Study On Inversion Method Of Resistivity Method In Geological Exploration

Induced polarization is the electrochemical process of charging and discharging current when the different rocks and ores are put in the external artificial produced field. The charging and discharging rate of the additional field is mainly dependent on the thickness of the rock particles. The more coarse ore particles are, the slower the discharge rate is. This physical phenomenon is the precondition of the advance detection of geological anomaly by the method of induced polarization. This topic is based on the theory of induced polarization method to detect the resistivity distribution of different underground abnormal bodies and complete the imaging of nonlinear joint inversion in geological exploration. It has important practical significance to the safety of tunnel construction.Geophysical inversion is a nonlinear problem, and the linear inversion algorithm used usually is more dependent on the initial model, so the main trend of geophysical inversion is the joint inversion. This paper mainly introduces the framework and steps of the combined algorithm by genetic algorithm and BP neural algorithm, ant colony algorithm and BP neural algorithm. As the BP neural network algorithm is easy to fall into local optimum, so this paper uses genetic algorithm and ant colony algorithm to optimize the initial weight and threshold matrix of the BP network, then the BP network uses these parameters to train and learn. The inversion speed of geological prospecting has been improved. In this paper, the simulated experiment platform of BEAM system is built in the laboratory environment. The system is mainly composed of an excitation power supply, a ring electrode, a data acquisition card PMD-1608 and a computer. Data acquisition is completed by the virtual collection platform compiled by LabVIEW. The data is transmitted to the computer to save and process for inversion through the USB port. Through MATLAB program, the three typical geological models of air, stone and water are calculated by the algorithm. It gets the parameters such as the shape and size of the geological anomaly, the distribution of the position and the resistivity value. It is proved that the two kinds of nonlinear combined algorithms are feasible in the geological advanced exploration.In order to highlight the advantages of the nonlinear algorithm, the same data is used by BP algorithm to inverse. Compared the inversion results, the conclusion can be obtained: Although the combined algorithm costs a lot time when it optimizes the initial weight and threshold value of the BP network, however, the training time of BP network in the late learning is greatly reduced, so the joint inversion algorithm is more fast and more accurate than the BP inversion to determine the properties of the geological anomaly. The algorithm can achieve the goal of advance warning of tunnel engineering.