@article{84_60de066b-0c03-462d-a713-49e5a8888b58,

title = {Application of Evidence Theory for Training Fuzzy Neural Networks in Diagnostic Systems},

author = {Ivanov V.K. and Palyukh B.V.},

url = {https://link.springer.com/article/10.1134/S1054661823030197 

https://disk.yandex.ru/i/hYt3FYBoxOo2hA},

issn = {1054-6618},

year  = {2023},

date = {2023-09-29},

urldate = {2023-09-29},

journal = {Pattern Recognition and Image Analysis},

number = {33},

pages = {354-359},

abstract = {The paper substantiates a method for creating training datasets for fuzzy neural networks, which can be used to promptly obtain probabilistic estimates for the causes of abnormal critical events or incidents in diagnostic systems. The rules for converting the hypotheses on potential incident causes into intervals of defect probability in a process chain at a certain stage of continuous production are considered using belief functions. We propose a procedure for converting these hypotheses into a database of fuzzy production rules automatically, which provides training an adaptive neural network based on the Takagi-Sugeno-Kang fuzzy inference system. This makes it possible to quickly calculate a relatively accurate probabilistic estimate of a malfunction in the process chain without using expensive computing resources. Ivanov, V.K., Palyukh, B.V. Application of Evidence Theory for Training Fuzzy Neural Networks in Diagnostic Systems. Pattern Recognit. Image Anal. 33 , 354–359 (2023). https://doi.org/10.1134/S1054661823030197},

keywords = {belief function, diagnostic system, evidence theory, fuzzy logic, fuzzy neural network, incident, membership function, multistage production process, process chain, production rule},

pubstate = {published},

tppubtype = {article}

}

@article{nokey,

title = {Демонстратор программной платформы для настройки гиперпараметров нечеткой нейронной сети},

author = {Иванов В.К. and Палюх Б.В.},

editor = {демонстратор, диагностическая система, инцидент, многостадийный технологический процесс, нечеткая логика, нечеткая нейронная сеть, продукционное правило, теория свиде-тельств, технологическая цепь, функция доверия, функция принадлежности, ANFIS, TSK, demonstrator, belief function, diagnostics, evidence theory, fuzzy logic, incident, membership function, multistage production process, neural network, process chain, production rule},

url = {https://disk.yandex.ru/i/uWiFPzjvqsyRvg

http://swsys.ru/files/2022-4/609-617.pdf},

doi = {10.15827/0236-235X.140.609-617},

year  = {2022},

date = {2022-12-31},

urldate = {2022-12-31},

journal = {Программные продукты и системы / Software & Systems},

volume = {35},

number = {4},

pages = {609-617},

publisher = {Программные продукты и системы / Software & Systems},

abstract = {В статье приводится описание исследовательского демонстратора для экспериментальной проверки и оценки вариантов применения нечетких алгоритмов и нейронных сетей в экспертной системе для диагностики сложных многостадийных технологических процессов. Цель разработки демонстратора – создание научно-технического задела для передачи готовых к внедрению решений на следующие этапы проекта.

Демонстратор позволяет оценить уровень системной готовности разрабатываемых компонентов, провести исследовательские испытания, проверить работоспособность и эффективность функционирования программных реализаций при различных значениях параметров и их сочетаниях. Диагностика состояния сложного многостадийного технологического процесса предполагает совместную обработку первичных данных для получения вероятностных характеристик аномальных критических событий или инцидентов в условиях неопределенности.

Авторами предложен вариант использования нечеткой нейронной сети, обучение которой происходит данными, сгенерированными с помощью функций доверия. Подход дает возможность значительно ускорить вычисления и минимизировать ресурсную базу. В статье основное внимание уделяется описанию функций управления моделями нейронной сети и обучающими наборами данных, обучения нейронной сети и проверки его качества, диагностики технологического процесса в различных режимах. Подробно описаны настраиваемые гиперпараметры нейронной сети. Приведены примеры реализации диагностических процедур в различных режимах. Показано, что при функционировании программной диагностической системы в условиях, близких к реальным, могут быть проверены и экспериментально обоснованы исходные предположения, касающиеся сокращения времени обнаружения и прогнозирования инцидентов, и более точно определены множества технологических цепей, являющихся причинами инцидентов.

 

A software platform demonstrator for configuring ANFIS neural network hyperparameters in fuzzy systems

V.K. Ivanov, B.V. Palyukh

This article describes the research demonstrator for experimental verification and evaluation of fuzzy algorithms and neural networks in an expert system for complex multi-stage technological processes. The demonstrator development purpose is to create a scientific and technical foundation for the ready-to-im-plement solutions transfer to the next project stages.

The demonstrator allows assessing the readiness level of the components being developed, conducting re-search tests, checking the operability and efficiency of the software implementations functioning proposed at various parameter values and their combinations. A complex multi-stage technological process state diagnos-tics involves the joint primary data processing to obtain probabilistic abnormal critical events or incidents characteristics under conditions of uncertainty.

The authors propose a way of using a fuzzy neural network, which is trained with data generated by belief functions. The approach makes it possible to significantly speed up calculations and to minimize the resource base. The article focuses on describing the neural network models and training datasets management, neural network training and quality control, the technological process diagnostics in various modes. The configurable hyper-parameters of the neural network are described in detail. There are examples of the diagnostic procedures implementation in various modes. It is shown that with the software diagnostic system functioning in condi-tions close to real, the initial assumptions concerning the time reduction for detecting and predicting incidents can be verified and experimentally substantiated. In addition, the technological chains sets that are the incidents causes can be more accurately determined.},

keywords = {ANFIS, belief function, demonstrator, diagnostics, evidence theory, fuzzy logic, incident, membership function, multistage production process, neural network, process chain, production rule, TSK, демонстратор, диагностика, инцидент, многостадийный технологический процесс, нечеткая нейронная сеть, нечеткая система, продукционное правило, теория свидетельств, технологическая цепь, функция доверия, функция принадлежности},

pubstate = {published},

tppubtype = {article}

}

@article{nokey,

title = {Generation of Production Rules with Belief Functions to Train Fuzzy Neural Network in Diagnostic System},

author = {Ivanov V.K. and Palyukh B.V. and Sotnikov A.N. },

url = {https://disk.yandex.ru/i/c59jQ0WhniInpw},

doi = {10.1134/S1995080222130169},

issn = {1995-0802},

year  = {2022},

date = {2022-10-31},

urldate = {2022-10-31},

journal = {Lobachevskii Journal of Mathematics},

volume = {43},

number = {10},

pages = {2853–2862},

abstract = {The article examines some algorithms for joint processing of raw data on the state of a complex multistage continuous production process to obtain probabilistic characteristics of abnormal critical events that can potentially lead to single failures or even emergencies. The article, thus, proposes and substantiates an approach to developing a technology to detect and predict malfunctions and determine their causes. The sequence of operations to process and convert diagnostic process data is considered essential. As a result, the article presents a general diagnostic model of a multistage production process. The model can formalize the main objects and processes in terms of the problem being solved. An incident is defined as an abnormal critical event described by non-normative values of diagnostic variables. Incidents are shown to be indicated by the corresponding membership functions. The hypotheses on potential incident causes are discussed to be built with belief functions being the basis of evidence theory or Dempster−Shafer theory. The hypotheses are characterized by an interval of malfunction probability in some process chain. The authors propose a procedure of converting these hypotheses into fuzzy production rules automatically. The automatical procedure is a prerequisite to using fuzzy neural networks to obtain a reliable estimate of the degree of belief in the incident cause. As a summary, the generated database of the production rules to train a neural network is substantiated to be used with the TSK architecture that makes possible to estimate a malfunction probability in the process chain quickly without resource-intensive computations.},

keywords = {belief function, diagnostic system, diagnostic variable, evidence theory, fuzzy system, incident, membership function, multistage production process, neural network, process chain, production rule, status-4},

pubstate = {published},

tppubtype = {article}

}