Research On Application Of Centralized Neurodynamic Algorithms In Home Energy Management

With the development of science and technology and the improvement of people's quality of life,traditional power grids are faced with the challenges of scarce traditional energy sources,rapid increase in environmental pollution,and increased demand for users' electricity to achieve reliable,safe,economical,efficient,clean and use of power grids The smart grid for safety has quickly become a hot topic in international research.Among them,demand response is an important part of the smart grid.It can guide the user's behavior to encourage the change of the user's original electricity consumption habits and reach the peak of reduction or transition.The power consumption during a period of time responds to the power supply,thereby improving user benefits,ensuring the stability of the grid system and suppressing the rise in electricity prices.The specific process is two-way communication between the power supplier and the user to determine the user's power demand,adjust the power price according to the power demand of all users,and the user receives power price information or incentive information to adjust the power strategy to avoid households in the power system during peak power consumption.Electricity consumption,increase household electricity consumption in the trough period to reduce electricity costs.In this paper,a smart house is used as a scenario to construct a demand response energy management problem that improves the user's power consumption efficiency.The centralized neural dynamics algorithm is used to solve the proposed problem,and an optimized power consumption strategy for home energy management is obtained.Convex optimization theory and stability are used.The theory of stability analyzes the stability of the algorithm.Finally,the MATLAB software platform verifies the rationality of the model and the effectiveness of the algorithm.First of all,this paper considers the application of the neural dynamics method based on augmented Lagrange in the household energy demand response scheduling model with the day-to-day electricity price mechanism,and considers the fixed load,elastic load and semi-elastic load with different operating characteristics And in the home scenario of energy storage equipment,a demand response energy scheduling model aimed at minimizing the cost of household electricity is established in combination with the previous electricity price information.Among them,because the energy storage equipment has charging and discharging characteristics,0-1 is introduced in the model,The variable represents the state of charge or discharge;the final optimized variable for this problem is the specific power consumption of the semi-elastic device in each time period and the charge and discharge state and specific power value of the energy storage device in each time period.In the subject,the neurodynamic optimization method based on the augmented Lagrangian function is considered to solve the problem,and the optimized solution for each time period of the user is obtained,and the stability of the algorithm is proved by mathematical theoretical analysis;finally,in the MATLAB software The platform conducts simulation experiments to find the optimal solution and analyze it in conjunction with the system model to verify the rationality of the problem and the feasibility of the algorithm.Then the subject proposed the application of feedback neural network algorithm in household energy management with demand response mechanism of real-time electricity price and incentive response.Constructed a family scenario with fixed equipment,reduceable equipment,transferable equipment,energy storage equipment and renewable energy,combined with real-time household electricity price information provided by the power supplier,and considered incentives that can be used to guide household users to change their electricity consumption strategies Mechanism,put forward a household energy management model for the purpose of reducing the cost of electricity for users,and then apply the feedback neural network algorithm to obtain the power consumption and energy storage equipment of each time period of the curable equipment,transferable equipment and energy storage equipment Optimized variables such as charge and discharge states,and proved the stability of the algorithm.A simulation process was carried out on the MATLAB software platform to prove the rationality and effectiveness of the model.