelligent systems.
Due to this new approaches to the artificial brain and the artificial nervous system of robots development which relate not only to artificial intelligence but also to its development in the form of an artificial mind.

Analysis of recent studies and publications initiating this task solving
Still, control systems differ significantly: forecasting, monitoring, diagnostic, development, and planning systems. Intelligent control systems' distinctive characteristic is their classification into the class of dynamic systems that work in real-time and include subsystems of interaction with the outside world (sensors, actuators) [1,2] .
Technologies and highly productive microprocessors development with a large amount of memory, the possibility of organizing multiple networks for the implementation of parallel calculations, on the one hand, and the necessity to process a large amount of data, use of knowledge databases for aimed activity generation, on the other one, led to intelligent systems development.
An intellectual system is considered united by an information process structure of technical instruments and software (including measuring instruments and sensors or special information and measurement equipment) working under the operator's control or independently. It is able to synthesize, generate decisions on actions, and solve tasks effectively based on measurement results, data, knowledge, and motivation [3,4,7,8] .
The general intelligent system structure is in Based on information about the environment and analyzing its condition, the system uses memory and motivation syntheses aim which is processed together with other data by a dynamic expert system. The latter, using a knowledge database, generates expert assessment based on which action decision is made, and results are forecasted.
According to the decision, a control mechanism is developed, so a particular algorithm or control law is synthesized and implemented by various executive bodies and directly affects the control object.
Obtained results of controlling effect are compared with the planned ones (feedback tools) [5,6] . In the case of obtained results, non-conformity based on repeated expert assessment system, performs actions to eliminate this non-conformity. If conformity is not obtained, then the aim should be adjusted. This structure is invariant to the control object and universal, Fig. 1.
A number of practical control tasks exist where information about the object can be incomplete, imprecise, or unclear. In this case, the application of classical computing algorithms is inefficient and lacks to obtain the necessary result [9] .
Moreover, the interrelation between input and output parameters may be complicated to model or even impossible. In such cases, a positive result is possible due to the use of neural network technologies.

Study goal (task) statement
Thus, the goal of this article is an analysis of intelligent systems' technical condition forecasting methods, including modern informational and measuring systems and scientific task statement aiming to improve technical condition forecasting reliability.

Determining ways for forecasting reliability improvement of intelligent systems technical condition
Researchers proved that modern tools for diagnostic and forecasting of intelligent systems' technical condition are based on features and characteristics of various physical phenomena and complicated technologies of collecting, processing, storing, transmitting, and presenting information realization [9,10] . and tools synthesis [4,5] .
It should be mentioned that extended implementation of forecasting means for intelligent systems technical condition is stimulated by constantly increasing requirements for operational reliability and safety, on the one hand, and related to the necessity of solving many problems with the insufficient study of operational failure algorithm and most heavily loaded information and measuring instruments damage, lack of necessary statistical data on operational failures characteristics complicated obtaining of adequate, complete, and reliable models for assessment and forecasting of intelligent systems components technical condition, etc. on the other [5,7] .
On the other side, the widest range of diagnostic equipment and systems is offered on the market, which differs in methodical, mathematical, algorithmic, software and hardware features and has different principles of operation, functional capabilities, metrological characteristics, areas of application, weight and size parameters, element base, price ratio, etc. [7,8] .
Experts in [9][10][11] suppose that it caused a change of methodics and instruments deficiency problem to the new problem of choosing a diagnostic support combination which is able to provide the required level of reliability of forecasting and diagnostic.
Solving this problem demands, as a rule, carrying out theoretical and experimental complex studies, which aim to analyze specific intelligent systems operating conditions, equipment design features, possible diagnostic features determination, diagnostic methods, and instruments development, capable to function in marine operating conditions [5][6][7][8]11] .

Scientific task statement
Scientific works analysis proves that special at- Therefore, a formal scientific task statement of improving existing and development of new forecasting special diagnostic methods, algorithms, and equipment adapted to the specific features of diagnostic objects implementation [9][10][11] .
Practical realization of this task should be regarded as equal by complexity to ensuring high-level automation of intelligent systems tasks.
It should be mentioned that the modern forecasting process of intelligent systems' technical condition is particularly related to computer information processing using advanced technologies, namely artificial intelligence (AI) [9,11] .
Key areas for artificial intelligence systems ap- intelligence» term was expanded to «computing intelligence», which is related to soft computing and soft knowledge concept [4,11] .
Authors consider that «soft computing» term in a mathematical and algorithmic sense is based on the following models and methods: possibility theory, fuzzy sets, fuzzy logic, fuzzy control, and related formalisms to model uncertainty, possibility-probabilistic modeled and decision-making methods, evolutionary modeling and genetic algorithms, and systems with chaotic dynamics [9,11] .
Thus, perspective directions for increasing the forecasting reliability of intelligent systems' technical condition in algorithmic sense are based on so-called soft computing (Fig. 2). of intelligent systems technical condition models and methods based on soft computing is given [1,5,11] .
It should be emphasized that these models and methods are implemented exactly in mathematical, algorithmic, and software components of intelligent model.

Conclusions from this study and further prospects of study in this field
1. Experts claim that the general problem of intelligent systems reliability, including informational and measuring ones, includes many components, which task of forecasting the technical condition is important and relevant. Many ways to solve this task exist in modern conditions, but the study showed the need to find new approaches, methods, and models exist nowadays. It is related to the absence of statistical or other a priori data on possible external influence and destabilizing factors in most cases. Providing a high level of forecast reliability requires effective approaches. The most appropriate one is based on artificial intelligence.

Existing science-based approaches to in-
crease the efficiency of complex, intelligent systems