The hydraulic pump is the power component of the entire hydraulic system. It converts the mechanical energy of the prime mover (electric motor, internal combustion engine) into the pressure energy of oil, providing oil with a certain flow rate and pressure for the hydraulic system. Axial piston pumps have their own advantages, such as compact structure, small radial size, small inertia, and high volumetric efficiency. The current maximum pressure can reach 40.0MPa or even higher. Therefore, they are generally used in high-pressure systems such as engineering machinery and presses in many occasions such as mining machinery, steel plants, etc.
A failure of the hydraulic pump will cause the entire hydraulic system to fail to work properly. Fault diagnosis of the hydraulic pump is of great significance to the hydraulic system and the entire production line. This article mainly establishes and discusses a model of a diagnostic method for faults in swash plate axial piston pumps.
For intelligent fault diagnosis of plunger pumps, methods such as fault diagnosis expert systems are generally developed to directly diagnose the location and cause of the fault, and then use the expert's knowledge to reason and propose solutions and measures for the fault. However, faults are rarely predicted, and the plunger pump cannot be effectively inspected in advance, which may delay the optimal inspection time, affect production, cause the pump to suffer irreparable damage, and shorten the service life of the pump.
Pump failures have many contributing factors. One factor can cause several faults, while one fault may result from the combined effects of several factors. In many cases, therefore, fault conditions cannot be described precisely. This paper uses fuzzy processing and fuzzy mathematics to fuzzify fault symptoms as the basis for diagnosis, while outputting fuzzy values for the fault causes. It diagnoses piston-pump faults to identify their causes and the affected components, and predicts piston-pump failures to provide early warning.
Fault classification is a comparison of different faults at the same time to identify the most likely fault. According to the fuzzy relationship between the obvious degree of the fault symptom and the fault cause, as well as the corresponding relationship between the fault cause and the difficulty of the fault source, the two paths are comprehensively considered to find the final cause of the fault.
Source: Public account "Hydraulic Learning"