当前位置：网站首页>MATLB | multi micro grid and distributed energy trading

MATLB | multi micro grid and distributed energy trading

2022-07-05 01:31:00 【Power system code】

Catalog

3、 ... and 、Matlab Code implementation

One 、 summary

In humans 、 Driven by the energy demand of industry and electric vehicles , Global energy demand is expected to grow steadily in the next few years ; More precisely , Expect to 2030 The annual growth will reach 40%. This demand is caused by the increasingly energy dependent lifestyle of mankind 、 The emergence of electric vehicles as major means of transportation and machines will promote the further automation of the process . In today's power grid , Energy is in centralized and large energy plants （ Micro grid energy generation ） Produced in ; then , Transmit energy to the final client , Usually through very long distances and through complex energy transmission grids . Such a complex structure , Less flexibility , It is difficult to adapt to the growth of demand , Thus, the probability of power grid instability and power failure is increased . The impact is huge , The recent power outages in Europe and North America caused millions of euros in losses, which proves this .

under these circumstances , Microgrid is becoming a promising energy solution , Among them, distributed （ reproducible ） Energy is meeting local needs . When local production cannot meet the requirements of microgrid , Purchase energy from major utility companies . Imagine that microgrid will provide many benefits , For example, the reliability of power transmission （ for example , Through the island ）, Improve efficiency and sustainability by increasing the penetration of renewable energy , Scalability and investment delays , And provide auxiliary services . Island is one of the outstanding characteristics of microgrid , It refers to the ability to disconnect the microgrid load from the main grid , And it is specially powered by local energy . The main power grid cannot support the total demand and / Or when the operator detects some major grid problems that may degenerate into blackouts , The expected island will be implemented . In these cases , Microgrid can provide enough energy , At least basic power services can be guaranteed . Once the whole system stabilizes again , The connection to the main grid will be restored . obviously , These are important functions that may lead to instability .

In order to improve the capacity of smart grid , A typical method is to consider the exchange of energy between several microgrids , Even if these micro grids are isolated , That is, disconnect from the main power grid . In other words , There is energy flow in a group of continuous microgrids , However, there is no energy flow between the microgrid and the main grid . In this context , The optimal power flow problem has recently attracted considerable attention . for example ,Ochoa and Harrison The power flow problem of coordinated voltage control is jointly considered . perhaps , The work in relevant literature focuses on Unbalanced Distribution Networks , A method to solve the three-phase power flow problem based on Newton descent method is proposed . Because these centralized solutions may be affected by scalability and privacy issues , Generally speaking , The optimal power flow problem is nonconvex ; therefore , The exact solution may be too complex to calculate . therefore , Suboptimal methods are often used . Some literatures use the so-called multiplier alternating direction method （ Participate in solving power flow problems in a distributed way .、

In this paper , A distributed convex optimization framework is developed for energy trading between island microgrids . More specifically , This problem consists of several isolated island microgrids , These grids exchange energy flow through any topology . Due to scalability problems , In order to protect the local information of the cost function , Explained a Cost minimization algorithm based on sub gradient , The algorithm converges to the optimal solution within the actual number of iterations . Besides , This method allows a very intuitive economic explanation , from “ Supply and demand model ” and “ Market clearing ” The numerical results of the iteration of the algorithm are explained from the perspective of, and the convergence speed of the algorithm and the implementation cost of different network topologies are given .

Two 、 mathematical model

Consider a reason for M An interconnected microgrid $\left(\mu \mathrm{Gs}\right)$ Multi microgrid system composed of , The microgrid operates in island mode . During each scheduling interval , Each microgrid $\mu \mathrm{G}-i$ produce $E_{i}^{(g)}$ Units of electricity and consumption $E_{i}^{(c)}$ A unit of electrical energy . Besides , It is permissible to $\mu \mathrm{G}-i$ towards $\mu G-j$ Sell electricity $E_{i,j}$ （ $j\neq i$ ）, And from $\mu G-k$ （ $k\neq i$ ） Buy energy $E_{k,i}$ . then , $\mu G$ The electric energy balance in the needs ：

3、 ... and 、Matlab Code implementation

This article only shows part of the code , All code points here ： We are shipping your work details

%%  visualization 
figure
plot(inputdata{:,end-2},uGcosts'/1000)
xlabel('E_4^{(c)} [MWh]')
ylabel(' cost [k$]')
grid
axis([0 12 0 1.4])
hold
plot(inputdata{:,end-2},C{1}(inputdata{:,1})/1000,'k--')
plot(inputdata{:,end-2},C{1}(inputdata{:,end-2})/1000,'k-.')
legend('uG1','uG2','uG3','uG4','disc. {1,2,3}','disc. 4','Location','northwest')
saveas(gcf, 'local_costs.png')

figure
hh = plotyy(inputdata{:,end-2},Sell4,inputdata{:,end-2},[Price Sell4.*Price]/1000);
xlabel('E_4^{(c)} [MWh]')
ylabel(hh(1),'[MWh]')
ylabel(hh(2),'[k$]')
grid
axis(hh(1),[0 12 0 3.6])
set(hh(1),'YTick',0:3)
axis(hh(2),[0 12 0 1.8])
set(hh(2),'YTick',0:.5:1.5)
legend(' Sold electricity ', ' The unit price ', ' earnings ')
saveas(gcf, 'trading.png')