A back analysis method based on multioutput least-squares support vector regression machine (MLSSVR) and improved differential evolution algorithm (IDE) is proposed to estimate multiple mechanical parameters of concrete dams. Based on the uniform design method, representative combinations of mechanical parameters are generated. Using these combinations, calculated hydrostatic displacement component differences are obtained through the finite-element method (FEM). The calculated hydrostatic component differences and mechanical parameters are then used to train MLSSVR models, with the model parameters selected by IDE. This allows for establishing the mapping relationship between dam displacements and corresponding mechanical parameters. The hydrostatic component differences separated from prototype observed data are then substituted into the model to obtain back analyzed mechanical parameters. If predefined terminating conditions are not satisfied, new combinations are added to the training set, and the training process continues until satisfied. The proposed back analysis method was successfully implemented to the Jinping I arch dam. Results indicate that the proposed method has high precision and strong generation ability.

Underwater cracks in concrete structures are often difficult to detect due to their complexity of the service environment. With numerical and experimental analysis of concrete beams immersed in water, an active monitoring system, based on a cement-based piezoelectric intelligent module array (CPIMA), was developed to locate and quantify the underwater cracks. Time reversal (TR) of the stress wave field is accomplished to focus on the crack area through the concrete beam specimen by the system. First, a piezoelectric actuator is applied to emit the initial propagating wave, which can be reflected, attenuated, and diffracted by the crack, transmitted through water filled in the crack, as well as diffracted by the coarse aggregates. To extract the damage waveforms associated with the crack and analyze the robust time-reversal invariance under the high-order multiple scattering effect, a pair of homogeneous and heterogeneous forward finite element (FE) models is established. Then, the damage waveforms are time-reversed and re-propagated in the inverse numerical model, where an optimal refocusing is achieved on the crack that behaves as an acoustic source. Finally, the damage area is obtained in the form of the stacked energy distribution of each time step. The focus results are represented by cloud images and compared with root-mean-square deviation (RMSD) values. Numerical simulation and experiments show that this method can identify and quantify underwater cracks effectively.

An optimization model for the complementary operation of a photovoltaic-wind-pumped storage system is built to make full use of solar and wind energy. Apart from ensuring the maximum economic benefit which is normally used as the only objective, the stable objectives of minimizing the output fluctuation and variation of load and output difference are added to form the multiobjective problems because of lack of study on access capacity of photovoltaic and wind power. The model aims to increase the power benefit and reduce the output fluctuation and variation of load and output difference under the constraints of station, output balance, and transmission limitation. In a case study, four schemes including single-objective independent operation, single-objective complementary operation, and multiobjective complementary operation are compared to discuss the effect of pumped storage station on economic objective and stable objectives. Furthermore, the opposite trend of the two objectives is proved and a compromise optimal solution is given. The results indicate that the pumped storage station can effectively increase power benefit and access capacity of photovoltaic and wind power. The study can provide references to the complementary optimization of the pumped storage station and the intermittent renewable energy.

Plastic optical fibers (POFs) have many advantages compared with conventional optical fibers for sensing and monitoring applications in civil engineering. The monitoring capability of POF sensors depends on the strain transfer characteristics between POFs and host materials to a great extent. Theoretical analysis is performed to develop the strain transfer relationship between surface-bonded POFs and concrete. Besides, finite-element method is conducted for verification, and the main influencing parameters are studied respectively. The calculated results demonstrate that the effect of strain transfer is more significant when the POF is bonded close to the surface of concrete and the length of the adhesive layer ought to exceed 300 mm simultaneously, and the width, thickness, and shape of the adhesive layer have slight impact on the strain transfer.