Aiming at the challenge of balancing the optimization between coverage and energy consumption of wireless sensor networks in the railroad monitoring scenario of train attitude measurement system, an adaptive backbone particle swarm based algorithm was proposed. A three-dimensional optimisation model for railway line monitoring was established. By comprehensively considering regional coverage, total system energy consumption, and coverage requirements for critical monitoring points, a multi-objective optimisation model more closely aligned with engineering practice was constructed. A dynamic inertia weight and backbone particle number adjustment mechanism were incorporated to balance global exploration and local exploitation capabilities. An ε-elite dimension-wise backpropagation learning strategy was employed to enhance the probability of escaping local optima. Simulation results demonstrate that compared to traditional PSO, APSO, and BPSO algorithms, ABPSO achieves higher network coverage and superior energy balance with faster convergence and lower energy consumption. In complex scenarios, it effectively safeguards monitoring quality in critical areas, providing an efficient and reliable solution for WSN deployment along railway lines.