To address the challenges of feature selection and the instability of single models in wind power forecasting, a novel ultra-short-term wind power forecasting method that integrates multiple LSTM networks with the Q-learning algorithm (Q_L-L-C-A) is proposed. This method employs the Maximum Information Coefficient (MIC) to filter the features of wind power data and uses Variational Mode Decomposition (VMD) to decompose the wind farm power data into multiple frequency modes as additional features. The filtered and decomposed data are then used as inputs for the LSTM, CNN-LSTM, and Attention-LSTM network models to make predictions. Based on this, the Q-learning algorithm is applied to dynamically allocate weights to the prediction results of the three models to achieve a more optimized combined forecasting outcome. To validate the forecasting performance of the proposed Q_L-L-C-A model, actual measurement data from a wind farm is used as the model input, and comparative experiments are conducted with six other models. The experimental results indicate that the Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) of the Q_L-L-C-A model are superior to those of the LSTM, CNN-LSTM, and Attention-LSTM models, demonstrating that the Q_L-L-C-A model exhibits higher accuracy and stability in ultra-short-term wind power forecasting.

Aiming at the problem that the basic wolf pack algorithm (WPA) converges too slowly in robotic arm path planning and easily falls into local optimality, a robotic arm path planning algorithm based on the improved WPA algorithm is proposed. The Tent initialization method using chaotic mapping improves the uniformity and ergodic characteristics of the initial population; the Levi flight strategy and the adaptive wandering behavior of changing directions are added in the wandering stage to improve the global optimization capability; the summoning and siege behavior of the WPA algorithm are Fixed step size replaces adaptive step size. comparing the same three-dimensional scene, a robot arm path planning simulation experiment of the improved WPA algorithm before and after was performed. The improved WPA The algorithm running time and path length were reduced by 21.44% and 16.21% respectively, proving that the improved WPA algorithm has better performance in convergence speed and global search capabilities when applied to robotic arm path planning.

With the application of artificial intelligence technology in production and life, safety behavior detection technology has become a research hotspot in the field of construction. Aiming at the existing helmet wearing detection algorithms with low accuracy and poor model robustness and other problems, this paper proposes a helmet wearing detection model based on the combination of Coordinate Attention (CA) mechanism and YOLOv5. The optimized model in this paper significantly improves the accuracy and has good performance in small target detection. The accuracy of the algorithm combining the improved CA mechanism and YOLOv5 reaches 90.2%, which is 1.6 percentage points higher than the average accuracy of the YOLOv5 model, and the algorithm outperforms the other comparison models. After comparison experiments, it is verified that the model has better detection effect and higher network accuracy, and can still maintain stable performance and show good detection effect under complex construction scenarios.

The campus navigation system studied in this paper is mainly an online navigation system for teachers, students and office staff. The system implements functions such as positioning, route query, route navigation, and live-view mode. The service scope covers the headquarters of Xinjiang University, and expands the guidance and local search for teachers and students that traditional navigation software lacks. The system is based on the design and implementation of Gaode Map. The campus navigation client system can obtain user location information, quickly search for the specific location and path of the teaching building or dormitory building, and mark the route on the map to provide convenient navigation services.

To address the system uncertainties and bilateral drive synchronization issues of a stepper motor-driven XY platform, a self-disturbance rejection control strategy based on deviation self-coupling compensation is proposed to resolve the contradiction between rapidity and overshoot. Based on the established mathematical model of single-axis drive, a self-disturbance rejection + PD dual closed-loop control structure is adopted; subsequently, a deviation self-coupling PID compensation strategy is proposed for the synchronization issue of bilateral drive in one direction. Experimental results demonstrate that the control strategy proposed in this paper has short response time, good disturbance rejection, and high tracking accuracy.

D-galactose was used to establish a rat model of oxidative liver injury, and to study the mechanism of Longdan Xiegan Decoction (LXD) in the treatment of oxidative liver injury. Rats were killed after 30 days of continuous administration, and biological samples such as liver tissue and blood were collected as planned. Gas chromatography-mass spectrometry was used to detect the related markers in liver homogenate and blood, and metabonomic was used to search the database, to analyze the differential metabolites of LXD in the treatment of oxidative liver injury in rats, and to analyze the metabolic pathways of the detected markers, finally, the mechanism of LXD in the treatment of oxidative liver injury in rats was determined. The experimental results show that, the biomarkers associated with the treatment of oxidative liver injury in rats were citric acid, 2-Aminobenzoic acid, hydroxylamine, α-ketoglutaric acid, cholesterol, sarcosine, nicotinic acid, nicotinamide, Pyridoxal Phosphate, itaconic acid, methyl phosphate, pyruvic acid, tyrosine, glyceraldehyde 3-phosphate, cisconite acid and oxalic acid. The mechanism of LXD may be related to three metabolic pathways: Gluconeogenesis, citrate cycle (TCA), acetyl entering Mitochondria, malate L-Aspartic acid shuttle and alanine metabolism.

Polysaccharides of Polygonati Rhizoma was prepared by aqueous extraction and alcohol precipitation method, and the effect of seven macroporous adsorption resins on the decolorization of polysaccharides was investigated. The static adsorption was used to determine the optimal resin, and the static and dynamic one-factor adsorption tests were used to optimize the decolorization process of the resin. The results shown that the D301G resin had good decolorization efficiency for polysaccharides of Polygonati Rhizoma; the optimal process parameters for static adsorption were: resin dosage of 1:20 (g/mL), decolorization time of 6 h, initial pH=4, and adsorption temperature of 40 ℃, and the retention rate of polysaccharides under these conditions was 93.79 %, and the decolorization rate of polysaccharides was 68.54 %. The optimal process parameters for dynamic adsorption were: polysaccharide solution concentration of 8 mg/mL, elution flow rate of 3.0 BV/h, diameter to height ratio of 1:8 (cm), polysaccharide retention of 94.40% and polysaccharide decolorization of 78.73%. In conclusion, D301G macroporous adsorbent resin can be used for the decolorization and purification of polysaccharides from Polygonati Rhizoma under mild and environmentally friendly conditions, which is helpful for the further development of Polygonati Rhizoma as an industrialized product.

Due to the poor pretreatment effect of aquaculture sewage in the previous past, a large number of large particles remain in the sewage, thus affecting the final effect of sewage treatment. Therefore, the aquaculture sewage treatment method of carbon nanotube composite film was designed. According to the actual situation of aquaculture sewage, through the calculation of large particles in aquaculture sewage, using chemical softener to soften sewage treatment, reduce the hardness of sewage, sewage pickling, remove the bacteria in sewage, according to the characteristics of carbon nanotube composite membrane, the preparation, using the preparation of carbon nanotube composite membrane aquaculture sewage filter, and then further treatment of filtered sewage, thus design the corresponding aquaculture sewage treatment scheme. Through the above design, the design of the aquaculture sewage treatment method is completed. The experimental results show that compared with the traditional aquaculture sewage treatment method, the total cost of the designed carbon nanotube composite membrane technology in practical application is only 491,600 yuan, which is much lower than the other 1,035,400 yuan and 1,405,700 yuan of the two methods. In addition, the ammonia nitrogen content in the sewage treated by this technology is as low as 4.8 mg/L, showing a better sewage treatment effect.

Lithium iron phosphate battery and ternary lithium battery were selected as the research object, the life cycle was divided into five stages: raw material acquisition, manufacturing, assembly, use and scrap recycling, the materials and energy consumption required for each stage were listed, the LCA model was established in SimaPro, and then the life cycle assessment list was output according to the database, and the impact was classified and characterized by the CML2000 evaluation method, and then normalized and weighted to complete the comparison of the life cycle of each stage of lithium iron phosphate battery and ternary lithium battery. The data shows that the greatest environmental impact of lithium iron phosphate batteries is the use stage,the proportion is reached to 34%; The biggest environmental impact of ternary lithium batteries is the raw material acquisition stage,the proportion is reached to 75.3%; The comprehensive environmental impact of ternary lithium battery is 2.66 times that of lithium iron phosphate battery.And with the increase of battery capacity, the environmental impact of lithium iron phosphate battery will gradually exceed that of ternary lithium battery.

Aiming at the problems of slow convergence speed, difficult parameter selection and easy to fall into local optimal solutions of the slimy mushroom optimisation algorithm (SMA), a multi-strategy fusion improved slimy mushroom optimisation algorithm (LTSSMA) is proposed. The Latin hypercube is used to optimise the population initialisation; the position of the population particles is perturbed using the adaptive t-distribution variation to improve the convergence speed of the algorithm; the positive cosine optimisation strategy is introduced and the adaptive weight improvement is performed to reduce the probability of the algorithm falling into the local optimal solution. The improved algorithm is tested using the CEC2005 test function, and through the analysis of the obtained images and data, it is obtained that the improved mucilage optimisation algorithm has improved in convergence speed and the ability to get rid of the local optimal solution.

In order to study the influence of different operating parameters on the peristalsis characteristics of the bionic reactor, a three-dimensional finite element model of the bionic reactor was established based on the Static Structural module in Ansys software and the peristalsis mechanism. The peristalsis process of the reactor was simulated by numerical simulation method, and the influence of different amplitude-frequency and material parameters on the deformation characteristics and equivalent stress of the reactor wall was analyzed, and the validity and reliability of the model were verified. The results show that in a certain range, the deformation characteristics of the reactor wall become more obvious with the increase of amplitude and frequency parameters. With the increase of frequency, the peristaltic period is shortened. The effect of Poisson's ratio on the deformation of the reactor wall is not obvious, and the equivalent stress decreases with the increase of Poisson's ratio.

Based on the work that won the first prize at the national level in the 8th China Undergraduate Engineering Practice and Innovation Ability Competition, a new type of solar energy-powered new energy vehicle was designed. UG and CAD were used for modeling, and with the help of 3D printing, the structural processing and manufacturing were carried out. Solar panels were employed to efficiently convert solar energy into electrical energy. EDA software was utilized to design the circuit and generate the PCB circuit diagram, and then the circuit board was fabricated to drive the vehicle to move. In order to obtain a more precise motion trajectory, Matlab and UG software were used to optimize the design of the cam mechanism and conduct motion simulation on the vehicle's trajectory. Data and parameters where the running trajectory fitted the designed trajectory were obtained, thus successfully realizing the functions of the solar energy-powered new energy vehicle and achieving good results.

In order to study the influence of gas supply pressure and water supply pressure on spray particle size, a double fluid spray device is developed and tested for atomization characteristics. Two tests were set at 1m,1.5m to calculate the medium-diameter VMD to show the size and distribution, and the effect of the spray distance on the spray particle size was investigated. The results show that the size of the gas supply pressure increases the size of the spray decreases with the increase of the spray distance. And the atomization characteristic test conducted by the independently developed double fluid spray device has achieved good results

With the rapid development of centralized heating systems, there is a growing phenomenon of severe hydraulic imbalance in the pipeline network, uneven heating and cooling among users, and poor targeted regulation. On the premise of learning and absorbing advanced heating technology from Finland, introduce building heat exchange units for smart heating in residential areas. Starting from the aspects of water, electricity, and heat, this study compares and analyzes the energy-saving effects of building unit heating and conventional heat exchange station heating, and investigates the application of building unit heating in centralized heating systems. Research has shown that using building units for heating, the heating heat consumption of underfloor heating users is 0.27-0.29 GJ/㎡, with an electricity consumption of about 0.52-0.6 kWh/㎡. The heating heat consumption of heat dissipation fin users is 0.33 GJ/㎡, with an electricity consumption of 0.56 kWh/㎡. Compared with conventional heat exchange stations, it can achieve a heat saving of about 20% and a power saving of about 27%, with great energy-saving potential.