SFSO: Surrogate-Based Fluid Supply Optimization Method for Improving Efficiency and Pressure Stability in Longwall Hydraulic Systems

In longwall mining, conventional centralized fluid-supply systems generally operate at a fixed flow rate and therefore cannot respond effectively to the highly time-varying and overlapping demands of multiple hydraulic supports operating in coordination with an automated shearer. This mismatch can cause excessive relief flow, insufficient supply, and substantial pressure fluctuations. To address this problem, this study proposes a surrogate-based fluid supply optimization (SFSO) method in which a continuous supply trajectory is parameterized by a limited number of time–flow control points and reconstructed using a radial basis function model. The control-point parameters are optimized subject to constraints on pump capacity and flow-rate variation, while pressure stability and supply efficiency are considered simultaneously. An AMESim hydraulic system model was validated against measurements obtained from a laboratory hydraulic support platform and was subsequently used to evaluate the proposed method under a predefined multi-support operating sequence. Under 200 independent test scenarios with demand perturbations of up to ±10%, SFSO increased the mean cycle-based volumetric efficiency from 30.16% to 60.10%, representing an increase of 29.94 percentage points. Meanwhile, the mean pressure fluctuation metric JP decreased from 0.625 MPa2 to 0.150 MPa2. Under the model, operating sequence, and demand perturbation range considered in this study, these results indicate that the proposed trajectory-surrogate formulation can improve demand supply matching and reduce the pressure fluctuation index relative to the constant-flow reference. Further work is required to evaluate the method under broader sources of uncertainty and in underground field applications.