This paper presents an energy-saving strategy that was applied to a parallel variable frequency pump system of a water circulation pumping station. Firstly, the mathematical model of shaft power consumption for the parallel pump system was established using quadratic polynomial fitting, with some constraints configured according to the system’s water supply demands. Then, the algorithm program was designed with the goal of minimizing the energy consumption through the application of an improved differential evolution algorithm. Additionally, the energy consumption model and constraints were integrated and simplified in order to adapt to the algorithm calculation. In the end, the algorithm was implemented according to the pump design parameters and supply targets of the pumping station. Meanwhile, a comparison was done between the differential evolution (DE) algorithm and the genetic algorithm (GA). Furthermore, an experimental test was conducted in an aluminum company in order to verify the applicability of the energy-saving algorithm in practice. The results demonstrated the feasibility of using the improved differential evolution algorithm in order to achieve a minimum energy consumption operation strategy; the results consequently manifested the superiority of the differential evolution algorithm in both computing time and optimal solution aspects.

At the Second Water Intake Pump Station of the Chenhang Reservoir in Shanghai, suboptimal pump scheduling resulted in electricity consumption cost attributable to pump-motor equipment accounting for an exceptionally large proportion of the total power expenditure. In response to the economical operation issues, a mathematical model of power consumption cost for the pump station was established by introducing time-of-use electricity pricing and constraint suppression terms. Taking the minimum cost as the research objective, the differential evolution (DE) algorithm was employed to establish a fitness function for electricity cost, aiming to find the most economical and reliable scheduling strategy. However, owing to its low computational speed and high complexity, machine learning was introduced to establish neural network surrogate models of the DE algorithm. By comparing three surrogate models, the Multilayer Perceptron (MLP) neural network model was adopted as the most appropriate surrogate model. It was optimized for robustness improvement and verified on site. The results demonstrate that implementing the surrogate model achieves over 25% savings in electricity cost per thousand cubic meters of water, while slashing the solution time by 88.53% compared to the standard DE algorithm. Furthermore, the overall power consumption is reduced by 2.20% under a cost-priority strategy and by 15.89% under a power-priority strategy, thereby directly mitigating the carbon footprint of the pump station. The proposed hybrid computational framework in this study bridges the gap between the computationally expensive heuristic optimization and the strict real-time control requirements in engineering, highlighting its significant contribution to the sustainable and low-carbon operation of water infrastructure.

A pump selected at one duty point may perform poorly over a full year because river level, pipe resistance, demand, and the number of running units all change the operating point. This paper compares discrete retrofit options for a water-intake pumping station using a documented steady-state hydraulic model, pump performance surfaces, and a representative hydrological year (RHY). The analysis distinguishes field observations from engineering calculations, reconstructed operation, and economic scenarios. Two pre-retrofit DCS snapshots give back-calculated pump hydraulic efficiencies of 73% and 65%. These values cannot be compared directly with the 87.8% factory efficiency because the operating conditions and test boundaries differ. The RHY comprises 35,136 modelled 15 min intervals and is not presented as measured SCADA. Using the engineering-record estimate of a 40 kW reduction per active pump, annual electricity use falls from 3232.2 to 2398.7 MWh, corresponding to 833.6 MWh (25.8%) of modelled savings. A 30–50 kW sensitivity range gives savings of 625.3–1042.1 MWh (20.7–30.3%). The redesigned pump has a reconstructed mean hydraulic efficiency of 86.72%, and 93.3% of delivered volume is conveyed at efficiencies above 86%. In the anonymous life-cycle screen, the retained-interface redesign has the lowest central cost at 62.36 RMB/t CO₂, while deeper replacement becomes competitive only at higher shadow carbon prices. The findings show the value of matching a retrofit to the annual operating spectrum, although full-year field data are still needed to verify the predicted savings.