Optimal Sensor Placement And Simulation Verification In Complex Nuclear Power Second-loop System Based On Signed Directed Graph

Under the background of the "dual carbon" strategy,nuclear power will occupy an important position in my country’s future energy structure.Ensuring the safe and stable operation of nuclear power units is the key to realizing the application and development of nuclear power,and rationally optimizing the arrangement of the sensors of the unit to realize the global fault monitoring of the unit is an important prerequisite for ensuring the safe and stable operation of the unit.The sensor arrangement mainly includes two aspects: the number of sensors and the installation location.The traditional sensor arrangement is mainly based on the qualitative requirements in the relevant standards,lacking quantitative analysis,it is difficult to ensure the diagnosability of the sensor to the fault,and the cost is high.Signed Directed Graph(SDG)can accurately describe the causal relationship between complex system faults and parameters,and then realize the causal relationship between fault types and optimal sensor placement and measurement points in the complex system of the secondary circuit of nuclear power.The advantages of simple modeling and flexible reasoning have attracted extensive attention in recent years.Based on SDG theory,this paper studies the optimal arrangement strategy of sensors for the complex system of the nuclear power secondary circuit.The main research contents and results are as follows:(1)Build the SDG model of the regenerative system and the fault-measurement point correlation matrix.Establishing an accurate SDG model is the basis and starting point of realizing the optimal arrangement of sensors on the basis of satisfying the diagnosability of system faults.A comprehensive modeling strategy for SDG model is proposed,the SDG model of the regenerative system is constructed,and the "fault-measuring point" correlation matrix of the regenerative system is established by analyzing the fault propagation path and the correlation between the fault and the measuring point..In order to make the model more accurate and the optimization results more reasonable,the actual fault detection ability of the sensor is considered in this paper,that is,the fault detection performance of the sensor is quantitatively described in the form of probability,and it is added to the fault-measurement point correlation matrix.As a source of parameters for subsequent optimization solutions.Finally,the accuracy of the SDG model is verified by fault simulation and SDG reasoning.(2)Build and solve the multi-objective optimization mathematical model of the sensor arrangement of the unit.Several optimization criteria and evaluation indexes of sensor arrangement are defined as objective functions and constraints,and a mathematical model of sensor arrangement considering fault diagnosis performance is established,thereby transforming the sensor optimal arrangement problem into a multi-objective optimization problem.The traditional optimization algorithm is prone to premature convergence,so the standard NSGA-II algorithm is improved,and the test function is used to test its performance.The results show that the improved algorithm outperforms the standard NSGA-II algorithm in all performance indicators.In order to facilitate decision-makers to select the appropriate layout scheme in the Pareto solution set according to different actual needs,a decision analysis model is also established by combining the combined weighting method and the TOPSIS method to realize the evaluation and ranking of the Pareto optimal solution.Taking the elements in the improved fault-measuring point correlation matrix as the solving parameters,the optimal layout scheme is solved for the sensor layout mathematical model,and it is compared with the traditional layout scheme.The results show that,on the basis of satisfying fault diagnosability,the optimized sensor layout scheme is better than the traditional layout scheme in terms of layout cost,fault detection unreliability and fault-measurement point correlation.(3)Build the regenerative system simulation model and verify the sensor arrangement.Based on the APROS integrated simulation platform,a high-precision simulation model is built for the secondary circuit regenerative system of the nuclear power unit.The steady-state check is carried out for 5 working conditions including 100%,90%,80%,70% and 60%,and the calculation results verify the high steady-state accuracy of the established model.In order to check the actual fault diagnosis ability of the optimized arrangement scheme,simulation verification was carried out for 7 typical faults,and the fault sources deduced by the reverse inference of the SDG model were consistent with the actual faults,indicating that the optimized sensor arrangement scheme has no effect on each fault.All kinds of faults can be effectively diagnosed,which verifies that the optimal layout scheme in this paper has good fault diagnosability and feasibility.