The Research Of Fault Diagnosis System Of Biological Oxidation Pretreatment In Oxidation Tank Temperature Control

Temperature is one of the important factors affecting the survival of bacteria in the process of biological oxidation.The speed and activity of the bacteria breeding can directly affect the oxidation of ore pulp in the pretreatment process,thus affecting the extraction rate,so it is very important of oxidation tank can maintain constant temperature.Once the system actuator or sensor failure caused by the temperature change in the oxidation tank,will lead to bacterial inactivation and even death,resulting in the loss of irreparable losses.Therefore,it is important to make fault diagnosis for the temperature control system in the oxidation tank.Based on the field production process,the oxidation tank temperature control system fault diagnosis method is studied.The main research work as follows:For discovering failure by manual operation in the oxidation tank at present,a fault diagnosis model based on BP neural network is proposed.The change of heat oxidation tank to do theoretical analysis,model of oxidation tank heat fault mechanism,then the model analysis,find the cause of temperature change and the reasons for the formation of the corresponding fault detection,determine the input output model and basic structure,laid the foundation for the improvement of the late model.In order to optimize the parameters of the fault diagnosis model,the learning strategy and the harmony search algorithm are used to optimize the parameters of BP neural network.Making use of the global optimization of LHS to make up the disadvantage that BP neural network is easy to fall into local optimum,the optimal parameter LHS-BP neural network fault diagnosis model is established.Compared with the basic HS algorithm,on the one hand,the changes in parameter settings become dynamic,used to increase the diversity of solutions and improve the convergence speed of the algorithm,on the other hand,the optimization of teaching and learning combined with two heuristic algorithm and harmony search algorithm,for location update to improve the calculation method of the ability to jump out of the local optimum.The test results show that the algorithm has better global optimization performance.In order to validate the feasibility and validity of the methods,build the oxidation tank simulation platform using the experimental equipment,to obtain the sample data and test data,the model parameters were determined by training the network on thesample data,the test sample test results into the model.The results show that the method improves the accuracy of fault diagnosis,and it has a certain guiding significance in theory and practice.