Fault detection and diagnosis can improve safety and reliability of gas turbines. Current studies on gas turbine fault detection and diagnosis mainly focus on the case of abundant fault samples. However, fault data are rare or even unavailable for gas turbines, especially newly-run gas turbines. Aiming to realize fault detection with only normal data, this paper proposes the concept of normal pattern group. A group of long-short term memory (LSTM) networks are first used for characterizing the mapping relationships among measurable parameters of healthy three-shaft gas turbines. Experiments show that the proposed method can detect all 13 common gas path faults of three-shaft gas turbines sensitively while remaining low false alarm rate. Comparison experiment with single normal pattern model verifies the necessaries and superiorities of using normal pattern group. Meanwhile, comparison between LSTM network and other methods including support vector regression, single-layer feedforward neural network, extreme learning machine and Elman recurrent neural network verifies the superiorities of LSTM network in fault detection. Furthermore, comparison experiment with four common one-class classifiers further verifies the superiorities of the proposed method. This also indicates the superiorities of data-driven methods and gas turbine principle fusion to some extent.

An integrated energy system interconnects multiple energies and presents a potential for economics improvement and energy sustainability, which has attracted extensive attention. However, due to the obvious volatility of energy demands, most existing integrated energy systems cannot operate in a totally self-sufficient way but interact with the upper grid frequently. With the increasingly urgent demand for energy saving and emissions reduction, renewable resources have occupied a larger and larger proportion in energy system, and at last they may be dominant in the future. Unlike conventional fossil fuel generation, the renewable resources are less controllable and flexible. To ease the pressure and guarantee the upper grid security, a more independent integrated energy system is required. Driven by that, this paper firstly reviews the optimal strategies considering both independence and benefit from perspectives of individual efforts and union efforts. Firstly, the general optimization process is summarized in terms of energy flows modelling and optimization methods to coordinate supply–demand side and realize benefit maximization. Based on that, handling with uncertainty of high-ratio renewable energy is reviewed from uncertainty modeling methods and multi-stage operation strategy perspectives to make the strategy accurate and reduce the adverse effects on the upper grid. Then, the hybrid timescale characteristics of different energy flows are explored to enhance operation flexibility of integrated energy systems. At last, the coordination among different participants is reviewed to reduce the whole adverse effect as a union. Remarks are conducted in the end of each part and further concluded in the final part. Overall, this study summarizes the research directions in operation optimization of integrated energy systems to cater for a renewable energy dominated scene to inspire the latter research.

Gas turbine hot component failures often cause catastrophic consequences. Fault detection can improve the availability and economy of hot components. The exhaust gas temperature (EGT) profile is usually used to monitor the performance of the hot components. The EGT profile is uniform when the hot component is healthy, whereas hot component faults lead to large temperature differences between different EGT values. The EGT profile swirl under different operating and ambient conditions also cause temperature differences. Therefore, the influence of EGT profile swirl on EGT values must be eliminated. To improve the detection sensitivity, this paper develops a fault detection method for hot components based on a convolutional neural network (CNN). This paper demonstrates that a CNN can extract the information between adjacent EGT values and consider the impact of the EGT profile swirl. This paper reveals, in principle, that a CNN is a viable solution for dealing with fault detection for hot components. Based on the distribution characteristics of EGT thermocouples, the circular padding method is developed in the CNN. The sensitivity of the developed method is verified by real-world data. Moreover, the developed method is visualized in detail. The visualization results reveal that the CNN effectively considers the influence of the EGT profile swirl.