Research And Development Of Energy Management System For Lentinula Edodes Logs Factory Production

This article addresses issues of low degree of information management,scattered data,difficult statistical analysis,and lack of energy consumption prediction methods in the factory production of Lentinula Edodes Logs.Based on the actual needs of energy management in the factory production of Lentinula Edodes Logs,a prediction model of electrical energy consumption for refrigeration process was constructed,and an energy management system for the factory production of Lentinula Edodes Logs in B/S mode was designed and implemented.The main research contents and results are as follows:(1)The classification of energy-using equipment in the production process of mushroom logs,the collection and management of equipment energy consumption data,and the design and implementation of energy consumption data retrieval services were completed.Based on the analysis of actual equipment and energy-related data characteristics of the factory,combined with the company’s administrative structure information,and based on the enterprise’s energy environment data unified platform and big data resource platform,the coding of various energyusing equipment and their data items in the production process of mushroom logs was completed,and the energy consumption data structure was unified.Three data acquisition APIs were completed,which realized the efficient management of factory energy-using equipment and the low-cost acquisition of energy consumption data.(2)A prediction model of electrical energy consumption for factory production of Lentinula Edodes Logs was constructed.By using the data retrieval service to obtain the electricity consumption data in the strong cold room of the refrigeration process in the mushroom log production factory,after pre-processing such as abnormal value processing,resampling,and normalization,the DABI-LSTM multi-variable multi-step energy consumption prediction neural network model based on bidirectional long short-term memory network and two-stage attention mechanism was inputted.The model is based on Encoder-Decoder LSTM,using the encoder-decoder structure,the encoder using Bi-LSTM,the decoder using LSTM,and the attention mechanism is introduced before the encoder and decoder to allocate weights to different sequences to optimize input data.After the model is trained,when the time window in the validation set is 40,the average absolute error(MAE)of the predicted results within ten steps is 0.142,indicating that the model has good fitting performance and achieves accurate prediction of electrical energy consumption data in the strong cold room of the refrigeration process.(3)An energy management system for factory production of Lentinula Edodes Logs was designed and implemented.Based on field research of the factory production process of Lentinula Edodes Logs,data characteristics,target users,functional requirements,and nonfunctional requirements were analyzed to complete the overall architecture,functional modules,and database design of the system.The system consists of 8 modules and 39 user pages,including user and permission management module,basic data management module,energy management module,real-time monitoring module,statistical analysis module,energy consumption prediction module,alarm module,and report management module.Based on the Spring Boot and Vue frameworks,using the development method of front-end and back-end separation,using the object-oriented software design method,and following the development principles of modularity,componentization,low coupling,and high cohesion,an energy management system for factory production of Lentinula Edodes Logs was developed,which realized intelligent collection of energy data for mushroom log production,electrical energy consumption prediction,standardized energy use management,and energy data statistical analysis.