Aiming at the problem of low prediction accuracy of single power prediction model due to the impact of photovoltaic power fluctuation, a combined photovoltaic power prediction model based on similar day clustering is proposed. Firstly, k-means clustering is selected to divide the original power data into three similar day sample sets of sunny, rainy and cloudy according to different weather types, and the variational mode decomposition (VMD) is used to decompose the similar day samples; Secondly, the convolution neural network is used to optimize the support vector machine (CNN-SVM) and bidirectional short-term and short-term memory (BiLSTM) neural network, respectively, to predict and superimpose the decomposed power data and combine the prediction results with weights, and the grid search algorithm (GS) is used to find the optimal combination weight to improve the performance of the combination prediction model. Finally, the validity of the PV power prediction model proposed in this paper is verified by the one-year measured data of a photovoltaic power station in Australia. The experimental results show that the model proposed in this paper can predict the photovoltaic power well and has strong adaptability no matter what weather type.

Test for large-size wind turbine rotor under static and rotating conditions have been carried out for rotor dynamic characteristis. Based on experimental modal and operational deflection shape, firstly this test obtained modal data using LMS TEST.Lab system by calculating FRFs and curve fitting. Secondly the resonate speed was obtained according to the vibration sweep test. Then modal parameters were analyzed through operational deflection shape test under the resonate speed. The results show that the resonate frequency under rotating condition is different from the resonate frequency under the static condition. And the modal shape of rotor under rotating condition is travelling wave while it is standing wave under static condition. In conclusion, the integral stiffness is decreased when the rotor is rotating which results in the decrease of the resonate frequency, the modal shape turns to be the travelling wave resulting from the rotating magnetic force. This research results can offer a reference and guidance for rotor dynamics evaluation, simulation model modification and parameter input, and optimization design of rotor structure .

A Printed Circuit Heat Exchanger (PCHE), with straight channels and semi-circle cross section, was fabricated and experimental studies on heat transfer and fluid flow were conducted, during which the flow regime was transition flow, water was working fluid, and flow rate of water was various. The results obtained from correlations of macro circular tubes had obvious deviations from the experimental results. Specifically, the f factor obtained from experiments are larger, and the changes of the overall heat transfer coefficient were more complex with various Reynolds number. The heat transfer and flow correlations in transition zone of PCHE was calibrated within corresponding application ranges. In order to obtain the heat transfer correlations, a numerical method was introduced to obtain one-sided average convective heat transfer coefficients under transition flow. The results showed that the average deviations of the overall heat transfer coefficient obtained from average convective heat transfer coefficients was 8.5% comparing to experimental results, while the maximum deviation reached 17.2%. However, in spite of that, a correlation to predicted the overall heat transfer coefficients through average convective ones still can be obtained, and the deviations comparing with experimental results was within 10%. It is recommended that obtaining one-sided average convective heat transfer coefficient with numerical method is feasible especially when it was transition flow in PCHE.

To solve the problems of uneven distribution of NO_x concentration at the outlet and excessive ammonia escape in the actual SCR operation of power plants, the SCR DeNO_x reactor was simulated numerically and a new ammonia injection optimization method for dual control of ammonia slip and nitrogen oxides emission was proposed. Taking the SCR DeNO_x system of a 660 MW unit as an example, a numerical model of visual flow field and DeNO_x reaction was established, and the change of the partitioned outlet NO_x concentration, ammonia slip and partitioned DeNO_x efficiency under different total ammonia nitrogen ratios were analyzed and compared. The results show that the relationship between partitioned DeNO_x efficiency, outlet NO_x and ammonia concentration, with the partitioned ammonia-nitrogen ratio is a two-parts linear relation separated by an inflection point, and the inflection point occurs when partitioned ammonia-nitrogen ratio is about 1.15. On this basis, the present study proposes a new optimization method for ammonia/nitrogen dual control optimization based on the combination of piecewise fitting function and optimization matrix equation. It is predicted that the 660 MW unit by using 5 zones and 42 nozzles for ammonia injection regulation, and the optimized total ammonia injection volume will decrease by 7.2% at the best compared with the uniform ammonia injection working condition. Relative standard deviation of outlet NO_x concentration and ammonia slip volume will decrease to 9.4% and 4.2% respectively, and the outlet uniformity will be significantly improved, and there is no locally over DeNO_x phenomenon.

During the operation of SCR flue gas denitrification system in coal-fired units, ammonium bisulfate (ABS) in flue gas causes ash scale slabbing at the cold end of the air preheater and increases the difficulty of purging and cleaning ash. To this end, ABS premixed ash samples were prepared and pressed and heated at different temperatures, and a new test method was designed to compare the changes in compressive strength of the samples and explore the influence law of ABS on the mechanical strength of ash scale. The experimental results showed that: 1) ABS premixed ash samples underwent physical agglomeration and chemical reaction during the heating of slabbing at 147-220 ℃, and the compressive strength was increased by about 95.50% at maximum, among which physical agglomeration played a dominant role with about 88%-89% influence and the influence of chemical reaction accounted for about 10%-12%; 2) ABS slabbed ash samples under heating at 220-300 ℃, ABS vaporization precipitation rate reached up to 96.43%, the ash sample from the slab state to loose, compressive strength from 195.50% of the blank sample to 110.17%. It is proved that the means of high temperature heating is feasible to reduce the ABS content in the blockage and create conditions for improving blowing and cleaning from the perspective of ash scale.

The function of query and statistics of data is always the basis of production management and decision-making businesses. According to the results of information construction and applications for many years, the mode, function and performance of data query and statistics still do not satisfy the actual needs. A component for data query and statistics based on supervisory graph software is proposed and developed. Through the design of data path architecture, the optimization of query and statistics mechanism, the developed component has achieved personalized management, lightweight analysis and flexible query of real-time data. The query and statistic component and data link mode developed in this study have been tested and applied in practice. The results show that the function of data query and statistics achieves loose coupling with database. The component can support both the operation modes of C/S and B/S, and provide unified query and statistics of data through supervisory graph from multi-data sources and multi-level organizations. The efficiency of data query and trend query reaches or approaches the level of database native management tools, and is better than that of conventional supervisory information system (SIS) websites. The integration mechanism of data components based on graph with SIS data query and report system is proposed, which helps to reduce the repetitive cost of data use.

The non-minimum phase plants with unstable zeros exists widely in the process of power production. Because of the non-minimum phase characteristics, the control system should ensure internal stability while completing output tracking, and improve response speed while overcoming the undershoot. The general PID control cannot meet the requirements of engineering applications. An engineering control and tuning method for non-minimum phase plants is proposed in this paper. Firstly, a robust PID controller is designed to ensure the stability of the closed-loop control system and overcome the under shoot of the system. Secondly, design a second-order filter that includes system position error, velocity error, and acceleration error to improve the response speed and dynamic performance of the control system. This method is simple, easy to tune, easy to configure in DCS, and has strong robustness to model uncertainty, which is worth promoting in engineering.

Combustion monitoring in large industrial furnace can be simplified to a radiation heat transfer problem within the enclosed cavity system, and precise quantification of its boundary radiation characteristic is the basis to carry out follow-up studyon the radiation inverse problem, but the coupled problem of wall radiation and media radiation need to be solved. A Monte Carlo priciple was involved to solve the radiation heat transfer equation in the enclosed cavity, and to decouple the shares of wall radiation and media radiation in the boundary detection information. The influence of temperature distribution and radiation properties on the share of wall radiation were discussed, at last the experiment verifies the feasibility of using the radiation information of boundary detection to retrieve the wall source term. This study will provide a reference to the exploration of physical field detection method of wall surface in industrial furnace.

During frequent long-term standby state of gas turbine generator unit, the surface of 08Al carbon steel of waste heat boiler economizer fin tube will have serious corrosion problems. In this paper, the macroscopic corrosion phenomenon and corrosion rate of 08Al carbon steel above the critical humidity were studied by the method of hanging piece and electrical resistance probe. The results show that the corrosion rate of 08Al carbon steel is the fastest in the first 5 days under the constant environment of 20 ℃ and relative humidity of 70%, and the corrosion depth reaches up to 0.85 μm, accounting for 47.7% of the total change in the whole process. However, the most obvious corrosion phenomenon, including the change of weight and the surface corrosion area, occurred from about the 19th to the 25th day. With the relative humidity gradually increasing from the critical humidity of 70% (ambient temperature 20 ℃), the corrosion evaluation indexes of 08Al carbon steel show a linear upward trend. Nitrogen filling maintenance strategy under long-term standby state was formulated, effectively alleviating the corrosion condition of economizer fin tube of waste heat boiler.

This paper proposes a multi-parameter collaborative monitoring system and method for degassed hydrogen conductivity based on the combination of double water membrane degassed method and electric regeneration ion exchange technology, which can quickly measure a number of key water quality indicators such as conductivity, hydrogen conductivity, degassed hydrogen conductivity and pH. It can make comprehensive evaluation of water quality, guide thermal equipment shutdown and startup. And it can adjust normal operating water conditions. It has a great significance for ensuring water vapor quality and thermal equipment corrosion, salt accumulation and scale formation. Double water membrane degassing technology is used to measure degassed hydrogen conductivity, which can effectively remove CO₂ from water. On this basis, a method of calibration degassed hydrogen conductivity measured by standard solution is proposed, so that the accuracy of the measured value can be effectively evaluated. This method has been applied to typical gas combined cycle units and heating units for on-site supervision. It proves that the method has high measurement accuracy, can quickly and comprehensively evaluate water vapor quality, and it can assist in solving various technical problems on site. Therefore, it has strong popularization and application value.