The advantage of the neural network model is that unlike traditional computer programs, it does not incorporate fixed mathematical algorithms to solve specific problems. It is intelligent programming. In general, when the data does not have a simple pattern, or if the appropriate equation is a higher-order polynomial, ordinary statistical modeling may have limitations in accuracy. Artificial neural networks use learning algorithms to find the optimal solution. The neural network system trains itself by learning samples. When each set of data is input, the difference between the actual output and the desired output of the network is used to modify the connection rights, modify the connection rights, or train the network. Until the actual output of all samples is consistent with the desired output within a certain limit, it can be said that the neural network has been trained and is ready to accept new inputs that need to be solved.

Artificial neural network data flow diagram 2 BP algorithm and application examples The BP neural network is used to estimate the service life of lubricating oil. BP neural network is currently the most widely used neural network. Theoretically, it has been proved that a 3-layer BP network can approximate arbitrary mapping relations with arbitrary precision. The use of BP network to estimate the service life of lubricating oil can be divided into three phases. First, the study sample is established. The study sample is established by using intermittent sampling of oil and testing various key characteristics such as viscosity, flash point, moisture content, and insolubility. The proportion of objects, TBN, etc. are completed; the second is the network training phase. The training of the network model is based on each characteristic. The input value of the network is taken as the test result of each characteristic, and the output value is taken as the corresponding test result of each characteristic. Sampling time, the training of the network has been carried out until the convergence; the third is the network work stage, that is, the estimated service life stage, when using the trained BP network to solve the service life of the lubricating oil, the service life of the lubricating oil can be defined as the network. The minimum value of all output values ​​(limit use time).