In view of the special environmental conditions and power grid characteristics in Central Asia， a decision model considering technical performance， environmental adaptability and economy for the selection of reactive power compensation equipment was constructed. Firstly， taking the modernization project of No.2 thermal power plant in Almaty， Kazakhstan as a case study， starting from the key indicators such as response speed， harmonic characteristics and environmental adaptability， the technical performance differences of two mainstream static reactive power compensation schemes， thyristor-controlled reactor （TCR） superimposed fixed capacitor （FC） and magnetically controlled reactor （MCR）， were analyzed. Then， a multi-level fuzzy comprehensive evaluation method was used to establish an evaluation system including 3 first-level indicators and 12 second-level indicators. According to the evaluation results， TCR+FC scheme was selected as the reactive power compensation scheme of the system. Finally， in order to ensure the rationality of the scheme， the simulation was carried out using PSCAD simulation software. The simulation results show that the TCR+FC scheme exhibits fast response speed， high regulation accuracy and good stability.

A calculation model of a subcritical reheat back-pressure steam turbine was established using the fixed-flow calculation method for steam turbine thermal systems. The equivalent enthalpy drop method was adopted to predict the variation curves of exhaust steam temperatures in the high-pressure and intermediate-pressure cylinders under variable load rate conditions， and the variation laws of heating and power generation performance were studied. The results show that with the decrease in load rate， the exhaust steam temperature of the high-pressure cylinder decreases， while the exhaust steam temperature of the medium-pressure cylinder shows an upward trend. As the unit load rate decreases from 110% to 50%， the thermoelectric ratio of the thermal system increases from 521.71% to 783.11%， the standard coal consumption rate for power generation increases from 152.9 g/（kW·h） to 165.4 g/（kW·h）， with an increment of 8.18%， and the average exergy efficiency of the thermal system is about 7 percentage points lower than the average thermal efficiency of the whole plant.

Taking a million-kilowatt ultra-supercritical unit with double reheat as an example， several factors affecting the economic performance of the unit were summarized， and the technical optimizations that have been made for low load operation of the unit were introduced. The influence of the opening point selection of the supplementary steam valve on the economic performance of the unit was emphasized. Different opening points of the supplementary steam valves were selected， including the turbine maximum continuous rating（TMCR） condition， the turbine heat acceptance（THA） condition， the 95% THA condition， the 90% THA condition， and the 85% THA condition. The changes in parameters and thermal efficiency of the units with different load rates under each supplementary steam valve opening point design were compared. The comparison results show that when the opening point of the supplementary steam valve is lowered， the economic performance of the steam turbine under high-load operation decreases， while the economic performance improves under low-load operation. This design approach is applicable to deep peak regulation units and can effectively enhance the peak regulation operational economic performance of coal-fired power units.

A comprehensive combustion optimization experiment was conducted to address the issues of insufficient reheat steam temperature at low loads， large temperature deviation at high loads， and low thermal efficiency during the operation of a 600 MW four-corner tangential firing boiler in a thermal power plant. Preliminary tests were first carried out to clarify the boiler's operating characteristics. The air and pulverized coal parameters of the coal mill were optimized to satisfy the full-load output requirements via adjustments to the pulverizing system， including cold and hot primary air balancing， pulverized coal fineness regulation， and maximum output verification tests. Through combustion system tests， including adjustments of the air distribution and overfire air swing angle， causes analysis and treatment of coking， and optimization of the reducing atmosphere， the steam temperature condition has been effectively improved， coking problems have been alleviated， high-temperature corrosion risks have been reduced， and boiler efficiency has been improved. After optimization， the main and reheat steam temperatures of the boiler have significantly increased within the load range of 45% to 100% boiler maximum continuous rating （BMCR）， the steam temperature deviation is controllable， and the wall temperatures of the four tubes are reasonable. The operational safety and economy are greatly improved， providing an important reference for the operational optimization of similar boilers.

Against the backdrop of the "dual carbon" strategy and the construction of a new power system， coal-fired power units are confronted with multiple challenges such as frequent deep peak regulation， clean and efficient operation， and fluctuating coal quality， which seriously threaten the flexibility， safety， and economy of unit operation. Boiler combustion optimization and adjustment have become the key means to address these issues. The four core issues in the current combustion optimization and adjustment of coal-fired boilers are the control of the flame center height， the suppression of NO_x emissions， the optimization of exhaust gas temperature， and the regulation of thermal load deviation. By analyzing domestic test cases and theoretical research， and combining engineering practice experience， the influence mechanisms of key parameters such as air distribution mode， primary air pressure， coal fineness， overfire air damper opening， and burner adjustment methods on the combustion process were summarized， providing reference solutions for combustion optimization and adjustment of coal-fired power units and their safe and stable operation. Case studies show that for opposed firing boilers， the positive pyramid air distribution offers the best economy and safety， and despite slightly higher NO_x emissions， it is comprehensively recommended. For tangentially fired units， the three-step adjustment of primary air balancing， overfire air de-swirling， and reducing overfire air flow can effectively eliminate thermal load deviation and overtemperature risk.

To address the abnormal temperature rise of the third-stage extraction steam in a 660 MW ultra-supercritical turbine， operational data from a representative faulty unit were compared and the internal structure was inspected， confirming that high-temperature steam leaking into the third-stage extraction steam system was the primary cause. Temporary operation strategies were implemented through methods such as component verification， cooling steam regulation， and operation parameter control. Combined with the results of the shutdown maintenance， it was discovered that the sealing structure of the middle-pressure inlet steam pipe had severe deformation and seal failure， resulting in reheat steam leaking into the extraction steam system. Further measures such as inserting a new pipe， reconfiguring the sealing ring and grinding the cylinder body were taken to effectively eliminate the steam leakage channels. Results indicate that， after fault handling， the third-stage extraction steam temperature falls from 527 ℃ to 471.6 ℃ and the pressure stays at 1.9 MPa， restoring design values. The research results provide an engineering reference for fault diagnosis and maintenance of similar ultra-supercritical units.

In response to the repeated leakage rate exceeding the allowable limit of a 1 000 MW ultra-supercritical hydrogen-cooled generator， the root cause was revealed through multi-stage progressive diagnosis and treatment measures were implemented. Before the first start-up of the unit， the leakage rate in the generator wind pressure test reaches 45.71 m³/（24 h）. After helium leak detection and location， it is found that the flange seal of the external circulation fan has failed. After treatment， the leakage rate decreases to 6.68 m³/（24 h）， meeting the manufacturer's requirement of less than 10 m³/（24 h）. During the second start-up operation of the unit， the hydrogen leakage rate abnormally soars to 105.7 m³/（24 h）. Experimental results show a significant negative correlation between the leakage rate and hydrogen temperature. Through structural analysis and disassembly inspection， it is confirmed that insufficient pre-tightening force of the seal ring pedestal bolts is caused by out-of-tolerance insulation sleeves. Thermal cycling stress relaxation results in a large gap at the joint surface of the seal ring pedestal end cover， aggravating hydrogen leakage under low-temperature conditions. Finally， by adjusting the sleeve dimensions and optimizing the installation process， the leakage rate stabilizes at 6.15 m³/（24 h）.

Failures of auxiliary equipment in thermal power plants not only directly impact electricity output and quality， but in severe cases they can even lead to serious equipment damage and personal safety incidents. Therefore， condition monitoring and fault diagnosis of auxiliary equipment hold significant practical importance. To overcome the issues of delayed maintenance and resource wastage inherent in traditional planned maintenance models， a condition monitoring approach based on vibration analysis was adopted. Utilizing Emerson's AMS 2140 vibration acquisition and analysis equipment， real-time data collection and spectral analysis were performed on the primary air fan to precisely pinpoint the root cause of the malfunction. Through this comprehensive remediation effort， the safety hazard of abnormal vibration in the primary air fan has been successfully eliminated， providing effective support for predictive maintenance of auxiliary equipment.

To address issues such as long service life， component aging， and difficulties in procuring imported spare parts for the power supply system of the control rod drive mechanism （rod power supply system）of the second-generation plus nuclear power plants in China， a control and protection system renovation plan with microcomputer protection devices as the core was proposed. The microcomputer protection devices have replaced the numerous original discrete protection devices. By integrating the built-in programmable logic controller （PLC） function of the microcomputer protection devices， adding a parallel current monitoring and protection mechanism for the rotor circuit， simplifying the hardware configuration， and optimizing the generator loss-of-excitation protection， the observability and protection level of the rod power supply system have been significantly improved， achieving an increase in system reliability and a reduction in operation and maintenance costs. Simulation and actual transformation verification show that the fault response time of the new system is shortened to within 48 hours， the types of spare parts are reduced by 45%， and the average annual maintenance cost is decreased by 52%.

In order to address the serious threat to the safe operation of large ultra-supercritical thermal power units caused by the tripping of important auxiliary machines under low load， which can lead to significant load fluctuations or even the shutdown of the unit， an automatic control method for the tripping of important auxiliary machines under low-load conditions （hereinafter referred to as LAT） was proposed. Combined with a 660 MW ultra-supercritical unit in Xinjiang， the LAT control strategy， functional design， triggering conditions and control loop were discussed. It focused on the dynamic tests of the LAT of coal mills， the LAT of forced draft fans and induced draft fans， and the LAT of primary air fans. The relevant functional designs were optimized. Results show that all functions during the test process are executed according to the design， the main parameters of the unit change stably， and the test is qualified， meeting the operation requirements.

To improve the operational economy and flexibility of a dual-turbine regenerative system， a balancing small generator and a converter were installed in a million-kilowatt-scale double reheat unit with a backpressure extraction steam turbine （BEST） at a plant. The control process of this unit under various operating conditions of the converter was analyzed， and the override reduction function of the total steam flow command for the small turbine during load rejection and the action logic of the quick closing of the regulating valve within 0.6 s were verified. Additionally， a control strategy was developed for the converter's transition to "constant power" mode， whereby the generator power was automatically ramped down to below 2 MW at a rate of 5 MW/min， based on its actual output and the small turbine's total steam flow command. The analysis and optimization of the relevant control logic can provide a reference for improving the operational reliability and control strategies of similar units.

In order to address the challenges of high training complexity of raw data， susceptibility to feature distortion during demodulation， and the difficulty of balancing accuracy and efficiency in wind turbine generator system vibration diagnosis， an intelligent diagnosis method integrating convolutional neural network （CNN） with statistical analysis was proposed. First， the vibration time-domain waveform and frequency spectrum were directly used as inputs for CNN to train a diagnosis model， eliminating preprocessing and avoiding demodulation distortion. Second， the CNN architecture synchronously extracted time-frequency features， enabling rapid single-sample diagnosis. Finally， the status of each measuring point was quickly determined by the proportion of statistical results. Field tests on two wind farms show that the accuracy of the optimal model training set obtained is approximately 98.5%， and the accuracy of the test set is approximately 94.5%. The research results are of great significance for improving the operational reliability of wind turbine generator system， reducing operation and maintenance costs， and ensuring power generation benefits.

With the deepened market-oriented reform of the on-grid price for new energy and the introduction of time-of-use tariff and other policies， higher requirements have been set for the power generation capacity of new energy power stations at different grid load periods. The design of different structural photovoltaic brackets is of great significance for tapping the potential of time-of-use on-grid price benefits and optimizing the power demand at the output and load ends. To deeply reveal the output characteristics and overall power generation capacity of different types of brackets at different time periods， a comparative analysis was conducted on the overall and time-of-use power generation， power distribution in different quarters， and cost calculation based on the annual operation data of a photovoltaic station in a high-latitude area. This objectively verified the actual operation and economic performance of different types of brackets under the typical climate scenario. The results show that there are significant differences in the output power curves， time-of-use cumulative power generation， and amortized costs of different types of brackets. By combining the distribution patterns of local power grid load and typical climate characteristics， and through reasonable selection of bracket equipment， the output curve can be further optimized， the power generation capacity during peak periods can be enhanced， and the electricity revenue can be increased， which can effectively adapt to the trend of market-oriented reforms in new energy on-grid price.

To address the growing flexibility demands of future new power systems dominated by renewable energy， coupling a molten salt thermal energy storage system with conventional thermal power units is an effective solution. Based on a 660 MW ultra-supercritical once-reheat condensing unit， three system configurations integrating molten salt and ultra-high-temperature heat pumps were designed. A thermodynamic simulation model of the unit was established using Ebsilon software， and the efficiency of the coupled system was analyzed. The results show that the technical scheme of extracting high-pressure steam as the heat source for thermal energy storage， with the condensed water returned to the feedwater system， exhibits a significantly higher exergy efficiency compared to the electric heating scheme. The pressure of the heating steam is a critical parameter affecting the temperature difference during heat exchanging for thermal energy storage， and using high-pressure heating steam can improve the exergy efficiency. The adoption of ultra-high-temperature heat pumps can enhance the thermal energy quality of the latent heat section of heating steam， which is an effective solution to improve the exergy efficiency of the thermal energy storage process. Using high-temperature molten salt to generate high-pressure and high-temperature steam is an effective way to reduce the heat exchange temperature difference and improve the exergy efficiency during the thermal energy release process.