[0077]Example
[0078]Appendfigure 1 As shown, this embodiment provides a home-load optimization scheduling method, including the following steps:
[0079]Step 1, according to the working characteristics of the electricity load and the user's electricity behavior, all household appliances are divided into rigid load and flexible load; wherein the flexible load includes an interrupt load and an uninterruptible load; the interrupt load is The work task can be completed within a predetermined period, and it is possible to shut down the home-electricity load that is working at any time; the uninterrupted load is maintained continuously in the working period to complete the household electricity load that is deactivated after completing the task.
[0080]Step 2, determine the control scheme, that is, determine the target function, control variables, and constraints;
[0081]The control variable is the operation of each interrupt load at each allowed runtime period; where the expression of the control variable is:
[0082]
[0083]among them, For the operating state of the household electricity load I in the second time period, b is the number of load operating time; 0 means that the home-loaded load I is in a shutdown state in the binary period. 1 means that the home electrical load I is in the working condition in the second time.
[0084]Constraint conditions include constraints that interrupt loads and constraints that cannot be interrupted loads;
[0085]Among them, the constraints of the interrupt load are:
[0086]
[0087]ta≤ βa-αa+1
[0088]among them, In order to interrupt the load a in the working state of the first load operation period; TaFor the operational time of the interrupt load A; αaIn order to interrupt load A, the start time of the allowable work; βaAt the end of the allowable operation of the interrupt load A.
[0089]The constraints for non-interrupt load are:
[0090]
[0091]
[0092]Where C is an uninterruptible load; αcIn order not to interrupt load C allowed the start time of the work; βcIn order not to interrupt the end of the load C allows the end of the work; At the time of the working opening of the work C. Do not interrupt the work of the load C; TcFor the operational time of the uninterruptible load C.
[0093]Optimization goals include family power cost and electricity power peak valley, and the expression of the target function is:
[0094]
[0095]
[0096]B = (T-1) × H + 1, T = 1, 2, λ, 24
[0097]
[0098]
[0099]Among them, COST is a cost of electricity for a day;tTaxi price; EtFor the amount of electricity in the day of day; Electrical load of electricity load I. Electrical power for household electricity load I in the second load runtime; eMax The maximum electricity amount of the first load runtime; divided into T period 24 hours according to the length H; preferably, the length of time is 30 min, and is divided into 48 times; A is a household electricity load total.
[0100]Step 3, establish a household load electrical model; family load electrical model is the relationship between home electricity load operation and household electricity cost and electric power peak valley; wherein the expression of the household load electricity model is:
[0101]
[0102]Where wCost WUND Different factors; COSTMIN Minimum electricity cost for family one day; CostMIN Δ is the best value for the minimum electricity cost of the family; UNDMIN For the minimum peak valley of electric power; UNDMIN Δ is the optimal value of the minimum peak valley with electric power.
[0103]Step 4, using improved adaptive weight multi-target particle group algorithm, solve the home load electrical model to obtain a household power load optimization scheduling result; specifically include the following steps:
[0104]Step 41, obtain an improved multi-target particle group optimization algorithm; specific:
[0105]Through constraint processing, individual optimal solution and global optimal solution, improve multi-target particle group algorithm;
[0106]Among them, the constraint processing adopts the adaptive function, the expression of the adaptivity function is:
[0107]
[0108]Individual optimal solution and global optimal solution:
[0109]Each particle searches for the optimal solution in the search space, and records it as the current individual value, and shares the individual value to the other particles of the entire particle group, find the optimal individual extreme value as the entire particle The current global optimal solution of the group, all particles in the particle group adjust their speed and position according to the current global optimal solution of the current individual extremum and the current global optimal solution shared throughout the particle group.
[0110]Step 42, using an improved multi-target particle group optimization algorithm to solve the home load electrical model to obtain an optimized scheduling result of the home load; the solution process specifically includes the following steps:
[0111]Step A, initialization
[0112]Initialize the population and external archives, set the particle group size psize, set the maximum number of iterationsITE, Given the initial value of inertia, learning factor C1Learning factor C2Each particle in the population is randomly distributed in the feasible solution space and impart initial speeds within the preset range; where the external archive is set to the storage unit for storing the solution.
[0113]Step B, determine the extreme value
[0114]Evaluate the initial population according to the adaptation function, initialize the individual of each particle, the global optimal solution of the population, the global Pareto optimal decoction and historical Pareto optimal university; calculate the adaptivity value of each particle And the optimal position in history is the individual value, and the global optimal solution of the particle group is found in the particle extract set, and the optimal solution is selected as the global extreme value.
[0115]Step c, determine the inertia weight of the current iteration
[0116]The expression of the current iterative inertia weight is:
[0117]
[0118]ΔW = WMax -wMIN
[0119]Where W is the current iterative inertia weight; WMax The maximum value of inertia weight, WMIN Minimum value of inertia weight; fmean The adaptivity value of the current particles, FvThe average adaptivity value of all particles; fMIN The minimum adaptivity value of the current particles.
[0120]Step D, update the speed and location of each particle
[0121]Among them, the speed and location of each particle is:
[0122]
[0123]Where VmSpeed for the mth particles, XmPBEST for the mth particlesmThe best position found for the mth particles until now, GBESTmThe best position obtained by the particle group; RAND is the random number on the interval (0, 1); Vm+1Speed for the m + 1 particles, Xm+1The position of the m + 1 particles.
[0124]Step e, update the inertia weight
[0125]Using non-linear change update inertia weight, where the update formula of inertia weight is:
[0126]
[0127]Among them, Q is the current iteration, WqRepresents the inertia weight value of the second iteration update, WMax Maximum value for preset inertia weight, WMIM The minimum value of the preset inertia weight.
[0128]Step F, update location and external files
[0129]According to the particle adaptivity value, update historical optimal PBEST and global optimal optimal GBEST, the optimal solution determined by the current particles and the previous generation of particles, if the current is more preferable, the individual extreme and the global optimal solution replace;
[0130]Keep the currently founded optimal decoction, then the most uploaded optimal decomposition and the current optimal solution collection, and then select the non-dominable optimal decoction.
[0131]Step g, determine whether the maximum number of iterations is reached, if not, the number of iterations plus 1, turn step d; if the algorithm terminates iteration, output non-dominable optimal decomposition to obtain an optimized scheduling result of the home power load.
[0132]Step 5, optimize the scheduling result according to the home electrical load, scheduling the household electricity load.
[0133]This embodiment is directed to the build-up electricity model of the family load, and the weight of the particle swarm optimization algorithm performs real-time update, and implements scheduling optimization for different types of electric load; multi-target particle group algorithm for improving adaptive weight, Strengthened the practicality of the optimization of the scheduling model in the smart home electricity period, which has strong practical significance; strengthening the local search ability of the algorithm, and saves the diversity of the population while avoiding the algorithm, and saves the diversity of the population. The quality of the particle solution under the optimum function; introduces non-dominant sorting ideas and method of Pareto optimal moderately dispensing, and is constantly updating the grain by using the collaborative relationship between particles. The optimal solution of particles, improve the robustness and stability of the algorithm.
[0134]In the present embodiment, in the case of a normal home user, the simulation schedule period of the home user is set to 1 day, the scheduling step is 0.5h; the amount of electricity given in each time period is limited to 3KW, the user side rated voltage Set to 220V, the rated current is set to 20A; the home-loaded electricity load including a plurality of work segments is respectively considered as a single flexible load, and the home electrical load optimization scheduling system is used in this embodiment. Differential information interaction on the user side and power supply side; wherein the flexible load comprises a rice cooker, a washing machine, a first water heater, a second heater, an electric vehicle charger, a first air conditioner, a second air conditioner and a sweeping robot; the basics of each flexible load As shown in Table 1 below, the parameter is shown in Table 2 below; wherein the same electricity load changes with the user, using an important indicator between the AHP acquisition electricity load.
[0135]Table 1 Basic parameter table of each flexible load operation
[0136]
[0137]
[0138]Table 2 Time electric price list
[0139]
[0140]In this embodiment, the multi-objective is optimized by the optimization of the electrical cost and fluctuation of the user, using improved adaptive weight multi-target particle group algorithm; where particle group size psize = 200 is set, and the maximum number of iterations is set.ITE= 200, given learning factor C1= 0.9, learning factor C2= 0.1; The operation results after the above flexible load scheduling optimization are shown in Table 3, which can be seen from Table 3, and the operation of the flexible load performs a large adjustment.
[0141]Table 3 Table 3 Optimize the flexible load operation results parameter table
[0142]
[0143]Appendfigure 2 As shown, withfigure 2 A comparison of home electrical power before and after optimization is given.figure 2 It can be seen that the electricity cost and the peak valley of the electricity system and the electricity system are used, only the user's own willingness is disorderly, and the electricity period is mainly concentrated at 6, when it is 17:00, the rest of the time is more Small, peak difference is more obvious, poorly stabilizing the electricity system; after optimization, consider the electricity cost, the electricity period is mainly concentrated in 2-7 hours and 18-23, and the electricity system peaks are small. There is no obvious peak or trough; therefore, the present embodiment uses the home electrical load optimization scheduling method and system, which plays a peak-filled entry, reducing the purpose of electricity cost.
[0144]Appendimage 3 As shown, withimage 3 A comparison of target values before and after optimization is given, from withimage 3 It can be seen that the flexible load will appropriately change the user's electricity behavior during dispatch, but the cost of electricity after optimization is significantly reduced, and the peak valley deviation is reduced; effectively guarantee the economics and comfort of electricity, in line with scheduling optimization The actual application situation.
[0145]Considering the optimization results of different situations, comparing the electricity cost, satisfaction, and electricity load, five different cases of optimization scheduling results are shown in Table 4 below; from Table 4, it can be seen When applied in the original plan, the user satisfaction is relatively high, but the use of electricity cost is increased; if the electricity fee is the least, the user's satisfaction is very low; it is, if only the user's satisfaction is most, the user is The electricity fee is relatively high, and it can achieve the best comfort and economy; compare the experimental results of the traditional multi-objective model, the improved model is reducing the use of electric habits, economic and satisfaction reaches relatively good As a result, the electricity bill is reduced, and the superiority of the proposed algorithm to solve the problem of resolving the construction of the household electricity load is further verified.
[0146]Table 4 Optimized scheduling results in different situations
[0147]
[0148]The home electrical load optimization scheduling method and system according to the present invention are established by establishing a home electrical load model. The model considers the comfort of life side, and the economics of home power use is also taken into account the power load fluctuation. Factors, more close to practical, practical, and the practicality is also high; the multi-target particle group algorithm for improvement of adaptive weight is proposed by comparing the number of inertial weight values of the existing population. Avoiding algorithms too early while saving a variety of populations; reasonably guiding users 'error-in-electricity, promoting the development of residents' electrical or orderly and intelligent development, effectively verifying the accuracy of the model and the superiority of the algorithm Sex.
[0149]This embodiment also provides a home-load-load optimization scheduling system, including division module, function module, modeling module, solving module, and adjustment module; where the module is used to different working characteristics according to the electricity load, will All household electrical loads are divided into rigid load and flexible loads; wherein flexible load includes interrupt load and uninterruptible load; function modules for determining target functions, control variables, constraints; modeling modules for establishing household load Modeling model; where the household load electricity model is the relationship between home electrical load operation and household electricity cost and electric power peaks; solicitation module for utilizing improved adaptive weight multi-target particle group algorithm Solving the household load electrical model to obtain a home-load-load optimization scheduling result; the adjustment module is used to optimize the scheduling result according to the household power load, scheduling adjustment.
[0150]This embodiment also provides a household electrical load optimization scheduling device, including a memory, a processor, and an executable instruction stored in the memory and can operate in the processor; Implement the following steps when performing the executable instruction:
[0151]According to the different working characteristics of the electricity load, all household appliances are divided into rigid load and flexible load; wherein the flexible load includes an interrupt load and an uninterruptible load; determining a target function, control variables, and constraint conditions; establishing a household load Electrical mode; where the household load electrical model is the relationship between home electrical load operating state and home electricity cost and electric power peak valley; use improved adaptive weight multi-target particle group algorithm for household load electricity model SOLVED, obtain the results of the home electrical load; optimize the scheduling result according to the home electrical load, scheduling adjustment.
[0152]This embodiment also provides a computer readable storage medium that stores a computer program in a computer readable storage medium, and the computer program can be implemented as follows:
[0153]According to the different working characteristics of the electricity load, all household appliances are divided into rigid load and flexible load; wherein the flexible load includes an interrupt load and an uninterruptible load; determining a target function, control variables, and constraint conditions; establishing a household load Electrical mode; where the household load electrical model is the relationship between home electrical load operating state and home electricity cost and electric power peak valley; use improved adaptive weight multi-target particle group algorithm for household load electricity model SOLVED, obtain the results of the home electrical load; optimize the scheduling result according to the home electrical load, scheduling adjustment.
[0154]The computer readable storage medium in this embodiment includes: a medium such as a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk, or an optical disk, can store the program code.
[0155]An explanatory description of the related part of the home electrical load optimization scheduling system, device, and computer readable storage medium, can refer to the detailed description of the corresponding portion of the home electrical load optimization scheduling method according to the present embodiment. Not described here.
