| With the continuous acceleration of the urbanization process in China,the number of power distribution rooms,which are the key nodes of electric energy distribution,is increasing day by day,and the locations are widely distributed,which causes difficulties for the existing manual periodic inspection work.Especially in the hot and rainy summer,due to the closed environment and poor permeability,the temperature and humidity conditions of power distribution room will be worse.Due to their great influence on the health of indoor power equipment,if they cannot be adjusted in time,short circuit or partial discharge will occur,affecting the normal operation of distribution network and greatly reducing the reliability of power supply.Therefore,the research on the automatic control of distribution room temperature and humidity has an important practical significance.Aiming at a 10 kV indoor dual-power distribution room with two dry-type transformers,this paper proposes a temperature and humidity control method based on model prediction.Therefore,based on the gray box theory and the black box theory,the temperature and humidity change prediction models of the distribution room first need to be established respectively.Then,based on them,each round of optimization is performed by predicting the future temperature and humidity status,and the optimal control sequence is obtained by rolling to achieve model prediction optimal control of temperature and humidity.Firstly,the external structure and internal configuration of the power distribution room are introduced,the temperature and humidity conditions are analyzed,and their main influencing factors are determined.Then based on this,the temperature and humidity changes in the power distribution room are modeled and studied.According to on the knowledge of heat transfer and the similar theory of thermoelectric,the heat transfer system of power distribution room is simulated as a network with capacitors and resistors to obtain an equivalent thermoelectric model.Then,the discrete random model of temperature is obtained through mathematical transformation,and the unknown physical parameters of this model are estimated by the maximum likelihood method according to the experiment data;secondly,the humidity time series prediction model of power distribution room is established using the black box method.The predicted output values of the above models are obtained through simulation,and are compared with the actual values,so the correctness of the two models is verified.Based on the power distribution room temperature and humidity prediction model which are established above,the model prediction control method of temperature and humidity is studied.Based on the given temperature and humidity reference values,a multi-step dynamic rolling optimization strategy is adopted.By making the system performance index to obtain a minimum value under the constraint conditions,the optimal action sequence of the controlled body—air conditioner or dehumidifier in the future control time domain is solved.And the feedback correction link is introduced to modify the current model predicted output value,so a closed-loop control system is formed.At the next control moment,the optimized time domain moves backward,and the above process is repeated.Finally,on the Matlab simulation platform,the simulation analysis of the example proves the effectiveness and flexibility of this method.Meanwhile,they are compared with traditional PID control effects.It is concluded that model predictive control has predictability and can make the control stability of the temperature and humidity better.Therefore,the power distribution room temperature and humidity control based on model prediction method can ensure the safe operation of power equipment and achieve the energy-saving effect. |
