Technical Diagnosis

the scientific and engineering discipline that studies and establishes indications of defects in technical equipment, as well as the methods and means used to search for and detect defects. The principal concern of technical diagnosis is the organization of an effective check on the proper working order and functioning of equipment—individual components, subassemblies, blocks, stock, devices, aggregates, and systems— as well as checks on processes for transmitting, processing, and storing material, energy, and information. Thus, technical diagnosis fundamentally deals with organizing the processes of diagnosing the technical state of objects during manufacture and operation, including before, during, and after use of the objects, and during preventive maintenance, repair, and storage. Technical diagnosis is one of the most important measures used to ensure and maintain the reliability of equipment.

Diagnosis is carried out either directly by a person, for example, visual inspection or inspection “by ear,” or with the help of equipment. The object being inspected and the diagnostic means used together form a diagnostic system and, by means of their interaction, produce a diagnostic algorithm. The result is a conclusion on the technical state of the object—a technical diagnosis, such as “the radio receiver is in working order,” “the machine tool is not in operational condition,” or “a frequency detector has failed in the television receiver.”

A distinction is made between off-line and operational diagnostic systems. The former are used during manufacture, repair, and storage of the object, in preventive maintenance, and before and after the object is used, at which times it is essential to test the proper working order of the object and to locate defects. In this instance, specially organized tests can be made of the object. Operational diagnostic systems are applied when the object is being used for its specific purpose, when there must be a check on the correct functioning of the object and defects that disrupt such functioning must be detected. In this case, the object can be subjected only to those actions provided by the algorithm of its functioning.

The development of a diagnostic system includes study of the object, of the object’s possible defects, and of the indications of such defects; compilation of mathematical models (formalized descriptions) of a properly functioning object and of the object in nonfunctioning states; construction of diagnostic algorithms; and adjustment and testing of the system.

In studying the objects to be examined, a classification of the objects according to various features is of prime importance: for example, according to the nature of the change in the values of the parameters or according to the type of energy required. Defects are studied in order to determine their nature and causes, the probabilities and physical conditions of their occurrence, and the conditions for their detection.

A mathematical model—either a deterministic or probability model—of the object under diagnosis is a description of the object in proper working order and in nonoperational condition. It states the formal relationship between the possible actions applied to the object and the object’s responses to the actions. Models used in technical diagnosis—even those describing an object in proper working order—may differ from models used in designing the same object. For example, models used in diagnosing the technical state of noise-producing objects may be noise or vibration curves, which are used for acoustical methods of technical diagnosis. Models used in microelectronics and welding may be X-ray images of the objects, which are used for nondestructive inspection.

A diagnostic algorithm provides for the execution of a conditional or unconditional sequence of specific experiments with the object. Each experiment is characterized by a test or operational action and by the composition of the monitored features that determine the object’s response to the action. A distinction is made between test algorithms and search algorithms. The former make it possible to detect the presence of defects that disrupt the proper working order of the object or its correct functioning. The results of experiments conducted by means of search algorithms indicate what defect or group of defects from among those examined is present in the object.

The diagnostic means are the carriers of the diagnostic algorithms. They store the object’s possible responses to actions, generate and apply the test actions to the object, read the object’s actual responses, and make the diagnosis by comparing the actual responses with the possible responses. Diagnostic means are divided into instrument, program, and program-instrument means; the last two categories are used to diagnose the technical state of computers operating according to a task program. Instrument means may be external (in relation to the object) or built-in. External instruments are primarily used in off-line diagnostic systems; built-in instruments are used in operational diagnostic systems. External instrument means may be automatic, automated, or manually controlled; they may also be general-purpose or specialized.

In its methodology, technical diagnosis has much in common with medical diagnosis. Technical diagnosis that determines the real-time technical state of objects is closely connected to technical prognostication and technical genetics, which determine both future and previous technical states from the probable evolutions of the present technical state and previous performance.