In order to enable those skilled in the art to better understand the technical solution of the present invention, its specific implementation will be described in detail below in conjunction with the accompanying drawings:
 Examples of the invention: see figure 1 , a distribution network maintenance decision-making optimization method, comprising the following steps: Step S1, read in the grid data of the distribution network and the information of each device in the distribution network;
 Step S2, initialize the position and velocity range of each particle, randomly generate a particle swarm, each particle represents a legal overhaul plan, that is, a plan that does not violate the constraint conditions, and initialize the global optimal and individual optimal;
 Step S3, updating the position and velocity range of the particle swarm to generate a new generation of particle swarm;
 Step S4, analyze the legality of each particle in the new particle swarm, that is, whether it violates the constraint conditions; if it is legal, that is, it does not violate the constraint conditions, then calculate its fitness value; otherwise, directly assign a larger value as its fitness value ; Then update the global optimal and individual optimal;
 Step S5, determine whether the maximum number of iterations is exceeded, if so, end the iterative algorithm, and the maintenance plan corresponding to the minimum fitness value in step S4 is the optimization result of the distribution network maintenance decision; otherwise, return to step S3;
 Step S6, outputting the optimization result of the distribution network maintenance decision.
 Constraints include power grid security constraints, maintenance relationship constraints and maintenance resource constraints.
 Grid security constraints include node voltage constraints and line power flow constraints, where:
 The node voltage constraints are shown in formula (1):
 u imin < U i < U imax (1)
 In formula (1), U i , U imax , U imin One-to-one correspondence is the voltage value of node i, the upper limit value of the voltage and the lower limit value of the voltage;
 Line flow constraints are shown in formula (2):
 S j jmax (2)
 In formula (2), S j , S jmax One-to-one correspondence is the power flow value of line j and the limit power flow value allowed for transmission.
The maintenance relationship constraints include simultaneous maintenance constraints and mutually exclusive maintenance constraints. Among them, the simultaneous maintenance constraints are to avoid repeated power outages caused by equipment maintenance and improve power supply reliability. Two equipments start maintenance at the same time, as shown in the following formula:
 t i = t j (3)
 In formula (3), t i is the maintenance start time of equipment i, t j is the maintenance start time of equipment j;
 Mutually exclusive maintenance constraints are to avoid avoidable load outages caused by the particularity of the interconnection relationship when some equipment is maintained at the same time, as shown in the following formula:
 t jt i +T i -1(4)
 In formula (4), t i is the maintenance start time of equipment i, t j is the maintenance start time of equipment j, T i is the maintenance duration of equipment i.
 For example, when the distribution transformer and distribution line of the terminal load are overhauled, the terminal load will be blacked out. Therefore, the transformer and the line should be scheduled for maintenance at the same time, which is applicable to the constraints of simultaneous maintenance. For example, two transformers in the same substation are applicable to mutually exclusive maintenance constraints.
 The maintenance resource constraints reflect the maintenance capability of the maintenance implementation unit, which is represented by the limit of the number of equipment to be maintained at the same time, as shown in the following formula:.
 Σ i = 1 N u i t m i ≤ S t - - - ( 5 )
 In formula (5), u it is the maintenance state variable, and it is recorded as 1 for maintenance, otherwise it is recorded as 0; m i Represents the resources required for maintenance of equipment i; S t Represents the upper limit of maintenance resources in period t.
 The distribution network maintenance decision-making optimization method of the present invention uses a particle swarm optimization algorithm for optimization, and each particle in the group represents a maintenance plan, and any change in the start time of maintenance of equipment is considered a new plan, so that the distribution network maintenance Decision-making is forward-looking and targeted.
 Those of ordinary skill in the art should recognize that the above embodiments are only used to illustrate the present invention, rather than as a limitation to the present invention, as long as within the scope of the spirit of the present invention, the above-described embodiments Changes and modifications will fall within the scope of the claims of the present invention.